ree pert Pptet er retet rere ti ceers etrteteee et ch PePiPrere tricia) tee Tavsgehe Hetetee Siiaisesteswes tee Tisttesss set eeeeeene eee bere Sstes sisyetuteee Sescerepress?: isteentrers reoeepr tite. | siaiiien ciate leneialede totes ose a +4 rps tetereee Ve seat faentatetetet laters sheveescese wate veteiere . dageenvats (ates qa loestete jemi sens serene eee setiee ee taleteeme +s fies tel ateteee ge eieeete tie erate Tele te tate qeeeeet teeaterer® sett st Sitetasee fe shellitinee preter i] Net ny ati ‘ if tyeee aloseres fatete fees #, peteene peers oer weet ee et eter errors Si lilolacece-ssece- bese scseleneeee Hitgtee Heit eit ae tH G TRANSACTIONS OF THE TRANSACTIONS OF THE Moto tre sO ClET Y OF EDINBURGH. EDINBURGH : PUBLISHED BY ROBERT GRANT & SON, 107 PRINCES STREET, AND WILLIAMS & NORGATE, 14 HENRIETTA STREET, COVENT GARDEN, LONDON. MDCCCCXVII. No. Ife Published ” September 23, 1915. September 22, 1915. November 19, 1915. November 29, 1915. November 30, 1915. December 10, 1915. December 14, 1915. January 17, 1916. January 31, 1916. December 31, 1915. December 28, 1915. January 15, 1916. August 14, 1916. August 11, 1916. KN XVI. De Vaile XVIII. XIX. XX. XX]. OCR 5 XXIII. XXIV. XXV. XXVI. XXVIL. Published August 14, 1916. April 25, 1916. August 21, 1916. September 16, 1916. August 29, 1916. November 6, 1916. November 8, 1916. November 29, 1916. January 17, 1917. February 27, 1917. April 21, 1917. February 2, 1917. March 9, 1917. CONTENTS. PART I (191415: ) NUMBER li ED. Ev. DY. VI. On the Zeolites and Associated Minerals from the Tertiary Lavas around Ben More, Mull. By W. F. P. M‘Lintocx, B.Sc., Royal Scottish Museum, Edinburgh. (Plates I-III), Spongiaires recueillis par la “ Scotia” dans Antaretique (1903-1904). Supplément. Par Emite Topsent, Professeur a la Faculté des Sciences de Dijon. Présenté par le Dr. W. 8S. Bruce, On Larve of Lingula and Pelagodiscus (Discinisca). By J. H. Asu- wortH, D.Sc., Lecturer in Invertebrate Zoology in the University of Edinburgh. (Plates IV and V), . The Temperatures, Specific Gravities, and Salinities of the Weddell Sea and of the North and South Atlantic Ocean. By Wrtttam 8. Bruce, LL.D., ANDREW Kine, F.1.C., and Davip W. Witton, . A Contribution to the Cranilogy of the People of Scotland. Part II. Pre- historic, Descriptwe and Ethnographical. By Principal Sir WILLIAM Turner, K.C.B., D.C.L., F.R.S., F.Soc.Ant.Scot., Emeritus Professor of Anatomy, . The Morphology and Development of the Free-Swimming Sporosacs of the Hydroid Genus Dicoryne (including Heterocordyle). By J. H. Asu- wortH, D.Sc., University of Edinburgh, and James kircuin, M.A., D.Se., Royal Scottish Museum, Edinburgh. (Plates VI-VIII), . PAGE 45 71 ae 257 v1 CONTENTS. PART II. (1915-16.) VII. Studies on the Development of the Horse. I. The Development during the Third Week. By J. Cossan Ewart, F.R.S., Regius Professor of Natural History, University of Edinburgh. (Plates IX—XVIII and Twenty-one Text-figures), VIII. Description of a Reconstruction Model of a Horse Embryo Twenty-One Days Old. By Professors ArtHUR Ropinson, M.D., and A. Gipson, M.B., F.R.C.S. (Plate XIX, figs. 54-63. Text-figs. 22-26), IX. A Contribution to the Study of the Scottish Skull. By Matrurw Youne, M.D., Senior Demonstrator in Anatomy, University of Glasgow. (With Three Plates), X. Skiagraphic Researches in Teratology. By Harry Rainy, M.D., F.R.C.P.E., and J. W. Battantyne, M.D., F.R.C.P.E. (With Fifteen Plates), . : XI. On a Small Collection of Terrestrial Isopoda from Spain, with Deserip- tions of Four New Species. By Watter EK. Cotiincx, M.S8ce., F.L.S., etc., Research Fellow of the University of St Andrews. (With Two Plates), XII. The Lateral Sense Organs of Elasmobranchs: The Ampullary Canals of the Genus Raia. By Auvcusta Lamont, B.Sc., Baxter Scholar in Natural Science, University of Edinburgh. (With Hight Plates and Nine Text-figures), ; PART III. (1915-16-17.) XIII. Contributions to the Geology of Benguella. By Professor J. W. GREGORY, F.R.S., D.Sc. (With Two Plates), ‘ XIV. A Contribution to the Petrography of Benguella, based on a Rock Collection made by Professor J. W. Gregory. By G. W. TyrrReE xt, A.R.C.Se., F.G.S., Lecturer in Geology, Glasgow University. (With One Plate), PAGE 287 347 455 467 495 NUMBER XV. XVI. XVII. XVIIL. XIX. XXI. XXII. XXIII CONTENTS. On some Cretaceous Brachiopoda and Mollusca from Angola, Portu- guese West Africa. By R. BuLtten Newron, F.G.S8., Geological Department, British Museum. (With Two Plates), . Note on an Algal Limestone from Angola. By Mrs M. F. Romans, late Harkness Scholar, Newnham College, Cambridge. (With One Plate), . j On some Cretaceous Echinoidea from the Neighbourhood of Lobito Bay. By Professor J. W. Grecory, F.R.S., D.Sc., . Contributions towards a Knowledge of the Anatomy of the Lower Dicotyledons. 1. The Anatomy of the Stem of the Papaveracee. By R. J. Harvey-Grason, D.L., M.A., Professor of Botany, University of Liverpool; and Minnie Braptey, M.Sce., Hartley Research Scholar, University of Liverpool. (With Three Plates), Apractocleidus teretupes: A new Oxfordian Plesiosaur in the Hun- terran Museum, Glasgow Unwersity. By Wititam R. SMELiIn, M.A., B.Sc. (With One Plate), . The Anatomy and Affimty of Platyzoma macrophyllum, R. Br. By Joun M‘Lean Tuompson, M.A., B.Sc., Chief Assistant in Botany and Robert Donaldson Research Scholar, Glasgow University. (With Four Plates), . > On Leaf-Architecture as Uluminated by a Study of Pteridophyta. By F. O. Bowser, D.Sc., F.R.S., Regius Professor of Botany in the University of Glasgow. (With One Plate), Contributions to owr Knowledge of British Palxozoic Plants. Part I. Fossil Plants from the Scottish Coal Measures. By Dr R. Kipston, F.R.S. (With Three Plates), A Revision of the British Idoteidxe, a Family of Marine Isopoda. By Water EK. Contiince, D.Sc., F.L.S., ete., Research Fellow of the University of St Andrews. (With Eleven Plates), vil PAGE 561 585 609 631 657 709 721 Vill CONTENTS. NUMBER XXIV. On Old Red Sandstone Plants showing Structure, from the Rhynie Chert Bed, Aberdeenshire. Part I. Rhyna Gwynne-Vaughani, Kidston and Lang. By R. Kipsron, LL.D., F.R.S., and W. H. Lane, D.Sce., F.R.S., Barker Professor of Cryptogamic Botany in the University of Manchester. (With Ten Plates), XXV. The Prothallus of Tmesipteris Tannensis. By Professor A. ANSTRUTHER Lawson, D.Sc., F.L.S. (With Three Plates), PART IV. (1915-16-17.) XXVI. The Ordovician and Silurian Brachiopoda of the Girvan District. By F.R.C. Reep, M.A., Se.D., F.G.8. (With Twenty-Four Plates), XXVIU. The Forest of Wyre and the Titterstone Clee Hill Coal Fields. (With Five Plates and Six Text-figures), ; Introduction. By Dr R. Kinston, F.R.S. Part 1: The Geology of the Forest of Wyre Coal Field. By T.C. CanTRILL, B.Sc., F.G.8S.—The Fossil Plants of the Forest of Wyre Coal Field. By Dr RB. Kinston. Part IL: The Geology of the Titterstone Clee Hill Coal Freld. By E. E. L. Drxon, B.Sc., F.G.8.—The Fossil Plants of the Titterstone Clee Hill Coal Field. By Dy R. Kipston. With an Appendix on The Fossil Plants collected from the Core of the Claverley Trial Boring. By Dr R. Kinston. INDEX, PAGE 761 785 795 399 . 1085 TRANSACTIONS OF THE THON AO yY OF “Eb DINBURGH. VOLUME LI, PART IL—SESSION 1914-15. CONTENTS. I. On the Zeolites and Associated Minerals from the Tertiary Lavas around Ben More, Mull. W. F. P. M‘Lintock, B.Sc., Royal Scottish Museum, Edinburgh. Communicated by J.S. Fuxrt, D.Sc., LL.D., F.R.S. (Plates I-III), . (Issued September 23, 1915.) If. Spongiaires recueillis par la “ Scotia” dans I Antarctique (1903-1904). Supplément. Par Emmis Torsent, Professeur a la Faculté des Sciences de Dijon. Présenté par le Dr. W. S. Brucr, (Issued September 22, 1915.) Ill. On Larve of Lingula and Pelagodiscus (Discinisca). By J. H. Asuworta, D.Sc., Lecturer in Invertebrate Zoology in the University of Edinburgh. (Plates IV and V), (Issued November 19, 1915.) IV. The Temperatures, Specific Gravities, and Salinities of the Weddell Sea and of the North and South Atlantic Ocean. By Wiuuram §. Bruce, LL.D., Anprew Kine, F,I.C., and Davin W. Witton, (Issued November 29, 1915.) V. A Contribution to the Craniology of the People of Scotland. Part Il. Prehistoric, Descrip- tive and Ethnographical. By Principal Sir Witu1am Turner, K.C.B., D.C,L, F.R.S., F.Soc.Ant.Scot., Emeritus Professor of Anatomy, (Issued November 30, 1915.) VI. The Morphology and Development of the Free-Swimming Sporosacs of the Hydroid Genus Dicoryne (including Heterocordyle). By J, H. Asaworrn, D.Sc., University of Edinburgh, and James Rircuiz, M.A., D.Sc., Royal Scottish Museum, Edinburgh. (Plates VI-VIII), (Issued December 10, 1915.) EDINBURGH: PUBLISHED BY ROBERT GRANT & SON, 107 PRINCES STREET, AND WILLIAMS & NORGATE, 14 HENRIETTA STREET, COVENT GARDEN, LONDON. MDCCCCXVI. Price Twenty-five Shillings. PAGE 71 171 257 ~~ ' - e - 4 ' - « ff a \ . aoa ‘ ; ~ . is ‘ ‘ » - ~ - . * 1 . F ‘ * ! bs . ‘ . ' o . . . ‘ j « ‘ . - « 4 ' ¥ LEAN SAO TT ONS. I.—On the Zeolites and Associated Minerals from the Tertiary Lavas around Ben More, Mull. By W. F. P. M‘Lintock, B.Sc., Royal Scottish Museum, Edinburgh. Communicated by J. 8. Fuerr, D.Sc., LL.D., F.R.S. (MS, received May 17, 1915. Read June 28, 1915. Issued separately September 23, 1915.) [Plates I-II.] INTRODUCTION. The special features of interest attaching to the Tertiary igneous rocks of Mull have been made known to geologists mainly through the classic descriptions of Professor J. W. Jupp and Sir ArcuriBaLp Geikis. The point of view of these two ‘investigators is, however, essentially geological, and they refer but little to the ‘occurrence of minerals and rarely give precise localities. It is clear, too, that the late Professor M. F.. Heppie did not make an exhaustive examination of many localities in Mull, for his collection in the Royal Scottish Museum contains very few specimens from that island. With a view to filling up this gap I have paid several visits to Mull, and this paper deals with some of the material which I collected. The officers of the Geological Survey who are at present mapping the island have helped me considerably with information and material; and to Mr J. E. Ricury, B.A., who has surveyed the area from which a large part of the material was obtained, I am particularly indebted for assistance in collecting specimens from somewhat in- accessible localities. The rocks around Ben More exhibit several peculiar characters which havé been noted by Sir ARcHIBALD GEIK1z,* and also by Professor Jupp,+ who gives a description of their petrographical characters and altered state, and notes the presence in them of veins and nests of green and pink epidote. Again, Mr James Currie has recorded } * Sir ARcHIBALD GerKin, The Ancient Volcanoes of Great Britain, 1897, vol. ii, pp. 184, 213. + J. W. Jupp, “On the Propylites of the Western Isles of Scotland,” Quart. Journ. Geol. Soc., 1890, vol. xlvi, pp- 368, 369. ¢ James Currm, “The Minerals of the Tertiary Eruptive Rocks of Ben More, Mull,” Trans. Edin. Geol. Soc., 1898, vol. vii, p. 223. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 1). 1 2 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS scolecite, epidote, heulandite, stilbite, prehnite, etc., from Maol nan Dambh, a spur of Ben More running down to Loch Scridain. Mr Curriz remarks upon the peculiar assemblage of lime-bearing minerals in the cavities of the lavas and the absence of soda-bearing ones; but, as I shall show later, albite is quite a common associated mineral at this locality and at many others in the vicinity. Apart from the points of interest already indicated, there is the further one that the lavas have been pierced by intrusive masses, and, as noted by Sir ARCHIBALD GrEIk1£, have suffered in consequence. considerable alteration—an alteration which is particularly prominent in the somewhat unstable minerals of the vesicles. In making a traverse of the country around Ben More one is forcibly impressed by the fact that for considerable distances around the plutonic centre the lavas differ entirely in character from those of the plateau-country far removed from the zone of intrusive rocks.* One of the most striking differences lies in the abundance of epidote in the cavities and veins of the central lavas, and its absence or extreme rarity in the basalts of the plateau. The central lavas, again, never show the spheroidal weathering to a brown loam so characteristic of the plateau-basalts, whilst there is the further difference, referred to by Mr Curriz,t that the mineral association in the cavities is entirely different from that found in the normal lavas of the plateau. Different interpretations have been placed on these facts by Professor Jupp and Sir ARCHIBALD GEIKIE. In the memoir already cited, the former observer groups the central lavas under the name of propylites, and he regards them as andesites of various types altered by solfataric action which “accompanied the intrusion of the acid masses.” { This action was widespread, and he distinguishes it from the contact metamorphism locally induced in the rocks lying near the margins of the intrusions. As regards the date of the solfataric changes in the lavas, it is stated that in many places they have preceded the action of contact metamorphism, and in others the opposite may have taken place.§ Sir ARCHIBALD GEIKIE, on the other hand, sees in the propylites merely the representatives of the plateau-basalts altered by contact metamorphism, || and he states that he was unable to find any trace of the solfataric action described by Jupp. The point is of considerable geological significance, for upon it rest questions of interpretation of a complicated series of igneous rocks. It occurred to me that an examination of the minerals in the cavities of the lavas might throw some light on the question of the cause and date of the alteration of the rocks, aud in the present communication I shall deal with a very well-defined zone of zeolite-bearing lavas which can be traced from areas free from contact metamorphism almost up to the margin of one of the large acid intrusions. This * Sir ARCHIBALD GEIKIR, loc. cét., p. 388. + JAMES CURRIE, loc. cit., p. 226. t J. W. Jupp, loc. cit., p. 382. § J. W. Jupp, loc, cit., p. 367. || Sir A. GEIKIR, Joc, cit., p. 185 (footnote), p. 388 (footnote). “| J. W. Jupp, loc. cit., p. 353. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 3 zone occurs on An Gearna and Beinn Fhada, two ridges running from Ben More in a north-westerly direction, and also on Maol nan Damh, a spur running south-west from Ben More towards Loch Scridain. Its distribution is indicated on the accom- panying sketch-map (fig. 1), and, as the geology of the region is complex, I include here short descriptions of the localities, kindly furnished by Mr RicHey, who surveyed An Gearna and Beinn Fhada, and Mr ANDErsoN, who mapped Maol nan Damh :— “The parallel ridges of Beinn Fhada and An Gearna are situated at the western ° ! 2 MILES Fig. 1.—Map showing the distribution of the zeolite-bearing zone (dotted) and the granophyre (crossed). border of the Mull plutonic centre. Tertiary plateau lavas form the country rock which is intruded by plutonics, sheets and dykes. The lavas are olivine basalts making thick cappings to the hills. A large granophyre mass (the Beinn a’ Ghraig granophyre) cuts through Beinn Fhada and ends in the valley below, not reaching to An Gearna. Its junctions with the lavas, seen a extenso on the hill-face, slope steeply outwards, while the valley stream-sections show a like relation in detail. On the south-east end of Beinn Fhada part of an earlier granophyre is cut off against the granophyre of Beinn a’ Ghraig. “There are two series of sheets both intermediate between the granophyres in age. The belt of thin inclined sheets which encircles the Mull plutonic centre crosses ~ 4 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS Beinn Fhada and cuts another series of thin basaltic sheets with a general westerly dip which are distributed throughout the district. Still earlier sheet-like masses of coarse dolerite form small scattered patches, of which some at least are later than the early granophyre. ‘‘Some north-westerly dykes are earlier than the Beinn a Ghraig granophyre, but the greater number of dykes belong no doubt to the north-west series proper and are the latest intrusions.” “The shoulder of Ben More known as Maol nan Damh is composed in the main of flat or very gently inclined lavas. Those forming the lower slopes of the hill are more basic, but the difference in character is not very marked and they are rather difficult to separate from the overlying flows, although the former are mapped as basalts and the latter as andesites.* The screes which flank the western slopes are in the lower more basic lavas. “The largest intrusive mass is a flat sill of mugearite some 200 or 300 feet in thickness. This comes some 500 feet above the screes referred to. There are also a large number of minor intrusions, including porphyritic and non-porphyritic dolerite, and one or two that are more acid in character. As a whole, however, they are not so numerous as those which intersect the lavas in the higher portion of Ben More.” THe MINERALS oF AN GEARNA. One of the most convenient places for studying the mineralogy of the lavas is the ridge of An Gearna referred to above. On the north-eastern slope at a height of about 1200 feet, large amygdales, filled chiefly with a white fibrous zeolite, make their appearance in the screes and become plentiful at heights of 1300-1500 feet. The size of these amygdales is phenomenal, measuring, as they do, as much as 15 cm. x 10 cm. It is noteworthy that the lavas of the upper part of the hill alone yield the zeolite-filled vesicles ; the amygdales in the lower lavas are small and are filled mostly with compact, massive albite. The fibrous zeolite is scolecite, and associated with it are epidote, prehnite, garnet, albite, and, much more rarely, hornblende, calcite, chabazte, and thomsonite. Scolecite has usually been considered one of the rarer Scottish zeolites,f but on An Gearna it occurs in profusion, whilst it is also common on Beinn Fhada and Maol nan Damh; on Coire Bheinn, a hill lying two miles south-west of An Gearna, it is much rarer, although good specimens have also been got from there. It occurs in white fibrous aggregates and crystals, which in some cases reach a length of 10 cm., but, although many cavities have been broken open, no terminated crystal has been observed. In some cases when an amygdale is broken across, the scolecite shows a series of perfectly sharp rectilinear cracks or veins with perfectly flat, lustrous * Microscopie examination has since shown that the lavas provisionally mapped as andesites are olivine basalts much like the underlying flows. + Of. J.G. Gooncuttp, “ Natural History of Scottish Zeolites,” Trans. Glas. Geol, Soc., 1903, vol. xii, suppt., p. 62. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 5 walls which, when examined closely with a lens, show a fine set of strize formed by two sets of lines intersecting at about 60°. There can be little doubt that these eracks were once filled with calcite, which has left its cleavage traces on the adjacent scolecite. The following analysis was made by Mr E. G. Rapizy in the laboratory of the Geological Survey :— Calculated for Formula CaAl,$i,0,, + 3H,0. SO ie . 46°10 per cent. 45°9 POgE mie ks oe, 25°05) 5 26°0 ResOs. ; : : Ot oss Be Cheney ee Po Maye. Wf, 14:3 MgO . : ‘ ' ee KO . : 103) * ;, Na,O . : ; . trace EEO at 10a: C._. : "13 per cent. Fe H,O above 105° C. ae BRAS Ps 13°8 100°13 per cent. 100°0 The mineral is thus a typical scolecite practically free from alkalies. The junction of the zeolite with the walls of the vesicle presents some interesting and unusual features. In place of the layer of green earth so frequently found under- lying the Scottish zeolites, there is often a confusedly crystalline layer of yellowish- green epidote which sends off small radiating groups of spear-like crystals into the scolecite. In addition to epidote there are also sometimes present in this layer prehuite, garnet, albite, hornblende, and chlorite. Fic. 2.—Crystal of epidote from An Gearna, Mull. Epidote occurs invariably, however, in more or less abundance, and ranges in colour from blackish-green, which is rather rare at this locality, through various shades of yellowish-green to pale pink or almost colourless. It occurs in radiate groups of erystals which sometimes show the combination a(100), ¢(001), r(101), b(010), n(111) (fig. 2). Besides occurring in intimate association with scolecite, epidote is also found in cavities where that mineral is absent. For example, at the north end of An Gearna, on the west slope near the summit, there is a pale, greenish- 6 MR W. F. P. M‘LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS gray lava the vesicles of which are lined with a beautiful pink epidote on the top of which is usually a layer of colourless, botryoidal prehnite. The epidote occurs in groups of divergent crystals which are sometimes green at the one end and grade off into pink at the other; it frequently rests upon a layer of albite. The following analysis, made by Mr Rap.eEy, shows this variety to be a lime- epidote, poor in iron, and owing its colour, doubtless, to the small amount of manganese present — Molecular Ratios. 810, 38°69 6405 TiO, ie sooth ee aewe Al,0, 28°54 2798) Fe,0, 637 agg} 8228. «Br FeO “22 ‘0030 MarO =s., eS) "0040 (CoNi)O nt. fd. ‘4437 4-05 CaO 23°78 "4246 MgO 49 “0121 K,O 03 Na,O ; : trace Li,O \ : .. trace Oat 105°C, ’ 5 ‘09 Bis mee Re H,O above 105° C.. ; "99 "055 "055 514 KOO Zi The above analysis gives the formula Ca,A],$i,0.;4H,O, in which the alumina is partly replaced by ferric iron and the lime by ferrous iron. The mineral obviously belongs to the somewhat rare series of epidotes poor in ferric iron referred to by Dr H. H. Tuomas in his description * of an epidote from Inverness-shire ; it is also peculiar in having only half the normal percentage of water. Owing to the fibrous and poorly crystallised nature of the material an accurate determination of the optical characters was impossible; but, by means of a solution of methylene iodide and benzene, the mean refractive index was found to be 1:720, which accords well with the values given by Dr THomas. Prehnate is a fairly common mineral in the vesicles, and occurs in the usual botryoidal form; sometimes it forms a dense white massive layer which is occasionally spotted with little yellowish or reddish masses of garnet. Typically it occurs in globular growths underlying the scolecite and seated upon epidote or albite. It is most plentiful at the locality previously noted in connection with the pink epidote. TheYe, the geodes found in the screes are usually dull and weathered on the surface, but when broken open show beautiful rosettes of epidote projecting into massive white prehnite. In cavities in the latter mineral there occurs occasionally a pale * H. H. Tuomas, “On an Epidote from Inverness-shire,” Mineralogical Magazine, vol. xiv, p. 109. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 7 ereenish clay-like substance which, when crushed and examined under the micro- scope, appears to consist of a mixture of prehnite and yellowish-brown isotropic chloritic material. Garnet.—This mineral is an unusual associate of zeolites, but on An Gearna it is of fairly frequent occurrence. Its presence has already been noted by Professor Jupp,* who mentions it as occurring with epidote in the volcanic agglomerates of the Hebrides. Mr Currin, in the paper referred to above, notes its absence on Maol nan Damh, but there can be no doubt that Jupp in his description had in mind the locality under consideration, although the rocks in which I have detected the mineral show none of the characters of volcanic agglomerates. The garnet varies in colour from pale yellow, or colourless, to deep brown, a beautiful wine-yellow tint being Fig. 3.—Crystal of garnet from An Gearna, Mull. commonest. It occurs usually in crystals—occasionally ‘5 cm. in diameter— studding the scolecite near the junction of that mineral with the underlying layer of epidote, prehnite, etc., although sometimes it is found well in the centre of the zeolite, The common form is the rhombic dodecahedron the faces of which are usually curved and irregular, but the form shown in fig. 3 has also been noted. In erystals of this shape it is of interest that the icositetrahedral faces show the unusual type of striation, parallel to the intersection of the icositetrahedron and _ the dodecahedron, noted by Professor SHanpt in his description of the garnets from Corsie Hill Quarry. Besides occurring in direct association with scolecite, garnet has also been found with prehnite and albite, forming a layer between the two, and, rarely, imbedded in calcite. Much of the massive white prehnite found on the hill is spotted with garnet, and it is noteworthy that whenever the latter mineral occurs with scolecite, prehnite occurs in direct association with the two. * J. W. Jupp, “On the Secondary Rocks of Scotland,” Quart. Journ. Geol. Soc., vol. xxx, 1874, p. 241. + S. J. SHanp, Proc. Perth. Soc. Nat. Sci., 1907, vol. iv, p. 210. 8 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS The following analysis was made by Mr Raptey, and shows the mineral to be a typical lime-alumina garnet :— SiG, 4 ‘ = BWACEIS TOs + : ; “12 AO, <> 5 pe Se HesOs « : : 4:07 FeO... : ; "34 MnO . ' : : 53 (CoNi)O. a) iit. std. CaO, : é : 33°06 MgO . ; "45 KOR. : ; : ers) NaOo : } ; Leah 74 iO : : Seeeb ehcle H5O%6 10540 =. Dail H,0 above 105° C. i 20 100°19 Sp. g. = 3°61 at 7° C. Albite is not infrequently found in the vesicles in association with the minerals already described. It rarely shows crystal form and occurs usually as a massive layer of variable thickness underlying scolecite or prehnite. When scolecite is present the layer of albite is separated from it by an irregular layer of epidote, whilst when prehnite overlies the felspar a zone of garnet occasionally separates the two. Vesicles also occur lined with albite upon which a layer of green epidote is seated, and albite-filled veins are of frequent occurrence in the lavas. The layer underlying the albite occasionally presents unusual features. In a few specimens it consists of white massive albite speared by long black fibrous crystals, measuring up to ‘5 em. in length and consisting of augite which, under the microscope, has the purple tint typical of the Tertiary basaltic lavas. These crystals occasionally wander into the albite of the vesicle proper, but the point will be more fully discussed when the microscopic characters of the vesicle-minerals come under consideration. When crushed and examined under the microscope the albite is easily identified by the usual twinning, the symmetrical extinction angles on twin-lamelle, and by its mean refractive index, which, tested in oil, is approximately 1°534; it is rarely clear and transparent, being usually filled with inclusions. The lavas underlying the vesicular zone under consideration are much more compact and are characterised by much smaller amygdales which are filled with dense white or pink massive albite associated with epidote. Calcite-—From the mineral association so far described, one might have expected FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 4) calcite to be of frequent occurrence in the amyedales, but, as a matter of fact, it is rare and has been noted on few specimens. On one of these it occurs in a fairly large hollow in scolecite as a white, crystalline, much-cleaved mass, studded with yellow garnets. It les near the walls of the vesicle and is separated from the scolecite by a space partially filled with yellow epidote. The surface of the scolecite adjoining this space is flat and shows the characters possessed by the walls of the rectilinear cracks already described as occurring in this mineral. Support is thus lent to the view that these cracks were once occupied by calcite, which has since been removed. Chabante and Thomsonite.—In addition to scolecite these are the only other zeolites which have been noted on An Gearna, and they are extremely rare. The former mineral has been found on only one specimen, where it occurs as simple rhombohedra lining a cavity in white, massive prehnite spotted with reddish-brown garnet; albite and epidote are also present. Thomsonite has been noted in sheaf-like growths intimately associated with prehnite, albite, and scolecite. It has also been found on the north-east slope of Beinn Fhada associated with albite and prehnite, and with like associates on Coire Bheinn. ~ Hornblende can be observed occasionally in hand-specimen as tufts or agoregates of acicular crystals always near the walls of the vesicle. It spears the scolecite and is intimately associated with epidote. As will be shown later, it is a common microscopic mineral in the amygdales. THE MINERALS OF Maot NAN DAmuH. To appreciate correctly the significance of the most unusual mineral association described above, it is necessary to study the occurrences on the neighbouring hills. In the paper already. quoted, Mr Currie records the presence of scolecite, epidote, ealcite, celadonite, heulandite, and, much more rarely, stilbite, and prehnite from the upper lavas of the south-western side of Maol nan Damh, a spur of Ben More running in a south-westerly direction towards Loch Scridain (see fig. 1). He very kindly presented specimens to the Royal Scottish Museum, and from an examination of them I came to the conclusion that what had been determined as heulandite was in reality albite. In the course of correspondence with me Mr Curriz very frankly stated that he, too, was now of the same opinion, and from an examination of specimens collected by Mr E. M. Anprerson and myself there can be no doubt that albite is of very frequent occurrence in the vesicular lavas at this locality. The zeolite-filled cavities are if anything larger and more abundant than those of An Gearna, and it is clear that here we are dealing with a similar set of rocks. Scolecite, as described by Mr Currin, is by far the most abundant mineral, and associated with it are epidote, chlorite, albite, calcite, and, more rarely, prehnite. Heulandite and stilbite are exceedingly rare and have been noted on only one specimen, where they fill up the spaces between a crystalline aggregate of quartz, albite, and epidote, whilst chabazite has been noted in a vesicle filled with an TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 1). 2 10 MR W. F. P. MLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS aggregate of scolecite and prehnite. Cubes of pyrites have also been observed enclosed in massive prehnite. One specimen of special interest shows a hollow vesicle lined with, apparently, dull, opaque icositetrahedra of analcite upon which are seated crystals of pale green epidote ; the icositetrahedra when crushed and examined under the microscope are seen to consist of albite which has replaced the zeolite. The typical vesicles filled with fibrous scolecite sometimes reach a phenomenal size, the largest one found measuring 15 cm. along its greatest leneth. They are usually closely packed with scolecite, terminated crystals of which are exceedingly rare: a few have been found, however, showing the combination b(010), m(110), d(101), o(111). In the simplest cases the scolecite is seated upon a layer of chlorite, but such specimens are rather exceptional. A more common association is (beginning from the wall of the amygdale) chlorite with deep green epidote, then a layer of crystalline albite, white or pale pink in colour, and of variable thickness, and, lastly, scolecite which fills the vesicle; a layer of prehnite sometimes intervenes between the scolecite and albite. Calcite occurs fairly frequently in crystalline, much-cleaved masses enclosed in the scolecite near the junction with the underlying minerals. Epidote is exceedingly abundant. When it occurs with scolecite it is always found near the walls of the vesicle in intimate association with chlorite. It forms groups of dark green radiating crystals which sometimes spear the scolecite and are occasionally completely enclosed by it. In a second type of vesicle, where scolecite is absent, it occurs in groups of divergent crystals seated upon albite, whilst in a third type the amygdale is packed with massive chlorite, in the centre of which is a kernel of fibrous epidote. ; Chlorite is universally present in the cavities and shows various interesting features, some of which have an important bearing on the question of the date of formation of the vesicle-minerals. It was obviously one of the first to be deposited, and forms a dark green, compact, and massive layer lining the walls of the amygdale. This layer varies considerably in thickness: in some cases, especially when the vesicle is small, it fills the cavity ; in others it forms a thin and patchy priming for the overlying minerals, which may be scolecite, albite, epidote, or prehnite. In many cases the chlorite is quite homogeneous, but it often contains crystals and grains of deep green epidote, whilst in a few specimens acicular crystals of black augite, measuring up to ‘4 cm. in length, can be detected. ‘These crystals occur usually at or near the junction of the chlorite with the rock, but occasionally they are found well within the chlorite layer and even wandering into the overlying epidote, albite, and prehnite. Sometimes they are arranged radially to the walls of the vesicle to which they are often attached, but in many instances they le tangentially to the boundary and have no visible point of attachment to the rock. On one specimen the layer of chlorite and augite can be seen to connect with a vein packed with similar material which pierces the lava. The crystals of augite in the centre FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. Lit of the vein are arranged roughly parallel to the walls. A few cavities have been found in which prehnite is specially abundant. The underlying layer consists of chlorite speared with acicular augite with which albite is intimately associated ; in such cases the augite is occasionally found enclosed in the immediately overlying prehnite. “Owing to the dark colour of the chlorite and the frequent occurrence of fibrous epidote, it is difficult to detect these acicular augites in hand-specimen. They are best seen on the weathered surface of the chlorite, which is pale, and they can be readily identified by an examination of a little crushed material under the microscope, when the purple colour, cleavage, and positive optical sign distinguish the mineral at once from epidote. They are a prominent feature in sections cut through the junction of the rock with the vesicle, and their significance will be discussed when the origin of the zeolites comes to be considered. In addition to vesicles, the lavas also carry veins filled with albite in which are, occasionally, clear crystals of quartz and rhombohedra of chabazite. Veins filled with albite, solecite, and prehnite also occur, and there are other minerals which can only be detected under the microscope, and which will be described later. Meanwhile it is interesting to compare the differences, obvious in hand-specimen, between the minerals of Maol nan Damh and those from An Gearna. First, we may note that the dark green chloritic layer underlying the zeolites is much sharper and better developed on Maol nan Damh than on An Gearna, where its place is usually taken by a confused zone of yellow epidote with which tufts of green hornblende are sometimes associated. The second notable difference lies in the colour of the epidote predominant at the two localities. At Maol nan Damh it is typically of a deep, bottle-green shade, whilst at An Gearna it is usually pale yellow, brown, or pink; the last variety has not been found at Maol nan Damh. A third difference lies in the abundance of garnet on An Gearna and its absence or extreme rarity at Maol nan Damh. On one or two specimens collected by myself I have noted very small crystals in microscopic section, whilst on another collected by Mr ANpERsoN it is visible to the unaided eye; Mr Currie notes its absence in the material collected by him. It is clear, therefore, that on An Gearna we are dealing with the somewhat metamorphosed representatives of the vesicle-minerals of Maol nan Damh, and it is interesting to trace this well-defined zone towards the large intrusive mass of granophyre which forms the centre of Beinn Fhada, a ridge running parallel to An Gearna at a distance to the north-west of about a mile. + THe MiInerRAts oF Beinn Fuapa. On this hil] the cavities are neither so plentiful nor so large as those obtained - at the localities previously described, but good specimens can be obtained from the 12 MR W. F. P. MLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS serees at the north-western extremity of the ridge at heights of 1200-1500 feet. There the vesicles are filled with scolecite underlain by a layer of pale green or yellow epidote intimately associated with garnet, prehnite, and, frequently, albite. The boundary of this layer with the zeolite is highly irregular, and every now and then it sends off growths into the scolecite, tufted aggregates of which can often be seen enclosed in the prehnite and epidote. On some of the specimens there is a white massive mineral which occasionally merges into opaque, white tufts of zeolite. When examined under the microscope the massive mineral proves to be prehnite, which is clearly replacing the scolecite originally in the vesicle. Beautiful specimens also occur in which the scolecite is sprinkled with groups of crystals of epidote and garnet, pale pink, yellow, or even red in colour. A peculiar vesicular, pale gray lava has also been noted in which the vesicles present rather unusual features. They are lined with a thin layer of black chlorite, which is succeeded by epidote or stilbite, sometimes well crystallised, or by a massive aggregate of the two in which the epidote occurs as grains and crystals enclosed in the zeolite. When the stilbite is absent the epidote is abundant, and vice versa. Some of the vesicles, again, are lined with dull brown botryoidal material which, on examination, proves to be epidote coated with prehnite. A study of the vesicles in this rock strongly suggests that some of the epidote at least has been developed at the expense of the lime-bearing zeolite. Thomsonite is occasionally found on this part of the hill. It is underlain by the usual zone of epidote and prehnite, spotted and veined, however, with albite. Along the north-east slope of the hill the zeolites and, indeed, amygdales get much rarer as the junction of the granophyre with the lavas is approached, and, when found, they show unmistakable signs of having been baked and altered. At a point almost 8.S.W. of the summit of Beinn a’ Ghraig prehnite is of common occurrence in the vesicles and, occasionally, white fibrous masses are found which look like scolecite, but, on examination, prove to be prehnite, which is replacing it. On one or two specimens cleaved masses of white calcite occupy hollows in massive white prehnite, and at the junction of the two minerals there are exceedingly minute crystals of white, perfectly colourless garnet. Still nearer the granophyre the original character of the amygdales is completely changed ; the junction with the walls loses its sharpness, and the material filling the cavities seems to merge into the surrounding rock. Prehnite veined and riddled with a pale yellow epidote and garnet, pale yellow to almost black in colour, is very common, whilst in other cases the vesicle is filled with a pale pink massive material which consists largely of a mixture of garnet and epidote. It is thus clear from the field evidence that. the zeolites were formed before the intrusion of the granophyre and have been metamorphosed by it. Microscopic exami- nation of a series of sections through the amygdales confirms this conclusion and reveals a number of interesting contact materials not visible in hand-specimen. But, FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 13 to understand correctly the course of the metamorphism, it is necessary to find out what were the original minerals in the vesicles, and, if possible, the date of their formation. Interesting light is thrown on these questions by a study of the micro- scopic characters of the amygdales of Maol nan Damh, which also shows the relation- ships of the minerals to each other and to the lava in which they occur. THE ORIGIN AND RELATIONSHIPS OF THE VESICLE-MINERALS. I. Introduction. The question of the origin of the zeolites so frequently found filling the vesicles of lavas is one which has excited very great diversity of opinion. In modern works of reference on mineralogy they are usually classed as secondary minerals which owe their origin to the decomposition of the minerals of the lava at a date subsequent to its consolidation.* This view was strongly supported by the late Mr J. G. Goopcuicp, who expressed the opinion that they were formed by the action of percolating surface waters upon the original minerals of the rock. He seems to have had doubts, however, about this explanation holding good for the occurrences under description, for he refers specially to the association in the centre of Mull as owing its origin perhaps to solfataric action. In this he no doubt was following Professor Jupp, who expresses that view in the memoir already cited. Dr A. Harker, in his description of the Tertiary igneous rocks of Skye, regards the zeolites as the products of the final phase of consolidation of the lavas,§ and states that the subsequent changes produced by the action of percolating meteoric waters are of a different order and readily distinguishable from the process of zeolite- formation. ‘This view has been strongly advocated by Mr James Srracuay, || who considers that zeolites, agates, green earths, etc., were formed during the last period of cooling of the lavas in which they occur. Similar explanations for certain occurrences of zeolites have been offered by BLumricu 4 and PrELiKan,** whilst Dr J. 8. Fuerr,sf in his account of the teschenites of the Edinburgh district, states that some of the analcite in these rocks may be primary. Directly bearing upon the question of the origin of zeolites are the occurrences in the syenite pegmatite veins of the Norwegian syenites ascribed by BroccEr ti to * Cf. Hintze, Handbuch der Mineralogie, vol. ii, p. 1658. + J. G. Gooncuixp, “On the Genesis of Some Scottish Minerals,” Proc. Roy. Phys. Soc. Edin., 1899, vol. xiv, p. 190. i Loe: cit., p» 200, § A. Harker, Mem. Geol. Surv., “The Tertiary Igneous Rocks of Skye,” 1904, p. 45. || JAmMes Srracwan, “The Carnmoney Chalcedony, its Occurrence and Origin (with a General Note on the Formation of Secondary Siliceous Minerals in Volcanic Lavas),” Proc. Belfast Nat. Field Olub, 1906, vol, 1i, appendices vii and viii, p. 336. “| J. Buumricn, Tschermak’s Min. Pet. Mittheil, 1892, vol. xiii, p. 482. ** A, PuLIKAN, thid., 1906, vol. xxv, p. 113, and Sttz-Ber. Wiener Akad., 1901, iii, p. 341. tt Mem. Geol. Surv., “The Neighbourhood of Edinburgh,” 1910, p. 296. {tt W. C. Broaeer, Zectschrift fir Krist. w. Min., 1890, vol. xvi, p. 168. 14 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS the action of magmatic solutions. C. N. Fenner* has given full descriptions of the mode of occurrence of zeolites and other minerals in the Watchung basalt, New Jersey. He shows how the minerals were deposited in a definite order from aqueous solutions and how the earlier anhydrous and slightly hydrous ones are succeeded and replaced by the later-formed, more hydrous compounds. He considers that the zeolites and their associates were deposited during the cooling of the basalt, but, from various facts connected with the occurrence, concludes that the water which caused the change was derived largely from the underlying sediments. II. Petrography of the Vesicles. Confining ourselves for the present to the well-defined scolecite-bearing zone on Maol nan Damh, we find that, under the microscope, the rock proves to be a typical olivine basalt. Fresh olivine is absent, but chloritic pseudomorphs after it can be detected. The augite occurs in somewhat large sub-ophitic plates and is of the purple colour so common amongst the Tertiary basalts; it is frequently altered to chlorite, but much of it is quite fresh. The felspar, which occurs in small laths piercing the augite, contains veins and inclusions of chlorite and has been albitised. Owing to the abundance of inclusions it is often difficult to apply the Becke test to the laths; but whenever a determination could be made the refractive index was found to be below that of balsam, and the mineral appears in most cases to be albite or, at least, a very acid plagioclase. In the coarser portions of the basalt the albitisation is very pronounced, and sometimes—presumably when the original felspar was highly basic—crystals are now represented merely by a chlorite pseudomorph with or without an albite rim. This point has been observed in some of the rocks from Devonshire.t The peculiarity of albitisation is of general occurrence in the rocks around the plutonic centre in Mull, and, as epidote is also characteristically present, an interdependence of the two phenomena is suggested. In addition to the above-mentioned minerals, black oxide of iron is also present in fair quantity in the basalt. Sections cut through the junction of the rock with the amygdales show interesting features which not only reveal the order of events during the deposition of the vesicle-minerals, but also suggest that they were deposited during the cooling of the lava itself. The identification of the minerals is a comparatively simple matter, and, owing to the abundant development of the various species on the specimens selected, the determinations can usually be checked by an examination of the various powders in oils of suitable refractive index. In general the zeolites are * ©. N. Fenner, “The Watchung Basalt and the Paragenesis of its Zeolites and Other Secondary Minerals,” Ann. N.Y. Acad. Sci., 1910, vol. xx, pp. 93-187. + Loc. cit., p. 106. t K. Busz, Neues Jahrbuch fir Mineralogie, 1896, vol. i, p. 59; J. 8. Fuerr, Mem, Geol. Surv., “The Geology of Newton Abbot,” 1913, p. 60. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 15 difficult to determine under the microscope, as in many cases they possess low refractive indices, weak double refraction, and anomalous optical characters which often make it impossible accurately to determine individual members by examination of thin sections alone. In the present case, however, the number of zeolites is limited and the individual species can be readily identified. Scolecite occurs in long fibrous crystals (refractive index, 1:5) which are weakly birefringent, extinguish obliquely, and have the zone of elongation negative. Trans- verse sections across the fibres have very low birefringence, and in many cases show the emergence of an acute, negative bisectrix with a Jow axial angle. These char- acters distinguish it at once from thomsonite (and natrolite), which shows much stronger birefringence, straight extinction, and the emergence of an acute, positive bisectrix perpendicular to the direction of elongation. The axial angle is large; and the optic axial plane is perpendicular to the zone of elongation, which is consequently sometimes positive and sometimes negative: in the former case the section shows low birefringence and is perpendicular to the acute bisectrix. The only other zeolite observed in the slides is heulandite, and it is of rare occurrence. It shows good cleavage and can be always determined by the fact that cleavage flakes show the emergence of a positive, acute bisectrix. The prehnite is characterised by good cleavage, high refraction and birefringence, straight extinction and positive optical character. It occurs in large, compact, colourless plates and spherules which usually show wavy extinction under crossed nicols. They are often built up of sectors which radiate from a point in the plate, and, on being rotated under crossed nicols, the section shows two dark hyperbole _ which close up and open out in directions at right angles to each other. Idiomorphic erystals with uniform extinction have not been found, but areas large enough to show the characteristic optical characters are quite common. The determination of the epidote and the albite offers no special difficulty, as they are quite typical. The albite, however, is sometimes so turbid that it is difficult to identify it, and in such cases the mineral was always separated from the hand-specimen, powdered, and examined in oil of refractive index 1°534 under the microscope. No attempt was made to identify the different kinds of chlorite present. The mineral varies in colour from pale yellowish-green to deep green, the latter variety being distinctly pleochroic. It occurs usually in small, vermicular growths, but is also found in fibrous and platy forms. With the help of the sections itis possible roughly to divide the vesicles into two main types :— (a) Those vesicles whose junctions with the rock are sharp, and which do not contain any of the igneous minerals of the rock. (b) Those vesicles whose junctions are not so sharp, and which contain, especially in their outermost zones, igneous minerals similar to those found in the rock. 16 MR W. F. P. MLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS (a) This type of vesicle tends to be common in the more finely grained portions of the lava, and, with exceptions, the larger specimens conform to it. In the simplest cases the amygdale is lined with a layer of chlorite containing an occasional lath of albite and succeeded by scolecite which is crowded by inclusions of chlorite for some distance from its highly irreeular junction with that mineral [73].* In the more complex types spherules of prehnite enclosing chlorite and, occasionally, minute greenish fibres, which may be augite, occur sporadically on the walls of the amygdale. They are succeeded by a zone of chlorite, enclosing erystals of turbid albite and an occasional irregular grain of yellow epidote, and this, in turn, gives way to a layer of cloudy albite; the rest of the amygdale is packed with scolecite into which some of the albite occasionally wanders [74]. The spherules of prehnite are roughly hemispherical in shape, and under crossed nicols the black bar travels round the section when it is rotated. Not infrequently the extinction direction is continued into the rock so as roughly to complete the circle, and close examination shows that the albite laths of the rock in the complementary hemispherical area are replaced by prehnite which is optically continuous with the prehnite of the amygdale. Under high powers minute crystals of purple augite can be seen projecting into the spherules of the vesicles and converted at their tips to a green mineral which may be pyroxene or amphibole. The replace- ment of albite by prehnite is also shown by the fact that occasionally the spherules contain small rectangular turbid areas exactly similar in appearance to the albite laths, but which now consist of prehnite [113]. The same relationship of scolecite to albite is also exhibited. The junction between the two minerals is highly irregular, and the scolecite frequently contains corroded crystals of albite which it is clearly replacing. The further the. albite is removed from the junction the more it is corroded, until finally it is entirely replaced by scolecite, the only trace left of it being a turbid patch corresponding in outline to the original lath. When epidote occurs it is always greenish-yellow in colour and shows the usual pleochroism. It is found chiefly in the outer zone of the amygdale, and is frequently studded with inclusions of chlorite. Occasionally it encloses sharp, idiomorphic crystals of albite somewhat clearer than usual [71], and its junctions with that mineral are often quite sharp and well defined, although sometimes the albite exhibits outlines suggestive of corrosion | 63]. The relationships of prehnite and scolecite to epidote offer some interesting features. In the case of prehnite one finds that the junctions between it and. the epidote are not infrequently sharp, but often there are traces of corrosion of the epidote crystal. Particularly is this the case when the epidote is enclosed in the centre of the prehnite spherules, where it often occurs as a patch showing irregular and corroded junctions with numerous minute fibres projecting into the prelnite. * The numbers refer to a collection of slides in the Royal Scottish Museum. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. Ali When the epidote is bounded by scolecite, this peculiarity is even better shown. Large crystals of epidote end against the zeolite in a series of fibrous tufts, whilst the outline of the crystal is frequently interrupted by little rectangular patches filled with clear scolecite; occasionally a large crystal has been dissected into a number of separate fragments which extinguish simultaneously under crossed nicols [113]. The junction of the prehnite with the scolecite is never a sharp one. The spherules of prehnite are separated from the zeolite by a band of fibrous, crystalline prehnite, which terminates against the scolecite in a highly irregular manner. Irregular patches of prehnite are also found well within the scolecite layer. The minerals in these vesicles were thus deposited in the following order :— (1) albite, (2) chlorite, (3) epidote, (4) prehnite, (5) scolecite, and there is distinct evidence that with changing conditions some of the material formed in the first stages has been partly replaced by later-formed minerals. A marked feature in some of the slides is that the rock for some distance around the vesicle is very much altered and shows a large development of chlorite and black oxide of iron. This altered zone is sometimes vesicular, the vesicles being filled with epidote or chlorite, or a mixture of the two [67], and it grades off into the normal basalt. (b) This type of vesicle is best developed in the coarser basalt, and is usually characterised by a considerable amount of prehnite, although there is a good deal of variety in the way in which the minerals occur. In the simplest cases small cavities are filled with prehnite, which encloses prismatic crystals of purple augite, fresh and unweathered and similar in all respects, except habit, to the ophitic augite of the surrounding rock. Frequently they are clearly attached to the walls of the amyegdale, but occasionally they are surrounded on all sides by prehnite and occur well in the centre of the vesicle. The mode of occurrence of these crystals and their difference in habit from the augite of the rock show clearly that the vesicles were the seat of igneous crystallisation after they had been formed [107]. Some of the larger vesicles show similar crystals round their junctions with the rock. When chlorite forms the first zone of the amygdale, the long prismatic augites project into it and are sometimes completely surrounded by it. Their boundaries are clear and sharp, although occasionally there is evidence of corrosion and alteration to hornblende and chlorite; but, generally, their fresh and unaltered condition contrasts rather forcibly with the decomposed state of much of the augite of the rock. The chlorite, which sometimes forms a layer of considerable thickness, may be succeeded by a confused zone of chlorite, albite, and yellowish-green epidote with a few patches of muddy scolecite. Upon this is seated a layer of turbid albite, and scolecite fills the central portion of the vesicle. Prehnite may be present as small spherules occurring here and there around the walls; it usually contains a fair amount of chlorite, and is frequently pierced by the large crystals of augite [65]. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 1). 3 18 MR W. F. P. MLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS In another type of vesicle the wall is lined with a more or less well-developed layer consisting of large acicular purple augites with idiomorphic laths of albite which sometimes pierce the augite. This layer may be succeeded immediately by scolecite intermingled with chlorite, which grades into a well-defined zone of chlorite upon which is seated scolecite, enclosing crystals of yellowish-green epidote. The purple augite usually projects from the augite-albite layer, but occasionally crystals are found surrounded by scolecite and at some distance from the walls of the vesicle ; a large projecting crystal has been noted in which a transverse fracture has been filled with scolecite. This augite is often quite fresh and unaltered, but sometimes it has been partially converted to green fibrous hornblende or yellow epidote similar to that found in the amygdales; crystals have been noted of which the one half is purple augite, and the other, yellow epidote. The laths of albite are full of inclusions, and they, too, wander into the overlying scolecite [70]. The augite-albite layer is usually thin and patchy, but specimens occur in which it reaches a thickness of 1 em. When it is well developed, chlorite is abundant in it and fills up the spaces between the large augite crystals which it frequently encloses. Sections across this layer show it to consist of crystals of purple augite, long skeletal crystals of magnetite, with albite, chlorite, and apatite. The augite is idiomorphic and has sharp junctions with the chlorite, although here and there it shows signs of corrosion and marginal alteration to chlorite and hornblende. When the long prismatic crystals are attached to the walls they project radially into the vesicle, but when they occur well within the layer they lie often with their long axes parallel to the walls. This layer may be succeeded by a zone of chlorite and epidote, the latter mineral showing the usual irregular junctions and full of chlorite inclusions; a layer of turbid albite is the next deposit, then a layer of prehnite, and finally scolecite [23], The prehnite in some cases is obviously replacing the turbid albite, for not only does it contain phantom crystals of that mineral but occasionally it occurs as veins cutting across the felspar, crystals of which are optically continuous on opposite sides of the vein. Some of the smaller amygdales show this phenomenon even better. The first layer is a well-defined zone of pegmatitic augite, albite, and chlorite, the augite showing the usual features of corrosion and, occasionally, occurring as detached erystals well towards the centre of the vesicle. This layer is succeeded by prehnite, more or less turbid and enclosing thoroughly corroded patches and phantom crystals of albite as well as a little chlorite. Scolecite comes next; it, too, contains corroded albite and also patches of prehnite with little phantom albites near their centres [111 }. Besides being highly vesicular, the lavas are sometimes traversed by white veins showing at their junction with the rock a very well-developed pegmatitic layer of augite, albite, and chlorite. Under the microscope this layer shows the features already described: namely, the long prismatic augites, often roughly parallel to the walls, the turbid albite, the magnetite, and the chlorite. Upon it is deposited a layer of turbid albite with some prehnite and small nests of scolecite. The latter mineral FROM THE TERTIARY IAVAS AROUND BEN MORE, MULL. 19 is clear and transparent and contains erystals of epidote, showing distinct signs of corrosion, and also fibres and crystals of pale green slightly pleochroic augite [109]. These veins resemble in some respects the contemporaneous veins described by Dr J. 8S. Fierr* as occurring in the teschenite of the Barnton railway cutting, Midlothian. A few exceptional vesicles have also been noted. Mention has already been made of one lined with albite pseudomorphous after analcite. Under the microscope the pegmatitic layer of augite, albite, magnetite, and chlorite is seen to be well developed and is succeeded by the albite-pseudomorphs. A smaller vesicle underlies the main one and is filled with albite secondary after analeite, chlorite moulded upon the pseudomorphs and purple augite which occurs in large crystals lying around the walls or stretching across the vesicle and projecting into the larger amygdale through a canal which connects the two. The mode of occurrence of albite in this amygdale points to an increase in temperature which may have been caused by local circum- stances during the period of vesicle-infilling or by the injection of some intrusion long after the zeolite was formed. ‘The absence of any other sign of metamorphism in the rock points to the former explanation being the probable one. Another specimen shows a fairly large vesicle filled with chlorite throughout which are scattered a number of white spots. In thin section the rock is a typical basalt showing the usual features, but the first mineral to be deposited in the amygdale was garnet. It occurs as a pale brown layer of variable thickness and exhibits sions of corrosion on its inner margin where it is overlain by scolecite or chlorite. In places signs of an hexagonal outline are visible in some of the grains lying away from the layer, but they are always corroded and tend to be circular. The further the grains are removed from the walls the more do they show signs of replacement, and in one or two instances they are represented by spots which are doubly refracting and which merge into the surrounding zeolite. The garnet is associated with yellow epidote, which seems to replace it in parts, and also with what — is apparently diopside. The latter mineral occurs as prisms showing good cleavage, high refraction and birefringence, and an extinction angle of about 33°; the mineral is positive. The scolecite occurs in the usual fibrous forms. It is sometimes quite clear, especially when it is adjacent to the garnet, but more often it is turbid; it contains corroded grains of garnet, fibres of diopside, which also show signs of replace- ment, and epidote. Patches of heulandite are also present in the scolecite ; they contain grains of garnet and epidote and their junctions with the scolecite are highly irregular [78 |. Grains of garnet have also been observed in a few other slides. In one of these [23] it occurs as minute crystals, six-sided in section, enclosed in clear scolecite or in chlorite ; in another slide there is a fairly well-defined band of greatly corroded grains lying between the rock and the overlying chlorite-scolecite zone [68]; whilst in a * Mem. Geol. Surv., “The Neighbourhood of Edinburgh,” 1910, p. 298. 20 MR W. F. P. MLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS third there is a small vesicle filled with chlorite, epidote, and clear albite with corroded grains of garnet enclosed in the albite and epidote [66 }. A difficulty arises in the interpretation of these occurrences, for, as noted above, garnet is typically developed in the vesicles which have undergone metamorphism due to the intrusion of the granophyre. As will be shown later, however, the development of garnet in these circumstances is accompanied by other changes in the vesicles, none of which is shown in the slides described above. It may also be noted that the garnet in the Maol nan Dambh slides always occurs practically at the walls of the amygdale, and shows obvious signs of corrosion and replacement. On the whole, the evidence points to the garnet and diopside being original vesicle-minerals, probably amongst the first to be deposited, which became unstable and experienced re-solution and replacement as the conditions changed and new minerals were de- posited. Frnner records the similar occurrence of garnet in the Watchung basalt, where it also shows signs of corrosion and replacement by later-formed minerals.* Thomsonite has not been detected in the vesicles on Maol nan Damh, but it occurs on the neighbouring hill of Coire Bheinn, which les about two miles to the west. A section through one of the vesicles shows a thick layer of albite, which is only slightly turbid, succeeded by yellow epidote (in places), then prehnite, and finally thomsonite. The albite shows slight traces of corrosion against the epidote, but its boundaries with that mineral, though often rounded, are sharp and well defined. The epidote ends against the prehnite in a series of tufted aggregates and, when a plate of that mineral adjoins both albite and prehnite, the contrast between the two junctions is most marked. Albite shows corroded junctions with prehnite, which, in turn, has irregular outlines with thomsonite. The sequence is obviously (1) albite, (2) epidote, (3) prehnite, (4) thomsonite, and the earlier-formed minerals show signs of having been unstable towards the conditions under which the later ones were deposited. III. Sequence of Events in the Vesicles. In his summary of the characters of the occurrence under consideration, Mr CurrtE dwells upon the peculiar facies of lime-bearing minerals—a facies not recorded from any other locality in Scotland nor, as far as I am aware, from any locality hitherto described—and ascribes the origin of the vesicle-minerals to pneumatolytic action. He did not examine the petrographical characters of the lava, but inferred from the minerals in the amygdales that it must have been originally a basalt containing a basic plagioclase, a pyroxene (either rhombic or monoclinic), and magnetite or olivine. teasoning from the abundance of green earth, he also suggested the original presence of biotite. It is clear from an examination of numerous sections that the rock is a typical, ophitic olivine basalt, now, of course, in an altered condition and, neglecting mean- * C. N. Fenner, loc. cit., p. 138. + Loc. cit., pp. 227-28, FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 21 while the exceptional cases, we may sum up the main features of the occurrence as follows :— 1. The rock is an olivine basalt with the purple augite typical of most of the lavas in Mull. 2. The felspars are albitised and veined with chlorite, and the olivine and augite also show alteration to that mineral. 3. The magma was rich in gases at the time of its eruption, and hence the vesicles are large and plentiful. 4, The cavities were sometimes the seat of igneous crystallisation subsequent to their formation, and, in the coarser portions of the rock, the vesicle-minerals have grown upon a pegmatitic layer of augite, albite, magnetite, and chlorite; crystals from this layer are occasionally enclosed in the contents of the amygdales. 5. With the exceptions of chlorite and albite, the vesicle-minerals have lime for their principal base ; zeolites containing soda are exceedingly rare. 6. There is evidence of a more or less definite sequence in the deposition of the minerals, namely :— (a) A layer of augite, albite, magnetite, and chlorite. (b) Albite. (c) Epidote. (ad) Prehnite. (e) Scolecite. Other sequences are :— (a) Garnet. (a) Garnet. (a) Albite. (b) Diopside. (b) Albite. (b) Hpidote. (c) Epidote. (c) Epidote. (c) Prehnite. (d) Chlorite. (2d) Thomsonite. (e) Scolecite. (f) Heulandite. 7. Calcite is of rare occurrence, and hydrous oxides of iron and other typical oxi- dation products of weathering are almost entirely absent. These facts suggest that the vesicle-minerals were deposited during the final period of cooling of the rocks, which seems to have been a long one. During the period of igneous crystallisation the normal pyrogenetic minerals, olivine, labradorite, and augite, were formed, the augite crystallising last. Consequently the magma, in the last stages of this phase, was rich in the augite-forming bases, and whilst the rock was solid, but necessarily at a high temperature, these formed the large, acicular augites occurring in some of the vesicles. When felspar occurs with them it is always albite in which inclusions are usually abundant. It forms turbid, lath-shaped crystals with much included chlorite, and also irregular patches much clearer and compara- tively free from chlorite, but containing frequently grains and crystals of purple 22 MR W. F. P. M‘LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS augite. It is difficult to determine whether the albite of the laths is primary or secondary in the sense that it is the albitised representative of a plagioclase originally more basic. There can be little doubt, however, that the clear compact patches were most probably deposited originally as albite, and, as they contain crystals of augite, no long period can have elapsed between the formation of the two minerals. The presence of this albite points to a transition from igneous crystallisation to a hydro- thermal stage which would favour the formation of alkali-felspar* and also of the chlorite present in the pegmatitic layer. During this hydrothermal stage the infilling of the vesicles was completed, the constituent minerals being derived partly from the residual material of the magma and partly from the breaking down of the igneous minerals which had already con- solidated. In some of the chlorite and epidote we see the hydrothermal representa- tives of augite, a conclusion supported by the occasional occurrence, as noted above, of augite crystals in the cavities partly replaced by epidote, whilst the labradorite finds its counterpart in some of the albite, scolecite, and prehnite. It is evident, however, that whilst hydrothermal crystallisation of the residual magma accounts for the formation of a portion of the contents of the vesicles it probably does not account for them all. Some of the vesicles, especially the smaller ones, are packed with chlorite containing a kernel of epidote, or albite, or both; but the great majority contain a proportion of scolecite much larger than that of any other con- stituent. This fact can be accounted for by the albitisation of the felspars and the partial conversion of the augite of the rock to chlorite and hornblende, each of which changes liberates lime. Albitisation is now known to be a fairly common occurrence in the igneous rocks of this country,f but the nature of the process seems to be imperfectly understood. In the cases cited by Battey and GRABHAM, it is a pneumatolytic phenomenon ascribed by them to the presence of carbon dioxide (or some other unknown constituent) in the original magma, which thus retained a large amount of soda in solution. The liquor then began to act on the felspars and albitised them. ‘They also record the association with the albite of chlorite and epidote, the latter mineral representing some of the lime derived from the felspar; in most cases, however, the lime has been carried off in solution. Drwry and Fuerr also ascribe the albitisation of the felspar of the spilites to pneumatolytic action.} At Maol nan Damh various considerations point to the albitisation having been a pneumatolytic change which took place during the cooling of the lava and marked the first step in the filling of the vesicles. This view gains support from the rarity of calcite and other oxidation products, which excludes the assumption that the change may be due to weathering; there is also the additional fact that rocks of * Of. BF. W. Crarke, Data of Geochemistry, 2nd edition, 1911, p. 348. + Cf. E. B. Batnry and G, W. Grasnay, “ Albitisation of Basie Plagioclase Felspars, ” Geol. Mua vol. vi, 1909, pp. 250-66 ; H. Dewey and J. S. Fert, “On Some British Pillow-Lavas,” zbd., vol. viii, 1911, p. 246. } Loc. cit., p. 204. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 23 similar petrographical character are common among the plateau-basalts, and their felspars have not been albitised. It must, however, be conceded that albitisation is of universal occurrence in the rocks around the plutonic centre, although, in this area, it is only in the localities mentioned that the peculiar association of minerals in the vesicles has been detected. The cause and date of this general albitisation can be settled only by an examination of the rocks over the whole area, but the frequent association of the vesicle-minerals with those of igneous origin in the basalt of Maol nan Dambh indicates that the change took place, in this case, during the last stages of cooling of the lava itself. The albitised felspars almost invariably contain veins and patches of chlorite, which, as noted above, sometimes replaces entirely the original plagioclase. There does not seem to have been much introduction of soda, and the process was probably earried out by solutions containing chlorite and capable of dissolving the anorthite of the plagioclase. These solutions also partially chloritised the augite, and subse- quently deposited their loads in the vesicles in the form of chlorite, prehnite, epidote, and scolecite, any surplus soda going to form albite, and, only very exceptionally, a zeolite containing soda, This rarity of soda-zeolites is an important point. From the amount of secondary albite in the vesicles it is clear that soda was present in considerable quantity in the solutions, yet in nearly every case it has been deposited as albite—a fact which indicates strongly that the vesicle-minerals were deposited at a fairly high temperature. Dok .TeR, in a suggestive paper,* has pointed out that, under laboratory conditions, a mixture of soda, alumina, silica, and water deposits analcite between temperatures of 190° and 420° C.; at lower temperatures natrolite is formed, and at higher temperatures either nepheline or albite. The formation of analcite seems to depend largely upon the temperature and to a very slight extent upon the relative pro- portions of the constituents, for it has been formed from solutions of very different compositions.f The conditions influencing the formation of albite are not so clearly understood, but, as shown by the work of Baur, and FriepEt and Sarasin,§ a temperature of 500° C. appears to be necessary. As regards the sequence of events in the vesicles it is certain that their infilling was a continuous process, and the succession of minerals is what we would expect from deposition under conditions of falling temperature. Chlorite seems to have been deposited during all the stages, a fact remarked upon by FENNER|| in his account of the Watchung basalt, but only very exceptionally does the vesicle-albite enclose it. It appears to have been formed in large quantities during, or slightly previous to, the deposition of epidote, which frequently encloses it and never shows sharp junctions * C. Dorner, Tschermak’s Min, Pet. Mittheil., 1906, vol. xxv, p. 97. + Loc. cit., p. 102. t E. Baur, Zetts. f. physikal. Chimie, 1903, vol. xlii, p. 570. § C, Frirpet and E, Sarasin, Compt. rend., 1883, vol. xcvii, p. 290. || C. N. Fenner, loc, cit., p. 174, 24 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS against it. Albite was in many cases the earliest vesicle-mineral, for it occurs in chlorite and epidote, against which it shows usually sharp, idiomorphic outlines. The succession, albite, epidote, prehnite, scolecite, is proved in many cases, and illustrates very well what happens during deposition under hydrothermal conditions. The albite was deposited first and is enclosed in places by the epidote which replaces it but slightly ; with falling temperature prehnite is formed, which encloses and frequently corrodes and replaces the albite, a phenomenon also observed by FENNER,* who figures an example exactly similar to many of those seen in the present instance. Finally came the scolecite phase, during which the albite was also replaced,-the epidote corroded, and even the prehnite attacked. The survival of much of the pegmatitic augite raises a difficulty, for, although it exhibits signs of corrosion and occasional marginal alteration to hornblende and epidote, typically it is fresh and contrasts somewhat forcibly with the decomposed condition of a good deal of the augite of the rock. The explanation is probably due to the fact that when the solutions welled into the cavities they were saturated with chlorite and silicates of lime and soda, and therefore incapable of attacking the pyroxene. This is supported by the fact that in the subsequent phases albite is the mineral which shows in the highest degree the phenomenon of corrosion and replace- ment, the reason being that, after it had formed, it was constantly under the action of lime-bearing solutions which were able to decompose it. The evidence that the vesicles were filled during the last stages of cooling may be stated shortly as follows :— 1. The vesicles sometimes contain pyrogenetic minerals. 2. The mineral association as far as the evidence goes is one demanding a temperature above the normal. 3. The silicates (with the exception of chlorite) were deposited in order of increasing hydration : — (a) Albite . Na,O.Al,0;. 6810, (b) Epidote . 4CaO. 3Al,0,. 68i0,. H,O; H,O= 1°98 per cent. (c) Prehnite . 2CaO.3AI,0,. 38i0,.H,0O; H,O= 4°4 per cent. (d) Scolecite . CaO. Al,0,. 88i0,.8H,O; H,O =13°8 per cent. 4. There is direct evidence that albite, epidote, and prehnite in the order named each passed through a phase of stability when they were deposited, and that they were more or less unstable to the succeeding phases. | 5. Three periods, which grade into one another, can sometimes be recognised :— (a) Period of magmatic consolidation. (4) Period of pegmatitic crystallisation. (c) Period of infilling of the vesicles. The last-mentioned of the five points suggests ‘analogies with the three phases 1 =D) * Loc, cit., p. 126. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 25 described by Bréccrr* in his account of the syenite-pegmatite veins of Norway, during the last of which the zeolites were deposited. Tue APPLICATION OF THE TERM “ PROPYLITE.” It is clear from the descriptions of Professor Jupp already cited (ante, p. 2) that the vesicular basalt of Maol nan Damh, An Gearna, and Ben Fhada belongs to his group of propylites. The term, however, has been reserved by Rosenpuscu f for a peculiar type of altered andesite, and was used by Jupp in that sense. The propylitic alteration is regarded by him as due to solfataric action connected with the presence of the acid intrusions, but quite distinct from the effects of contact metamorphism (ante, p. 2), to which cause Sir ARCHIBALD GEIKIE ascribed the peculiar condition of these rocks (ate, p. 2). The lava of Maol nan Damh is a typical olivine basalt, and the evidence goes to show that its present characters and its peculiar vesicle-minerals arose during the formation of the rock itself. A propylite, according to RoseNnsBuscH,! is an andesite altered by solfataric action, and it is clear that in this restricted sense the term cannot be applied to the basalt under discussion. But the pneumatolytic changes which it has undergone bear a strong resemblance to the alteration produced by the conversion of an andesite to a propylite, and, as basalts with these peculiar characters have a wide distribution in Mull, it might be advisable to extend the term propylite to them also. It is natural to inquire into the cause of this peculiar condition of the basalts around the plutonic centres, a condition not observed so far in the plateau-ground; but the question can only be answered when the volcanic history of the district is made out. Meanwhile it is safe to say that the basalt of Maol nan Damh and some of the surrounding spurs must have been rich in vapours at the time of its eruption and was probably kept at a high tempera- ture for a prolonged time. This suggests that the lava now occupies a site near the focus from which it was erupted. THE METAMORPHISM OF THE VESICLE-MINERALS. I. Introduction. The subject of the metamorphism of the vesicle-minerals of lavas by intrusive masses of rock has been dealt with by Drs Harker and Marr,§ who note that the minerals of the amygdales are the first to show the effects of metamorphism.|| * W. C. Broaenr, loc. cit., pp. 160-81. q Jal: RosENBUSCH, RGattemahe Physiograplie der Massig gen Gesteine, 4th eaten, 1908, pp. 1102-05. { Loe. ctt., p. 1105. § A. HARKER and J. E. Marr, “The Shap Granite and the Associated Igneous and Metamorphic Rocks,” ane Journ. Geol. Soc., 1891, vol. xlvii, p. 292. || Loe. cit., p. 296. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 1). 4 26 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS The changes induced in the surrounding andesites by the Cheviot granite have been described by Mr H. Kynaston,* whilst Dr. J. J. H. Teaun has described the meta- morphic effects produced in the Arenig lavas by the Galloway granites. More closely related to the subject under consideration are the Tertiary basic lavas of Skye which, as described by Dr Harxrr,} have been considerably altered by the large intrusions of granite and gabbro. He notes. that the first minerals to be affected are the unstable contents of the vesicles, and records the interesting. conversion of lime-soda zeolites to lime-soda felspars ; epidote is also formed, and hornblende and biotite are developed at the expense of chlorite. The metamorphosed amygdales of An Gearna and Beinn Fhada afford most instructive material for study because, on account of their large size, they usually contain portions of the material originally filling them, and it is only in close proximity to the granophyre that all traces of their original contents are obliterated. It is, therefore, possible to follow with confidence the course and nature of the metamorphism. How far the metamorphism induced in the lavas is due to the granophyre and how far to the minor intrusions so common in the district, it is impossible to say. It is, however, a fact that, despite local variations, the intensity of the changes increases as the margin of the granophyre’ is approached. Il. Lhe Petrography of the Metamorphosed Amygdales. It has been remarked already that one of the differences, obvious in hand- specimen, between the amygdales of An Gearna and those from Maol nan Damh consists in the somewhat paler appearance of the outermost layer of specimens from the former locality. Sections through this layer show that it varies considerably in character. In some cases, especially when scolecite is abundant, it consists of long, bladed crystals of a pale green, pleochroic hornblende sometimes partially enclosed in large plates of albite full of inclusions. Associated with these are rod-like growths of black oxide of iron showing partial conversion to sphene. The hornblende shows good cleavage and frequently contains erains of magnetite and kernels of epidote. The albite encloses grains of epidote and crystals of hornblende. An occasional rounded mass of chlorite showing partial conversion to hornblende is also present. This layer is overlain by epidote, succeeded by scolecite, both of which enclose numerous fibres of hornblende. It is probable that the layer of hornblende and albite represents the pegmatitic zone of common occurrence in the vesicles of Maol nan Damh. ‘The conversion of augite into hornblende liberates lime which has * H. Kynaston, “Note on Contact Metamorphism round the Cheviot Granite,” Trans. Hd. Geol. Soc., 1899, vol. vill, pp. 18—26. + Mem. Geol. Surv., “The Silurian Rocks of Scotland,” 1899, p. 647. {| A. Harker, Mem. Geol. Surv., “The Tertiary Igneous Rocks of Skye,” 1904, p. 50. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 27 gone to form epidote and also to convert the chlorite to hornblende and the titaniferous oxide of iron to sphene [112]. In other amygdales, however, in which the chief contents are albite, prehnite, and epidote, the underlying layer consists of albite and augite which shows conversion to epidote and not to hornblende. Under the microscope the augite crystals are paler than those in the unaltered vesicles, and they contain kernels of epidote which, in some cases, replaces entirely the purple augite. A dull, turbid chloritic substance sometimes surrounds the augite crystals. It is not clear whether this change of augite to epidote is to be ascribed to metamorphism or to pneumatolysis during the original filling of the vesicle, since, as noted above, a partial conversion of augite to epidote has been observed in the unaltered vesicles [111]. Other instances occur in which the vesicle-minerals have been completely metamorphosed and the pegmatitic augite at the junction is practically unaltered, the only change being a slight marginal corrosion and alteration to hornblende. In such cases, however, the augite is enclosed by albite and was not originally in intimate association with an unstable zeolite which would give rise to steam with increase of temperature [26 |. Some of the most interesting points connected with the metamorphism are shown by the zeolites, scolecite and thomsonite. In the case of scolecite the first sign of alteration consists in the production of a slight turbidity, and the subsequent changes seem to depend to some extent on the minerals with which the zeolite was in intimate association. Alteration to epidote and prehnite is exceedingly common. _ The secondary epidote occurs sometimes as a network of interlocking crystals with the spaces filled up with altered scolecite, and it can be distinguished in several ways from the primary epidote of the amygdales. Thus, it is usually colourless and shows no pleochroism, and it sometimes occurs as colourless rims surrounding a yellow kernel which often shows the original corroded outline and the inclusions of chlorite so typical of the primary epidote; the rim occasionally possesses a zonal arrangement parallel to the boundary of the core, and it shows weaker bire- fringence [15, 17]. The prehnite produced by metamorphism can be readily distinguished both by its appearance and mode of occurrence from the original prehnite of the vesicles. It forms confused aggregates of small crystals which ramify into the zeolite and sometimes replace it fibre by fibre [100, 50, 81]. ‘he crystals are often large enough to permit of accurate determination under the microscope, but sometimes the fibres building the secondary aggregate replacing the zeolite are too small for that purpose, and, in such cases, the material was always crushed and the refractive index tested in a suitable oil. The secondary prehnite resembles in some respects the corroded rim which, as noted before, often separates the original prehnite from the scolecite ; but it can be~ distinguished by its much greater abundance in the 28 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS . metamorphosed amygdales, by the manner in which it encloses dissected scraps of scolecite which are sometimes optically continuous, and by the fact that it not infrequently forms true pseudomorphs after that mineral. A colourless pyroxene showing good cleavage and normal optical properties is a common mineral in the slides. It seems to have been formed around the inclusions of chlorite so common in the zeolite, for it is often dusted with turbid chloritic material. It occurs in fairly large crystals and plates and sometimes encloses scraps of scolecite [15]. It is identical in appearance to the pyroxene already described as a rare constituent of the vesicles on Mao nan Damh, but from its mode of occurrence in the present instance there can be no doubt that it is a product of metamorphism. Thus, when the epidote shows the core-and-rim structure already described, the pyroxene is frequently enclosed in the secondary epidote, especially near the junction with scolecite where the crystal often ends in a confused aggregate of the two minerals. Again, when the amygdale originally contained albite and scolecite in intimate association, the pyroxene has picked out and partially replaced the scolecite, whilst it never occurs in the albite. It sometimes occurs in prehnite [15, 16]. Garnet has also been formed at the expense of scolecite. It occurs either as idio- morphic dodecahedra, not infrequently showing pronounced zonal structure and optical anomalies, or as irregular patches traversing the zeolite. Inclusions are usually plentiful and consist of epidote, prehnite, and dull turbid fibres arranged in a sub- parallel manner towards the centre of the crystal [58]. Sometimes it forms true perimorphs consisting of a shell of garnet, pale brown in colour and with idiomorphic outlines, which encloses a spongy mass of secondary prehnite [43. | Thomsonite, though rare, affords an instructive example of the metamorphism, of a zeolite containing both soda and lime. It has not been detected on Maol nan Damh, but, as previously mentioned, it occurs on Coire Bheinn, where the sequence is albite, epidote, prehnite, thomsonite, the various minerals showing the relationships already described. On An Gearna and Beinn Fhada the thomsonite shows conversion to prehnite and albite. In some of the slides the fibrous crystals of thomsonite are partially replaced by crystalline aggregates of finely fibrous secondary prehnite in intimate association with albite which forms irregular, turbid masses full of in- clusions. ‘To distinguish this secondary metamorphic albite from the original albite of the vesicle is a matter of some difticulty; but its absence of idiomorphism, its constant association with” undoubted secondary prehnite, which it not infrequently encloses, and the occasional presence in it of fragments of thomsonite, all go to prove that it has been developed from the zeolite [50, 81]. Dr Harksr,* in his account of the metamorphosed amygdales of the Skye basalts, mentions that the lime-soda zeolites give rise to lime-soda felspars. So far, | have failed to detect in the vesicles felspars with a refractive index above that of balsam a fact which is most probably due to the presence of steam in the vesicles during the metamorphism. Under such conditions * A Harker, loc, cit., p. 51. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 29 basic plagioclases appear to be unstable, their representatives being albite and a lime- alumina silicate, such as prehnite, containing more or less combined water. Hornblende is a common mineral in the metamorphosed amygdales, and has been formed by the alteration of the chlorite. It varies considerably in colour, being some- times deep green and strongly pleochroic and sometimes very pale yellowish-green [9, 55, 59, 101]. It occurs in prehnite, epidote, and scolecite, and is always fibrous in habit. The change from chlorite to hornblende or biotite is quite a well-known effect of contact-metamorphism,* and, as lime is plentiful in the amygdales, and alkalies relatively scarce, hornblende has always been formed in preference to the mica. . : These changes, which can be followed in detail in the less altered amygdales, become much more intense as the granophyre is approached. The solid primary prehnite, as well as the fibrous secondary material derived from the zeolites, is con- verted to garnet and epidote. The garnet varies considerably in appearance. It may be white to dark brown in colour and occurs either as idiomorphic crystals in the prehnite or as large irregular masses traversing and enclosing it. The mineral is frequently isotropic, but optical anomalies are also common; in some cases large rounded crystals consist of a brown, isotropic centre and a colourless, birefringent rim. Inclusions of epidote and prehnite are fairly common [26, 32]. The epidote is colourless, and crystals show pronounced zonal structure and re- markable variation in birefringence; occasionally sections show indigo-blue polarisa- tion colours with positive optical sign, resembling in this respect zoisite rather than epidote. Inclusions of prehnite are sometimes present [25]. Sections through the amyedales collected from the screes lying nearest the grano- phyre margin on the north-east slope of Beinn Fhada show the most intense type of alteration noted. ‘The zeolites have completely vanished, and their place is taken by a dull turbid substance which preserves their radiate structure. It is isotropic, and under high powers is seen to consist of minute yellowish grains of garnet. Any prehnite which has survived is studded with small hexagonal sections of garnet which is also replacing the epidote. ‘he latter mineral occurs in crystals part of which consists of garnet, or it forms a corroded patch surrounded by garnet which sends off growths into the epidote [31, 40, 41]. Sphene is frequently present in the slides and forms grains and rounded crystals enclosed in the epidote or in the adjacent prehnite, whilst the chlorite has gone to hornblende, which sometimes forms compact green patches showing good cleavage and strong pleochroism | 46 |. Ill. Summary and Conclusions. It is clear from a study of the metamorphosed amygdales that they were filled ‘prior to the intrusion of the granophyre and that they participated in the meta- * A, Harker, loc. ctt., p. 50. 30 MR W. F. P. MSLINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS morphic changes which it produced. "These changes resulted in the production of the following minerals :— Prehnite. Garnet. Epidote. Sphene. Pyroxene. Albite. Hornblende. With the exception of sphene, all of the above minerals occur as original constituents of the cavities, and the result of the metamorphism has been to recrystallise them in the reverse order to that in which they were originally deposited. Thus with the lime-bearing silicates the metamorphic changes progress as follows :—scolecite > prehnite > epidote > garnet, whilst the original order of deposition was :—garnet > epidote > prehnite > scolecite. Similarly with the lime-soda zeolites the order with rising temperature is :—thomsonite + albite and prehnite > epidote > garnet, whilst the original order of deposition was :—albite > epidote > prehnite > thomsonite. The pyroxene and hornblende have been developed from the chlorite and scolecite or prehnite, whilst the sphene owes its origin to the titanium originally present in the augite and iron oxides of the rock, which found its way into the epidote (see ~ analysis) and probably also into the chlorite of the vesicles. The above evidence confirms in a striking way the view that the vesicles were originally filled under conditions of falling temperature, for, with the rising tempera- ture occasioned by the intrusion of the granophyre, the only change has been to obliterate the minerals formed originally ‘during the last and lowest temperature stages of the hydrothermal phase, and to form at their expense those minerals which crystallised in the first and highest temperature stages of that phase. Not only so, but the metamorphic minerals, considered from the point of view of their general distribution, have been developed in the reverse order to that in which their original representatives were laid down in the cavities. An interesting point arises from the abundance of garnet in the metamorphosed amygdales and its rarity in the unaltered ones. This is probably to be ascribed to the great difference between the pressures under which the two sets of minerals were formed. Dr L. L. Fermor* has pointed out that garnet is a high-pressure mineral, and—as recrystallisation of the vesicle-minerals under the influence of the intrusion must have taken place at a considerable depth and, consequently, at a considerable pressure—the tendency would be for it to form at the expense of the other, less dense, lime-alumina silicates. In other words, the heat of the granophyre, combined with the high pressure, was steadily making the vesicles less hydrous and more favourable to the formation of garnet. Under the hydrothermal conditions, on the other hand, the temperature was falling, the pressure was never great, and the vesicles were becoming more hydrous and less favourable to the formation of garnet, or the pre- servation of any that might have crystallised. To judge from the vestigial appear- *L, L. Fermor, Records Geol, Surv. India, 1918, vol. xliii, pt. 1, p. 42. FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 31 ance which the garnet presents in the vesicles where it does occur, it is quite possible that it may have been present more generally during the first phase of deposition and that it became obliterated during the succeeding stages. ACKNOWLEDGMENTS. I take this opportunity of expressing my best thanks to Dr J. 8. Fierr and Mr E. B. Barry, B.A., of H.M. Geological Survey, for, the interest which they have taken in the investigation and for the many valuable suggestions which they have made during the course of the work. To Dr Roperr Campse.t, of the University, and Mr F. D. Mirss, B.Sc., of the Heriot-Watt College, I am greatly indebted for help in the photographic part of the work. My gratitude is also due to the cauepie Trust for defraying the cost of illustrating this paper. EX PLANATION OF PLATES. Puate I, Fig. 1. Ophitic Basalt [107] with Vesicles, magnified 27 diameters.—The light area to the right is a large vesicle containing prehnite and chlorite ; at the bottom an acicular augite projects into the prehnite. The small vesicle to the left is filled with prehnite, enclosing a little chlorite ; the dark crystals projecting into the prehnite are augite. Maol nan Damh, Ben More, Mull. Fig. 2. Ophitic Basalt [70] with Vesicle, magnified 27 diameters.—To the right at the bottom is the rock, much decomposed, with numerous grains of magnetite. The large crystal projecting into the vesicle is purple augite pierced by laths of albite. To the left it shows a fracture filled with white scolecite, which occupies the rest of the cavity. The darker patches in the scolecite are chlorite. Just above the fracture in the augite is a crystal consisting of purple augite (dark) and pale yellow epidote (light). Maol nan Damh, Ben More, Mull. Fig. 3. Ophitic Basalt [2 3] with Vesicle, magnified 12 diameters.—The rock, at the bottom, is succeeded by a pegmatitic layer consisting of lath-shaped crystals of purple augite (white), patches of chlorite (dark), and albite. At the junction with the rock the augite projects radially to the wall, but towards the vesicle they lie tangentially. ‘This layer is succeeded by a zone of chlorite and epidote, above which lies albite more or less turbid. Maol nan Damh, Ben More, Mull. Fig. 4. Ophitic Basalt [110] with Pegmatitic Layer underlying Vesicle, magnified 27 diameters.—The layer on the right consists of lath-shaped crystals of augite, magnetite, chlorite, and albite. The clear, irregularly shaped patches are albite enclosing a few crystals of augite. Maol nan Damh, Ben More, Mull. Fig. 5. Ophitic Basalt [111] with Vesicle, magnified 27 diameters.—The rock, to the left, is a dense aggregate of augite, chlorite, and magnetite. The pegmatitic layer is well developed and consists of laths of augite, albite, magnetite, and chlorite. The white mineral in the vesicle is prehnite into which some of the underlying augite has wandered. Maol nan Damh, Ben More, Mull. Fig. 6. Ophitic Basalt [111] with Vesicle, magnified 27 diameters.—The dark patch at the foot is the rock, which here consists of augite and chlorite. The turbid patches with black outlines which stretch on the left from the rock across the vesicle are albite which is being replaced and corroded by prehnite. Just above the middle patch of albite are two lath-shaped crystals of purple augite. The remaining clear and turbid areas consist of prehnite which, when it is turbid, has replaced albite. To the left of the central patch of albite there is a clear area consisting of prehnite surrounded by a turbid zone consisting of rectangular phantom crystals of albite replaced by prehnite optically continuous with the clear central portion. Maol nan Damh, Ben More, Mull. 32 MR W. F. P. M'LINTOCK ON THE ZEOLITES AND ASSOCIATED MINERALS Puate IT. Fig. 1. Ophitic Basalt [114] with Vesicle, magnified 27 diameters.—The dark area at the foot is the rock showing the outline of the vesicle. The light-coloured area filling up the hollows at the junction is a mixture of purple augite, magnetite, and prehnite. This is succeeded by a zone consisting of turbid albite, at the left towards the top; irregular patches of epidote, just above the centre; augite, in idiomorphic, lath-shaped crystals ; and prehnite enclosing chlorite, which shows white in the photograph. A large augite crystal is seen at the top, on the left, projecting into albite ; others, associated with magnetite, are seen on the right towards the top. Maol nan Damh, Ben More, Mull. Fig. 2. Ophitic Basalt [109] with Vein, magnified 19 diameters.—The rock at the foot is succeeded by a layer consisting of large crystals of augite, one of which lies with its axis parallel to the wall of the vein, magnetite, albite, and chlorite. Above this is turbid albite which, at the top, towards the right, is succeeded by scolecite enclosing corroded scraps of epidote. Maol nan Damh, Ben More, Mull. Fig. 3. Amygdale [113] in Ophitic Basalt, magnified 27 diameters.—The clear white material is scolecite. In the centre is a large, dissected crystal of epidote consisting of four fragments. The first is the irregular patch with tufted margins in the centre; the second is the rhomboidal piece lying directly beneath ; the third lies immediately to the right of the second; and the fourth lies above the third. The four frag- ments have their cleavages parallel and extinguish simultaneously. The spaces between them are filled with scolecite enclosing rectangular turbid crystals of albite, in the centre and to the left connecting the third and fourth fragments of epidote. Near the foot on the left a turbid albite is in process of being replaced by scolecite. The dark areas at the left and right margins are epidote ; beneath the latter one is a corroded albite. Maol nan Damh, Ben More, Mull. Fig. 4. Ophitic Basalt [104] with Vesicles, magnified 27 diameters. —The small vesicle underneath is filled with analcite replaced by granular albite. The dark patches are chlorite moulded upon the analcite, and the long lath-shaped crystals which lie round the junction and also pierce the analcite are purple augite. A large one projects through the orifice connecting the small vesicle with the overlying main one. This vesicle is filled with a layer of augite, magnetite, and albite, seen on the right of the canal between the two vesicles ; chlorite, forming the dark patch on the left ; and albite, pseudomorphous after analcite, at the top. Maol nan Damh, Ben More, Mull. Fig. 5. Ophitic Basalt [78] with Vesicle, magnified 27 diameters.—The dark area to the left is the rock. At the junction with the vesicle is a thin layer of garnet with little rounded and dissected masses isolated from the layer. Near the top is a projecting knob of garnet in process of dissection. Overlying the garnet is a mixture of scolecite and heulandite enclosing laths of diopside and epidote. The dark bar consists of diopside, showing signs of corrosion, and a dull birefringent material similar to that which surrounds much of the corroded garnet. he dark crystal on the right is a corroded piece of epidote. Maol nan Damh, Ben More, Mull. Fig. 6. Metamorphosed Amygdale [17], magnified 27 diameters.—The amygdale is filled with scolecite showing incipient turbidity. Traversing the scolecite are crystals of epidote with yellow primary cores, enclosing chlorite and showing the original corroded outlines, and secondary, colourless rims. North-east slope of An Gearna, Ben More, Mull. Puate III. Fig. 1. Metamorphosed Amygdale [49], magnified 27 diameters.—The turbid patch on the extreme left and the two areas extending from the top and bottom respectively towards the centre are scolecite in process of replacement by prehnite, which forms the white areas in the photograph and, in the case of the upper and lower central masses of zeolite, forms little fibres traversing and replacing scolecite. The dark crystals to the left of the central zeolite are garnet, the one near the centre showing a core of secondary prehnite. The fibrous crystals, on the right towards the foot, are secondary epidote traversing secondary prehnite. Beinn Fhada, north end, east side, below the summit, Mull. Fig. 2. Metamorphosed Amygdale [58], magnified 27 diameters.—The turbid fibrous material, extending from the right towards the centre and forming the area at the bottom towards the left, is scolecite in process of replacement by prehnite, which forms the white areas. The mass of prehnite extending from the centre FROM THE TERTIARY LAVAS AROUND BEN MORE, MULL. 39 to the top encloses remnants of turbid fibres of scolecite, the various portions of which are optically con- tinuous. The dark inclusions at the centres of several of the prehnite areas are small garnets. The large crystal near the top towards the right is epidote ; other crystals are seen at the margin on the right, and at the margin towards the bottom on the left. North-east slope of An Gearna, Ben More, Mull. Fig. 3. Metamorphosed Amygdale [15], magnified 27 diameters.—The clear area to the right is primary prehnite enclosing diopside which is turbid and full of chloritic inclusions. The large crystals at the bottom and on the left are epidote, showing the core-and-rim structure. The turbid inclusions of diopside are con- fined to the colourless, secondary material of the rims, North-east slope of An Gearna, Ben More, Mull. Fig. 4. Metamorphosed Amygdale [50], magnified 27 diameters.—The clear areas are prehnite, mostly secondary after thomsonite, whilst the turbid areas at the top and towards the right are secondary albite enclosing fragments of secondary prehnite and thomsonite. The corroded scrap in the prehnite at the centre is a zeolite, either scolecite or thomsonite. The crystals towards the bottom showing cleavage are epidote. Beinn Fhada, north end, east side, Mull. Fig. 5. Metamorphosed Amygdale [41], magnified 27 diameters.—The turbid fibrous material is garnet secondary after a zeolite. The clear area at the top is also garnet enclosing and replacing prehnite. The colourless areas in the turbid material consist of garnet showing weak birefringence ; it probably replaces epidote. Beinn Fhada, north-east slope, near the granophyre, Mull. Fig. 6. Metamorphosed Amygdale [41], magnified 55 diameters, nicols crossed.—The large crystal cross- ing the field is epidote, secondary after prehnite some of which it occasionally encloses. The dark areas are garnet replacing prehnite and epidote. Beinn Fhada, north-east slope, near the granophyre, Mull. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 1). 5 rans, Roy. Soc. Edin‘ Vou. LI. W. F. P. M‘Linrock: On THE ZEoLITEs, Etc., FROM Tertiary Lavas In Muui.—P uate I. M‘Farlane & Erskine, Edin. ¥, PF, M‘Lintock, photo. Trans. Roy. Soc. Edin‘ Vous LL W. F. P. M‘Lintocx: On THe Zeouites, Erc., rrom Tertiary Lavas in Muti.—P arte II. W. F. P. M'‘Lintock, photo. M'Farlane & Erskine, Edin. Trans. Roy. Soc. Edin Vou. LI. W. F. P. M‘Lintock: On THE ZeEouitEs, Erc., From Tertiary Lavas in Mouui.—Puate III. W. F. P. M‘Lirtock, photo. M‘Farlane & Erskine, Edin. wr BR an ‘ ee r - ‘ i 7 La {ie , . ‘ $ ‘ i ‘ t vA . i. a ‘ ae ni \ r y , j u} +e wis tatah ( 35 ) I].—Spongiaires recueillis par la ‘‘Scotia” dans l’Antarctique (1903-1904). Supplément. Par Emile Topsent, Professeur 4 la Faculté des Sciences de Dijon. Présenté par le Dr. W. 8. Bruce. (MS. received June 1, 1915. Read June 28,1915. Issued separately September 22, 1915.) Depuis la publication de mon mémoire sur les Spongiaires de U Expédition antarctique nationale écossaise (9), j'ai regu de M. le Dr. W. 8. Bruce quelques Eponges antarctiques faisant aussi partie des collections de la Scotia mais dont le triage s était trouvé retardé. Indépendamment de fragments en double d’especes mentionées dans mon mémoire, ce second lot contenait plusieurs formes quil me parait* utile de signaler ou de deécrire. Dendrilla arctica, Topsent. Station 325, avril-aoit 1903; Scotia Bay, Orcades du Sud, 60° 43’ 42” lat. S., 44° 38’ 33” long. W.; 9-10 brasses. Un rameau. Cette Dendrocératide a été recueillie plus X Vouest, aux Shetland du Sud et au-dela, dans les campagnes du Francais (8, p. 11) et du Pourquor Pas? Elle parait étre répandue dans la région américaine de |’Antarctique, au voisinage des terres et par des profondeurs médiocres. Eumastia attenuata, n. sp. (Figs. 1 et 2.) Janvier 1903, Port Stanley, iles Falkland, oreve. Un seul spécimen, incomplet, sans support, long de 7 centimetres, large de 3 centimetres en son milieu. Mince sur l’un de ses cdtés qui, de contours doucement arrondis, représente son bord naturel intact, il est, sans compter les papilles, épais de 10 millimetres du cété opposé, qu’un instrument contondant a tranché nettement. L’outil a certainement laissé en place une partie plus ou moins étendue du corps, mais il a, pour ce qu il en a détaché, suivi de tres pres le support. Le spécimen est donc un morceau d’une Eponge en plaque, X bords libres, et de quelque épaisseur en son milieu. I] revét un aspect particulier parce quil souleve toute sa surface en ‘processus digitiformes. Hautes de 5 49 mm., épaisses de 1°2 mm., en moyenne, ces papilles sont sensiblement cylindriques, droites ou un peu tordues; généralement simples et indépendantes les unes des autres, elles deviennent souvent plus ou moins concrescentes par deux ou trois, tout en restant distinctes sur toute leur longueur ; quelques unes se divisent en deux branches. Leur nombre est tel que les intervalles qui les séparent restent étroits. Elles sont lisses, ainsi, d’ailleurs, que la surface générale 4 nu dans leurs intervalles. Aucune d’elles, de celles dont lintégrité est TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 2). 6 36 EMILE TOPSENT: SPONGIAIRES RECUEILLIS PAR LA “SCOTIA” certaine, ne parait percée d’un orifice 4 son extrémité; toutes, au contraire, se montrent en ce point opaques et plus fortement teintées que le reste. L’absence de papilles ouvertes au sommet au moment ot elle a été recueillie, n’empéche pas de considérer l’Kponge comme représentant une espece du genre Eumastia. Outre que — certaines des papilles peuvent avoir normalement joué le rdle d’oscules et se montrer contractées, il est évident que toutes ont fonctionné comme organes aquiferes. Sous Fic. 1.—Humastia attenuata, n. sp. x4. leur ectosome spiculeux, mince, transparent, rampe un systéme de canaux qui occupe la majeure partie de leur intérieur, le reste contenant un axe ramifié, irréeulier, de place en place relié aux parois par des brides ténues. A part cela, comme chez EHumastia sitiens, la structure est celle des Halichondria, tant dans le choanosome que dans l’ectosome. Le choanosome est assez dense, quoique fragile; l’ectosome se laisse détacher en lambeaux translucides. Le tout, SSS SS EEE a | ie a AGO MD! © 053 Jo 3 Pe eRe) teen 1 li aed jt Seni tom ane ils). 5 The maximum dorso-ventral diameter of each of the last four specimens is ‘2 mm. The peduncle shows a marked increase in size during this stage; in the first two specimens (a, b) it is about 80y long, but in the others it is 200-250 long, ON LARVZ OF LINGULA AND PELAGODISCUS (DISCINISCA). 51 and is bent upon itself near the middle of its length, so that the terminal part is dorsal and its tip directed to the left side. The other organs are similar to those of the preceding stage, except that the two ccelomoducts (nephridia) are now very clearly seen (fig. 10). Three of the four larger larve of this group have been cut into serial sections, and a few notes on these may be given here (see figs. 8, 9, 10). The shell was found to be in most parts about 20 thick,* increasing to 30m near the posterior margin, and to consist of a chitinoid substance, a thin outer layer of which forms a not very clearly differentiated periostracum. The statocysts (fig. 8) are closed sacs the external antero-posterior and transverse diameters of which are about 40-50, and the dorso-ventral diameter 10-15p. Several minute statoliths in motion ft were observed in each statocyst in the living specimens, but are not now visible. Hach statocyst is situated in the dorsal body- wall at the point where the latter is joined by the lateral body-wall, and the appearances suggest that the statocyst was formed by an invagination of the epithelium in the lateral angle of contact of the two portions of the body-wall, but the statocyst now exhibits no trace of connection with the exterior. The three ganglia of the central nervous system are situated in the body-wall ; the ventral or sub-cesophageal (fig. 10) is much larger than the two lateral ganglia, which lie immediately in front of the anterior occlusors. Muscle-fibres, almost longitudinal in direction, have appeared in the peduncle between the outer and inner cell-layers, but they are really formed in the latter. The median tentacle bears long cilia at its tip on the ventral side. The tentacle is hollow, and the lumen,} which may not extend far into the distal half, contains cells and muscle fibres, the latter serving to bend or contract the tentacle. Sur- rounding the lumen in a zone of supporting substance of homogeneous nature, and around this is a layer, apparently nervous, which is in contact with the base of the sensory external epithelium. Each cirrus is also hollow, and the lumen contains a bundle of longitudinal muscle- fibres which lie chiefly towards the medial side of the cirrus. The thick, external epithelium, which is ciliated on its median aspect, has very deeply staining nuclei. In the base of each cirrus there is homogeneous supporting substance, usually thinner or absent on the median aspect, which is continued into the two arms of the lophophore, where it forms, just proximal to the insertions of the cirri, a continuous curved band (fig. 9) supporting this basal portion of the arm-apparatus. The right and left bands are linked together by a small mass of the same kind of supporting substance situated in the mid-ventral wall of the lophophore near its * The shells of these larvee with 13 pairs of cirri are about twice as thick as those figured by Dr Yarsu from his larvee with 15 pairs of cirri. It may be noted that his larve, from the stage with 7—9 pairs of cirri onwards, were reared in captivity. + Ciliated cells were not observed in the wall of the statocyst. { The lumen opens posteriorly into the arm-sinus, which is connected with the general body-cavity. 52 DR J. H. ASHWORTH posterior margin. The skeletal substance of the lophophore is therefore approxi- mately U-shaped. This substance is, in these specimens, almost homogeneous and contains neither cells nor nuclei; it has evidently been secreted by one or both of the cell-layers in contact with it. The mouth-cavity and cesophagus are lined with ciliated columnar cells, and the anterior and posterior walls of the stomach with high and narrow flagellated cells, those of the posterior portion of the stomach being especially striking. The intestine is apparently not ciliated. The digestive glands or “liver”-lobes are thick- walled diverticula of the mid-gut. There are two anterior dorsal diverticula—the smallest and last formed “liver’-lobes—situated in the dorsal and anterior part of the body-cavity ; they are bounded laterally by the occlusors, and extend backwards as far as the gastro-parietal bands and forwards to the level of the anterior edge of the occlusors (fig. 8). These lobes unite ventrally and open into the anterior dorsal wall of the stomach by a very short median duct; dorsally the lobes are separated by the dorsal mesentery. The large posterior dorsal lobes of the digestive gland— one right and one left—open separately into the dorso-lateral portion of the stomach about the middle of its length (fig. 9). Hach of these lobes presents an indication of subdivision into anterior and posterior portions. The ventral lobe of the “liver” is a large, A-shaped diverticulum (fig. 10)—that is, it is subdivided into right and left portions, which open by a common aperture at their anterior ends into the stomach near the middle of its ventral wall. The cells composing the “liver” are highly vacuolated—the vacuoles contained in life digestive secretion and oil globules, —cell-outlines are not distinguishable, and the nuclei are comparatively small. When the contents of the vacuoles have been removed, the walls of the “liver”- lobes have a spongy appearance, and embedded in them, here and there, are the unicellular algee * on which the larva principally feeds. The two ccelomoducts (nephridia) could be clearly seen in the living larve, and are well.seen in some of the whole mounts and in the sections (fig. 10). Their funnels, the apertures of which are directed postero-medially, lie in the ccelom near the posterior end of the mid-gut, and are on the ileo-parietal bands. Hach ccelomo- duct, the lumen of which is narrow, runs forwards in a curved course, and opens into the mantle-chamber on the antero-ventral wall of the body proper, immediately lateral to the anterior occlusor muscle—that is, ventral and a little postero-lateral to the mouth. (v) Larvve with 14 or 15 pairs of cirri.—Four large specimens were found in the Red Sea, two of them on June 21 at locality (i), the others on October 22 at locality (1). (The measurements of their shell-valves are— Length. Breadth. a, October 22, 1914 ; . 1°47 mm. 1:22 mm. b, June 21,1914 . : ay AVA as 2s c, October 22, 1914 : » Alb2e5: 122815 d, June 21,1914 . . Pr tel clee As US Pe te * A Radiolarian was also noticed in the gut. ON LARVA OF LINGULA AND PELAGODISCUS (DISCINISCA). 53 Growth in length of the shell-valves is evidently taking place at this stage much more rapidly than growth in width, as the later growth-lines show (fig. 4). The shell has now become elliptical, the longer axis being antero-posterior. In specimens a and b there are 14 pairs of cirri; in ¢ and d the number of pairs is apparently 15, but as the cirri are closely packed together it is difficult to be certain of their number. Additional cheetz have been developed in the postero-lateral region of the mantle, where there is now a group of about six or eight long chzetze on each side, among which the two original ones may generally be distinguished by their slightly greater length and thickness (fig. 6). Other shorter chetze have been added at intervals round the lateral and anterior margins of the mantle. The posterior occlusor is now a round or oval cylinder of muscle about 80-100. in diameter. The peduncle is from ‘4 to ‘8 mm. in length,* and is bent into one or two loops, its terminal part lying dorsal to the proximal part, and its tip pointing to the left (fig. 6). The terminal portion of the peduncle is dilated, and the epithelium covering this region is much higher than elsewhere, but it has not yet proceeded to form the secretion which later envelops it. The peduncle, which is colourless, is in each specimen enclosed within the shell-valves, no part of it being extended. The median sensory tentacle does not exhibit the least sign of reduction; it is still large, about 160y long, and is evidently fully functional. These specimens are the largest and latest free-swimming stages recorded for any species of Lingula (s. lat.). Probably these larvee would soon have settled down, the peduncle of each would have been extended beyond the valves, and the high epithelium of its terminal portion would have produced secretion by means of which the larva would have fixed itself to the substratum. A note of the colours exhibited by the living larvae may be given here. The shell-valves, even of the largest larvee, are transparent, there being no calcareous matter present. As already mentioned, the posterior and neighbouring lateral margins of the valves are yellowish brown, and in these regions the zone just within the margin is a bright green colour. The mantle-margin is usually brownish, especially posteriorly. The basal half of the tentacle, particularly on the dorsal side, and the distal portions of the*cirri, are yellowish brown, and there is a patch or spot of deeper tone at the tip of most of the older cirri, and a similar spot about the middle of the dorsal surface of the tentacle. There is yellow pigment on each side of the mouth at its postero-lateral margins. The “liver”-lobes are lemon-yellow, and near their periphery a small amount of brown pigment is present. * The largest specimen seems to be abnormal in this respect, for the peduncle is comparatively short—about ‘2 mm. in length,—but it is not possible to state its length exactly, owing to its being much foreshortened, as seen in the preserved specimen. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 3). f 9 54 DR J. H. ASHWORTH CoMPARISON OF THE LARVA WITH THOSE PREVIOUSLY RECORDED. The larvee described in the foregoing account fall into a single series, and evidently all belong to one species. On comparing them with those described by Dr Yarsu, there can, I think, be no doubt they belong to the same species—Lingula anatina Bruguitre,—which is, so far as I am aware, the only species recorded from the Indian Ocean. In structure and colours my larvee agree closely with Dr Yartsv’s, and special points of agreement are: the hinge-line is practically the same length in both series of larvee, and the cheetee evidently arise in the same manner, for it is clear from Dr Yarsv’s figure (e.g. fig. 86) that in his larvee, as in mine, the two cheetze first formed in the postero-lateral region remain for a considerable time distinguishable from those formed later. Captain SpwExL concluded that his larvee also were those of L. anatina. There are certain differences between my larve and those of Dr Yarsu and Captain SzwELL which may be now considered. (a) Size-—The following table summarises the average dimensions of the shell- valves of the larvee of different stages :-—— No. of Pairs of YATSU. SEWELL. ASHWORTH. Cirri present. | Length. Breadth. Length. Breadth. Length. Breadth. 5 ‘31 x °39 mm. 6 "AA 4 7-8 66 x ‘61 ,, fe: 52 x °62 mm. (December Series.) 9 aie ‘65 x 66 mm. 10 x SOO Pale lee “76, % . "Ol = 11 . er 188) xe KD) gg, "Si Xa OA 12 ee WOO 3 IO), 3 25h sh Ome olll yee (February.) 14 fae Det So el lems UAB MODs 5 55 ; F Tx ol oe. 15 8 x 64mm. ee me iene Dr Yatsv’s larvee with 7 or 8 pairs of cirri were rather larger than mine, but his later stages were much smaller—compare them when they have 15 pairs of cirri. This disparity is no doubt due to the fact that Dr Yarsv’s larvee, from those with 8 or 9 pairs of cirri onwards, were kept in aquaria, and the growth of the shell was retarded, for it may be noted that the shell-valves of his specimens were not only smaller but much thinner (see p. 51 and footnote). Captain SrwELt’s larva with 14 pairs of cirri is considerably smaller than mine, but as he suggests that the conditions at the time of the year when it was taken were unfavourable, the difference in size may be due to this cause. (b) Change in the Shape of the Shell.—The table serves to show that the change in the shape of the shell-valves—that is, when they become for the first time longer ON LARVZ OF LINGULA AND PELAGODISCUS (DISCINISCA). 5d than broad—took place at different stages in each of the series of larvee under con- sideration. In Dr Yarsu’s this change occurred when 7 or 8 pairs of cirri were present, in Captain SEWELL’s specimens collected in December the change took place when 9 pairs of cirri were fully formed, and in my larve when 12 pairs of cirri were present. Captain SEWELL’s single specimen taken in February had a shell as long as broad ; he attributes the delay in change of the shape of the shell* to the less favourable time of the year when development was taking place. (c) Chxtx.—Cheetee appeared in Dr Yarsv’s larve at an earlier stage than in mine, but in both series the time of appearance of the cheetze coincided with the change in shape of the shell-valves. (d) Peduncle.—In Dr Yarsv’s larvee the peduncle appeared when 6 pairs of cirri were present, and was protruded by the time 10 pairs of cirri had been formed, at which stage his larvee, kept in aquaria, became fixed. Perhaps the artificial von- ditions brought about precocious fixation. In Captain Sewe.t’s larvee the peduncle began to develop in the stage presenting 9 pairs of cirri, but there was no sign of its protrusion in his specimens with 11 pairs of cirri. In my larvee the first trace of peduncle did not appear until 11 pairs of cirri were present, and the peduncle was entirely internal in the largest larvee, which had 15 pairs of cirri and were still free-swimming. (e) Tentacle.—In my largest larvee, with 14 or 15 pairs of cirri, the median sensory tentacle was not in the least degree reduced, but was fully functional, whereas in Dr Yatsu’s examples with 15 pairs of cirri—the specimens being then attached to the bottom of the aquaria—the tentacle was either absent or reduced to a small papilla) Dr Yarsu regards the tentacle as a larval organ, and therefore considers its disappearance in his fixed specimens as a natural consequence. There is, however, the possibility that the conditions of captivity had determined the reduction and eventual loss of this sense-organ in his larve. A similar tentacle is present in Gilottidia ; and in Brooxs’s figure (fig. 7) of a young specimen, with 16 pairs of cirri, soon after it had become sedentary, the tentacle is shown as long as the neighbouring cirri, and Brooks states definitely (p. 73) that it is persistent. Most of the differences between the various series of larvee discussed above, particularly in reference to the size of the shell-valves, the size and protrusion of the peduncle, and perhaps the persistence or loss of the tentacle, are doubtless dependent almost entirely on the environmental conditions. Variations in one or more of these characters may be expected in larvee of the same species obtained in widely separated areas and under diverse conditions. There are, however, certain features which appear to be more constant for the species, at any rate they present a fairly close agreement in Dr Yarsu’s larvee and mine t-—namely, the length of the hinge-line and * The peduncle of this specimen was also in a backward state, being “ still only a small rudiment.” + In Captain Sewetv’s larvee there was apparently a rather greater range of variation in the length of the hinge, but its average length in his twelve larvee was 3 mm., practically the same as in Dr Yarsv’s and mine. 56 DR J. H. ASHWORTH the antero-posterior diameters of the valves of the protegulum,* and probably also the form of the chetz. These features may therefore be found useful in the deter- mination of the species of larvee, but a definite decision as to their taxonomic value can only be reached after they have been examined in further series of specimens. Judging from the large size attained by the oldest examples, the larvee found in the southern portion of the Red Sea developed under very suitable conditions. It will be observed that these larvee form a regular series, there being no aberrant members, and they probably afford a fairly reliable criterion of the normal course of development in a favourable environment. BREEDING SEASONS. Dr Yarsu states that the breeding season at Misaki is very short, and is certainly restricted to a month and a half of the summer—from the middle of July to the end of August. He estimates that the period which elapses from the onset of develop- ment to the stage with 15 pairs of cirri is about six weeks. Assuming that my » larvee had developed at about the same rate, the oldest specimen, with 15 pairs of cirri, taken on June 21, 1914, in the southern portion of the Red Sea, would be one of the products of a spawning which had taken place in the first half of May. Spawning evidently continued until, or was resumed about, the end of May, for a much younger ijarva, only half as long as the preceding, was taken at the same time and place. The finding of two large larve, with 14 and 15 pairs of cirri respec- tively, in the southern portion of the Red Sea on October 22, is evidence that a spawning occurred in the first half of September. The specimen in Dr ANNANDALE'S material was taken on March 25, 1901, in the Strait of Bab-el-Mandeb, and probably had developed from an egg fertilised about a month previously. The evidence available in regard to the southern end of the Red Sea indicates that there is a succession of spawnings extending at least over the period from the heginmine of March to the early part of September. Captain SEWELL records the occurrence of larvee during the winter months, December and February, in the plankton off the south coast of Burma. I found one young larva, which was probably from two to three weeks old, in the Indian Ocean, about 3° N. lat., and 80° KH. long., on October 14, 1914. CoMPARISON OF THE LaRvV& OF LINGULA ANATINA WITH Brooks’s Larva OF GLOTTIDIA AUDEBARTI. As Brooks's memoir (1879) on the development of Glottidia is not easily accessible, a few notes on his specimens are given here, and their characters compared with those of the available examples of Lingula anatina. * The primary shell is laid down while the larva is enveloped by the egg-membrane, and hence the protegula and the hinge-line are determined under relatively uniform conditions ; later the larva is subject to more diverse conditions, which cannot but react upon such characters as the secondary shell and the peduncle. ON LARVZ OF LINGULA AND PELAGODISCUS (DISCINISCA). 57 The three free-swimming larve figured by Brooks have shell-valves with the following dimensions :— Shell-Valves. No. of Pairs of Cirri. Length. Breadth. 5-6 . ; A . 23 mm. ‘27 mm. Tea. : ; ; ae oes 290) om: 9-10. ‘ , : Si eos 5, "ale ys These larvee, and especially the two last, are much smaller than those of Lingula anatina with the same number of cirri. The distance between the teeth at the ends of the hinge-line on the ventral valve in Brooks's larve is ‘15-16 mm., as compared with about °3 mm. in LZ. anatina; and the antero-posterior length of the protegulum of the dorsal valve is about ‘065 mm. in Brooxs’s larvee (figs. 1 and 3), and ‘12 mm. in L. anatina. Cheetze are formed in the larve of Glottedia when about 9 pairs of cirri are present, and are shown (Brooks’s fig. 3) all of the same size and sparsely scattered round the margin; there is apparently not a larger first-formed pair on each side as in L. anatina (see p. 50). It will be noticed that the cheetee appear about the same time as the change in the shape of the shell-valves takes place, as was also the case in DL. anatina. The posterior occlusor and the peduncle are formed when 7 pairs of cirri are present, and by the time there are 10 pairs of cirri the peduncle has become long and looped, and Brooxs expressed the opinion that the larvee became sedentary soon after this stage, but the change was not actually observed, as it was found that “the larvae could not be made to thrive in confinement.” The youngest sedentary example of Glottidia found has 16 pairs of cirri. The shell-valves of this specimen are about 2 mm. long and 1 mm. broad, and the peduncle—now fully protruded—about 5-6 mm. long. In the dorsal and in the ventral mantle-lobe there are four pigment-spots near the anterior margin; there are no pigment-spots in my larve of Lingula 1:52 and 1°6 mm. long respectively, nor do they appear to have been present in Dr Yatsu’s specimens. The most noticeable differential characters of Brooxs’ specimens of Glottidia are: in the free-swimming stages, the small size of the shell-valves, the short hinge- line and consequently the small protegulum’; in the sedentary specimen, the narrow shell-valves * and the presence of pigment-spots near the anterior mantle-margin. Brooxs’s larvee of Glottidia were obtained in Chesapeake Bay, and as young stages were found in the middle of July, while only older larvze were taken in the middle of August, the breeding season was probably short, as was also found by Dr Yatsu for Lingula at Misaki, which is in nearly the same latitude as Chesapeake Bay. * In this connection it may be remarked that full-grown examples of Glottidia audebarti do not attain so large a size as those of Lingula anatina, and the shell-valves of the former are narrower in proportion to their length. 58 DR J. H. ASHWORTH THE LARVAD OF PELAGODISCUS (DISCINISCA). Previous RECORDS OF THE LARV# OF DISOINID&. (a) Larve of Pelagodiscus (Discinisca) atlanticus (King).—Fritz MUR (1860, 1861) was the first to describe a larva of an ecardinate Brachiopod. He recognised that his larvee were those of some Brachiopod, but in his published accounts did not refer them to any genus, though he evidently not long afterwards * reached the conclusion that they belonged to the genus Discina.t He probably saw about a dozen living larvee, which he collected in the years 1859 and 1860 near Desterro on the southern portion of the coast-line of Brazil. The only other observer who has hitherto seen a living larva is Dr Yatsu (1902, p- 105) who found, in the plankton near Misaki (Japan), a specimen which is probably referable to the species P. atlanticus. Professor BLocHMANN (1898) studied ten preserved examples taken in the plankton off the island of Bintang, about 45 miles south-east of Singapore. (b) Larve of other Discinide.—MULiER obtained another larval form of a species of Discina, which he stated, in a letter to Professor Morsz,* belonged to the species D. radiata Dunker, but I am unable to find any further reference to this larva, no description of which seems to have been published. Professor SimrotH (1897, pp. 8-6) has described two larve obtained by the Plankton Expedition. The smaller one, from Palmas roadstead, has a transverse diameter of '22mm., and only 3 pairs of cirri It is further remarkable for possessing on each side four very long chete. The larger larva, taken in lat. 5° 9’ N., and long. 20° 3’ W., at a depth of 1000-1200 metres, has a transverse diameter of 42 mm. This larva has 4 pairs of cirri, which are much more slender and elongate than those of P. atlanticus, and the median tentacle is also ‘elongate and terminates in a bulbous dilatation. There are only two chetz on each side, which appear to be equal in length and simple. Both these larve are referred by Professor StmrorH to the genus Discina. KrcHLER (1911) has given a short account of two larve obtained by the German South Polar Expedition near the winter station of the Gauss (lat. 66° S., long. 90° E.), from a depth of 3000 métres: The specimens were ‘787 and ‘825 mm. respectively in transverse diameter, and had 4 pairs of slender cirri of unusual length. On each side there were four stronger cheetee, all of similar form and * Professor MorsE (1673, pp. 356, 357) states that he received “a letter from Herr MULuER, accompanied with a sketch of another larval form of Discina, in which he describes features similar to those above mentioned [referring to his abstract of the account of MULLER’s larvee from Desterro], and states that the species has been defined by Professor DuNKER as 1). radiata.” By the courtesy of Mr E, A. Surrg I have been able to see DunKER’s description of D. radiata (Malak. Blitt., vol. viii, p. 39, 1861) : there is no reference to the larva. + The genus Discina was subdivided by Dau in 1871 into Discina (sensu stricto) and Discinisca, and the latter was again divided by’ Datu in 1908 into Discinisea (s. str.) and Pelagodiscus. Discina atlantica King falls into the last-named section, and its correct designation is therefore Pelagodiscus atlanticus (King). ON LARVA OF LINGULA AND PELAGODISCUS (DISCINISCA). 59 apparently simple, 7.c. not spinulose, and a series of finer cheetze on the lateral margin of the mantle. ErcaHLer and Biocumann consider that these larvee probably belong to Pelagodiscus (Discinisca) atlanticus; but there is no justification for referring them to this species, from which they differ in the following characters: their much greater size, the form of the median tentacle, their cirri (which are about three to four times as long as those of P. atlunticus), and in their possession of only four principal chzetz, the longest of which is not specially stout. These larve differ so strikingly from those already referred on good grounds to P. atlanticus that they cannot be regarded as belonging to the same species; they may be larvee of some larger species of Dascinisca. THe Larva or PrLacopiscus (DIScINISCA) ATLANTICUS. Six free-swimming larvee were taken on October 16, 1914, in the Indian Ocean a few miles west of Cape Comorin, where the chart shows depths of about 40 fathoms. Adult specimens of P. atlanticus have been found almost entirely in deep water ; there are records from 200 and 690 fathoms, but with these exceptions specimens have been found only at depths greater than 1000 fathoms, and there are four records from more than 2000 fathoms, the deepest being 2425 fathoms. The occurrence of the larvze of Pelagodiscus off Cape Comorin in shallow water was therefore rather unexpected. The fact that half a dozen larve were obtained together indicates the probable close proximity of the parent forms. There is, however, the possibility that the adults lived in the deep water to the west, for the depth increases rapidly in that direction, and there was deep water—700-800 *fathoms—only about thirty miles away. It is known that the larve may remain free-swimming for five or six days,* a period which would have been sufficient for their transportation by the strong currents from the area of deep water to the locality where they were found. Although the conditions suggest that the adults of P. atlanticus occur in shallow water near Cape Comorin, a definite conclusion cannot be reached on the evidence available. I have recently looked over chartst of the other areas from which larve of Pelagodiscus have been recorded. Around Desterro, where Mixuer found his larvee, the sea is shallow, the nearest water of 100 fathoms depth being some sixty miles eastwards. Off Bintang, where Professor BLocHMANN’s ten larvee were taken, the water is very shallow (0-24 fathoms), the nearest water of 100 fathoms depth being about four hundred miles away. At Misaki, where Dr Yatsu took a single larva, there is deep water close inshore. It is evident from a consideration * MU wer (1861, p. 54) observes, regarding his larve in aquaria: ‘“‘ Die Dauer dieses Schwarmstadiums tiberstieg bei den eingefangenen Larven nie 5-6 Tage, meist schon friiher setzten sie sich fest, am Boden oder an den Seiten des Glases.” ScnucHeErT (Bull. Geol. Soc. Amer., vol. xxii, 1911, p. 272), however, states that the larvee “are known to live in the free and floating condition for nearly a month,” but there is no evidence in support of this statement, which is erroneous. + I am indebted to Dr W.S. Bruce for giving me access to the charts in the Scottish Oceanographical Laboratory, 60 DR J. H. ASHWORTH of the conditions in the areas where the larvee recorded by MUtLier and BLocuMann were taken that the adults of P. atlanticus ave not confined to deep water, and the occurrence of the larvee in shallow water near Cape Comorin, while not conclusive, lends support to this view. Description of the Larve (Plate V). The six larvee agree closely in size and structure, and are evidently all about the same stage of development. The shell-valves (fig. 11), which are unequal, are free from each other all round their margins, there being no hinge; they are connected only by the muscles— especially the occlusors, and the body-wall. The dorsal valve, which is almost like a watch-glass in form, is sub-cireular and varies in the different specimens from : ‘39-43 mm. in length, and -48-"47 mm. in breadth. The almost flat ventral valve, which has a peduncular sinus in the posterior middle line, is 32-34 mm. long, and 39-42 mm. wide. The maximum dorso-ventral diameter of each of the three larvee in which it was measured is ‘(075 mm. ‘These larvee are about the same size as those described by Professor BrocuMann and that by Dr Yarsu, and slightly larger than MULLER’s original examples. Both valves are chitinoid and transparent, and exhibit no trace of calcareous substance. The edge of the dorsal valve, and to a less extent that of the ventral valve, are yellowish brown in colour. The shell- valves vary in thickness in different portions from about 3 to 8. The chitinoid shell, which cuts easily in paraffin, is covered on its external surface by a definite, though very thin periostracum, which is continued over the edges on to the inner face of each valve, where, overlying the thin margin of the mantle, it extends inwards until it reaches the thickened zone of the mantle by which it is evidently formed. In surface view the periostracum presents a shagreened appearance. The mantle is for the most part colourless, but in its postero-lateral margins exhibits some yellowish-brown pigment, and along the margin of the peduncular notch bears darker brown pigment-granules. The mantle is very thin over the greater part of its extent, but presents a thickened zone in its lateral and anterior portions in both valves, but especially in the dorsal valve (fig. 17). The posterior mantle-margin is thin in the dorsal valve, and in the ventral one forms a narrow strip posterior to the peduncle. In the thin portions of the dorsal and ventral mantle-folds, the outer surface of which is, of course, applied to the shell, there are numerous gland-cells containing large granules; these cells no doubt secrete the shell. In the thickened zone of the mantle there are, besides gland-cells, more numerous epithelial cells and some muscle-fibres; the chzetze are implanted in this region, which, as already stated, secretes the periostracum. There are five pairs of principal chzetze, as described by Mituer (1860). Addi- tional details regarding them may be given here. The two anterior cheetee of each side are slender and flexible, and are about ‘15-16 mm. long. Each has a simple ON LARVA OF LINGULA AND PELAGODISCUS (DISCINISCA). 61 joint situated a short distance proximal to the middle of its length, and the flattened and tapering distal portion exhibits faint indications of several joints, and bears along one margin minute spinules, which, however, do not extend to the trp of the cheeta (fig. 13). When at rest these chzetee usually point antero-medially, but in preserved specimens they may be found directed antero-laterally owing to the contraction of the muscles attached to their bases. The third cheta is usually gently curved, about ‘09 mm. long, and is directed postero-laterally. It bears on its anterior edge minute, regularly arranged, pointed processes, which begin a little proximal to the middle of the length of the cheeta, but do not extend to its tip (fig. 14). The fourth cheta is the largest, being ‘3-33 mm. long,-and much thicker than the others. It issues between the shell-valves postero-laterally and then usually curves towards the middle line, as shown on the right side of fig. 11, but it may be turned so as to point outwards, as on the left of fig. 11. Its proximal portion—about one-fifth—which is almost uniform in diameter, merges into a dilated region where the cheta issues between the two shell-valves; distal to this the cheeta tapers gradually to its tip. The greater part of the tapering portion is beset, except on its medial side, with numerous pointed teeth or thorns, the tips of which are directed postero-laterally (fig. 15). The thorns do not extend quite to the tip of the cheeta ; a terminal portion about 154 long is without them and is very fragile. The four pairs of chzetze described above arise from the ventral mantle-fold. The fifth cheeta on each side, which arises from the dorsal mantle-fold, is about ‘18-2 mm. long. It is slender and strongly curved so that its tip points postero- medially, and it bears, on its distal half, minute spinules similar to, but smaller than, those of the fourth cheta. Muscles are attached to the bases of each of these five cheetze. There is also on each side a series of about thirty cheete, all of similar form, regularly arranged in close order in the thickened zone of the dorsal mantle-fold, and extending from the region of the fourth principal cheeta forwards almost to the middle line. These cheetz are almost uniform in diameter along the greater part of their length, but taper rapidly at their tips. They exhibit nodes at intervals, and are flexible ; they are usually directed postero-laterally, and their free ends are bent under the edge of the ventral valve. Only the two most anterior and the most posterior of these cheetze (Ca. 8.) are shown on the right of fig. 11. On each side are two or three thicker chete situated in the ventral mantle- fold, about midway between the second and third principal cheetee; and directed almost laterally (fig. 11, right side). Each of these bears longitudinal striations and is subdivided by nodes so that it resembles a bamboo (fig. 16). The peduncle, as seen in the living or stained specimen (fig. 11), is an oval mass, the transverse and antero-posterior diameters of which are, on an average, about ‘13 and ‘04 mm. respectively. In vertical section (fig. 17) the peduncle appears as an elevation, triangular in section, of the inner wall of the ventral mantle-fold close to TRANS. ROY. SOC. EDIN., VOL. LI, PART I ‘NO. 3). 10 62 - DR J. H. ASHWORTH its posterior margin, so that in its origin it is similar to that of Lingula (see p. 49). The broad base of the elevation is ventral, the posterior or dorsal side of the ridge lies parallel and close to the dorsal mantle-fold,* and the anterior side is almost parallel to the posterior wall of the body proper. The epithelium covering the peduncle is continuous with, but much thicker than, the inner epithelium of the ventral mantle, which unites with the epithelium of the body-wall proper just anterior to the front margin of the peduncle. The central portion of the peduncle consists of strong curved muscle-fibres, which have a general ventro-dorsal direction. A narrow transverse cavity 1s present in the posterior portion of the peduncle, but this does not appear to have a defin'te epithelial lining, nor is there any visible connection in the preserved specimens between this cavity and the body- cavity, though there is probably a connection in life, as in the case of Langula (p. 50 and fie. 6). The narrow strip of mantle immediately posterior to the peduncle is strongly pigmented, and the pigment extends forwards right and left into the epithelium which covers the sides of the peduncle. The dorsal and anterior faces of the peduncle are almost completely covered with a thin film of periostracum. The peduncle is in condition to be extended ready for fixation; the small lip at the posterior margin of the ventral valve, which consists only of periostracum, would be bent ventrally, and the neighbouring region of the ventral valve would yield to some extent so as to permit the exit of the peduncle between the shell-valves. During extension the peduncle would be so bent as to bring its flat, previously dorsal, surface into contact with the object to which fixation was about to take place, and the granular cells forming the thick epithelium of that face of the peduncle would secrete the necessary cementing substance. The median tentacle is an elevation on the epistome about 30-40 in height and 50p in width. Its apex is traversed by a shallow median groove, the cell of which are pigmented. The apical portion of the tentacle is solid, but the basal half contains a narrow axial lumen, which opens proximally into the arm-sinus. The lumen is lined by cells and contains longitudinal muscle-fibrils by means of which the tentacle can be contracted or bent laterally or ventrally towards the mouth. Supporting tissue, like that found in the tentacle of Lingula (see p. 51), is not present, but there is a nervous layer between the lumen and the epidermis. The tentacle is ciliated along two tracts (indicated by the darker areas in fig. 11), one on each side, near its base. The discoidal lophophore bears the mouth near the middle of its ventral surface, and on each of its lateral margins there are, in all the specimens except one, four approximately equal cirri arranged as shown in fig 11. In one of the larvae (not the smallest) there are four cirri, all of normal size, on the right side, but on the left there are only three, the fourth or most anterior cirrus being absent. The cirri * The peduncle is not in any way connected with the dorsal valve ; its posterior surface may touch the dorsal mantle-fold, but remains quite independent of it. ON LARV OF LINGULA AND PELAGODISCUS (DISCINISCA). 63 are hollow, and the comparatively large cavity, which opens into the arm-sinus, is lined by an epithelium and contains longitudinal muscle-bands. The cells on the medial face of each cirrus bear long cilia. As in Lingula, the deeply staining epithelium of the cirri is thrown into annular folds by contraction of the muscles. Supporting substance is not recognisable in sections of the cirri, but is present in each half of the lophophore as a narrow band lateral to the arm-sinus, and extending along the region of attachment of the cirri. The greater part of each cirrus, the distal region of the tentacle and the ciliated tracts, right and left, in its basal portion were yellow in life. There was also a considerable amount of yellow pigment in the lip which overhangs the mouth in front, and a narrow transverse band of brownish pigment in the ventral body-wall immediately anterior to the sub-cesophageal ganglion. There is a deep depression on the ventral side between the posterior margin of the lophophore and the anterior edge of the wall of the body proper (fig. 17). In this depression, and situated on the anterior surface of the wall of the body proper, just ventral to the cesophaeus, is the large ventral nerve-ganglion. The two lateral ganglia, which are also large, are situated in the body-wall in front of the anterior margins of the occlusors.* The outline of the wall of the body proper is almost semicircular or bluntly conical, its form and height depending on the condition of expansion or contraction of the animal. In the lateral wall of the body proper, rather nearer the dorsal than the ventral surface, is on each side a large black, or brownish-black, oval pigment-spot or “eye,” about 20u long. Similar pigment-spots were present in the larvee examined by Frirz MULLER, but not in those described by Drs BLocHMANN and Yatsv. The epidermis of the body-wall in the region of the “eye,” especially ventral and posterior to it, is higher than elsewhere, and forms a regular columnar epithelium. This elevation, which is readily seen in most specimens mounted whole, probably represents a sensory area. The “eye” is like a shallow cup in shape (fig. 18), and consists of masses of brown pigment-spherules deposited in the distal ends of several of the epithelial cells. Immediately below the epithelium of this region of the body-wall is a nervous tract extending backwards from the lateral ganglion, which supplies the columnar epithelium described above, the “eye,” and the statocyst. In the dorsal body-wall on each side is a large statocyst situated immediately posterior and slightly lateral to the dorsal margin of the massive anterior occlusor muscle (fig. 12). The maximum internal diameter of the statocyst is about 25p ; the external diameters are: autero-posterior, 30-404; transverse, 25-304; dorso- ventral, 10-15u. In surface view the statocyst is ovoid, and the narrower end, * That is, the anterior occlusors ; there are no posterior occlusors at this stage. Professor BLocHMANN has given an account of the muscles of the larva, to which I have nothing special to add. 64 DR J. H. ASHWORTH which is directed anteriorly, lies near the lateralis muscle. Several minute statoliths were present in life, forming a small mass in the centre of the statocyst. The wall of the statocyst consists of two layers: an epithelium with large nuclei similar to the epithelium of the neighbouring region of the mantle, and a thin enveloping layer of coelomic epithelium (fig. 18). The statocyst is in contact with the body-wall at the angle of union of the latter with the dorsal mantle, and in this angle is a distinct - depression in the external epithelium suggesting that the statocyst had been formed by invagination at that point. Professor BLocumann holds that the organs inter- preted as statocysts are really the funnels of nephridia,* and describes and figures an ‘ Ausfiihrungsgang.” There can, however, be no doubt that the organs in question are statocysts,.and they are closed sacs. Dr Yarsu (1905, p. 568) states that the statocysts ‘must become smaller at the time of attachment” of the larva, but his fig. 2 from a young attached Discinisca levis shows a statocyst considerably larger than that of a free-swimming Pelagodiscus atlanticus.t The mouth is situated immediately behind the epistome by which it is overhung ; its cavity is lined by ciliated cells. The cesophagus, which has a high columnar ciliated epithelium, passes at first dorsally and, after a short course, turns through aright angle to run posteriorly. It opens, at the level of the body-wall, into the mid-eut by a somewhat constricted aperture. The mid-gut is widest posteriorly ; it represents the stomach and the digestive gland or “liver,” the lobes of the latter not having yet been formed. The cells of the wall of the mid-gut are, however, sharply differentiated into two kinds: high epithelial flagellated cells which will form the wall of the stomach, and vacuolated cells which will be included in the future lobes of the “liver” (fig. 17). The high epithelium characteristic of the stomach is present in (1) the anterior wall of the mid-gut, but only for a short distance around the point of entrance of the cesophagus; (2) in the middle and posterior portions of the ventral wall; and (3) in the ventral half of the posterior wall of the mid-gut. The remainder of the mid-gut, comprising the entire dorsal and lateral walls, the dorsal half of the posterior wall, and the anterior portion of the ventral wall, will form the “liver”-lobes.. The epithelium of the stomach, especially that of the posterior ventral region, is remarkable for the height and slenderness of its cells, which are 11-15 long, but not more than 1 wide at their distal ends. The elongate, deeply staining nuclei are situated in the proximal portions of the cells. Each cell bears only a single flagellum about 8-12y long, which arises from a well-marked basal granule in the distal region of the cell (fig. 19). Many of the “liver”-cells are highly vacuolated ; the remaining cells are filled with spherules or less regular masses of secretion. The nuclei are at the bases of the cells, and cell-outlines cannot be distinguished. The “ liver ”-tissue is similar in structure to * See also footnote, p. 48. + In Lingula the statocysts do not diminish in size when the animal becomes sedentary, as may be seen from the figure of a statocyst of a sedentary example with shell 65 mm. long given by Dr Yarsu (1902a, fig. 22). This statocyst is about three times as large as those of the larve described on p. 51. ON LARVZ OF LINGULA AND PELAGODISCUS (DISCINISCA). 65 that of Zingula, but the contents of the cells, as seen in life, are not so refringent. The intestine, which is not ciliated, issues from the ventral wall of the mid-gut near the middle line and a little anterior to its hinder margin (fig. 11). It runs towards the right and is at first narrow, but becomes wider as its enters the postero-lateral corner of the body-cavity, where it is dilated into a sub-spherical sac from which the narrow terminal portion runs anteriorly and slightly laterally to open on the body- wall ventral to the pigment-spot and: near the postero-lateral margin of the right anterior occlusor. The mid-gut is pale yellow in colour, and scattered through the ‘“liver”-cells there is a considerable amount of brownish pigment. The larva feeds on unicellular alge; spherical algee and a few diatoms are present in the gut. : The ccelom is not spacious; there are, however, obvious portions of the ccelom in the postero-lateral regions of the body and a connecting transverse cavity posterior to the mid-gut. Unfortunately, I have no observations on the ccelomic fluid or its contents in life. All the specimens possess a pair of ccelomoducts (nephridia), which have not hitherto been observed in the free-swimming larve of any Discinid. The funnel of each ccelomoduct is situated on the ileo-parietal band about the level of the posterior margin of the mid-gut; its opening is directed posteriorly and rather towards the middle line (fig. 11). The funnel leads into a slender tube which can be traced forwards a little more than half way along the lateral margin of the occlusor muscle to its opening into the mantle-chamber. The total length of the ccelomoduct is. about 70-90. ‘The funnel appears in section asa double series of cells with deeply staining nuclei, representing the two lips, between which is a narrow lumen. The dorsal lip is: slightly longer than the ventral. The rest of the duct is lined witha flat epithelium and ensheathed with a very thin ccelomic epithelium. The lumen of this portion is widest about the middle of its length, where it may attain a diameter of 54, but is usually much narrower (2—8,). The larvee described above correspond exactly with those studied by Frirz Mutter, and they agree with those described by Drs BLocuMann and Yarsu except that their specimens had no pigment-spots (“eye-spots”). This difference, in view of the close agreement in other respects, should probably not be considered as significant ; it may be merely a local variation. Professor BrocuMaNnn has adduced cogent evidence for regarding the larve described by Mitiuer and himself as belonging to Pelagodiscus (Discinisca) atlanticus. The identity of my larve with those of MULLER and BLocHMANN seems so clear that they may safely be referred to the same widely distributed species, the range of which is now extended into the Indian Ocean. ‘There is, so far as Iam aware, no record of adults of P. atlanticus from the Indian Ocean, and the nearest record of the species is that of Professor BLocHMANN, whose larvze were taken off the island of Bintang. 66 DR J. H. ASHWORTH It is interesting to notice that all the known larvee of Pelagodiscus atlanticus— about thirty altogether—have 4 pairs of approximately equal cirri.* MULLER was able to keep his larvee under observation in aquaria, and it is evident from his accounts that they became fixed at the stage with 4 pairs of cirri. Dr Yarsu (1905) found a young attached example of a related species, Disciisca levis, which had 6 pairs of cirri, so that in this species also attachment of the larva occurs at or not long after the stage with 4 pairs of cirri. These facts account for the absence of free- swimming stages with more than 4 pairs of cirri, but there is no satisfactory explanation why earlier lJarvee have not been found. MUiimr (1860, p. 79) suggested that the younger stages may be retained in the shell of the parent, but I am not aware of any evidence in support of this view. It may be that the younger stages live on the sea-bottom and hence have hitherto escaped capture ; all the larvee of Pelagodiscus recorded have been taken at or near the surface. Breeding Season.—Few data are available on this point. Professor BLocHMANN’s specimens were collected off. Bintang on July 3, 1899, and mine were taken off Cape Comorin on October 16, 1914. Dr Yarsu does not state when he found his. specimen, but, as it occurred among the larvee of Lingula at Misaki, it must have been in the period from the latter part of July to the middle of September. MULLER states that at Desterro, Brazil, the larvae occurred in late summer, from February to April, a season which corresponds approximately with the period in which the larvee have been found in the northern hemisphere. LIST OF LITERATURE. Biocumany, F., “ Die Larve von Discinisca,” Zool. Jahrb., Abt. Anat., Bd. xi, pp. 417-426, Taf. 31, 1898, BiocuMmayyn, F., “ Unters. tiber den Bau der Brachiopoden,” ii. Jena, 1900. Brooks, W. K., “The Development of Lingula,” Chesapeake Zool. Lab. Se. Results of 1878, pp. 85-112, pls. 1-6. Baltimore, 1879. ; Ercuter, P., “ Die Brachiopoden der deutschen Siidpolar-Exped.,” D. Siidp.-Exped., Bd. xii, Zool., iv Bd., pp. 384, 385, Taf. 44, Fig, 22, 1911. M°Crapy, J., “On the Lingula pyramidata described by Mr W. Stimpson,” Amer, Journ. Sct. and Arts, 2 ao vol. xxx, pp. 157, 158, 1860. Morse, E. S., ‘‘ Embryology of Zerebratulina,” Mem. Boston Soc. Nat. Hist., vol. ii, VP. 249-264, 1873. Morse, E. S., Proc. Boston Soc. Nat. Hist., vol. xix, p. 266, 1878. Mutter, F., “ Beschreibung einer Beacirepedenine Archiv f. Anat, Physiol., Jahrg. 1860, pp. 72-80, Vat, i, ; Miter, F., “ Die Brachiopodenlarve von Santa Catharina,” Archiv f. Naturgesch., Jabrg. xxvii, Bd. i, pp. 538-56, 1861. Semper, C., “ Reisebericht,” Zettschr. wiss. Zool., Bd. xi, pp. 103, 104, 1861; Bd. xiv, p. 424, 1864. Sewer, R. B. S., “ Note on the Development of the Larva of Lingula,” Records Indian Mus., vol. vii, DP. 88-90, 1912. % meant the abnormal specimen mentioned on p. 62, which has only three cirri on the left side. ON LARVA OF LINGULA AND PELAGODISCUS (DISCINISCA). 67 Simrotn, H., “Die Brachiopoden der Plankton-Exped.,” Zrgcbn. Plankton-Exped., Ba. ii, F.f., 19 pp. ie Wart., LSoir. Yarsu, N., “On the Development of Lingula anatina,” Journ. Coll. Sci. Imp. Univ. Tokyo, vol. xvii, art. 4, 112 pp., 8 pls., 1902. Yarsu, N., “ Notes on Histology of Lingu/a anatina,” op. cit., art. 5, 29 pp. 2 pls. 1902a. Yatsv, N., “ Notes on the young Discinisca,” Zool, Anz., Bd. xxix, pp. 561-563, 1905. DESCRIPTION OF THE PLATES. List of Reference Letters. A. Anus. A.S. Arm-sinus. B.M. Brachial muscle. B.M.T. Transverse brachial muscle. B.W. Wall of body proper. C. Cirrus. Cl, C4, Cll, C13. First, fourth, eleventh, thirteenth cirrus. C.D. Coelomoduct. Cu. Cheta. Cu. I, Cu. II, Cu. III, Cu. IV, Cu. V. First to fifth large cheetre of Pelagodiscus. Cu. 8. Slender cheete. Co. Celom. Co. E. Ccelomic epithelium. D.B. Dorsal blood-vessel. D.M. Dorsal mesentery. EK. Epistome. G.C. Gland-cells in mantle. G.L. Lateral nerve-ganglion. G.P. Gastro-parietal band. G.V. Ventral nerve-ganglion. H.L. Hinge line. In. Intestine. L.A.D. Anterior dorsal lobe of “ liver.” L.M. Lateralis muscle. L.P.D. Posterior dorsal lob of “liver.” L.V. Ventral lobe of “ liver.” M. Mantle. M.G. Mid-gut. M.M. Mantle-margin. M.M.D. Margin of dorsal mantle. M.M.V. Margin of ventral mantle. Mo. Mouth. M.T. Thickened area of mantle. N.T. Nervous tract. di. Gsophagus. O.A. Anterior occlusor muscle. O.I. Obliquus internus muscle, O.P. Posterior occlusor muscle. P. Pigment. P.D, Dorsal protegulum. Pe. Peduncle. Pr. E. Epithelium of peduncle. Pe. M. Muscles of peduncle. Pr. Periostracum. P.S. Pigment-spot (“eye”). P.V. Ventral protegulum. S. Shell. Sc. Statocyst. S.M. Shell-margin. S.M.D. Margin of dorsal shell-valve. S.M.V. Margin of ventral shell-valve. 8.8. Supporting substance. Sr. Stomach. T: Tentacle. T.M. Transverse muscles. Puate LY. Lingula anatina Bruguiére. Figs. 1, 2. Outlines of ventral (fig. 1) and dorsal (fig. 2) shell-valves of youngest larva obtained (52 mm. long). The protegulum, or primary shell, is seen in each valve. See p. 48. (x 39.) Fig. 3. Outline of ventral shell-valve of a larva 1:1 mm. long (larva ¢, group iii, p. 49). (x 39.) 68 : DR J. H. ASHWORTH Fig. 4. Outer surface of the ventral valve of the largest larva, 1°6 mm. long (larva d, group v, p. 52); to show the protegulum and the principal growth-lines on the secondary shell. See p. 53. (x 39.) Fig. 5. Larva with 11 pairs of cirri, ventral aspect ; drawn from a preserved specimen. The anterior and posterior ocelusors and the lateralis muscle are shown, but the other muscles have been omitted. The coelomoducts were present, but have not been shown ; they are more clearly seen in later stages (see fig. 10), The cirri are contracted, and the mantle, which in life extends to the margin of the shell-valves, has also been withdrawn. The right and left lobes of the ventral “liver” are seen to unite in front and to open into the stomach by a common portion (see also fig. 10). Between the posterior margin of the Jophophore and the anterior edge of the ventral nerve-ganglion in the body-wall there is a deep depression shown in light tone in the figure. Note the first indication of the peduncle. See p. 49. (x 90.) Fig. 6. Posterior end of a Jarva with 14 pairs of cirri (larva a, group v, p. 52), ventral aspect. Note the peduncle arising from the ventral mantle-fold, turning to the animal’s right, then dorsally and left, and terminating in a knob lying between the dorsal and ventral mantle-folds. The narrow canal connecting the peduncular cavity with the ccelom is indicated. In the mantle on each side are cheete and a series of gland-cells. Two of the cheetae—those first formed—are rather larger than the rest. The dark central area indicates the extent of the body proper. (x 90.) Fig. 7, The distal half of one of the large chete of a larva with 15 pairs of cirri (larva d, group v, p. 52). (x 500.) Fig. 8. Horizontal section of the dorsal region of a larva (f, group iv, p. 50) with 13 pairs of cirri. In the posterior part the section contains the dorsal body-wall (which is intact) in the anterior region of which are the two statocysts. Further forward the dorsal body-wall has been cut away. The two antero-dorsal “liver ”-lobes are coming together in the middle line preparatory to opening into the stomach, The dorsal blood-vessel is seen through the dorsal body-wall, and a small portion of the vessel is cut again further forward in the dorsal mesentery. (x 100.) Fig. 9. Horizontal section, 204 thick, of the same larva, taken nearly midway between dorsal and ventral surfaces. The section passes through the esophagus and stomach and the postero-dorsal “liver ”- lobes. The cirri are drawn only in outline ; note in the lophophore on each side a portion of the arm-sinus and of the band of supporting substance lateral to it. Only the basal half of the tentacle is present in the section. (x 100.) Fig. 10. Horizontal section, 50% thick, of the body proper of the same larva, taken in the ventral region, showing the posterior portion of the stomach (cut tangentially), the intestine, the ccelomoducts (nephridia), and the ventral “ liver.” The median anterior portion of the “ liver ” leads into the stomach, which, in that area, lies immediately dorsal to the plane of the section. Note also the ventral nerve-ganglion in the anterior wall of the body. (x 100.) Puate V. Pelagodiscus (Discinisca) atlanticus (King). Fig. 11. Larva, ventral aspect. Between the posterior margin of the lophophore and the anterior edge of the body-wall there is a deep depression (cf. fig. 17). The dark tone on the mid-gut indicates the area occupied by the flagellated cells, i.e. the area which will form the chief part of the stomach. Note the intestine, celomoduets (nephridia), and peduncle. The five principal chete are shown on both sides ; but on the right of the figure only are represented the three jointed chetze which are situated in the ventral mantle-fold, and the first two and the last of the series of about thirty slender chet present in the dorsal mantle-fold. The mantle, which extends to the margins of the shells, is extremely thin over the greater part of its extent, but its thickened zone (cf. fig. 17) is represented by the deeper tone. (x 180.) Fig. 12. Dorsal aspect of the same larva, showing only the body proper. Note especially the statocysts, and the “eyes.” The thickening of the body-wall in front is due to the presence on each side in that region of the lateral nerve-ganglion. (x 180.) : Fig. 13. Distal end of the first left cheta. See p. 60. (x 800.) Fig. 14. The third principal cheta. The processes are on its anterior margin. Seep. 61. (x 800.) Fig. 15, The distal end of the fourth principal cheeta ; note the absence of spines on its median margin. See p. 61. (x 800.) ON LARVA OF LINGULA AND PELAGODISCUS (DISCINISCA). 69 Fig. 16. Proximal portion (about two-fifths) of one of the three jointed chetz in the ventral mantle-fold. See p. 61. (x 800.) Fig. 17. Sagittal section of a larva, Note that at the margins of the shell-valves the periostracum is turned in and continued for some distance over the mantle-fold. The surface of the periostracum is studded with minute points and is therefore shown dotted. The mantle is in contact with the inner surface of the shell, and, except over the body proper, is composed of two lamelle; these are represented diagrammatically as two thin nucleated membranes, over the greater part of its area the mantle is very thin. See pp. 61-65. ( x 450.) Fig. 18. Horizontal section of a larva passing through the middle of the pigment spot (“eye”), and through the statocyst of the right side. A portion of the outline of the right anterior occlusor muscle is shown by dotted lines. See pp. 63, 64. (x 700.) Fig. 19. Three cells from the postero-ventral wall of the mid-gut, i.e. cells of the future stomach. See p. 64. (x 1800.) TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 3). 11 wt — 4 7 , ' » ; yf ee i ‘ | { Lt! . . . 44 4 R re -“ . 5 a) ‘ - ‘ ‘ 4 2 - me w i ‘ ~ + 4 : » f 2 : } * ] . } ; we . ’ ~ 4 . ~ as ans. Roy. Soc. Edin. Ashworth. Lingula. Wolk il Sriave Vv so ek ee ae Cr. H.Ashworth and A.K.Maxwell del. André Sleigh & Anglo L¥# Trans. Roy. Soc. Edin. Ashworth: Pelagodiscus. Welle. Fietem: WET ——— = eniraasiiin D i ——é a aa IN. “ d ‘CHV BP PE. CHS Co "4 I / aa ee ere emacs NEG: OA... I1H.Ashworth and A.K.Maxwell del. André Sleish & Anglo L* (71>) IV.—The Temperatures, Specific Gravities, and Salinities of the Weddell Sea and of the North and South Atlantic Ocean. By William 8. Bruce, LL.D., Andrew King, F.I.C., and David W. Wilton. (MS. received March 15, 1915. Read May 3,1915. Issued separately November 29, 1915.) ’ Intropuction By W. 8. Brucs, LL.D. After my return from my first voyage, namely, the Scottish Antarctic Expedition of 1892-93, for which I was chosen as naturalist, I had the advantage of coming closely in touch with Mr J. Y. BucHanan, who then trained me in his Edinburgh laboratory in the use of his hydrometer. During the past. twenty years I have been in constant touch with him regarding problems relating to the physics of the ocean. As a consequence, before my departure for a voyage to, and wintering in, Franz Josef Land during 1896-97, I was well trained in the use of his hydrometer, and obtained during that expedition observations of considerable interest. I also carried out all the hydrometer work on board the Prince or Monaco’s yacht Princess Alice during his Arctic voyage of 1898, under the direction of Mr J. Y. Bucnanan, who was also on board during that cruise, and also during the Prince or Monaco’s second Arctic cruise during 1899, when I was solely responsible for that work. Consequently, I was thoroughly familiar with hydrometer work when I set sail in the Scotea. To meet modern criticism, it may be well to state that I was well trained in physics and chemistry under Witiiamson, Crum Brown, and Tart, and that up to the time of my departure to Franz Josef Land I was for nearly two years on the summit of Ben Nevis. On the voyage of the Balena to Antarctic seas in 1892-93, I only used the usual hydrometer supplied by the British Meteorological Office ; but prior to my departure I had received special instructions from the late Mr Rosertr Irvine of Royston and Dr H. R. Mitx in methods for collecting samples of sea water and taking sea temperatures. I had also reduced most of the physical observations of the Scottish Fishery Board for a period of ten years. Meruops anp Instruments. By W. S. Bruce, LL.D., anp AnpREW Kina, F.I.C. It was Mr D. W. Witton who earried out all the hydrometer work in the deck laboratory of the Scotia, which was well fitted for the purpose. Mr Witron had been a student of chemistry and physics in the University of Edinburgh, and had passed his theoretical and practical examinations—both class and TRANS. ROY. SOC. EDIN., VOL, LI, PART I (NO, 4). 12 72 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE degree examinations—with credit. Subsequently he was an observer at the summit and base, Ben Nevis Observatory, and was put in charge of, and was responsible for, the initiation of the observatory at an altitude of 2200 feet—a position half way between the summit and the base observatories. Before his departure with the Scotea he received special instruction in hydrometer work with Mr Bucuanan and Mr Kine in Mr Bucwanayn’s laboratory in Edinburgh. On board the Scotia Mr Wixron sub- mitted his observations to Mr R. C. Mossman and to Dr Brucr, who checked them from time to time by counter observations. These check observations in every case compared most favourably with Mr Writon’s. Mr Wivron’s observations may there- fore be taken as absolutely reliable. The collection of the water samples and the recording of temperatures and the general conduct of the concurrent work on the scientific bridge was undertaken by Dr Bruce, with the assistance of Dr J. H. Harvey Prris, who handled the Lucas sounding machine; the late Mr Attan Ramsay or Mr Gravitt, first and second engineers, who handled the special high-speed winch ;* and in turn one of the three mates of the Scotia, namely, Mr Joun Fircurm, Mr Roserr Davison, or Mr Roserr MacDoveauy. Captain THomas Ropertson handled the ship during the operations, while Dr Brucs was in such a position on the flying platform of the forward scientific bridge as to be in sight and hearing of Captain Roperrson on the nautical bridge, the engineer at the winch, and Dr Pirie at the sounding machine, while one of the mates was with him at the flying bridge. During these operations Mr Wiuron was on the main deck, and the door of the laboratory was here just below the scientific bridge. Consequently, as soon as Dr Brucs recorded a deep-sea sample on the bridge his reading of the temperature was checked and recorded. He then emptied the contents of the deep-sea water-bottle into the special bottle which Mr Witton had in readiness for its reception, and thereafter Mr Witton duly labelled it and put it in the place definitely assigned to it in the laboratory. So systematic was this arrangement that when Dr Bruce gave the order, or rather expressed the wish, that a deep sounding and physical observa- tion should be taken, everybody was in his right place and knew exactly what was required of him without any delay or confusion—the crew cheerfully assisting and showing increasing interest in the work. The only regret is that a vessel so well fitted out for deep-sea research in any part of the world, and a body of chosen men who had become such adepts at the work, should not be able to keep together permanently for further research. Alas! the ship had to be sold, and scientists, officers, and crew scattered, never again to continue work that they had faithfully and ably carried out during almost two years; and it ought to be mentioned. for Dr Pirie and the officers and men who assisted on the scientific bridge in all * Vide special description, “ Scientific Equipment,” “ Scotia” Reports, vol. i. 1t is interesting to note that this winch has been subsequently used on Sir Doucuas Mawson’s expedition, and is at present being used on board the Aurora by Sir Ernest SHACKLETON, to which two expeditions Dr Brucs has been glad to lend it. TEMPERATURES, SPECIFIC GRAVITIES, AND SALINITIES OF CERTAIN SEAS. 73 weathers and temperatures down to zero Fahrenheit, and especially for those on the flying platform, that the work was not carried out without great discomfort, because the rapidly incoming sounding-wire carried with it a continuous spray, which in low temperatures congealed as it fell and covered them from head to foot with ice. These conditions were intensified owing to the fact that all such work has to be carried out on the weather side of the ship. In spite of the high speed which the Scotia’s scientific winch was able to attain, all these operations meant long and arduous deck work, as will be seen bysomeactual records here quoted, and more fully dealt with under “ Scientific Equipment.” * The usual time for winding in was 80 to 90 fathoms per minute (although in a speed trial the great speed of over 140 fathoms per minute was attained), which for the Scotia's deepest sounding of 2900 fathoms meant continuous heaving in for thirty-four minutes. This, when added to necessary halts for reading the temperatures, collecting the water samples, and taking the instrument aboard, would amount to fully an hour’s deck work for heaving in alone. In addition to this there was fully another hour’s work “ letting go” and attaching and setting the sounder and deep-sea water-bottles. Repeating the operation once or twice more for lesser depths, in order to obtain a more complete series, furthermore propor- tionately added to the time. For this work there was always used a special three-stranded wire on the Lucas sounding machine. This wire naturally was not only more reliable than the single wire, but was very suitable for holding a temporary seizing which gave instruments a better hold and prevented them from slipping down the sounding-wire. The water-bottle generally used was the Buchanan-Richard water-bottle, and the thermometer the most recent pattern Negretti & Zambra reversing thermometer. The sounder was the Buchanan sounder, and immediately above it was fixed a reversing thermometer. Only two, or at the most three, Buchanan-Richard bottles were fixed on the wire at the same time, and usually at a distance of 500 fathoms apart. This practice risked the loss of fewer instruments in the event of the wire carrying away, and also had the advantage of giving the use of the same instruments for the next series of observa- tions, thus making the observations more directly comparable. For the same reason it also was endeavoured, as far as possible, to use the same instruments on every occasion. Occasionally the Pettersson-Nansen insulated water-bottle was used with the direct-reading Richter thermometer. This instrument is a very beautiful one, but there are serious objections to its general use, especially in low temperatures. These objections are— 1. Its cost. 2. The manipulating of several metal levers and fine screws—a serious handicap when working with numb fingers. * “Scientific Equipment,” “ Scotia” Reports, vol. i. 74 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE 3. The question of its reliability when used during low air temperatures, and the fact that the whole instrument may be solidly frozen up immediately it comes out Hoh the water (this actually occurred on board the Scotia). 4. The fine scale of the Richter thermometer, which is difficult to read on the deck of a ship in bad weather and poor light. 5. The fact that it is not possible to use it with reliable results at greater depths than 1000 fathoms. The Buchanan-Richard water-bottle is more desirable than the Pettersson-Nansen— 1. Because it is inexpensive. 2. Because it is easily manipulated. 3. Because it does not jam by freezing up during cold weather. 4. Because the Neeretti-Zambra reversing thermometer scale is easily read under conditions of weather and light when the Richter thermometer cannot be read at all, or with difficulty. 5. Because it can be used at any depth desired. It has been contended that the point at which the mercury breaks off in the reversing thermometer is not constant. Errors due to this will probably be found to be less frequent than those which occur with an instrument with such drawbacks as have been mentioned in the case of the Pettersson-Nansen bottle with the Richter thermometer. From what has been already said, it is unnecessary to emphasise that the hydrometer observations taken by Mr Witton were those of a trained physicist,andsuch methods* as using “ towels or wash-leather ” that “ were not generally quite clean,” or “ the surface of the water samples” being “ more or less contaminated during the determinations with the hydrometer,” were not practised on the Scotia. Mr W1xTon had all his instruments, silk cleaning-cloths, vessels, and fingers, and surroundings generally, scrupulously clean before commencing observations. Besides this, he was most careful in maintaining the laboratory and the water samples as nearly as possible at the same temperature. In this he was in the highest degree successful, even in weather conditions that were on many occasions very difficult to deal with. The laboratory was fitted with a special’serving-table- for this work ; and there was also constructed a special swinging platform, which on some occasions, but not as a rule, was more convenient than the table. The table proved most useful for this and other delicate operations. The form of hydrometer used by Mr Witron in determining the rEeGe of the sea-water was the Buchanan absolute-weight hydrometer. One of these hydrometers (No. 25) was used for Samples 1 to 199, but unfortunately it was broken on 24th February 1903. Thereafter hydrometer No. 14 was used. This form of hydrometer is fully described in Mr J. Y. Bucnanan’s elaborate monograph on “ Experimental ° The Norwegian North Polar Expedition, 1893-96, vol. iii, pt. ix, pp. 147, 148, “The Oceanography of the North Polar Basin,” by Frrptsor NANSEN, TEMPERATURES, SPECIFIC GRAVITIES, AND SALINITIES OF CERTAIN SEAS. 75 Researches on the Specific Gravity and the Displacement of some Saline Solutions.” * It was used by Mr Witton in the same way as during the Challenger Expedition, 7.e. with the help of previously determined constants. These constants are (1) the weight im vacuo of the hydrometer in grams; (2) the volume in c.c. of the body of the hydrometer at 0° ©. up to the lowest division of the stem; (3) the rate of expansion of the body per degree C. in c.c.; (4) the volume per myn. of the divided stem in c.c. Of late years Mr Bucuanan has introduced vast refinements into the method of using the instrument on shore for chemical research, and these are described in the monograph referred to above. However, it was not very practicable to apply these delicate methods on board the Scotia, so the original method of use was adhered to during this Expedition. This form of hydrometer has a volume of about 180 c.c. Its divided stem is 100 mm. long and displaces from 0°8 to 1'0 cic. The maker, before sealing it up, loads it to float at the bottom of the stem in distilled water of ‘from 30° C. to 40° C. (when it is intended for use in aqueous liquids). Its weight in vacuo is once for all determined with the greatest care. A set of weights of such form as can easily be placed on the top of the stem is provided. The heavier weights used by Mr Witton were made of brass wire, and the lighter ones of aluminium wire. The probable error of each weight was not more than 0'1 mg. With these weights the effective weight of the hydrometer can be increased at will, and they are so adjusted that the successive addition of each weight can increase the total weight by 0°05 gram. The stem is divided into 100 mm., and its displacement is 0°1 c.c. per from 10 to 12mm. Thus when the hydrometer is floating in distilled water, or in an aqueous solution.of not very high specific gravity, the addition of 0°1 gram to the effective weight sinks the stem by from 10 to 12 divisions. Hydrometer 25.—This hydrometer was used in: the testing of the density of samples 1 to 199. Its weaght im vacuo is 186'2944 grams. The volwme of the body up to the lowest division of the stem at 0° C. is 186°8427 c.c. The rate of expansion of the body per degree ©. is 0°00567 cc. The volume per mm. of the stem is 0°00887 c.c. These constants were determined in Edinburgh before the start of the Scotia Expedition. Let t’ be the temperature in degree C. of the sample of water whose density is to be determined. R the stem-reading at the place of immersion. W the weight in vacuo of the hydrometer in grams. w the total weight of the added weights in grams. Then (W + w) is the total effective weight of the hydrometer, and consequently is the weight in grams of the sea-water displaced ; while (186°8437 + 0°00567t’ + 0°00887R) ¢.e. is the volume of the immersed part of the hydrometer and consequently of the displaced sea-water. * Trans. Roy. Soc. Hdin., vol. xlix, part i, 1912. 76 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE Hence the density of the sample at t’ is : (W +w) (186'8437 + 0°00567t' + 0°00887R) Thus if this particular hydrometer (No. 25) floats at division 31°8 of the stem with anadded load of 5°80 grams, and the temperature of the water be 21°2°C., the density ate 2 C. As z 186°2944+4 5-80 ee SD 186°8427 + 0:00567 x 21°2 + 0°00887 x 31°8 re Hydrometer 14.—This hydrometer was used in the testing of samples 200 to 578. Its weight in vacuo is 181°0189 grams. The volwme at 0° C. of the body of the hydrometer up. to the lowest division of the stem is 181°5471 cc. The rate of expansion per degree C. of the body is 0°00624 c.c. The volume per mm. of the stem — 18/0 009046 cc. 7 = | > a3 “J As in the case of hydrometer 25, these constants were determined in Edinburgh before the start of the Expedition. Thus in determinations of density made with this instrument, the density of. the sample is found by the formula W+w (181°5471 +0°00624t' + 0:009046R)’ W, w, R, and t’ having the meanings assigned above. Mr Witton never took less than four separate readings of the stem by successive addition of weights so as to increase the amount of immersion. In most cases, lhowever, he took considerably more, generally about eight. From each of these readings an independent value for the density could be calculated ; but as this would have involved much laborious calculation, he calculated the density from the mean effective weight and the mean stem-reading. ‘The temperature of the sample during experiment (t’) was read before immersing the hydrometer in the liquid and after removing it, and the mean of these two readings was taken as the “ temperature during experiment.” This was used in the calculation of the density by means of — the formula given above. The two readings of temperature never differed by more than 1°70 C., but generally varied by very much less.* It was very exceptional to have so large a difference as 1°°0 C., and in those cases the accuracy of the density determination may be open to doubt. In the tables the column T’ gives the temperature of the air at the time when the determination of density was being made, and the column t’ gives the temperature of the sample of sea-water whose density is being determined. The nearer these temperatures are to each other, the less alteration in the temperature of the sample during the experiment is to be expected. * Mr Bucwanan would not be satisfied in his chemical work with an experiment in which the temperature of the solution during experiment varied by more than 0:01° C., but under the circumstances and conditions such a degree of accuracy as this could not be expected. TEMPERATURES, SPECIFIC GRAVITIES, AND SALINITIES OF CERTAIN SEAS. 77 The column w gives the actual additional weights, and the mean of these numbers (which is also incorporated in this column) gives the mean added weight. On addition of this mean number to the weight of the hydrometer in vacuo we obtain the total effective weight (W + w). Column R gives the readings of the stem of the hydrometer corresponding to each successive addition of weight, and the mean of these readings is also incorporated in this column and represents the mean portion of stem immersed. Column V gives (1) the volume of the body of the hydrometer at 0° C. up to the lowest division of the stem, the figures 18 being omitted; (2) the expansion of the body of the hydrometer due to. the particular temperature of the sample during experiment ; (3) the volume of the mean portion of the stem immersed. (1) is of course a constant number, while (2) and (3) have been obtained by the use of the formula stated above. The sum of (1), (2), and (3) which is given in the column expresses the total volume in c.c. of the immersed portion of hydrometer, and consequently of the displaced sea-water. Column ,S,. The numbers in this column represent the density of the sample at the temperature of experiment (t’), and are obtained by dividing the total effective weight of the hydrometer by the sum of the numbers in column V, which expresses the volume of the immersed part of the hydrometer. Column ,Si5.¢ gives the density of the sample at 15°°56 C. The numbers in - this column are calculated from those in column 4S; by the use of the tables con- tained in Dittmay’s “Report on Researches into the Composition of Ocean Water,” pe. 70." Column ,S; gives the density of the sample at (t), which is the temperature in 0°C. of the sea at the time the sample was collected. It thus gives the actual density of the water im situ. The numbers in this column are derived from those in column Siss6 by means of the tables in Dittmar’s Report in the case of those in which the value of t is above 0° C. In those cases in which the value is below 0° C. recourse was had to Knudsen’s Tables published as No. 11 of the ‘ Publications de Circonstance” of the Conseil Permanent International pour |’Exploration de la Mer. N.B.—In the tables numbers which are not actually observed quantities, but which have been obtained by calculation, are printed in italics. Thus in columns (t) and (T”’) the temperatures were actually observed in ° F., and their equivalents in °C. are printed in italics. Also the figures in the last three columns are all in italics. * Phys. Chem. “ Chal.” Eup., pt. i, 1884. [ OBSERVATIONS. 78 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAYVI OBSERVATIONS OF THE TEMPERATURE AND DENSITY OF SEA-WATER MADE ON BOARD THE §&.Y. “SCOTIA,” 1902-1904. By Davip W. Witton. HyproMeterR No. 25. WericuT=186:2944 Grams. - Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the § " a Depth | eS | Vol i Date (I). Position (L). 4 Dp Temperature | Temperature ohehaks Density in Fathoms. at time of during Weights | of im- (De alo Collection of | Experiment. | added | Read- | mersed Ww s & ) D:. d. Sample. A ma see ; 3 Sg ee | Colon of Water atime) | ainnne anna Elycdio- As 2 a | of oe T t Current, and and Ae tgs meter Bors. Bye Ob- |Reg = | |Month.|Day.|Hour. Lat. | Long. 58! ‘the _——|——— Hema: Date. ieee Ca (cub. | aa o | | . El¢ / Posi-|S@™ple) of of of of | icemtamaed | 3 E tion Mere the the | the the A |G L. | jected, | Ait. | Water. | - Air. |Sample.| w. R. V. Se. | l i f 1902. oak Shee | | 2UBY cab °F. G5 | 5 | 1) Nov. | 24 | 12.20) 3014N./1813 W.! .. | Surface 67-7 | 70:0 Bright blue atc (67-6 20-4 5-70 17-9 | 21-1 C. ‘19-8 C. 5:75 23-4 5:80 29-1 6:8437 il. | 5-85 | 344 | 1168 i | 20:8 5:90 39-2 +2554 y bel 20:6 | 5:80 | 288 | 7.2159 | 1-02606) 1 6| 2 > Nov. | 25 | 13.45 | 2814N./1915 W.| .. | Surface! 69-2} 71-0 Bright blue cit I71-0 | 212 5:70 21:8 | ) 21-7 C. 21:7 C.. 5:15 26:8 “a 5-80 31:5 6-8437 5°85 37-5 +1202 | | 21-2 5:90 415 +2820 21-2 5:80 31-8 7:2459 | 1-0258: 7| 3) Nov. | 26 | 12.10 | 2623 N.|2020 W.| .. | Surfacel 70-5 | 72-0 Bright blue 20 72-2 21:8 5:70 24:0 ; | 22:2 C. Z 22-3 C. 5:15 27-9 5:80 34:0 6-8437 5:85 39-0 -1242 | | 22:0 | 5:90 | 445 | +3007 21:9 | 5-80 | 33-9 | 7-2686 | 1-02577\ 1 . |} 10; 4) Nov. | 27 | 12.0 | 2421 N.|2120W.| .. | Surface| 72-3] 73:0 Bright blue tH 73-9 22-6 5:50 3:0 ! 22°8 C. 23.3 C. 5-60 15-0 . 5-70 25-0 6:8437 ; 5:80 36:0 +1287 | 22:8 5:90 46:0 2217 ; 22:7 5:70 25-0 7-1941 | 1:02564) 1 | 13 | 5] Nov. | 28 | 18.15 | 2158 N.|2226 W.| .. | Surface! 72:0} 73-9 Bright blue 5th 75-2 23-4 5:45 5:0 | 23-3 C. 24:0 C. 5:50 10:9 ) 5-55 | 165 5:60 20-0 5:65 25:5 5-70 31:5 } 5°15 36:5 6:8437 | 5:80 42-0 +1338 | 23-8 5:85 48:0 +2324 | 23-6 | 5-65 | 26-2 | 7-2099 | 1-02629| 1 17| 6} Nov. | 29 |12.0 | 2020N.|2321 W.| .. | Surface| 73-0] 74:5 Less bright blue, na 76-2 23-8 5:35 5:0 23:60.) S.W. 24-6 C, 5:45 15:5 ; 5:55 25-2 6-8437 5-65 36:0 +1349 ; 23:8 5:75 47-0 +2279 ; 23-8 5:55 | 257% | 72065 | 102477) 04 | | 7 AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 79 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. Jate (E.) Position (L.) in De, Temperature Temperature Volume Density of Sample. ‘| at time of during Weights of im- (Density of Distilled Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. d. Sample. a ing | Portion aise F SS f£ of = Colour of Water,} Time Hydro- e of | from T t Current, and | and ave t’. meter yer i Bie Ob-_ |Reduced) Reduced Sea | Which i 5 Remarks. Date. (grams). : served to to. fora. |Day.|Hour.; Lat. Long. ak CNC ea === (co at t’. |15°-56C. t. Posi- poe of of of of ae tion wae the the the the L. | jected. | Ait: | Water. Air, |Sample.| _w. R. We aS, |4Sisse| 49 t. Ba ae TE ee one | s2C. 30 |12.0 |19 7N.| 2414 W.| .. | Surface} 74:1 | 174-7 Deep blue, S.W. ve 78-9 24-9 5:30 4-0 23:7 C. 26-1 C. 5:40 15-0 5:50 24:5 6:8437 5:60 34:5 -1417 25:2 5:70 45-5 +2190 : 25:05 5:50 24-7 7:2044 | 1:02452| 1:02704| 1:02491 1 | 7.50 |3 miles N.| E. of Bull| 1970) Surface] 74:0] 75:8 Deep blue, N.E. ce 19:9 26-1 5:25 5:5 Rock, C.| Verde Is. 24-3 C. 26-6 C. 5:35 16-0 5:00 38-2 6-8437 5:65 49-5 +1480 26:2 5:15 60:5 +3007 26-15 5:51 339 7-2924 | 1-02409| 1-02694| 1-02464 1 | 9.50 | Crossing |Salamasa | 465) Surface] 75:5 | 76-1 Deep blue, N.E. ac 80-0 26-2 5:25 6:0 Bay 24-5 C. 26-7 C. 5:30 11:5 ; 5:35 17-0 5:50 33:2 5°55 38:8 6:8437 5:60 45-0 -1486 + 26:3 5:70 56-0 +2625 e ee | eee ee 26-25 5:46 29-6 7:2548 | 1:02402 | 1:02690 | 1:02454 2/12.0 | 1519 N. | 25 6 W.| 2340] Surface] 75-5 | 76:0 | Lighter blue than a0 79-9 26-2 5:20 2:5 24-4C.| yesterday, S.W. 26-6 C. 5:30 13-8 5:40 24-8 5:00 36-5 5:60 46-5 5:10 57-0 5°15 62-0 6:8437 5:80 68-0 +1486 i 26-2 5°85 73-5 +3787 26-2 5-57 42-7 7:3710 | 1:02398 | 1:02685 | 1:02451 3/12.0 |13 7N.| 25 9W.| .. | Surface] 77-3] 78-1 Bright blue ve 79-0 26-2 5:20 8-0 25-6 C, 26-1 C. 5:30 18-8 5:50 41-0 6-8437 5-70 62-8 +1486 26-2 5:80 73:0 +3610 26-2 5:00 40-7 7:3533 | 102370 | 1-02656| 1-02388 4/12.0 | 1118 N. | 2523 W.| .. | Surface! 77-3] 79-1 Bright blue ae 79-0 26-0 5:10 6:0 26-2 C 26-1 C. 5:20 17:0 5:25 22:5 5°30 27-5 5:35 33-2 5:50 50-0 5:50 55:0 5-60 60-0 6:8437 5:65 65-5 +1477 26-1 5-70 71:0 3619 26-05 6-42 40-8 77-3533 | 1-02328 | 1:02609 | 1:02323 5 | 13.0 | 928 N.| 2539 W.| .. | Surface| 79:0| 80-7 Bright blue c 79:9 27-0 5:05 35 27-1 C 26-6 C 5-10 8-2 5:20 19-5 5:25 25:5 5-30 30:3 5°30 37-0 5:40 42-0 6:8437 5:60 63-2 *1531 27-0 5-70 74-8 2998 27:0 5:33 33-8 7-2966 | 1:02311| 1-:02621| 1-02308 | ROY. SOC. EDIN., VOL. LI, PART I (NO. 4). 80 Number of Station. wo a 38 39 41 43 Number of Sample. 14 15 16 17 18 19 20 21 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITIE Date (E.) / Month. | Day.) Hour. | 1902. Dec. 6 | 9.5 Dec. | 6 | 12.0 Dec. 6 | 22.0 Dec. 7 | 6.20 Dec. io, 2:0 Dec. 7 | 12.0 Dec. 7 | 20.0 Dec. | 8 | 6.0 Data Relating to the Collection of the Sample. | Position (L.) Lat. Long. oe / 758 N. 25 27 W. ° , 7 37 N. , 25 27 W. | 6 39 N. | 25 41 W. 5 59 N. | 25 50 W. 5 44 N. | 25 55 W. 5 27 N. | 26 -1 W. 442 N. 26 17 W. 3 46 N. | 26.17 W. Depth in Fathoms. D d. from ou, | which an the Posi-| Sample tion 1 i col- * | lected. | Surface | Surface Surface Surface Surface Surface! Surface Surface, Temperature at time of Collection of Sample. MM t. of of the the Air. | Water. SieBya| aks 77-0 | 80-7 27-1 C. 78:5 | 80:9 27-2 C. 76:2 | 81:6 27-6 C. 79:8 | 80-4 26:9 C. 80-7 | -81-1 27:3 C. 770} 81-1 27:3 C 78:1} 81:0 27:2 C. 79:3 80-9 27-2 C. Data Relating to the Determination of the Density of th Temperature during | Experiment. Colour of Water,| Time Current, and and 4 t’. Remarks, Date. eas of of the the Air. |Sample Bright blue 80-1 26-5 26-7 C. 26-6 26:55 Bright blue 80-2 26-9 26-8 C. 27:0 26-95 d 83-0 27-6 F 28-3 C. 27:8 27-7 Deep blue 83-0 27-8 28-3 C. 28-0 27:9 Bright blue 83-6 27-9 28-7 C. 28-1 28-0 Bright blue 83-7 28-2 28-7 C. 28-2 28-2 A 81-7 27-2 27-6 C. 27-2 27-2 Bright blue 82:0 | 27-4 27-8 C. 27-4 | 27-4 Volume A of im- eres Read- | mersed Fi ing Portion of of Hydro- | Fydro- | Hydro- (grams) meter. meter (cub. centims.) Ww. R. Vv. 5:05 1:2 5-15 10:8 5:30 27:8 5-40 38:8 5:45 44-0 6°8437 5:60 59:8 1505 5:70 715 +3220 5:38 36-3 7:3162 | 5:05 5:0 5-25 26-0 5:30 31:5 5:35 37-0 5:50 53:0 6:8437 5:60 64-0 +1528 5:70 73-8 3681 5:39 41-5 7:3646 5:05 6-5 5:15 175 5°35 39:5 6-8437 5:50 56:8 1571 5:70 78-5 3530 5°35 39-8 7-3538 5:00 14-0 5:10 24:5 5:20 35:0 6:8437 5:30 47-0 +1582 5:40 57-0 +3148 5:20 B55 73167 4:95 6-5 5-00 13-0 5-10 22:8 6:8437 5:20 33:8 +1588 5:25 39-0 2040 5:10 23-0 7:2065 4-95 6:8 5-00 13-5 5°20 34:8 6:8437 5:25 42-0 +1599 5:50 67-5 -2918 518 32-9 72954 5:05 2:8 5:15 13-0 5:20 19-0 68437 5:25 30:5 +1542 5:50 52-5 +2093 5-23 23-6 72072 5:10 12:8 5:20 23°5 5:30 35-0 6:8437 5:40 46-0 +1554 5:50 56:5 +3086 5:30 34:8 73077 ay 1-02237 | 1-025 1-02231}. 1-02306 | 1-026 date (E.) 1. |Day.| Hour. 9.0 19.0 12.10 18.28 12.0 12.0 12.0 Data Relating to the Collection of the Sample. Position (L.) Lat. ° ‘4 ° , 3 29 N. | 26 23 W. 312 N. | 26 29 W. 2 46 N. | 26 40 W. 2 2N.|2717 W. 1 42 N, | 27 29 W. 1 25 N. | 27 59 W. 0 56 N. | 29 24 W. St Paul’s} Rocks 018 S. |30 37 W. Depth in Fathoms. D d. from ot | which at the Posi- Sample tion| Was L col- * | lected Surface Surface Surface! Surface Surface Surface Surface Surface AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 81 Data Relating to the Determination of the Density of the Sample. Temperature Temperature Volume Density of Sample. at time of during Weights of im- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) Sample. i mg Portion Colour of Water,} Time Hydro- or e Se T t Current, and "| ‘and 14 t’. meter Hye Eyer Ob-_ |Reduced| Reduced ‘ Remarks. Date. (grams). 4 (cub served to to ri centims.) abt’. |15°-560.| t. of of of of el oae rh the the the the Air. | Water Air. |Sample w. R. We 494, |4Sisese.| 4Se caple Woe SOE. 2G, 80-5 | 80:7 Bright blue 82-0 27-4 5:00 2:5 27-1 C. 27-8 C. 5:10 12-5 6:15 18-0 6-8437 5-20 24-0 +1559 27-6 5:50 56:5 2013 27-5 5-19 22:7 72009 | 1:02288) 1:02613 | 1-02300 80:4 | 80-9 Bright blue 82:3 27-4 5-00 18 27:2 C. 28-0 C. 5:10 12-0 5:20 23-0 5:25 28-0 6-8437 5:35 40-0 +1559 ‘ 27-6 5:50 55'5 2368 27-5 5:23 26:7 7-2364 | 1:02290| 1-02615 | 1-02299 80-0] 80:3 Bright blue 82-7 27-4 5-05 5:0 26-8 C, 28-2 C. 5:10 10:3 5:20 21-0 6-8437 5:25 27-5 +1556 27-5 5-50 55:0 2111 27-45 5:22 23-8 7-2104 | 1:02299| 1-02623| 1:02318 78:5 | 79-7 Bright blue 81-9 27-3 5-05 3-0 26:5 C, 27-7 C. 5:15 15-0 5:25 25:5 6-8437 5:50 53-0 +1551 27-4 5-60 65-0 *2865 27-35 5:31 32-3 72853 | 1:02306)| 1:02627 | 1-02332 79:3} 80-1 Bright blue 82:0 27-2 5:10 8-0 26-7 C, 27-8 C. 5:15 13:3 5:35 35:0 6-8437 5:50 51-2 +1542 27-2 5:70 72-0 3184 27-2 5:36 359 7:3163 | 1:02316| 1-02633)| 1-02331 79:0| 80:0 we 81-1 26-8 5-10 7-0 26:7 C, 27-3 C. 5:20 17-5 5:25 22-0 6:-8437 5°35 33-0 ~1520 26:8 5:40 37:8 -2084 26-8 5:26 23-5 7-2041 | 1:02324| 1-02629| 1-:02327 79:5 | 79:8 Bright blue 80-9 27-0 5-10 5:8 26-6 C 27-2 C. 5-20 17-7 5:25 23-0 6-8437 5°30 28-0 +1525 26:8 5:50 49-0 +2191 26-9 5:27 24-7 7:2153 | 1:02323 | 1-02631| 1:02332 78:6 | 78:9 Bright blue 80-8 27-0 5-20 11-2 26+1C. 27-1 C. 5°30 22-0 5:35 26-5 5°50 44:5 5:60 55:5 5:25 16-0 5-40 32-0 6-8437 5:45 37:0 +1531 27-0 5:55 49-0 +2891 27-0 5:40 32-6 72859 | 1:02354| 1-02665| 1-02382 82 Number of Station. oO = 61 61 61 61 61 ‘al The following four s Number of Sample. 30 31 31 31 31 | 31 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVI TE Data Relating to the Collection of the Sample. Position (L.) Date (E.) i ee Month. Day.| Hour. 1902. | / Dec. | 12 | 12.0 Dec. | 13 | 12.0 Dec. | 13 | 12.0 Dec. | 13 | 12.0 Dec. | 13 | 12.0 Dec. | 13 | 12.0 Lat. Long. o 7 Go 4 1598S. |32 9 W. 3 39S. | 33 21 W. 3 39S. | 33 21 W. 3 39S. | 33 21 W. 3 39S. | 33 21 W. 3 39S. | 33 21 W. This sample was tested for checking purposes five times on 17th Dec. 1902, each observation being perfectly independent of the other. Two observations each were made by W. S. B. and myself and one by R. C. M. The conditions during the last three observations were more favourable than during the first two, chiefly on account of the Ww motion of the ship b ioe eing gentler.—D. W. | | amples tested at Fort Stanley on the 10th Jan. 1903 :— Depth in Fathoms. gi ee Collection of D d. Sample. re F Colo of Bett f sj 1 urrent, an Sea ui * eet t. Remarks. at Posi Panis of of tion aD the the iTS 1eAteAs Air. | Water. ° ‘ ° F. Surface 78-7 | 79-1 Bright blue . 26-2 C Surface 78-9 | 79-0 Bright blue 26:1 C | Surface 78:9 | 79-0 Bright blue 26-1 C. Surface) 78-9 | 79-0 Bright blue 26:1 C Surface 78-9 | 79-0 Bright blue 26-1 C. Surface 78-9 | 79-0 Bright blue | 26-1 C. | | MRE SG | | No. 49 ROM) = Gc Nol « 5 ¢ Data Relating to the Determination of the Density of the Samp Tempera uring Weights Experiment. ico Read- to. | ite Time Hydro- and Ts ti", meter Hide Date. (grams). i of of cl the the Air, |Sample.} w. R. cee SiGe 80-7 26-9 5:15 5-0 27-1 C. 5:20 10-0 5:25 16-0 5:30 20:8 26-9 5:35 26-0 26:9 5:25 15-6 81-6 26:3 5:20 4-0 27-6 C. 5:25 9:8 5:30 15:2 5:35 20:5 26-4 5:40 26-5 26-35 5:30 15-2 81-6 26:5 5:20 6:0 27-6 C. 5:25 11-0 5:30 16:5 5:35 22-2 26:5 5:40 27-0 Wall). en) 2Ore) 5:30 16:5 a 81-6 27:0 5:30 18-5 27-6 C. 5:35 22:5 5:40 28-7 5-45 34-0 5-50 39-8 5-55 45:0 5-60 51-0 5:65 56:2 27-1 5:70 62-0 27-05 5:50 39-7 81-6 26:8 5:30 19-0 27-6 C. 5:35 24:5 5-40 29:5 5:45 34:5 5-50 41-0 5:55 46:0 5-60 51:0 5-65 56-5 27-0 5:70 62:5 W.S.B. 26-9 5-50 40:5 81-6 27-2 5-30 18-5 27-6 C. 5:35 22:5 5:40 28:0 5:45 35:5 5-50 40:5 5°55 AT-5 5:60 51:0 5-65 56:5 27-2 5-70 62-0 R.C.M. 27-2 5:50 40-2 “ 55:2 11-95 6:10 37-0 . |65°6 12:45 6-10 48-7 Dae MOO 12:4 6-10 47-1 . |66-0 12:45 5:80 15-1 Volume of im- mersed Portion (cub. centims.) V. 6-8437 -1494 1348 6-8437 +1503 1463 68437 1534 +3521 73492 6-8437 +1525 +3592 17-3554 6-8437 -1542 +3566 7-3545 7:2397 17-3462 7-3318 70482 71-1279 7-1403 Density of (Densi 102357 1-02387 | 1-026 1-02380| 1-026) 1-02373| 1-0268 1-02369 | 1-026; 1:02370 | 1-0268 1-02753\ 1-086 1-02695 | 1-0263b 102702) 1-02636 102698 | 1-026 Data Relating to the Collection of the Sample. Depth ate (E.) Position (L.) oo. Temperature ( in Fathoms. at time of ollection of Dz d. Sample. Ao Surface] 60:8 | 58:3 : 14-6 C. Jan. 2 | 15:0 ae 5 Surface] 59-6 | 55:0 12:8 C. Jan. 2 | 16.0 5s ae Surface] 57-7] 54-9 12:7 C. Jan, 2117.0 or ; Surface] 58:0 | 55-2 12:9 C. Jan. | 2/18.0 nn fe Surface 57-4 | 56-2 13-4 C. Jan. 2 | 19.0 54 as . | Surface} 57-5 | 57-7 14:3 C, Jan. 2 | 20.0 an Ac .- | Surface| 58-0; 56-9 13-8 C. ___| Colour of Water, Current, and Remarks, Olive green Slaty blue Olive green Olive green Olive green Light slaty blue Olive green More intense green Same as last hour Time and Date. 14.30 2/1/03 15.0 2/1/03 15.30 16.0 5.45 3/1/03 6.24 3/1/03 7.5 3/1/03 13.30 3/1/03 14.15 3/1/03 Data Relating to the Determination of the Density of th Peuapensinre uring Weicht Experiment. Wiel Bees to £ Hydro- oe ANG ih meter pee = (grams) auc of of the the Air. |Sample Ww R. 2195 26; 73-7 | 19-8 | 5-20 7-0 23-2 C. 5-25 12:5 5:30 18-2 5°35 24-7 20:3 | 5:40 | 29-3 20-05 5-30 18-3 74:5 20:0 | ° 5:20 4-0 \23-6 C. 5-25 10-0 5:30 16-0 5:35 21-5 20-4 | 5:40 | 27-3 20-2 5-30 15-8 748 | 18:2 | 5-30 9-7 23-8 C. 5:35 15-5 5:40 21-2 5-45 27-2 18-4 5:50 33-0 18-3 | 5-40 | 213 74-2 17:8 5:30 5-6 23-4 C. 5-35 11:7 5-40 17-0 5-45 23-0 18-0? 5:50 28:8 17-9 5:40 172 60-1 | 160 | 5-35 3-0 15-6 C. 5-40 75 5-45 13-0 5:50 18-2 15:8 5:55 23-7 15:9 5:45 13-1 64:2 | 15:8 | 5-40 9-0 17-9 C. 5-45 13-8 5:50 18-8 5:55 24:8 15:5 5:60 30-0 15:65 | aco) | wags 58-7 15-7 5-40 9:8 14:8 C. 5-45 15-0 5:50 20-2 5:55 25:5 16-1 5:60 31-5 15:9 5:50 20-4 62-1 | 164 | 5-40 | 128 16:7 C. 5:45 19-0 5:50 24-0 5:55 30-0 16-4 5:60 35-0 16-4 | 5-50 | 242 61:5 16:4 5:30 2-5 16:4. 5:35 15 5:40 12:5 5-45 18-0 16-4 5-50 23-7 16-4 5:40 12-8 | 7.0983 Volume of im- mersed Portion 6-8437 +1137 1623 7-1197 | 1-02391| 1-025 6-8437 +1145 -1401 1-02403 | 1:02 6-8437 +1038 -1889 7-1364 6-8437 +1015 +1525 70977 6-8437 0902 +1162 7-0501 6-8437 0887 1712 7-1036 6-8437 0902 +1810 7-1149 6-8437 -0930 +2147 7-1514 6:8437 0930 +1135 7-0502 1-02483 | 1:02 27 27 28 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. sho Depth | Volume i Position (L.) : Temperature Temperature : Density of Sample. in Fathoms. at fim of during Weights of im- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) D. d. Sample. i ing | Portion — : f of Colour of Water,]} Time Hydro- Y of | from T t Current, and "| and TM t’ meter Eyre ee Ob-_ |Reduced) Reduced Sea | Which i Remarks. Date. (grams). : served to to ap |) Lous, "| the |—— sr aa (cup. | att’. [Is°56C| 4, Posi- panple of of of of by poate) tion Seat the the the the | L. | jected, | Ait: | Water. Air. |Sample.} w. R V. 498¢, | 4915-56 | 454 ORs ore als 2 Be Sai aC: 20.5 A Surface] 58:0 | 55-9 Same as last 15.0 |60-8 16-4 5:40 12-0 13:3 C 3/1/03 |16-0 C. 5-45 | 17-5 5:50 23-0 6-8437 5-55 28:5 0921 16-1 5:60 33-5 -2031 16-25 5:50 22-9 7-1389 | 102488) 1:02504| 1-02552 12.0 Surface] 57-2 | 55-1 Slaty blue a.m. |52-4 11-6 5:50 4:7 12:8 C 10/1/03 |21-3 C. 5:55 | 10-0 5-60 15-2 6:8437 5:65 20-9 0658 11-6 5-70 26-2 -1366 11-6 5:60 15-4 77-0461 | 1:02592) 1-02511 | 1-02569 8.0 : Surface] 53-6 | 48-9 Greenish blue a.m. |53°6 11:8 5:50 5:9 9-4 C. 10/1/03 |12-0 C. 5:55 11-2 5-60 16-7 6-8437 5:65 22-1 0669 11-8 5-70 27-7 +1481 11-8 5:60 16-7 7:0587 | 1-:02585 | 1-02507 | 1-02627 12.0 Surface] 53-0} 49-2 Very dark green a.m. {52:9 11-6 5:60 16-2 9-6 C. 10/1/03 |11-6 C. 5:65 21-9 5°10 27:2 6-8437 5:75 32-8 0663 11-8 5:80 38-1 -2413 11:7 5-70 27-2 7-1513 | 1:02588| 1-02509 | 1-02625 12.0 Surface] 49-7 | 48-3 Turquoise a.m. |53:9 11-6 5-60 17-5 9-1. 10/1/03 |12-2 C ; 5:65 | 22-9 5:70 28-3 6-8437 5-75 33-6 0658 “ 11-6 5:80 39-0 +2510 11-6 5:70 28-3 7-1605 | 1:02583) 1:02502| 1-02626 9.0 Fort §} tanley Surface] 59-2 | 53:0 Very rich green a.m, [54:2 11-4 5:50 8:3 11-7 C 10/1/03 |11-3 C. 5:55 13-7 5-60 19-0 6-8437 5-65 24:3 -0649 11-5 5:70 30-0 1694 11-45 5:60 19:1 70780 | 1:02575 | 1-02492| 1-02570 12.0 | 5255S. | 55 OW. Surface] 46:5 | 44-6 Light greenish a.m. |42-5 5-7 5:70 11:5 7-0 C. blue 31/1/03 | 5-8 C. 5:15 17-0 5:80 22-5 6-8437 5:85 28-7 0335 6-1 5:90 34-5 2022 59 5°80 22:8 70794 | 1:02681| 1:02511| 1-02667 20.0 | 5328S. | 5355 W. “| Surface] 45-7} 43-8 a.m. |44-2 6-2 5:60 1:2 6-6 0. 31/1/03 | 6-8 C. 5-65 65 ~ 5-70 12-0 6-8437 5:15 17:8 0357 6-4 5-80 23-0 1073 6-3 5:70 12-1 6-9867 | 102678) 1:02514| 1-02674 8.0 | 5422S. | 52 20 W. Surface] 45-0} 43-4 Greenish blue 10.40 |44-7 6:3 5:60 2-2 6-3 C. 31/1/03 | 7-1 C. 5°65 8-0 5:70 13:5 6-8437 5:75 18-5 0369 6-7 5:80 24:8 -1188 6-5 5:70 13-4 6:9994 | 1-02671)| 1-02509 | 1-02673 OY. SOC. EDIN., VOL. LI, PART I (NO. 4). 14 88 DR W. S. BRUCE,-MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT | | Data Relating to the Collection of the Sample. a Depth Date (E.) Position (L.) : Temperature in Fathoms. at es of s 2 == = pana, Collection of & e D d. Sample. Te BS aro mn | a from aig of | which | T. | t bs 4 | Month. Day. Hour. Lat. Long. am Le = 4|3 Posi-/S2™Ple! of | of =) WE tion a the the Zz |e | i L. leetod: Air. | Water. 1903. | shalt oe oR. Bah 127| 71} Jan. | 28 | 12.0 | 5435S. | 5150 W. Surface] 44:9] 43-1 é 6:2 C. 128 | 72} Jan. | 28 : 16.0 | 5155S. | 51 27 W. Surface| 43:2 | 40-6 4:8C, 129| 73 | Jan. | 28 20.0 | 5512S. | 51 2 W. Surface] 42:1] 40-6 : 4-80. 134 | 74} Jan. |} 30/12.0 | 5628S. | 4752 W. Surface] 38:7 | 35-9 2-2C. 135 | 75} Jan. | 30 | 19.45|57 2S. | 4715 W. Surface] 37-2 | 35-0 170. 136 | 76 | Jan. | 30 | 24.0 | 57 22S. | 4653 W. Surface] 37:0} 35-1 ‘ 17 C. 137 | 77 | Jan. | 31 | 4.0 | 57 42S. | 46 33 W. Surface] 36:5 | 35-0 L7C. | 138 | 78 | Jan. | 31} 8&0 | 57578. | 46 5 W. Surface| 36:7 | 35:3 i 1:8 C. / ; > Bl 139/79) Jan. | 31/12.0 | 5814S. | 4550 Ww. Surface 37-0 | 35-4 1:9 C. | | / Data Relating to the Determination of the Density of th ‘Penuperaitie Velo Densi y net ae Weights Read- miensed om Experiment, oe ane) ll Donioet Wate Colour of Water, | Time Hydro of of ia Current, and ’| ‘and aN, i eater BEG By a Ob- Remarks. Date. | (grams). | ™©te? (enka served 7 i centims.) atm the the Air. |Sample.| w. R. We Se Soaks caCe Bluey green 11.20 |44-9 6-4 5:60 2:5 31/1/03 | 7-2C. 5-65 8-0 5-70 13-5 6:8437 5-715 19-0 +0369 6-6 5:80 24:5 +1197 65 | 570 | 13:5 | 7-0003 | 1-02671 Light greeny. 11.40 |45-2 6-4 5:60 7:3 xl ~ plue 31/1/03 | 7:3 C. 5-65 12-8 5:70 18:3 6:8437 5-75 24-0 0374 6:8 5:80 29-7 +1632 6-6 5-70 18-4 | 7-0443 | 1-02646 0 15.50 [44-4 6-2 5:60 5:0 31/1/03 | 6-9 C. 5-65 10-5 5:70 16°5 6:8437 5:75 22-0 0363 6-6 5:80 27°5 +1446 6-4 | 570 | 16-3 | 7-0246 | 1-02657| 10: Light greeny 14.0 {43-0 4:8 5-60 6-7 blue 1/2/03 | 6-10. 5:65 | 11-7 5:70 17:5 6°8437 5°15 22°8 0284 5:2 5:80 28°5 +1543 5:0 5-70 17:4 7:0264 | 1:02656 14.40 |43-7 5:0 5:60 7:3 1/2/03 | 6-5 C 5:65 | 12-5 5:70 18-2 6-8437 5-75 24:5 0295 5:4 5°80 29:5 +1632 5:2 5:70 18:4 7-0364 | 1:02651) 14.45 |44:0 5:0 5:60 7:0 1/2/03 | 6-7 ©. 5-65 | 13-0 5°70 18-5 6-8437 5:75 24-0 0295 5-4 5:80 29:8 +1641 5:2 5:70 18-5 70373 1-02651| 1:02: re 15.10 [44-6 5:3 5-60 TT : 1/2/03 | 7-0 C. 5:65 | 12-8 , 5:70 18-2 6-8437 5:75 24-7 ‘0315 5:8 5:80 30-0 +1659 5:55 5:70 18:7 70411 | 1:02648 Reeve’s French | 15.40 (44:8 56 | 5-60 8-5 blue 1/2/03 | 7-10 5:65 | 14-7 5:70 20:0 6-8437 5:75 25:0 0329 60 | 5:80 | 309 +1756 58 5-70 19-8 | 7-0522 | 1:02642| 10 Reeve’s French 16.15 /44:0 57 5:60 8-7 blue 1/2/03 | 6-7 0. 5-65 | 14:5 5:70 20-0 6:8437 515 25:5 +0329 5-9 5:80 31-0 +1765 . 58 | 570 | 19-9 | 7-0531 | 1-02642) 14 AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 89 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. nn Depth Volum i Position (L.) : Temperature Temperature pour Density of Sample. in Fathoms. | “3+ time of during Weights of im- (Density of Distilled Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. d. Sample. eG ing | Portion a sae ' f of Colour of Water Time Hydro © ; | of | from T t Current, and ; and 1; t’. meter She ; ete, Ob- |Reduced|Reduced Seq | Which : Remarks. Date. (grams) ; b served to to Lat. | Long. |‘, |_ the == to | abt’. |15°560.| t. Posi- pee of of of of wee tion wae the the the the L. | jected, | Ait: | Water Air, |Sample.| w. R. V. a5t, | 4Sisese.| 4S. Stat or! NN tle Se Os ilies SCOR : Surface] 36-5 | 35-9 Reeve’s French | 16.40 |44:3 6-0 | 5-60 9-9 220. blue 1/2/03 | 6-8 C. 5:65 15-0 ; 5:70 20-5 6-8437 5:75 26-3 0346 6-2 5:80 31-5 -1827 6-1 5:70 20-6 7:0610 | 102637 | 1:02471| 1:02677 Surface] 37-1 | 35-4 Reeve’s French 17.43 |43-8 5:9 5:60 10-0 : 1-9 C. blue 1/2/03 | 6-6 C. 5:65 15:2 5:70 20:8 68437 5:75 26-5 0340 6-1 5°80 315 +1845 6:0 5:70 20-8 7:0622 | 1:02637 | 1:02470 | 1-02679 Surface] 33-1] 35-8 ‘ 18.0 [44-2 5:8 5:60 8-0 2-10. 1/2/03 | 6:8 C. 5-65 14-0 5-70 19-5 6-8437 5°75 25-0 0343 6:3 5°80 30:3 1721 6:05 5-70 19-4 7:0501 | 102643) 1:02477 | 1-02684 Surface] 30:3 | 34-6 9.47 |48-7 4-7 5:60 14 1-40. 2/2/03 | 6-5 C 5:65 13-0 5°70 18-0 6-8437 5-15 23-7 0275 5:0 5:80 29-5 +1632 4-85 5-70 18-4 7:0344 | 1:02652)| 1-02471| 1:02683 Surface} 30:0] 33-1 Reeve’s French | 10.11 (|44-4 5:2 5:50 7-2 0-6 C blue 2/2/03 | 6:9 C. 5:55 12-7 5:60 18-0 6-8437 5-65 23-7 0303 5:5 5°70 29-3 -1614 5:35 | 5:60 | 182 | 7-0354 | 1-02598| 1-02424| 1-02639 Surface] 31:0] 34-0 Reeve’s French | 10.40 /|45-1 5:3 5°50 4:3 1-1. blue 2/2/03 | 7-3 C. 5:55 10-0 5:60 15-0 6-8437 5:65 21-0 -0312 5:7 5-70 26-7 +1366 ’ 5:5 5:60 15-4 70115 | 1:02611| 1-02438 | 1:02651 Surface] 30-7 | 33:3 Reeve’s French | 11.10 /45-2 5:3 5:50 7-3 0:7 C. blue 2/2/03 | 7:3 C. 5:50 12-7 5:60 18-0 6-8437 5:65 23-5 +0309 5-6 5:70 29-0 -1605 5:45 | 5:70 | 181 | 7-0351 | 1-02652| 1-02481| 1-02698 Surface} 30:0 | 32-6 Reeve’s French | 13.11 [44-8 5:8 5-50 8-2 0-3 C. blue 2/2/03 | 7-10. 5-55 13-9 5:60 19-5 6-8437 5:65 25:0 0335 60 | 570 | 303 | -1721 fp a9 5:60 19-4 70493 | 1-02590 102423 1:02640 : Surface] 30:5 | 33-2 a 13.31 |447 | 61 | 550 | 12-0 | 0:7 C. 2/2/03 | 7-10. 5-55 17:6 | 5-60 23-0 6-8437 5:65 28:3 0346 Gl | 5-70 | 33-7 | -2087 / 6-1 5:60 22-9 7:0814 | 1:02573 | 1-02407 | 1-02627 90 Number of Station. 150. 151 152 153 154 155 156 157 158 Number of Sample. 90 91 92 96 97 | DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITIE Data Relating to the Collection of the Sample. Date (E.) Position (L.) Month. Day.) Hour. 1903. | Feb. Feb. Feb. Feb. Feb. Feb. Feb. Feb. bo bo 8.0 | 12.0 16.0 20.0 12.0 16.0 20.0 8.0 Lat. Long. Depth in Fathoms. Dz. d. from Ss which lected. Surface Surface Surface Surface Surface Surface’ Surface Surface Surface Data Relating to the Determination of the Density of the Sar Temperature at time of Collection of Sample. a ne of of the the Air. | Water Sih |) “See 33:8 | 30:8 —0-7 0. 32-4 | 29-6 —1:30C, 32:3 | 33:0 0-6 C. 31:0} 29-0 —170C. 31:1] 31-1 —0-5 C. 29-2 | 32:6 0:3 C. 27-4] 29-0 —170., 27:0] 29-9 —1:20C. 28-7 | 30:2 —1:00C. Colour of Water, Current, and Remarks. Reeve’s French blue Reeve’s French blue Reeve’s French blue Amongst ice Reeve’s French blue Amongst ice Reeve’s French blue Steaming alon edge of pac amongst small bits of ice Reeve’s French blue Amongst ice Reeve’s French blue Amongst ice Reeve’s French blue Amongst ice Reeve’s French blue Time and Date. 13.45 14.0 10.25 3/2/08 10.15 3/2/03 9.50 6/2/03 10.16 6/2/03 10.48 6/2/03 11.0 6/2/03 11.39 6/2/03 centims.) my Volume of im- mersed Portion (cub. 6-8437 0295 -2270 6-8437 0184 2226 7-0847 | 1-02571 6-8437 0235 1694 7-0366 6-8437 0247 +1889 70573 Hempeialule uring Weights Experiment. ae Read- to | te Hydro- a ee tie meter bare (grams) exe of of the the Air. |Sample.} w. R. O18, HOF 44-7 5-1 5:50 15:0 7-10. 5:55 20:2 5-60 25:5 5-65 30:9 5:3 5:70 36-5 a2 5:60 25-6 44-5 3-0 5:50 13:8 6-9 C, 5:55 19-8 5:60 25:2 5:65 30-6 3-5 5:70 36:1 3°25 5:60 25-1 40:9 4-1 5:50 8-2 4:9-C. 5:55 13-6 5-60 19-0 5:65 24:8 4-2 5:10 30-0 ial eco) ez 40:8 4-2 5:60 10-1 4.90 5-65 15-7 5:70 21-4 5:15 26:8 4:5 5:80 32:5 4-35 5:70 21-3 40:7 4-4 5:50 14-2 4-80 5:55 20-0 5:60 25-6 5:65 31:0 4-5 5:70 36:3 4-45 5°60 25-4 41:2 4-7 5:50 11:0 510. 5:55 16-6 5:60 22:0 5:65 27:5 4-9 5:70 33-0 4:8 5:60 22-0 42:0 4:8 5:60 8-7 560. 5:65 14:3 5-70 19-9 5°75 25:1 5:0 5:80 30:8 4-9 5-70 19:8 42-6 4:8 5:60 5:0 5:9 C. 5:65 11:0 5:70 16-2 5:75 21:8 5-1 5:80 27:2 4-95 3:70 16:2 42-8 4-9 5-60 10:0 6:0 C. 5-65 16:0 5:70 21:7 5:75 27-0 5:0 5:80 32:7 4:95 5:70 21-5 6-8437 0252 +2202 70941 6-8437 0272 1951 7-0660 6-8437 0278 1756 70471 6-8437 +0280 +1437 70154 6:8437 0280 +1907 70624 4| 20.0 5] 8.0 Fek 5 | 12.0 Fe 5 | 16.0 Fe’ 5 | 20.0 Fek ; 5 | 24.0 Fe 6] 8.0 Feb} 6] 12.0 Fe 6 | 16.0 Feb} 6 | 20.0 Position (L.) Depth in Fathoms. ee _; Sample d. from which the was col- lected. Surface Surface Surface} Surface Surface Surface Surface Surface Surface! Surface Data Relating to the Collection of the Sample. ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. Sl Data Relating to the Determination of the Density of the Sample. Temperature Temperature Volume Density of Sample. at time of during Weights! p of wae (Density of Distilled Collection of Experiment. | added ead- | mersed Water at 4° C.=1.) Sample. to ee owas Colour of Water, | Time Hydro- le t Current, and "| and AM td meter ees Bye Ob-_ |Reduced|Reduced eis Remarks. Datos || eae (grams) 7 (cab served to to “ae oe centims,)| tt’. [15°56C.) t. of of of of ae the the the the Air. | Water. Air. |Sample.| — w. R. Vv. aS, | 4Si556] «St | OIE wy out. CEO 26:3 | 29-8 Amongst ice 11.56 |43-2 4:8 5:60 9:9 —1-2 6/2/03 | 6-2C. 565 | 15-0 5-70 20-3 6-8437 5-15 25-5 0278 5:0 5:80 30:9 +1800 4-9 5:70 20-3 7-0515 | 1:02643 | 1-02462| 1-02687 27-4 | 29-0 Amongst ice, in 12.23 (48-8 4:8 5:60 5:0 —1:70C. open pool 6/2/03 | 9-3 C. 5:65 10-7 : Reeve’s French 5:70 16-3 6-8437 blue 5:75 22-0 0278 5:0 5:80 27-5 +1446 4-9 5:70 16:3 70161 | 1:02662) 1-02482| 1-02707 27-4 | 29-8 In open pool, sur-f 13.50 (|49-0 6-0 5:60 8-7 | -1:26 rounded by ice} 6/2/03 | 9:4 C. 5-65 14:6 Reeve’s French 5-70 20-2 6-8437 blue 5:15 26-0 0363 6:8 5:80 31:8 +1800 6:4 5-70 20-3 7:0600 | 102637 | 1-02475 | 1-02700 29-0} 30-0 | Inlarge open pool —1-:1C.| Reeve’s French blue 30°3 | 31-2 eeve’s French 15.21 55-2 6:8 5:50 14-0 —0-4C blue 6/2/03 |12:9 C. 5:55 20-0 5-60 25-6 6-8437 5-65 31-0 -0408 7-6 5:70 37-0 “2262 72 5:60 25-5 7-1107 | 1:02557 | 1:02406 | 1-:02625 28-3 | 31:5 —0-3 C0. 32:0} 32-0 Reeve’s French 16.30 [53-7 168) 5:50 13-2 0-0 C. blue ? 6/2/03 |12-1 C. 5-55 18-8 5:60 24:5 6-8437 5:65 30-0 -0462 8-4 5:70 36-0 2173 8-15 5:60 24-5 7:1072 | 1:02559 | 1:02420 | 1:02638 33:0 | 33:6 | Farfromthe pack} 13.50 {51-0 78 5:50 10-0 0-9 C.| Reeve’s French | 1/2/03 |20-6 C. 5:55 15:2 blue 5:60 20:6 6:8437 5:65 26-0 -0451 8-1 5:70 31-7 1836 7-95 5-60 20-7 7:0724 | 1:02578 | 1:02438 | 1-02661 33-0 | 31:9 | Farfromthe pack} 14.7 [52-1 7-6 5-50 11-1 —9-1C.) Reeve’s French | 17/2/03 |11-2 C. 5-55 17:0 blue 5:60 22-5 6-8437 5:65 28-0 ~-0448 8-2 5:70 33- -1987 7-9 5:60 22-4 7:0872 | 1:02569 | 1:02429| 1-02647 31:8 | 30-1 | Farfrom the pack} 14.30 [53-7 7:6 5:50 17:0 —110C.| Reeve’s French | 7/2/03 |12-1 C0. 5:55 22:5 blue ; 5:60 28-0 6-8437 5-65 33-6 -0448 8-2 5:70 39-5 -2492 7-9 5:60 28-1 7-1377 | 1:02542 | 1-02402)| 1:02623 92 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAV. Data Relating to the Collection of the Sample. Depth | | Date (E. Position (L. : _ | Temperature ate (E.) MI (L.) in Fathoms. ALO GR oa ees a) Collection of s oe D. d. Sample. 3/8 _'____| Colour of Water, n|n ¢ | from T t Current, and «14 Sea | Which | *- : Remarks. = | = |Month.|Day.) Hour.) Lat. Long. aA the 3 2 Posi eparle of of E 5 tion bie the the ZA |a L. iectad: Air. | Water. ie } | / | 1903. | ae? os OB Lescalt 169 108, Feb. 6 | 24.0 Surface 31:3 | 30:6 —49:8C 170 | 109| Feb. | 7] 8.0 | Se .. | Surface] 30:0} 29-8 |Steaming amongs —1:2C.| loose pieces of ice Reeve’s French blue 171 |110| Feb. 7 | 12.0 ne .. | Surface} 30-2} 30-4 |Steaming amongst —0:9C.| loose pieces of ice Reeve’s French blue 172 | 111) Feb. 7 | 16.0 Su .. | Surface 31-4 | 29-9 |Steaming amongst) —1:2C.| loose pieces of ice Reeve’s French blue 173 | 112] Feb. 7 | 20.0 be .. | Surface] 32:3] 29-9 Ac —1:2C 174 |113| Feb. 7 | 24.0 sis es .. | Surface} 32:0] 31-7 Hc —0:2C 175 | 114] Feb. 8 4.0 ac .. | Surface] 31:3] 31:1 —0-5C. 176 | 115} Feb. 8 8.0 On .. | Surface! 28:6] 30:6 Reeve’s French —0-8 C. blue 177 116) Feb. 8 | 12.0 Surface} 30-2 | 31-0 Reeve’s French —0-6 C. blue Data Relating to the Determination of the Density of tie Time and Date. 14.45 7/2/03 15.8 7/2/03 15.25 7/2/03 14.0 8/2/03 14.14 8/2/03 14.30 8/2/03 14.50 15.10 15,25 Temperature during Experiment. 2B oaG: 54:3 7:8 12:4 C. 8-5 8:15 54:6 TA 12:6 C. 7-9 7-65 54-7 6-5 12-6C.) (al 6:8 50:2 9-6 12:9 C. 9-9 9:75 57-2 9:8 14:0 C. 10-2 10-0 57-4 9-5 14:1 C. 9-8 9:65 56-7 9-6 13-7 C. 10-1 9-85 56:5 9-4 13-6 C, 9:9 9:65 56:6 8:3. 13:7 C. 8-8 8-55 5:50 5:05 5:60 5:65 5:70 5:60 5:50 5:55 5-60 5:65 5:70 5:60 5:50 5:55 5:60 5:65 5:70 5:60 5:50 5:55 5:60 5:65 5:70 5:60 5:50 5-55 5:60 5:65 5-70 5:60 5:50 5:55 5:60 5:65 5:70 5-60 5:50 5:00 5:60 5:65 5-70 5:60 5:50. . 5:55 5:60 5:65 5-70 5:60 25-0 5:9 11-0 16-5 22:5 27-7 16-7 15-2 20-5 26-1 32:0 37-7 26-3 18-5 24-0 29-7 35-4 41-2 29-8 17-6 23-0 28-6 34-1 39-9 28-6 18-2 23-8 29-1 34-9 40-3 29-3 20:0 25:7 31-2 37:0 42-8 31-3 15-7 21-0 26-7 32-1 38-0 26-7 Volume of im- mersed Portion 6-8437 0462 +2386 71285 6-8437 0434 2217 7-1088 6-8437 0386 +1481 7-0304 6-8437 0553 +2333 * 1.02547) 1 . Data Relating to the Collection of the Sample. ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 93 Data Relating to the Determination of the Density of the Sample. Depth Volume i : Temperature Temperature ; Density of Sample. in Fathoms. | “34 time of during of im- (Density of Distilled Collection of Experiment. mersed Water at 4° C.=1.) | D. d. Sample. ; Portion i Colour of Water,} Time H oF | ] of | from t Current, and and ie te =e nee Ob-_ |Reduced|Reduced | Sea which ‘ Remarks. Date. served to to Day.) Hour at |q the = Cee || abt’. lisse. t. Posi pap of of of of centims. tion} Gop | the the the the , L. Tented. Air. | Water. Air. |Sample.| w. R. Ne 49¢. |4Sis-s6.{ 4S, 1908 Cite nt oR oR oR, AC. ; Feb; 8 | 16.0 26 we .. | Surface] 31:5] 30-4 Reeve’s French | 14.16 [47-7 7-4 5:50 16-9 : —0-9C.| blue (lighter) 9/2/03 | 8-7. 5:55 22:3 5:60 27-9 6-8437 5°65 33°5 -0425 7-6 5°70 38:9 -2474 7-5 5:60 27-9 7-1336 | 1:02544 | 1-:02399| 1-02619 Feb} 8 | 20.0 a4 ai .. | Surface] 31-0] 30-0 Reeve’s French 14.35 |48-4 75 5:50 10-7 Sri. blue 9/2/03 | 9-1. 5-55 15-7 5-60 21:0 6-8437 5:65 26:3 04354 4 78 5-70 32:0 1871 7-65 5:60 21-1 7:0742 | 1:02577 | 1:02433 | 1-02655 Febs| 8 | 24.0 a so -- | Surface] 31-0} 31-0 Sailing all day | 14.50 |48-9 T5 5-50 5-9 —0-6C.| along edge of | 9/2/03 | 9-4C. 5:55 11:0 | pack 5:60 16-2 6-8437 | Reeve’s French 5:65 21-9 -0434 | : blue 7-8 5:70 27-2 +1455 7-65 5:60 16:4 7:0326 | 1:02599| 1-02454) 1-:02675 Feba} 9 4.0 ae a .. | Surface] 30-6] 29-8 ae 15.6 49-7 7-4 5:50 a7 : —1:2C. 9-8 C. 5:50 10:5 5:60 16:1 6-8437 5:65 21:4 -0431 . T8 | 5-70 | 26-7 | -1428 7-6 5:60 16-1 7:0296 | 1-02601| 1:02456 | 1-02680 Febg} 9 | 8.0 a5 Surface} 30:9 | 30-0 |Steaming amongst] 15.25 |49-6 T1 5:50 5:0 —11C. bay ice 9-8 C. 5:55 10:0 Reeve’s French 5:60 15:3 6-8437 blue 5:65 21:0 0411 7-4 5-70 26-2 1375 : 7-25 5:60 15-5 7-0223 | 1:02605 | 1:02455 | 1-02678 Febs| 9 | 12.0 5a 50 .. | Surface} 30-6] 30:0 | Amongstlightice} 15.45 |49-2 5:8 5:50 3:3 —1:1(C.| Reeve’s French 9-6C 5:55 14-0 blue 5:60 8-7 6-8437 5:65 19-5 0343 6-3 5-70 25-0 1251 6-05 5:60 14-1 7:0031 | 1:02616)| 1:02450 | 1:02673 Feb. |.9 | 16.0 On Surface} 30:5 | 29-9 Reeve’s French | 13.52 [48-0 7-1 5:50 5:6 —1:2C, blue 10/2/03 | 8-9 C. 5:55 11-0 5:60 16-8 6-8437 5:65 22-1 -0408 7-3 5:70 27-7 1472 7:2 5:60 16:6 7:0317 | 1:02600| 1:02449 | 1-02673 Feb. | 9 | 20.0 Surface] 29:2} 29-7 Reeve’s French 14.10 |47-9 6-9 5:50 3:0 —1:3C. blue 10/2/03 | 8-8 C 5°D0 8-0 5:60 13-6 6-8437 5:65 19-0 0400 7-2 5:70 24-4 1206 7-05 5:60 13-6 7:0043 | 1:02615 | 1-02462| 1:02687 Feb.) | 9 | 24.0 5 ze Surface] 28-2 | 29-4 we 14.32 47-5 6-5 5:50 0-5 —1-46. 10/2/03 | 860. 5-55 5-4 ; 5-60 1-1 6-8437 5:65 16-6 0374 6-7 5-70 22-0 0984 6-6 5:60 11-1 6-9795 | 1:02629| 1:02469 | 1-02694 | 94 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA} ITIR Data Relating to the Collection of the Sample. ee eS Data Relating to the Determination of the Density of the § Date (E.) Position (L.) ud Barth Temperature Temperature a bi ae trae Or during Weights = OF D q Collection of Experiment. | added Ss at . : Sample. to 3 5 aa a _| Colour of Water,} Time Hydro- m | 7 of ae Tp t Current, and and ue 6% meter 5 = Month. Day.) Hour.| Lat. Long sea ther | Homeuks. eS (ota) & 2/3 Posi-|Sample| of | of of | of Bs tion] co. | the | _ the | the | the 2/24 L. lected Air. | Water. Air. |Sample. w. | 1903. % 4 eb IAS orn: Sas 187 | 126; Feb. | 10 4.0 Surface] 28-3] 29-2 14.50 [47-3 6:6 5:55 —1-6C. 8-5 C. 5:60 5:65 5:70 6-6 5:75 6-6 5:65 188 | 127} Feb. | 10] 8.0 Surface) 28-6 | 29-4 Amongst loose | 15.10 |46-7 6-2 5:55 —14C.| thinice 8:2. 5:60 Reeve’s French 5-65 blue 5:70 - 6-5 5:75 6-35 5:65 189 | 128} Feb. | 10 | 12.0 Surface] 31-8 | 30-2 Amongst loose 15.30 |48-0 4:8 5:60 —1:0C.| thinice 8-9 C. 5:65 Reeve’s French 5:70 blue 5:75 5:2 5:80 ; 50 5-70 190 | 129; Feb. | 10 | 16.0 Surface] 33:2 | 29-7 Ice touching ship} 13.50 |49-5 7:8 5:50 —1-3C.| Reeve’s French | 12/2/03 | 9-7 C. 5-55 blue 5-60 5:65 8-1 5:70 7:95 5:60 191 | 130} Feb. | 10 | 20.0 Surface] 28-3 | 29-2 Tce closely 14.10 |50-1 7:8 5-50 —16C.| packed, touch-} 12/2/03 |10-1 C. 5-55 ing ship 5:60 Reeve’s French 5:65 blue 8:3 5:70 ; 8-05 5:60 193 | 131} Feb. | 11 | 12.0 Surface] 30-2} 30-0 Ice near 14.35 [51:5 (G2) 5:50 —1:1C.| Reeve’s French | 12/2/03 |10-8 C. 5:55 blue ; 5:60 5:65 8-3 5-70 8-1 5:60 194 | 132} Feb. | 11 | 16.0 Surface} 30-8 | 29-8 Ice near 14.50 [52-5 7-9 5:50 —1:2C.| Reeve’s French | 12/2/03 |11-4 C. 5:55 blue 5°60 5°65 8-1 5-70 8-0 5:60 195 | 133) Feb. | 11 | 24.0 | Surface] 29-7 | 29-8 15.7 {50-2 7-6 5:50 —120C, 12/2/03 |10-1 C. 5:55 . 5:60 . 5:65 | TT 5:70 ) 765 | 5-60 196 | 134) Feb. | 12 5.0 Surface) 29-8 | 29-2 15.30 /48-7 7-2 5:50 ) —1:6C. 9:3. C. 5:55 5:60 5:65 74 5:70 See) B86 Ne NMR e Mae SS [SEE WORDS W] CASWH Nee NAA Volume of im- mersed Portion (cub. jcentims.) Via 6-8437 0374 +1437 6-8437 -0360 +1508 6-8437 0283 +1925 6-8437 0451 1685 6-8437 0456 +1561 6:8437 -0459 +1375 68437 0454 1357 68437 0434 +1215 6-8437 0414 +1261 70248 70305 71-0645 7-0573 71-0454 70271 70248 71-0086. 7-0102 102612 f 1-02 = iND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 95 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. 7 aye Depth | : | Volume ; te (E. Position (L. ¢ Temperature Temperature : Density of Sample. (2) ; ori in Fathoms. | “4 dane of . during Weights of im- (Density of Distilled = a7 ns == a neem OOllectron Of Experiment. | added Read- | mersed Water at 4° C.=1.) | 2 D. d. Sample. : ce ing Foxton a : SS f to} = Colour of Water,] Time ! Hydro- e | 1 of | from T t Current, and "| and | 1’. t’. meter pay ee Ob-_ |Reduced)Reduced Sea | Which Remarks. Date. (grams). Bae b served to to Month Day.) Hour Lat Long. a the = = alias ) att’. |15°-56C. t. Posi- pemyle of of of of é é tion ee the the the the L. | jected, | Ait: | Water. Air. |Sample.| w. R. Vv. av, |4Sis-s6,| «Se 19 Set cae” eau oor; PAB S1Ck Fel} 12 | 8.0 oh s6 .. | Surface) 30-4} 29-3 Reeve’s French | 15.45 |47-8 6:8 5:50 3-7 —1:5C. blue ; 8:8 C. 5:55 9-0 5:60 14-1 6-8437 / 5:65 19-2 0388 6-9 5-70 24-7 “1251 6-85 5:60 14-1 7:0076 | 1:02613 | 1-02456 | 1-02682 Feb.| 12 | 12.0 | 5949S. | 3132 W.|, .. | Surface| 31-4| 29-9 Reeve’s French | 16.15 /|46-6 5:5 5:60 10-3 s —1:2C. blue 8-1C 5-65 16-0 5:70 21:7 6-8437 5:15 27-1 0326 6-0 5-80 33-0 1916 5:75 5:70 21-6 7-0679 | 1:02634| 1-02464| 1:02688 Febs} 12 | 16.0 36 5 .. | Surface] 29-7 | 29-4 No ice touching 13.40 |43-5 6-6 5:50 6-5 —1-4C.| ship, though | 14/2/03 | 6-4C. 5:55 12-0 , near 5-60 17-2 6-8437 Reeve’s French 5:65 22-7 0380 blue 6-8 5:70 28-0 1534 6-7 5-60 17-3 7:0351 | 1:02598 | 1:02440| 102664 Febgj 12 | 20.0 56 ae .. | Surface] 30-1] 29-8 Ice touching shipj 14.4 |45-0 7-2 5:50 9-9 —1-2C.| Reeve’s French } 14/2/03 | 7-2 C. 5:55 15-0 blue 5:60 20-8 6-8437 5-65 26-0 -0414 T4 5-70 31-3 -1827 7-3 5:60 20-6 7-0678 | 1-02580 | 1-02430| 1-:02654 Febg| 13 | 12.0 né ae .. | Surface} 32-8} 30:0 | Some pieces of } 14.25 |45-3 7-0 5:50 9-0 —11C.| iceafewyards} 14/2/03 | 7-4 C. 5:55 14-5 from ship 5:60 19-8 6-8437 Reeve’s French 5-65 25-4 0403 blue 7-2 5-70 30:8 1765 Ct 5:60 HOE) 7:0605 | 1-02584| 1:02432| 1-02654} Feb, }13 | 16.0 se ae .. | Surface] 32-7) 30-1 | Ice about 100] 14.45 |46:0 7:3 5:50 11:5 —1-1C.| _ yards away 14/2/03 | 7-8 C. 5:55 17-0 Reeve’s French 5:60 22-2 6-8437 blue 5:65 28-0 0420 7-5 5:70 33-0 +1978 7-4 5-60 22-3 7-0835 | 1-02572| 1:02425| 1:02647 Feb, |13 | 20.0 ob “eRe, .. | Surface] 32-0} 30-2 | Amongstice, but} 15.15 |47-0 7-6 5-50 9-3 —1-0C.| _ none touching } 14/2/03 | 8-3 C. 5:55 14-0 Reeve’s French 5:60 19-7 6-8437 blue 5:65 24-5 0437 i 7-8 5:70 30-2 “1729 er 5:60 19-5 70603 | 1:02584| 1-02440| 1:02663} Feb, |13 | 24.0 oe oe -. | Surface] 32-2 | 30:0 nD 15.40 |46-7 TA 5:50 6:5 = 700. 14/2/03 | 8-20 5:55 | 15 5-60 17-0 6-8437 5:65 22-0 0425 7-6 5-70 27:5 -1499 7:5 5:60 16-9 7:0361 | 1:02598 | 1:02451| 102674 Febs|14 | 4.0 ate aA -. | Surface] 32-0} 30:2 In pack ice 16.0 46-8 7-2 5-50 6-0 —1-:0 C. 8-20 5:09 11:0 5:60 16:0 6-8437 * 5:65 22-0 0414 14 5-70 27-2 +1455 7:3 5:60 16-4 7-0306 | 1-02601| 1:02451| 1:02674 96 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITIES Data Relating to the Collection of the Sample. Depth Date (E.) Position (L.) : Temperature in Fathoms. at ime of a Collection o. § a D qd. Sample. 3s|& c Colour of Water, nm | of ae T. t Current, and © | 2 |Month.|Day.|Hour.| Lat. Sea | ‘the Hemerks, oles at | Sample a : Posi x of of : 5 tion Bi the the 2/14 L. techedt Air. | Water. 1903. abi ou Ar 206 | 144) Feb. | 14 8.0 Surface] 32-0 | 30:3 Open water —0:9C.| Reeve’s French blue 207 |145| Feb. | 14 | 12.0 Surface| 32-0] 30:3 | Very small —0:7C jeces of ice oating about, no pack to be seen Reeve’s French blue 208 | 146) Feb. | 14 | 16.0 Surface] 31-9 | 30:5 No ice anywhere —0:8C near 209 | 147) Feb. | 14 | 20.0 Surface] 31:5 | 30-2 No ice anywhere —1-0C. near 210 | 148) Feb. | 14 | 24.0 5 Surface] 31:0} 29-1 Amongst ice —1:6C.| touching ship 211 |149| Feb. | 15 4.0 Surface] 30-4} 29:8 Near pack ice —1:2C.| Reeve’s French i blue 212 |150) Feb. | 15 8.0 Surface] 30:0 | 30:0 Amongst loose —11C pack ice Reeve’s French blue 213 |151| Feb. | 15 | 12.0 Surface] 30-4} 31:0 Ice about 100 —0-6C yards off Reeve’s French blue Data Relating to the Determination of the Density of th 14.0 15/2/03 14,22 15/2/03 15.0 15/2/03 15.20 15/2/03 15.40 16.0 16.30 Temperature during Experiment. ue tite of of the the Air. |Sample. SoH malic: 52:8 8-3 11-6 C. 87 85 54:0 8-6 12-2 C. 9-0 8-8 53:2 8-6 11-8 C. Ont 8-85 52-1 8-8 11:2 C. 9-0 8-9 50:8 8:5 10-4 C. 8:8 8:65 51:0 TT 10:6 C. 8-0 7-85 50:5 7-6 10-3 C. (Gs) 7-75 49-0 6:8 9-4C, 6-7 6-75 Ww. 12:8 18-5 24-0 29-9 35:3 ys 10:0 15-7 20:8 26-0 315 20-8 centims.) 7:1056 70781 70627 70159 6-9967 6:9991 Volume Densi of me (Densi merse a Portion Water ae Hyd ydro-| op. motes served (cub. Vv. 454) 7 6-8437 0482 +2137 6-8437 0499 +1845 1-02581| 1-0 6-8437 0505 +1685 6-8437 -0490 1774 i "7.0701 | 1-02579| 105 6:8437 3 0445 1277 102609 6-8437 0439 +1091 6-8437 0383 ‘1171 AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN, 97 Data Relating to the Collection of the Sample, Data Relating to the Determination of the Density of the Sample. eh Depth | Vol te (E.) Position (L.) in Fathoms. | Le™perature Temperature olume | Density of Sample. "| at time of durin : of im- (Density of Distilled = : ing Weights | My Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) D. de Sample. ee ine Bowron ° ees Colour of Water, | Time Hydro- ie 2 of from T Current, and and ne te mater Hy: dro- Hydro- Ob-_ |Reduced/Reduced “| Seq | Which t. Remarks. Date (grams) gaieher |) meter) carved to to MoniiijDay.|Hour.| Lat. | Long. |“? | the |——|——— lar ae (cub. | att’. |IB°56C.| + Posi- peel of of | of of Coniuee) tion eal the the the the L. Atetad. Air. | Water. | Air, |Sample. Ww. RE Vv. a9v, | 481-56.) 45¢, 196 2 Kot ‘ ° 4‘ Chas ORY Ong 5h Cay Feb} 15 | 16.0 aa ie .. | Surface] 30:3] 31-0 Ice about two 16.20 |47-3 7-9 5:55 4:0 —0:6C.| — miles off 16/2/03 | 85 C. 5:60 9-2 Reeve’s French 5:65 14:8 blue 5:70 22:0 5:75 26:0 5:80 31:5 5:85 37:0 6:8437 5:90 43-0 -0462 8-4 5:95 48-0 +2324 8-15 5:75 26-2 7-1223 | 1:02630| 1-02491| 1-02714 | Feb,| 15 | 20.0 | 6158S. |26 3W.| .. | Surface] 30-2] 31-0 Ice about three 5.00 |49-0 8:3 5:50 2-0 | —0-6C. miles off 16/2/03 | 9-4 C. 5:55 6:8 5:60 12-0 5-65 18:0 5:70 23-0 5:75 28:2 5:80 33°5 6-8437 5:85 39-0 +0482 ie 5 8:7 5:90 45:0 -2049 85 5:70 23-1 7-0968 | 1:02618)| 1-02485| 1-02707 Febg| 15 | 24.0 ag 30 .. | Surface] 30:2] 31-1 Only afew small} 6.15 |49-4 7:8 5:50 1:0 —0-5C.| distant bergs, | 16/2/03 | 9-7 C. 5:55 6-0 no pack ice 5:60 11:5 5:65 16-8 5:70 23-0 5°15 28-0 5:80 33-5 6-8437 5:85 39-2 -0456 oe 5 8:3 5:90 45:0 *2013 “ 8-05 5:70 22-7 7:0906 | 1-02621| 1:02482| 102704 Fem) 16); 40}; .. sé .. | Surface} 29-3) 30-9 Only a few dis- 5.00 |49-1 6-7 5:55 3:5 —0-6C.| tant bergs, no | 17/2/03 | 9-5 C. 5:60 8-0 pack 5:65 14:0 5-70 20:2 5-75 26-0 5:80 31:0 5:85 36:5 6:8437 5:90 42-0 0405 47-0 - 7-6 5°95 48-5 +2262 8-3 C, SSS 7-15 O75 25°59 71104 | 1-02637 | 1:02485| 1202707 Again| tested at 6.25 on | 17/2/03 |47-0 76 5:55 5:0 8-3 C. 5:60 10-5 5:65 16-0 5:70 21:0 ? 5-15 27-0 5:80 33:0 5:85 38:0 6-8437 5:90 43-7 +0442 478 8-0 5:95 49-0 +2395 8-8C SS_ SS 78 575 27-0 7:1274 | 1:02628 | 1:02485| 1202707 30-4} 31:0 Only a few dis- 7.35 |38-6 3-8 5:60 5:5 —0-6C,| tant bergs, no | 18/2/03 | 3-7 C. 5-65 11-0 pack 5-70 16-2 Reeve’s French 5:75 22-0 blue 5:80 27-0 5:85 33-0 5:90 38-2 6-8437 5:95 44-0 +0244 43-0 4:8 6:00 49-5 +2430 6-10. =| 4-3 5°80 27-4 71111 | 1:02663\ 1:02476| 1-02698 98 Number of Station. 220 221 222 223 224 225 226 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAY. Data Relating to the Collection of the Sample. oa Depth Date (E. Position (L.) : Temperature (E.) in Fathoms. at time of a i lmenanee Collection of — D. | d. ! Sampie. 5 aqoenl | ae ie ra ae which | T. | © t. me Month. |Day.| Hour. Lat. Long. at they |———— 2 } Posi secu of of = tion eel the the 7 7. v2) L. Tecreds Air. | Water. 1903. ° é ° ° ne ° F. 157| Feb. | 16 | 12.0 on NE Surface| 30-7 | 30-9 —0-6C. 158| Feb. | 16 | 16.0 Ae it Surface] 30:9] 30-5 —0-8 © 159| Feb, | 16 | 20.0 Surface] 31-0 | 30-4 —0:9 C 160| Feb. | 16 | 240 Surface] 31:0 | 31:0 —0:6 C. 161; Feb. | 17 4.0 1 Surface} 31:4} 31-1 —0:5C. 162} Feb. | 17 8.0 oe rie Surface| 31:0 | 31-4 —0:3 C6 163} Feb. | 17 | 12.0 | 6418S. | 23 9 W. Surface} 31:5 oy “ ; —0:20C. Colour of Water, Current, and Remarks. Only a few small bergs in sight, no pack Reeve’s French blue Very considet- able number of bergs a- round, and some broken off berg ice, but none very near - Iee about 100 yards off No ice at 5.15 Dull greenish blue 4 bergs on hori- zon Slaty blue Berg about } mile off Slaty blue Time and Date. 9.40 18/2/03 10.10 18/2/03 14.50 18/2/03 15.20 18/2/03 15.45 18/2/03 16.20 18/2/03 11.15 19/2/03 Temperature during Weights Experiment. | added to Hydzo- ones t’. meter (grams) of of the the Air. |Sample. Ww SEs 2G5 52-0 6:3 5:55 11:1 C. 5:60 5-65 5:70 5-75 5-80 5:85 5:90 50-0 7:0 5-95 10:0 C. 6:65 5:75 49-0 6:0 5:60 ~ 9-4 C. 5:65 5:70 5:75 5:80 5:85 5:90 5:95 50:6 7-0 6:00 10:3 C, 6-5 5:80 43-8 5:8 5:55 6-6 C. . 5:60 5:65 5:70 5:75 5-80 5:85 ' 5:90 6-0 5-95 5:9 O75 43-8 5:8 5-60 6-6 C. 5:65 5-70 5°75 6:0 5-80 5:9 5-70 44-2 5:5 5:60 6-8 C. 5:65 5-70 5:75 5:80 5-85 6-0 5:90 O75 5:75 44-2 OT 5:60 6:8 C. 5°65 5-70 5:75 5:80 5:85 6-1 5-90 59 O75 47-7 4-9 5:70 87C 5°75 5-80 5:85 5-1 5-90 5:0 5:80 | Sewewwve Fe All eee eas CU CRE SANSsd S&S! KROMnUSONSAS Whe CO ae Data Relating to the Determination of the Density of the Volume of im- mersed Portion (cub. centims.) We 6-8437 0377 +2297 7-1111 6-8437 0369 +2953 7-1759 6-8437 0335 *2341 6-8437 0335 +1685 7:0457 6-8437 0326 +1880 7-0643 6-8437 0335 -1694 7-0466 6-8437 0283 1898 7-0618 1-02637 | 1:0 f I 1-02662: 1-02672| 1-025 a 102690) 10: ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 99 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. oe Depth etn | Volume ; ie (E.) Position (L.) : | Temperature Temperature | A Density of Sample. in Fathoms. ! atitime of during Weights of im- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) D. d. Sampie. to ing | Portion 7 | Colour of Water,} Time | } "| Hydro- | ae la at i of from T t Current, and and Ne t’. | meter ut 2 y aa Ob- |Reduced|/Reduced Sea | Which 5 : Remarks. Date. (grams).| Meret | meter | served to to [onthPay.| Hour.| Lat ieee ra | thes a lee 5 ree bite W15°56.Cs| i be Posi- Bamiple of of of of — cutee.) : : tion] gop | the the the the ibe | lected. Air. | Water. Air, a Ww. Re We 4 NS) t’. 4 Ss 15°-56, 4 Ss te 1903 Omg cae SAae | ae O13) ois Febg}17 | 16.0 ac as .. | Surface] 31:0 | 32:0 No ice in sight 11.50 |44:0 5:3 5-70 12-0 0:0 C.| Light slaty blue | 19/2/03 | 6-7 C. 5°75 17-5 | 5:80 23-0 6-8437 5:85 28-5 0309 5-45 5°80 22-9 70777 | 1-02682| 1-02507| 1:02728 Febg|17 | 20.0 a is .. | Surface} 29-5 | 31-7 | One ortwobergs} 14.10 /|47-7 5:6 5:70 12:8 —0-20C.) on horizon 19/2/03 | 8-7 C. 5-15 | 18-0 5°80 23-2 6:8437 5:85 28-8 0320 57 5:90 34:5 2084 5:65 5:80 235 7-0841 | 1-02678| 1:02506| 1:02728 Feb, }17 | 24.0 a0 ee .. | Surface}/ 29-6 | 31-2 |Distant bergs only} 14.25 |43-2 5-7 5-70 14-0 —0-40. 19/2/03 | 6-20 5:75 20-0 5:80 25-8 | 6-8437 5:85 31:5 0329 a9 5:90 37-0 2279 5-8 | 5:80 | 25-7 | 7-1045 | 1-02667) 1-02496) 1-02718 Feb.gus | 4.0] .. A .. | Surface] 29-2] 31-1 No ice in sight 14.50 |44-0 5:7 5-70 15-0 —0-5 C. 19/2/03 | 6-7 C 5-15 | 20-0 5:80 | 25-0 | 68437 5:85 30:0 0329 5:9 5:90 35:8 *2235 5-8 5-80 20-2 7-1001 | 1:02669 | 1-02498 | 1-02721 8.0 an Sc -. | Surface} 30:0} 31-0 No ice in sight 15.0 |44:5 57 5:60 4:5 —0-6C.| Slaty blue 19/2/03 | 6-9 C 5:65 10-0 5:70 15-5 6-8437 5:75 20:5 0329 59 | 5:80 | 260 | -1357 58 | 570 | 15-3 | 7-0123 | 1-02664| 1.02493) 1-02716 12.0 an us .. | Surface} 30:0} 31-0 No ice in sight 13.30 [44-7 5:6 5:70 13-0 —0-6C.| Green 19/2/03 | 7-1 C. 5:75 18-0 5:80 24-0 5:85 30:0 5:90 30°5 5:95 40-0 6-00 45-0 68437 6-05 50-2 0332 6-1 6-10 55:5 -3069 5:85 5:90 34:6 7-1838 | 1:02677 | 1-02506| 1-02730 #8 | 16.0 | 66 5S. |23 9W.| .. | Surface] 30-7] 31-0 No ice in sight 10.20 {52-9 7-6 5-60 8-0 —0-6 0. 20/2/03 |11-6 C. 5:65 | 13-5 5:70 19-5 5:75 25:0 8-0 5:80 30:5 5:85 36°5 5:90 42-2 6-84.37 5:95 47-0 0456 8-5 6-00 53:0 “2714 ; | | 8:05 5:80 30-6 7-1607 | 1:02636 | 1:02496 | 1:02719 100 Number of Station. 234 236 237 238 239 240 | 241 Number of Sample. 171 172 173 174 175 176 117 | 178 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITH Data Relating to the Collection of the Sample, se Depth Date (E.) Position (L.) ca Hathoeaes D. d. from Month. |Day.| Hour.| Lat. Long. at the Posi-| Sample tion’ | mes iy col- * | lected. 1903. ° , ° Ld Feb. | 18 | 20.0 a ai Surface Feb. | 18 | 24.0 oe AG Surface Feb. _| 19 4.0 AG 50 Surface Feb. | 19 8.0 ae ae Surface! Feb. | 19 | 12.0 | 6833S, | 2431 W.| . Surface Feb. | 19 | 16.0 ay Pic Surface Feb. | 19 | 20.0 Surface Feb. | 19 | 24.0 =a ee Surface } eee Peuipetaiite at time o urin, ight Collection of Bxperiacuty Ween Read- Sample. to a Colour of Water,] Time Hydro- T t Current, and *| ‘and 1. te meter Bye : Remarks. Date. (grams) : of of of of i the the the the Air. | Water. Air, |Sample.| w. R. ine OTE CARs “ACs 31:0] 30-9 No ice in sight 11.0 {54-1 8-2 5:60 9-8 —0-6C.| Green 20/2/03 |12-3 C. 5:65 15:0 5:70 20:8 5:75 26-5 8-6 5:80 32-0 5:85 37:8 5:90 43:3 5:95 48-0 9:0 6-00 53:5 8-6 5:80 31:8 30-4 | 30-7 ae 11.30 |55-0 8-4 5-70 25-2 —0-7 C, 20/2/03 |12-8 C. 5:75 30:5 5:80 36-5 5°85 42-0 8-9 5-90 47-5 8-65 5-80 86-3 30-0] 30-2 | No ice insight .. ([47-8-|- 48 5-60 11-0 —1:0C 8-8C . 5:65 16-2 5-70 21:8 5°75 27-5 5-6 5:80 32-8 5-2 5:70 21-9 20:0] 30:0 No ice in sight : 50:2 3:5 5-60 9-5 —1-1C.) Light blue 10-1 5:65 15-0 5-70 20-0 5-75 25:5 3:7 5:80 30:8 3:6 5:70 20-2 29-5 | 29-7 Several small] 12.0 55-9 9-0 5:70 30:0 —1:3C.| broken pieces | 20/2/03 |13-3 C. 5:15 35°8 of ice floating 5:80 41-0 about 5:85 46-5 Reeve’s French 9-5 5:90 52: blue —— —-=-—— 9:25 5:80 41-1 29-8 | 29-6 No ice in sight 13.55 |55°3 9-4 5-60 20-0 —1-3C.| Reeve’s French | 20/2/03 |12-9 C. 5:65 26-0 blue = 5:70 31:2 5:75 36-8 9-6 5:80 42-2 9-5 5:70 31:2 27-1] 29-1 Pieces of ice} 14.10 [55-3 10-0 5-70 33:5 —160C.| floating about | 20/2/03 |12.9 C 5:15 38-9 Reeve’s French 5-80 44-0 blue 5-85 49-5 103 5:90 54:5 10-15 5:80 44-1 23-1 | 29-0 Amongst 14.30 |548 10:2 5-50 10:5 —1:7C.| brackish ice 20/2/03 |12:7 C. 5:55 158 5:60 21-2 5:65 26-5 10:6 5:70 31:8 5°75 37:8 5:80 42:8 5:85 48-0 10:8 5-90 53:5 10-5 5-70 32-0 Data Relating to the Determination of the Density of the Samp vou Density o} orim- (Densi ai mersed ‘ag Portion Waters at 7 ydro-|} op. meter | served (cub. at t” centims.) v. Sy, 6:8437 +0488 +2820 | SSS | | 7:1745 | 1:02629| 6-8437 -0490 +3219 7-2146 | 1-02607| ¥ 6-8437 0295 1942 7-0674 | 1:02634 10 6-8437 . +0204 +1792 70438 | 1-02647| In 6-8437 0524 +3645 7-2606 | 1-02581| 1-02 6:8437 0539 +2767 7-1743 | 1-02575| 1 58437 0576 | +3911 A 7:2924 | 1:02564| 1 \ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 101 Position (L.) Depth in Fathoms. 190 . . oD? ° fe ; 20 | 4.0 os ae : Febs|20 | 8.0 ee e 12.0 | 69 39S. | 22 58 W. 16.0 ee es 21 50 W. d. Surface Surface Surface Surface Surface Surface Surface 22-1 Data Relating to the Collection of tne Sample. | Temperature | attimeof | Collection of Sample. Colour of Water, Current, and Remarks. of the Water. Fay SE. 28-9 Amongst brash —1:7C ice Reeve’s French blue 23-0} 29-0 Amongst brash —1:7C.| _ ice Slaty blue 24:1] 29:0 | Allday steaming —17C.| through light ice; pancake ice} on the water Freshly formed ice in sample Slaty blue (dark) 21-6) 29:0 | Amongst loose —1:7C.| _ pieces of ice Dark slaty blue 22:0} 29-1 ar —1-6C. 21:1) 29-0 ss —170C. 21:0| 29-0 —170C. Data Relating to the Determination of the Density of the Sample. Time and Date. 15.0 13.30 16.15 11.40 21/2/03 11.7 21/2/03 11.50 Temperature Wot Density of Sample. uring Weights 5 (Density of Distilled Experi : Read- | mersed °C —]. xperiment. ta aa Bactnn Water at 4° C.=1.) ay f of Hydro- e ’ ve ie meter Bate Hye Ob-_ |Reduced Reduced Pa (grams). 3 (cub served to to ey. : ceatims)) att’. |15°-56C. t. the the Air, Sample w. R. Vi 4 Ss +’. 4 Ss 15°'56, 4 Ss t. ats TAD: 55°0 9-9 5:50 10-0 12-8 C. 5:55 | 14:8 | 95:60 19-5 | 5:65 23-8 10-1 5:70 28-8 | 5:75 33-5 | 5:80 38-8 6:84.37 | 5:85 44:8 0571 10-2 5:90 52:2 2625 10-07 5-70 29-6 7-1633 | 1-:02581| 1:02473) 1-02699 52-5 9:0 | 5-50 6-5 11-4 C. 5:55 11-8 5:60 17-5 5:65 22:5 9:2 5:70 27-5 5:75 33-5 5:80 38-5 6:8437 5°85 43-8 0519 9-3 5:90 49-2 2474 9-15 5:70 27-9 7:1430 | 1:02592| 1-02469| 1-02695 50:0 7-3 5:70 33-5 10-0 C. 5:15 39-0 5:80 44-2 68437 5:85 49-5 0417 TA 5:90 55-0 3920 7:35 5:80 44-2 72774 | 1:02572| 1-02424| 1:02648 48-9 TT 5:00 15-6 9-40. 5:55 22-0 5:60 26-5 5:65 31-2 7-8 5:70 36-5 5:75 42-0 5:80 47-0 6-8437 5:85 52-3 +0442 7-9 5:90 57-9 +3264 7-8 5:70 368 7:2143 | 1-02553| 1:02411| 1-02634 48-6 7-6 5-50 4:8 92 C. 5:55 9-2 5:60 14-0 5°65 19-8 TT 5:70 24-8 5:75 30-2 5:80 35-9 6-8437 5°85 41-2 0437 7-8 5:90 46-7 2235 7-7 5:70 25-2 7-1109 | 1:02610 | 1:02466| 1-02692 48-4 7-5 5:70 25-9 9-1 C. 5-75 31-0 5:80 36-5 68437 5:85 41-2 0425 7-5 5:90 46-5 +3211 7:5 5:80 | 36:2 7:2073 | 1-02611)| 1-02464| 1-02691 48-5 7-0 5:70 25-8 9-2 C. 5:75 30:8 5:80 36-2 6-8437 5-85 41:3 0403 7-2 5:90 46-8 *3211 7-1 | 5:80 36-2 7:2051 | 1-02612 | 1:02459)| 1-02686 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITIE Number of Station. 187| Feb. 251 | 188| Feb. 189| Feb. 190| Feb. 191} Feb. Data Relating to the Collection of the Sample. Depth in Fathoms. d, from (a) P f which 2) the Sample was col- lected. . | Surface Surface Surface Data Relating to the Determination of the Density of the Surface Surface Surface Temperature during Experiment. Colour of Water, | Time | Current, and ial jp Ne t’. Remarks. Date. | | i of of the the Air. |Sample. sek Cs Slush in sample 9.45 50-0 7-6 .| Sailing through} 22/2/03 |10-0 C. pancake and brash ice Reeve’s French 7-6 blue 7-9 WE Steaming through} 10.20 |49-6 78 pancake andj 22/2/03 | 9-8C. brash ice Pancake ice in sample 8-1 Dark slaty blue 83 8:05 Sounding a- 11.30 |46-3 7:8 mongst ice 22/2/03 | 7-9 C. Dull slaty blue 7-9 7-85 Steaming through} 11.45 |48-6 7-6 ice 22/2/03 | 9-2 C. Ice spicules in sample Dark slaty blue (cs) 775 Sample sofullof} 12.0 [49-7 7-2 ice that it was] 22/2/03 | 9:8 C. left in lab. for 3 hours before being poured 7-4 into glass bottle 7-9 7:55 Ship stationary 14.0 {50-4 8-0 amongst ice 10-2 C. 8:3 8-8 8-4 Weights added | UN VOT EN OVEN VEN Ww OD O10 & Go OO bo bo ONS 119 O11O 1S S a AM NTE OO Od Od OF OF Ee CO WO WSO 1S O10 O11 5:70 RAKMANAN MOAADW|D Ua WS VO O19 O11 Sulipos DOAAIAAAK | 0 So ANN NAN NN NOMWNMIOnTo no alo ~] SO S|o LSS SSS |) POLO CNS) Soh &| ONO SHSOHNH ao or whore Plea CPSU SI Cpls ONSONSNS Volume of im- mersed. Portion (cub. centims.) V. WNNRE S| RSASRSSor9 ~ MNOADOWOOO ~o cS Data Relating to the Collection of the Sample. Day.|Hour.| Lat. 22 | 8.0 es 22 | 12.0 | 7021S. 22 | 24.0 3 | 8.0 3 | 12.0 | 6557S. 4 | 12.0 | 69 52S. e(E.) Position (L.) ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 103 Data Relating to the Determination of the Density of the Sample. Depth Tem: Vol ‘ : perature Temperature Qe Density of Sample. in: Hathomss)| at time af during —_| Weights ofim- | (Density of Distilled | Collection of Experiment. | added mersed | ‘Water at 4° C.=1.) d. Sample. ey Portion ; Colour of Water, | Time Hydro- a . Ahiat T t Current, and and me. tf meter y ae Ob-_ |Reduced/Reduced Sean aye Remarks. Date. grams) meter ' served to to at’) === Ge | abt’. [15°56C.| t. Posi- ea os of of of : tion| Wa | the | the the | the lected. | Alt: | Water Air. |Sample w. Wo Sv, | 4Sises6,| 45¢, ORES 10) eH SIDER Rae Surface! 18-7} 29-0 | Lyingtoamongst} 14.30 |54-5 78 5:50 —170C. pack ice touch- 250) _ 5:55 ing ship 5:60 5°65 8-1 5:70 5:15 5-80 6:8437 5°85 -0459 8-4 5:90 1978 : 8-1 5:70 7-0874 | 1:02623 | 1-02484| 1-:02710 Surface} 19:6 | 29-0 Reeve’s French 14.55 [53-7 6-0 5:50 —17C.| _ blue 12:1 C. 5:55 Slush ice in 5:60 sample 5:65 6:0 5:70 5:75 5:80 6-8437 5:85 0337 5:9 5:45 +3326 5-95 5:65 7-2100 | 1:02529| 1:02364| 1-02585 Surface] 20:0} 29-0 | Lyingtoamongst} 11.30 (52-5 74 5:50 —170C.| ice; none in } 23/2/03 |11-4 C. 5:55 sample 5:60 5-65 78 5-70 5-75 5:80 6-8437 5°85 0439 8-1 5:90 +2182 7:75 5-70 71058 | 1:02613 | 1-02469 | 1-02695 Surface] 20-1 | 29-0 | Steamingthrough}] 12.12 {51-8 6-9 5-70 —1:7C.) _ bay ice 11:0 C. 5°75 Reeve’s French 5:80 blue 5:85 12 5-90 5:95 5:65 6-8437 5:60 0408 | T5 | 5-55 2891 72 575 7:1736 | 1:02602) 1:02451| 1-02676 Surface] 21:0} 28-9 | Steaming through} 14.30 |48-5 10-1 5:50 —1:7C.| thinice; some} 25/2/03 | 9:2.C 5:55 in sample 5:60 Reeve’s French 5-65 blue 5:70 9:9 5-75 5-80 6-8437 5:45 -0564 9-8 5-40 +3645 9:95 5:60 7-2646 | 1:02472| 1-:02364| 102585 Surface} 23-8 | 29-0 |Steamingthrough} 15.0 |47-2 10-1 5-40 - —1-70.| pack 25/2/03 | 8&4. 5-45 A few ice spicules 5:50 in sample 5:55 Reeve’s French 10-1 5-60 blue 5:65 1-02565 | 1:02459 | 1-02685 104 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITI Data Relating to the Collection of the Sample. Date (E.) Position (L.) tt perth Temperature Temperature . ‘ at time of during Weights alee a Collection of Experiment. | added ee ar Sample. to 3|§ Colour of Water, } Time Hydro D s of see T t Current, and and He. t’. meter 2 ‘ Month. Day.| Hour.| Lat. Long ve thers Remarks. a eee (erauis) o = = Posi Sample f of of of 5 3 tion] Gop | the the the the 2a2|4 L. |y poked Air. | Water Air. |Sample. w. 1903. eZ, oe Patel S185 O18) Ge 260/198) Feb. | 24 Surface} 22-1 ats A piece of ice} 18.0 46-8 2-6 0-80 : (hard __ blue) | 25/2/03 | 8-2 C. 0-85 about 8 inches 0:90 thick 0:95 3-1 1-00 1-05 1-10 1-15 45-5 3:3 1:20 7:5 C. 2-95 1-00 | 261 |199| Feb. | 24 | 24.0 Surface] 22:0) 29-0 18.40 52-6 10-4 5:50 —17C. 25/2/03 |11-4 C. 5:55 5:60 5:65 After testing the 5-70 last sample 5°75 (199) I unfor- 5:80 tunately broke 5:85 } the hydro- 11:0 5:90 | | meter 10-7 5:70 Hyprometer No. 14. WericotT=181:0189 Grams. 262 | 200) Feb. | 25 | 16.0 Surface] 22:7 | 29-2 Dull blue 9.0 [58-7 9-2 5:50 | —1-6C 28/2/03 |14-8 C. 5:55 | 5:60 5-65 9-9 5-70 5:75 5-80 5:45 10-2 5-40 | 9-7 5:60 264 |201| Feb. | 26 | 12.0 | 6936S. | 20 20 W. Surface] 23-2} 29-1 Light dull 10.0 [58-7 9-9 5-40 —16C blue 28/2/03 |14-8 C. 5-45 5:50 5-55 10-4 5:60 5:65 5:70 5:15 10-8 5:80 | 10-35 | 5-60 265 | 202! Feb. | 26 | 20.0 Surface] 27-0} 29-1 Sailing through | 10.40 |63-6 13-2 5:50 | —16C pancake ice 28/2/03 |17-6 C. 5-55 Light greeny 5:60 blue 5:65 13-6 5-70 5-75 | 5:80 / 585 14-0 5:90 13-6 5:70 Data Relating to the Determination of the Density of the § 54-0 32-5 27-0 32-8 38-0 43-2 48-8 54:5 59:0 22:8 17:8 38-2 . 16-0 22-2 27-8 33:5 38-5 44-0 49-0 54:0 59-0 38-2 25:5 31-0 35-5 40:0 45-0 50:0 56:0. 62-0 68-0 45-9 Volume of im- mersed Portion (cub. icentims.) V. 6-8437 0167 +3121 7-1725 6-8437 0607 “2882 7-1926 1-5471 0605 +3456 1:9532 15471 |. -0646 3456 1-9573 1-5471 0849 4152 2-047 2 1-02566 10 Data Relating to the Collection of the Sample. a Depth Position (L.) ae Temperature in Fathoms. mi RAS OR Collection of D. d. Sample. _from oe which | T t Lat. Long. rik ihey Posi-|S2™Ple) of | of tion} "4° | the | the L. rected, Air. | Water. ° ’ ° fs ° 10 ° F, Feb,.§ j26 | 24.0 | Surface; 28-0 | 29-1 —1-6C. Feb Surface] 28-9 | 29-2 —1-6 C. Feb. Surface] 29:0] 29-1 —1-6C. Feb, Surface} 28-7 | 28-9 —1-7 C. Feb. Surface] 27-3 | 29-0 —1-7 0. Feb, Surface] 26-5 | 29-0 —1-7 C. ND SALINITIES OF THE WEDDELL SHEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 105 Data Relating to the Determination of the Density of the Sample. | Temperature ee Density of Sample. during Weights 3 (Density of Distilled Sana Read- | mersed (ple Experiment. ae sae pec Water at 4° C.=1.) F f of ; Colour of Water Time Hydro- g : Current, and : and Ie ii meter pai a es Ob-_ |Reduced|Reduced Remarks. Date. (grams) “ (cub served to to cae Mims.) abt’. |15°56C.) t. of of centims.) the the Air. Sample. Ww. 1a We 4 Ss t/. 4 Ss 15°'56. 4 Ss te SRG aC: an 11.45 [62-8 11-7 5:50 30:0 28/2/03 |17-1 C. 5:55 35:0 5:60 40:0 5-65 42-0 11:5 5-70 46-0 5-75 51-0 5:80 56-0 15471 5:85 61-0 0724 11:5 5:90 65-8 4288 11-6 5:70 47-4 2-0483 | 1-02566 | 1-02486 | 1-02712 Sailing through} 12.10 {55-0 10-6 5-40 14-5 bay, and old} 28/2/03 |12-8 C. 5-45 19-8 ice touching 5:50 24-0 ship 5:55 28:8 Dull blue 10-6 5-60 34-0 5:65 38-5 5-70 42-8 1-5471 5:75 47-0 0658 10:5 5:80 - 52:0 3032 10-55 5-60 33-5 1-9161 | 102585 | 1-02485 | 1:02712 Lying to for} 13.40 |53-5 | 10-7 5-50 29-0 sounding, ice} 28/2/03 |11-9 C. 5:55 34-5 touching ship 5:60 39-8 1-5471 5:65 45-0 0674 10-9 5:70 50:0 3591 10:8 5:60 39-7 1:9736 | 1:02553 | 1-02458 | 102685 Sailing through} 14.0 (55-0 10-8 5:00 9-5 mostly pan-} 28/2/03 |12-8 C. 5:05 13-5 cake ice; some 5:10 18-5 in sample 5:15 22-5 11-2 5-20 27-5 5:25 33-5 5:30 39:5 1-5471 5:35 44:5 -0696 11:5 5-40 49-5 2596 11-15 5-20 28-7 1:8763 | 1:02388 | 1-02301| 1-02516 14.40 |55-6 11-2 5:30 21-2 28/2/03 |13-1 C. 5:35 26-5 5-40 31:8 5-45 37:0 11-4 5:50 42:0 5:55 AT-8 5:60 53-0 1-5471 5:65 57°8 0708 11:5 5:25 15-5 3347 11:35 ee 37-0 1:9526 | 1:02482| 1-02398 | 1-02619 15.10 54:5 10-2 5-50 22-8 12-5 C. 5:55 27-8 5:60 32-0 5-65 37-0 10-4 5:70 41:0 5:75 50-0 5:80 54-0 1-5471 5:85 58-0 0646 10-5 5:90 62-8 3872 10-35 5-70 42-8 1:9989 | 1:02593| 1:02490 1-02717 106 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVITI Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Vg a | ta Depth ; Volume Date (E.) Position (L.) : Temperature Temperature ‘ De in Fathoms. at time of during Weights] pond. oF ia (Dens g o4 | 2 (ee eee Poa bi th | Papas ee of Experiment. | added ea Portion Wa e|& [i ___| Colour of Water, | Time mM | of ae T t Current, and ae Te t’. = = Month. ‘Day.| Hour.| Lat. Long. os the - Heranaks, nee 2/2 | Posi-| Sample of of of of E E tion Lie the the ; the the AL ee | L. | tected, | Ait. | Water. Air. |Sample.| w. R. Mic 4Su, 1903. oe a 4s cane aah, oR 2 C, 272 | 209| Feb. | 28 | $.0 is as .. | Surface] 24-9 | 29-6 | Sailing through] 16.0 [51:5 11-1 5:50 —1:7C.| bay and pan- 10:8 C. 5:55 cake ice ; some 5:60 in sample 5:65 Reeve’s French 11-1 5:70 blue 5-45 5-40 5:35 11-1 5:30.. ~ @ Wt | 5-50 1-02519) 10 | 273 | 210| Feb. | 28 -- | 6922S. | 2636 W.]| .. 2587,| .. 38-66 | Buchanan-Rich- | bottom 3:7C.| ard reversing thermometer and bottle Memot thermo- meter 65,743 273 | 211| Feb. | 28 | 15.30 fe a3 .. | Surface] 24-6 | 29-4 Lying for stop-}] 16.30 [50:5 8:3 5:50 —14C.| ping in open 10-3 C. 5:55 pool 5:60 Slush in sample 5:65 : Reeve’s French 8:3 5:70 blue (light) 5:75 5-80 5°85 8-4 5-90 8:35 5:70. 1:0259 273 |212| Feb. | 28 | 16.0 ae 45 30 50 | 24-7 | 29-40 |Nansen-Pettersson} 10.0 (59-0 8-5 5:50 —1-44C.| _ water-bottle 1/3/03 |15-0 C, 5:55 Thermometer No. 5:60 18,725 5-65 8-8 5-70 5-75 5:80 5-85 9-1 5:90 8-8 5:70 1-02674 273 | 213| Feb. | 28 | 17.0 OF ee ae 500 | 25-0 | 32:6 Nansen-Pettersson}] 10.35 [55:9 8-2 5-50 0-33 C.| water-battle 1/3/03 |13-3 C, 5:55 Thermometer No. 5:60 18,725 5-65 8-5 5:70 5:75 5:80 5:85 8-8 5:90 85 5:70 1:02695) 273 |214a| Feb. | 28 | 17.30 5c s 3¢ 1000 | 25:0} Not |Nansen-Pettersson} 15.25 |49-2 9-0 5:50 3:3 recorded! _water-bottle 1/3/03 | 9-6 C. 5:55 8-0 Thermometer No. 5:60 13-5 18,725 5:65 19-0 5:70 24:5 9-2 5:75 29:5 5:80 35-0 5-85 40-1 9-3 5-90 46-0 | 9:15 | 5-70 24-2 | 1-8240 | 1-02692) Lb ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN, 107 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. | re Depth : i> (E.) Position (L.) : D Temperature Temperature Volume Density of Sample. | in Fathoms. at pea ae ONG anne Weights of im- (Density of Distilled ae ——7_., | Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) | D. qd. Sample. AG ing | Portion a Colour of Water,}| Time | | | Hydro- oe o | of from T t Current, and and ue t’. Butes eee Hy es Ob- |Reduced|Reduced Sea which y Remarks. Date. (grams). GMO RUE I rani to to fon ay.| Hour. Lat. Long. ait the a Peer tone att’. |15°-56C. t. Posi-~ Sample of of of of centims.) ' tion ied the the the the L. | jected. | Ait | Water. Air. |Sample.| w. R. Vv. 494. +Sieae,| 494. 1903 ; Bats a a cas Spy ||. 200; Feb. |28 | 20.0 ae ys o8 1000 | 24:0 | 32:05 ‘|Nansen-Pettersson) 11.0 {54-8 8-4 5:50 3:8 0:03 C. water-bottle 1/3/03 |12-7 C. 5:55 8-9 Thermometer No. 5-60 14:5 18,725 5:65 19-8 8-7 5:70 25-6 5:75 30-2 5:80. 35:8 1:5471 5:85 41-0 -0540 8-9 5:90 46-0 “2271 8-65 5-70 25-1 1-8282 | 1-02690 | 1:02559| 1-02782 Feb.) |28 | 20.30 ae 50 20 1500 | 24:0 | 31:52 | Nansen-Pettersson) 11.35 [51-5 8:7 5-60 15:8 —0:29 C. water-bottle 1/3/03 |10-8 C. 5:65 21-0 Thermometer No. 5:70 26-5 18,725 5:75 32-0 8-9 5:80 37:8 5:85 43-2 5-90 48-7 1-5471 5:95 54:0 0552 9-0 6-00 59-3 -3401 8°85 5:80 37-6 1-9424 | 1-02681| 1:02552) 1:02775 Mar.@j} 1} 8.0 ne ac -. | Surface] 25-0 | 28-9 Steaming through) 14.25 {47-5 11:3 5-50 31-7 { —1:7 C. pancake ice 8-5 C. 5-55 36-5 | Slush (little) in 5-60 41-7 i sample 5:65 44:8 Reeve’s French 10-8 5:70 49-9 blue 5:75 53-9 5-80 58-5 15471 5:85 62-5 -0690 10-8 5-45 21-0 4025 11:05 5:65 44:5 2:0186 | 1:02555 | 1:02466| 1-02690 Mar, |1 |12.0 | 69 3S. |28 2W.| «. | Surface] 25-0 | 29-2 Steamingin open} 15.0 |48-7 8-3 5-40 6:9 —1:6C. pool of water 9-3 C. 5:45 12: Dull blue 5:50 17-0 i 5:55 22-2 | 8-4 5:60 27-0 } 5:65 31:8 | 5-70 36-5 | 1-5471 | 5-75 41-0 0527 ; 8-6 5:80 46-2 +2415 ; 845 | 5-60 26-7 1-8413 | 1:02627| 1:02493| 1-02720 > . Mar, }1 | 16.0 oe Br .- | Surface] 25-6 | 29-0 Steaming through}! 10.20 /47-5 1) 5°50 25-5 i i —17C. loose ice 2/3/03 | 8-6. 5:55 30-5 . A little slush in 5-60 36-0 sample 5:65 41-0 | Reeve’s French 8-0 5:70 46-5 blue (light) 5-75 51:5 5:80 | 57-0 | 1-5471 5-85 62-0 0502 8-2 5-90 67-0 -4188 8:05 5:70 46-3 2:0161 | 1:02584| 1:02446 | 1-02669 Mar. }}j1 | 20.6 28 ne -. | Surface) 25-8 | 29-0 Steaminginopen}| 10.45 |50-0 8-2 5:70 38-0 —1:7 C. pool sur-} 2/3/03 |10-0 C. 5-75 43-5 rounded by ice 5°80 49-0 1:5471 Turquoise ? (light 5-85 | 540 | -0524 bad) 8-6 5:90 59-0 4405 8-4 5:80 48-7 2-0400 | 1-02625| 1°02490| 1-02717 108 | | Date (E.) Position (L.) an ae, - he ri ig i] Pep ae | ot | 08 ¥ © | Month. (Day. Hour.|} Lat. Long. oe the 2/2 Posi- | 52™Ple 2 5 tion | Was Z1e L, |, cok * | lected. | | 1903. | re og 278 220) Mar. | 1 | 24.0 Surface ) ) . 279 221] Mar. 2| 40 Surface | 280 222) Mar. | 2| 8.0 Surface 280 | 223) Mar. 2 | 11.0 2511 280 | 223!) Mar. 2 |11.0 2511 280 |224| Mar. | 2 | 11.50 500 | ; | 280 |225' Mar. | 2 | 12.0 | Surface | | | | Temperature Temperature at time of during Weights Collection of Experiment. | added Sample. hese i) — Colour of Water, | Time Hydro- 7p t Current, and and es t’. meter : : Remarks. Date. (grams). of of of of the the the the Air. | Water. Air. |Sample. w. 7 roe = ak PALMS a8 Gp 25-4 29-0 11.30 {51-5 9-4 5-70 —17C. 2/3/03 |10-8 C. 5-75 5:80 5:85 9:8 5:90 9-6 5:80 25:6 | 28-9 Ice crystals inf 12.0 55-5 8-9 5:70 —170C. sample 13-1 C. 5:75 y. 5:80 5:85 9-4 5:90 ¢ 9:15 5:80 25-7 | 28-9 | Steaming through] 14.30 |47-2 7-2 5:50 —1:7C.| _ pancake ice 840. 5:10 Ice crystals in 5-15 sample 5:20 Reeve’s French 5:25 blue TA 5:30 5:35 5-40 | 7-6 5:45 7:4 5°30 25-7 | Thermo-) Buchanan-Rich-} 10.30 55-9 9-6 5:45 meter ard sounding-} 6/3/03 |13-3C 5:50 came up tube 10-2 5:55. twisted, | Thermometer No. not 102,504 9:9 5:50 reliable Tested again on} 11.0 {50:3 7-6 5:50 10/3/03 at 11]- 10-2 C. 5-60 hours 5-65 7:8 5:55 v7 5575 25-0 | 32:65 |Nansen-Pettersson] 7.20 [55-3 7:8 5-60 - 0:36 C. water-bottle 3/3/03 |12:9 C. 5-65 Thermometer No. 5:70 18,725 5:76 8-0 5:80 | 5:85 5-90 5:95 8-4 6-00 8&1 5:80 25:0 | 28-9 Stopped for} 15.50 |48-3 5-9 5:50 —1-70C.) sounding in 9:1 C. 5:55 open pool, 5:60 amongst ice, 5-65 mostly pan- 6-4 5:70 | cake 5:15 . Reeve’s French 5-80 blue 5:85 | ) 67 5:90 6-3 5:70 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the § DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT centims.) Volume ‘of im- mersed Portion (cub. 1-5471 -0462 +3428 1-9361 15471 0618 0678 16767 1-5471 0480 0923 1-6874 k 1-5471 0505 | +3085 | " 1-9061 | 1-02701| 1 1-5471 0393 | 2822 | 1-8686 | 1:0266 | I YD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 109 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. Position (L.) Depth in Fathoms. Long. d. from which the _|Sample was col- lected. Colour of Water, Current, and Remarks. Time and Date. 1000 1500 400 300 200 100 Temperature at time of Collection of Sample. 40 t of of the the Air. | Water. Oe Sekt, 26-2 | 31-84 —0:08 C. 26:2 | 31-71 —0:17 C. 26:2 | 32-82 0-45 C. 26:2 | 33-1 0-61 C. 26-2 | 32-72 0-40 C, 26:2 | 32-30 0-17 C. Nansen-Pettersson Nansen-Pettersso Nansen-Pettersso Nansen-Pettersso Nansen-Pettersso water-bottle Thermometer No. 18,725 water-bottle Thermometer No. 18,725 water-bottle Thermometer No. 18,725 water-bottle Thermometer No. 18,725 Nansen-Pettersson| water-bottle Thermomteer No. 18,725 water-bottle Thermometer No. 18,725 9.35 3/3/03 10.10 3/3/03 10.45 3/3/03 11.10 3/3/03 11.50 3/3/03 12.15 3/3/03 | | Volume Density of Sample. of im- (Density of Distilled rc Water at 4° C.=1.) Pol of < Boe Ob- |Reduced|Reduced cab served to to leenitanna ) att’. |15°-56C. t. | eae i) \%, 4S, | 4Sisess,| 4 St, 5:50 5:55 5:60 5:65 5-70 5:15 5-80 15471 5-85 -0568 5-90 -2343 Temperature during Experiment. AW t’. of of the the Air, |Sample. Pane SAGs 52-5 8-8 11-4C. 9:1 9-4 9-1 53-6 8-9 12-0 C. 9-2 9-4 9-15 52:9 9:0 11:6 C. 9:3 9-6 9-3 53:0 9-4 11-7 C. 9-6 9-8 9-6 52:8 9-4 11:6 C. 9-6 10:0 97 | 52:3 9-2 11:3 C. 9-4 9-6 Ss Doo nw MOORE NOS ASANO 1-8382 | 1-02684| 1:02559| 1-02782 1-5471 0571 +2370 1-8412 | 1-02682| 1:02558| 1-02782 1:5471 0580 2397 1-8448 | 1:02680| 1-02559) 1-02784 1-5471 0599 -3465 1-9535 | 1-02674| 1-02558| 1-02783 1-5471 “0605 +3447 1-9523 1-02675| 1-02560| 1-02785 1-5471 | -0587 +3302 | 1:9360 1:02684) 1:02564| 1:02787 110 Number of Station. 280 | 2 281 282 282 282 282 282 Number of Sample. 233 234 235 236 228 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVI Data Relating to the Collection of the Sample. Depth Data Relating to the Determination of the Density of the Date (E.) Position (L.) : Temperature Temperature ae EO at me of during Collection of Experiment, D. d. Sample. Colour of Water, | Time of an T t Cantal acd and it ti. Month. Day.| Hour.| Lat. Long. os the scam Date ile ¥ Posi-)5@™Ple| of | of of | of tion} Go. | the the the the L. ientede Air. | Water Air. |Sample. 1903. ie ee PIne on. Ss aCe Mar. | 2 | 16.40 50 | 26-2} 29-02 |Nansen-Pettersson} 13.35 48-7 8-9 —1:65 C water-bottle 3/3/03 | 9:3 C. : Thermometer No. 18,725 9-1 9-2 9:05 Mar.'| 3 | 8.0 Surface] 25-9 | 28-9 | Steaming through} 14.15 {50-9 12-0 —170C mostly pan- 10-6 C cake ice Ice crystals in sample 11:9 Reeve’s French blue 11-9 11:95 Mar. | 3 |12.0 | 6835S. | 3156 W. Surface] 26-6 | 29-4 | Steaming through} 14.50 [51-5 8:8 —1:4C.) _ pancake ice 10-8 C. Reeve’s French blue 9-2 9-5 9-15 Mar. 3 | 15.50 oe AG 2452 | 27-4 | 31:89 | Thermometer No.J 14.35 [56-8 11-7 —0:05C.) 102,504 4/3/03 |13-8 C. 12-0 12-4 12-05 Mar. 3 | 17.30 ne fe 1952 | 27-0 | 33-29 Buchanan-Rich- 0-71.) ard bottle Thermometer No. 102,514 Mar. 3 | 17.45 2442 | 27-0 | 30-12 Buchanan-Rich- —1:05 C, ard bottle Thermometer No: 102,514 Mar. | 3 | 20.0 Surface} 28-0 | 29-2 Steamingin open} 10.20 |46-9 58 —1-60C. pool 5/3/03 | 3-3 C. Reeve’s French blue 6-0 6-1 Volume : of im- R Wea Read. mersed pe . ing Portion of Ww. 185 Vi Sy. 4 5:50 6-2 5:55 11:8 5:60 17-0 5:65 22:5 5:70 28-0 5:75 33-0 5:80 37:8 5:85 42-8 5-90 48-2 5:70 27-5 1-8524 | 1:02676 5:50 33-0 5-55 38-0 5:60 43-2 5:65 48-7 5:70 53:8 5:45 27-0 5:40 21-7 5:35 16-0 5:30 10:6 5-50 32-4 1:9148 | 1:02531 5:50 18-0 5:55 23°5 5:60 29-0 5:65 34-0 5-70 39-0 5:75 44-0 5:80 49-6 5:85 54:6 5:90 59-6 5:70 39-0 I-:9570 | 1-026 5:50 11:5 5:55 16-7 5:60 22-4 5:65 27:8 5-70 33°5 5-75 | 39-0 q 5:80 44-5 E 5:85 50-0 | 5:90 55:5 ; 5-70 33-4 1:9244 ol b 5:50 12:0 5:55 17-0 5:60 22-5 5:65 27-7 5:70 33-0 5-15 38-0 5-80 43-5 1:5471 5-85 49-0 0371 5-90 53°8 2976 5:70 32:9 1-8818 | 1:02658 Data Relating to the Collection of the Sample. AD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 111 Data Relating to the Determination of the Density of the Sample. (E.) . Position (L.) i 2 2 ae Temperature Temperature 4 3 at time of during Weights 5) | Collection of Experiment. | added | Read- D d S ing a : ample. ord, to ze | Colour of Water,} Time Hydro- | wydro- | of foe T t Current, and and 1%, t’. meter ee onthwihy. Hour. Lat. | Long. ea the: |—— RAGES, Ete eee feranis). | Posi-/S2mple) ot | of of | of | | tion an the the the the . L. lected, | Ait: | Water. Air, eaapl w. R. 903. ° ‘ ° , ° FR. | oR ° F, ° (Gh Jar, Wj3 | 24.0 f Surface; 28:0} 29-0 11.35 |49:3 6-7 5-50 12-5 17 C, 5/3/03 | 9-6 C. 5-55 18-9 5:60 23:5 5-65 28-7 Teil 5-70 34:0 | 6-9 5:60 23-3 far, Wi4 | 12.0 Piece | 28-9 Temperature 12.0 (49-9 T1 1-20 15-1 of ice about 3 ins. in} 5/3/03 | 9-9C. 1:25 20-1 from a ice=—2-0 C.: 1-30 25°5 berg not sample 1-35 31-0 Temperature of 7:3 1-40 36-1 water with ice 1:45 41-7 =0°C.: sample, 1:50 47-5 first melting 1:55 52-7 7-6 1-60 57:8 7:35 1-40 36-4 Same as} 240a Second melting 13.50 |46-7 8-8 0-70 5-7 5/3/03 | 8-2C. 0:75 10:7 0-80 16-0 0:85 21:8 8-8 0:90 27-0 0-95 31-9 1:00 37:5 1-05 43-6 8-9 1:10 49-0 8-85 0-90 27-0 Surface| 28-8 | 29-0 Too dark 14.20 47-5 7-4 5-40 1:5 —17 5/3/03 | 8-6 C. 5-45 6-2 5:50 11:5 5:55 16-0 7-6 5:60 21-6 7:5 5:50 11-4 Surface] 28-5 | 29-2 14.45 [47-1 7-0 5:50 9:3 —1-6C. 5/3/03 | &4C 555 | 14:5 5-60 19-7 5:65 24-8 5:70 30-0 5:75 25:1 5:80 40-1 5:85 45:0 7-2 5:90 50:0 7-1 5:70 29-8 Surface] 23-3 | 29-0 Turquoise 15.20 46-5 8-0 5:40 2-0 E76 5/3/03 | 810. 5-45 7-0 5-50 12-5 5:55 17-6 8-1 5-60 22-5 8-05 5:50 12-3 2473 | 21-9 | 31-49 Buchanan-Rich- —0:29C.| ard reversing thermometer and bottle Thermometer No, 102,504 | | Volume | of im- mersed | 5 | Portion (cub, V. centims.) Density of Sample. (Density of Distilled | Water at 4°C.=1.) | Ob- served at t’. Reduced Reduced to to "| 15°-56C.; t. | aSe. | 4Sises6) 4S. 1:5471 0431 +2108 1-5471 0459 +3293 1:9223 1-5471 0443 2696 1:8610 1-5471 0502 1113 1:7086 1:8010 1-02650 | 1-02494| 1-02721 100273 | 1-00178)| 1-00278 1:00041| +99962) 1:00048 1:02654 | 1:02506 | 1-02733 1:02671 | 1:02517 | 102744 1:02647 | 1-02507 | 1-02734 17 112 1. Number of Station. Number of Sample. | ' | | 286 |2448 286 |244B 287 | 245 288 | 246] Mar. 290 | 248 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the 289 | 247| Mar. Time and Date. 9.20 7/3/03 11.0 6/3/03 11.30 6/3/03 12.0 11.30 ae Depth Position (L.) . Temperature _ in Fathoms. Br i Oh a _ | Collection of d. Sample. Colour of Water, from Current, and which | 7. t Remarks. Lat. Long. thes | ==> ae souk of of col. | the the ‘lected. Air. | Water. == , ° , ° F. ° F. 2488, |Same) as 2444 | Buchanan’s bottom sounding-tube sf 0 «. |Tested again jat 11.20 on the | 10/3/03. 55 3 Surface] 23-6 | 28-8 Pr —1-8C. Surface] 24:0 | 29-0 —1:70C. Surface] 21-9.| 28-9 Spicules of ice —1:7 C. in sample 67 39 S. | 3610 W. Surface) 24:0] 29-2 Taken whilst —1:6C.| — soundin Reeve’s French blue DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT | Temperature during Experiment. 1s We of of the the Air, |Sample. 285 aC, 50-9 8-2 10-5 C, 9-3 875 50:0 7:8 10-0 C. 8-4 8-1 54:7 8:8 12:6 C. 9:0 9-2 9:0 52-9 8-6 11:6 C. 8:8 9-2 8-9 52-5 8-4 11-4 C. 8-6 8:8 8-6 52:6 8-2 11-4 C. 8-6 9:0 8-6 AA WBOAAMWAODon AOnonone COP PWWRDH eH R| ESESESQow ~ OCONNDHOOMo So 1-5471 0546 +1760 pra 1-5471 0505 +1547 1-7523 1-5471 0562 +3302 1-9335 1-5471 0555 +3085 1:9111 1-5471 0537 3266 "71-9274 1-5471 0537 +2840 1-8848 102630] YD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 113 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ope Depth atten Volume ' Daj (E.) Position (L.) ‘ Temperature Temperature ; Density of Sample. : in Fathoms. | “3¢ time of during | Weights ofim- | (Density of Distilled = | i as Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) | Dz. d. Sample. is ing Heusen : a 4 ; we Sa f (0) es | Colounof Water, “lime Hydro- or ‘ of | from T t Current, and and LN ti meter ay ee Bae Ob- |Reduced| Reduced Sea which : ‘ Remarks. Date. (grams). eae b served to to nth@py.| Hour.; Lat. Long. at CHEN eres 7 ce ¢ Water abt’. |15°-56C. t. : entims. | Posi- peube of of of of 2) : 4 tion} Go. | the the the the L. Teated® Air. | Water. Air. |Sample.} — w. R. V. 4a9¢. | 4S i5-56,| 4 Se 03. °o ’ ° , ° Fr. ° F. ° F. ° (on lar. }|6 | 9.45 ae AG -. | 2000 | 25:4 | 31-44 Buchanan-Rich- —0-31C.| ard reversing thermometer and bottle Thermometer No. 102,509 lar. H|6 | 9.53 , - 2485 | 25-4) 31-42 Buchanan-Rich- —0-32C.) ard reversing bottle Thermometer No. 102,504 lar. $5 | 9.56 r 2500, |Same} as 2504 | Buchanan’s 11.55 {51-5 8-4 5:50 5:2 bottom sounding-tube J] 10/3/03 |10-8 C. 5:55 10-2 5:60 15:6 15471 5:65 20:8 0534 : 8-7 5-70 25°8 +1402 } a ee Pose ae 8:55 5:60 15:5 1-:7407 | 102684) 1-02551] 1-02775 ar. jes | 8.0 - | Surface} 27-0 | 28-9 Steaming through} 10.40 [56-3 8-3 5:00 7-0 —17C. closely packed J 11/3/03 |13-5 C. 5:10 18-0 pancake ice 5:20 28-7 . Ice in sample 5:30 39:3) | 21-5471 Reeve’s French 5:40 50:0 0543 | blue 8-8 5:50 60:2 3067 | —_——_$$_ | | —__ —___] 8:55 5°25 33-9 1:9072 | 1:02398) 1:02270| 1:02485 ar. (ep | 12.0 - Surface] 29-5 | 29-2 Steaming through} 11.10 [52-7 8-4 5:35 5:0 —1-6C. loose pack 11/3/03 |11-4 C. 5:45 15:8 Ice crystals in 5:55 26-5 sample 5:65 37-0 i Reeve’s French 5:15 4T-5 1:5471 blue 5:85 57:8 0537 8-8 5:95 67:8 +3329 ; 8-6 5:65 36:8 1:9337 | 1:02603 | 1-02473)| 1-02696 ar. | 20.0 Surface] 29-2 | 29-3 Ice touching ship} 11.30 |52-2 8-5 5:40 3-5 : —1:5C. on all sides 11/3/03 |11-2 C. 5:50 14-2 5:60 25-3 5:70 36-0 5:80 46-8 1-5471 5-90 | 57-2 | 0549 9-1 6:00 67- +3238 8-8 5°70 35°8 1-:9258 | 1:02635 | 1:02506| 1-02732 ‘ar, 8.50 1925 | 28:0 | 31-46 Buchanan-Rich- | P —0-30C.| ard water-bottl | Thermometer No. 102,509 (ar, 9.0 2410 | 28:0 | 31-37 Buchanan-Rich- —0:35 C.) ard water-bottl Thermometer No. 102,504 lar 9.5 2425, | 28-0 | Same as} Buchanan’s 14.0 (535 | 104 | 5-40 2:9 bottom 2554 sounding-tube } 11/3/03 |11-9 C. 5-45 8-7 5:50 14-2 1-5471 5-55 20:0 0668 11:0 5:60 25-6 1293 10:7 5-50 14:3 1:7432 | 1:02628 | 1-02531| 1:02753 114 Data Relating to the Collection of the Sample. Position (L.) DR W. 8S. BRUCE, MR A. KING, AND. MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVII Data Relating to the Determination of the Density of th Date (E.) os Ee ee Temperature Temperature Volume . "| at time of during | Weights ofa a |] 9 Collection of Experiment. | added Read~ | mersed B=} ee D. d. Sample. re ing Poe ee ae oS Colour of Water, } Time Hydro- e yi 7 opus of es t Current, and 4 and ia Dated pies the Ob- {I 3 2 Month. |Day.| Hour.| Lat. Long. pee Fi Remarks, Date. (grams) (cub. served eS Posi Sample| o¢ centims.), "= | Els OSI- | was a of of ——a = ie tion} Qj. | the the the a a L. lected. Air Water Sample. Ww. R. Vv 454 | __| 1903. loans ela PP oF, | oR, °¢, 295 | 256) Mar. | 10 | 10.0 | oe me Surface] 28-9 | 29:3 Just after sound-] 12.10 8:3 5:50 13:3 —1:5C.| ing, ice quite} 11/3/03 5:60 24-2 near 5-70 34:9 | 1:5471 Reeve’s French 5:80 45-8 0524 blue 8:5 5:90 56-2 +3157 | 8-4 5-70 34-9 1:9152 296 | 257) Mar. | 11 | 8.0 te ae Surface] 25-0 | 29-1 In open pool,} 14.40 13-6 5:50 24:5 ; —16C.| ice about 3} 5:60 34:8 mile away 5-70 45:0 | 1:5471 Reeve’s French 5:80 53:0 0842 blue 13-4 5:90 62-8 +3980 13-5 5:70 44-0 2-0293 | 1:02576 297 | 258) Mar. | 11 | 12.0 | 6622S. |4220W.| .. | Surface] 25-8| 29-2 Very little ice} 15.0 11:2 5:50 15:0 —160C.| about 5:60 25:3 Reeve’s French 5-70 35°7 1-547 1 ; blue 5-80 46-0 0702 11:3 5-90 55°6 +3211 11:25 5:70 35:5 | 1:9384 298 | 209) Mar. | 11 | 16.0 oe “A .» | Surface] 26-0} 29:0 | Amongst ice,] 11.35 |46.5 8-5 5:20 5:7 —170C. some in sample} 12/3/03 | 8-1. 5:30 15-8 Reeve’s French : 5:40 26:0 blue 5-50 36-2 8-6 5:60 47-0 5:70 57-9 5:80 68:0 | 1:5471 5:90 76-4 0534 8-6 6: 86-5 +4215 8:55 | 5-60 46-6 2:0220 299 | 260} Mar. | 11 | 20.0 an \- -» | Surface] 25-0] 28-8 Sailing through} 12.0 6-7 4-10 8-5 —1-8 C. pancake ice;]} 12/3/03 | 8-0C, 4:20 19-0 so much ice in 4:30 30:0 sample that it 4-40 41:0 had to be 6:8 4:50 52:0 melted before 4-60 62:8 pouring in 4-70 73-2 | 1:d471 bottle 4-80 83-2 0427 Too dark to see 7-0 4-90 94:0 4659 6:85 | 4:50 61:5 2-0557 | 1-0190. 300 | 261) Mar. | 12 | 9.10 “ Surface) 20-5] 29:0 | Taken whilst} 13.50 4-7 5:30 20 —1:7C.| _ sounding 5:40 12-8 Ice in sample 5:50 23-5 15471 Reeve’s French 5-60 34-0 0306 | blue (light) 51 5:70 45:5 +2135 =. | |. ail 49 5:50 23-6 | 1-7912 | 1-02601) 300 | 262) Mar. | 12 | 10.15 2000 | 19-7 | 31-48 | Large Buchanan} 14.0 [49-0 5:3 | 5-70 2:5 —0:29 C, Richard water- 9-4 5:80 12-7 bottle 5:90 23-2 Thermometer No. 6:00 33-0 102,504 5:6 6-10 43-0 6-20 54-2 6-30 63-6 | 1-d477 6-40- 71-9 0349 / 5:9 6:50 80:7 +3872 | | 5-6 6:10 42-8 | 1-9692 | 1-0 ' . } YD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 115 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ane Depth De: (E.) Position (L.) . Temperature in Fathoms. at time of — | —_ 7, 4 | Collection of ; : D. d. Sample. anae vice of weet of ) T urrent, an : which : t. onthijay.| Hour.| Lat. Long. See the. |= Bete Posi-|52™P Ie of of tion bikes the the L. feted. Air. | Water. L903. ° , ° ’ ° F. ° F. Mar,! {12 | 10.25 ae ae -. | 2485 | 19-7} 32-15 Buchanan- 0-08 C.| Richard water- bottle Thermometer No. 102,509 Mar, {12 | 10.30 ae Ris .. | 2500 |Same|as 2634 | Buchanan’s : sounding-tube Mar.gil2 | 19.0 ve on .. | 1500 | 14-6 | 32:00 |Nansen-Pettersso 0:00 C.| _ water-bottle Thermometer No. 18,725 Mar. §il2 | 19.40 -@ os Ap 100 | 14:2 | 32:01 |Nansen-Pettersso: 0-01 C. water-bottle Thermometer No. 18,725 Mar, 42 | 20.0 a aie -- | Surface} 14-2} 28-8 | Sample about hal —1-8C.| full of ice i | Mar.| #3 | 8.40 Surface} 7-4 ne e Bay ice near —14C.| _ ship | No ice in sample Reeve’s French blue | ————— Time and Date. 11.20 14/3/03 9.0 13/3/03 9.30 13/3/08 13.30 13/3/03 14.0 Temperature during Experiment. Tes t’. of of the the Air. |Sample. ANE i ICR 61-0 10:8 16-1 C. 12-0 11-4 60-3 12:8 15-7 C. 13-0 13-4 13+] 61:8 12-9 16:6 C. 12:8 12-9 12-9 57-3 11:8 14-10. 11:8 12-2 12:0 59:9 11-2 15:5 C. 11-4 11-8 11:5 | Nol ne Density of Sample. Weights Read- asa Density) of Distilled oe ing Deere Water at 4° C.=1.) Hydro eae! Bits weds meter aioe 3 as ee Ob-_ |Reduced' Reduced (grams) 4 (cub served to to centims.) att’. |15°-56C. t. | a ‘ Ww. k. Vv. sSy. | 4Sisse.| 45e, 5:40 4:0 5-45 9-2 5:50 15-0 5:55 20:8 | 1:5471 5-60 26-3 0711 5°65 31:8 1610 5-525 17-8 =| 1:7792 | 1:02621| 1-02536| 1:02757 5:40 0-0 5:50 11-0 5:60 22:0 5-70 32-1 5:80 43-0 5:90 52:9 6-00 62-8 1:5471 6-10 73-0 0817 6:20 83-6 +3825 6:80 42-3 2:0113 | 1:02641| 1-02589| 1-02814 5-40 2:8 5:50 13-5 5-60 24-2 5:70 34-4 5:80 45-1 5:90 56:0 6-00 64:8 1-5471 6:10 75:0 0805 6-20 86-0 4034 5-80 44-6 | 2:0310 | 1:02630| 1-02574| 1-02797 5:00 1:0 5:10 11-7 5:20 21:8 5:30 32-4 5:40 43-2, 5:50 54:0 5:60 63-9 15471 5:70 74-1 0749 5:80 84-2 3881 5:40 42-9 20101 | 1:02422| 1-02349| 102569 5:40 4:8 5:50 15-6 5:60 26-0 5:70 37-0 5:80 47-2 5:90 58-0 6:00 68-0 1-5471 6-10 79-0 0718 6-20 88-2 4261 5:80 47-1 2:0450 | 1:02622| 1:02539 1-02767 Number of Station. 301 301 301 301 301 301 Number of Sample. 268 269 270 271 272 273 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT Date (E.) Data Relating to the Collection of the Sample. Position (L.) / Month. Day. Hour.} Lat. Long. 1903. oa its Mar. | 13 | 9.10 or Mar, | 13 | 9.25 es Mar. | 13 | 9.50 an ar Mar. | 13 | 10.0 ee A Mar. | 13 | 10.10 Mar. | 13 | 10.15 Depth in Fathoms. D. d. from ee which on the Posi Sample tion aera L col- * | lected. 500 1000 300 400 : 50 200 Temperature at time of Collection of Sample. the Air. 7-4 | 32-17 0-09 C. 1P 4 TT 0-50 C. TT | 29-43 —1-42C. TT | 33-14 0-63 C. Colour of Water, Time Current, and and Remarks. Date. Large Buchanan-} 10.0 Richard water-} 14/3/03 bottle Thermometer No. 102,504 Large Buchanan-+} 10.40 Richard water- bottle Thermometer No. 102,509 11.40 14/3/03 Thermometer No. 102,504 12.10 14/3/03 Large Buchanan- Richard water bottle Thermometer No. 102,509 14.10 14/3/03 Thermometer No. 102,504 14.50 14/3/03 Thermometer No. 102,509 Temperature during Experiment. TY. t’. of of the the Air. |Sample.!. or. °C. 57-2 11-2 14:0 C. 11-4 11-7 11-45 12-0 57-0 14/3/03 |13-9 C 12-3 12-7 12:35 13-1 13-1 13-1 13-1 13-0 13-0 12-9 12-95 12-1 12-2 12-5 12-3 12-2 12-6 Data Relating to the Determination of the Density of the | Volume “4 of im- Weenie Read- | mersed a a ing Fontieg of (0) — edie: Hydro- | Hydro- (grams). ou an | served centims.), w. R. Vi | 5:45 3:0 | 5:55 14-2 5:65 25:0 { 5-15 | 36:0 585 | 465 5:95 57:6 6-05 | 67-5 | 1-5471 | 615 | 78:0 | 0714 6:25 88-9 4188 At 5:85 46-3 2-0373 | 102654) 5:45 4-8 | 5-50 | 10-2 i 5:55 16:0 ' 5:60 21:0 | 5:65 26:8 5-70 31-9 | 575 | 37-0 | 15471 | 5:80 42-3 0771 | 5:85 | 47-4 | +2388 | 5:65 26-4 1-8630 102643) 5:40 2:7 | 5:50 | 136 | 560 | 24-2 | 5:70 | 35-0 | 5:80 | 45-5 | 5:90 56:0 | 6:00 66:5 | 1:5471 | 6:10 76:6 -0817 | 6:20 86:8 +4089 i) 5:80 | 45-2 | 20377 1.02628) 5:40 2-9 il 5:50 13:9 | 5:60 24-7 { 5:70 35°6 ; 5:80 46:0 j 5:90 56:7 1-5471 i 600 | 66-0 | -0808 6-10 | TT 3645 at a — | ‘iz 5:75 40-3 1:9924 | 1:0262 | 10 5-40 3-0 } 5:50 13:9 | 5:60 24-6 i 5-70 | 35:8 | 5:80 | 46-0 5:90 56-2 6-00 65:8 1-5471 6:10 1T-0 0768 afl 6-20 87-2 +4116 : 5-80 45:5 20355 | 1-:02628 5:45 55 5-55 165 5-65 27:2 5:75 38-0 5:85 48:8 5:95 | 60-0 4 6-05 69-5 1:5471 k 6:15 80-9 0780 6:25 90-8 +4396 | 5-85 48:6 2-0647 | 1:02639) SD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN, Position (L.) Lat. Long. ae Depth Data Relating to the Collection of the Sample. é Temperature in Fathoms. areas ake Collection of D. d. Sample. ___| Colour of Water, of from t Current, and Sen ye 4 : Remarks. | a at Posi-|Sa™Ple| of | of tion hers the the L. ieee. Air. | Water. SUE. oF, Surface} 9-4 | 29-2 Trawl down, —1:6C. steaming slow ly through bay ice No ice in sample Reeve’s French blue 2470 | 10-9 | 31:25 Large Buchanan- —0:42 C. Richard water-| bottle Thermometer No. 102,509 > 2485, |Same] as 2754 | Buchanan sound bottom ing-tube Surface} 20-3 | 29-0 —1-7 C. Surface] 25-1 | 29- nS 26 2007 | 24-9 | 31:33 —0:37 C, Ice crystals in sample Reeve’s French blue Stopped for sounding a- mongst pan- cake ice No ice in sample Buchanan-Rich- ard water- bottle Thermometer No. 102,504 Data Relating to the Determination of the Density of the Sample. Time and Date. 15.15 14/3/03 10.40 15/3/03 |14-9 C. 11.10 15/3/03 10.0 15/3/03 10.20 15/3/03 12.10 15/3/03 Temperature during Weights Experiment. | added to Hydro- ls ine meter (grams) of of the the Air. |Sample.|— w. | SER Cs 57-5 12-1 5:35 14-2 C. 5:45 5:55 5:65 12:5 5:15 5:85 5:95 6:05 12-6 6:15 12:35 5:75 58:9 18-1 5:20 5:30 5:40 5:50 17:9 5:60 5-70 5:80 5:90 17-7 6:00 17:9 5:60 58-8 18-3 5:20 14:9 C. 5-25 5:30 5:35 5:40 17-6 5:45 17:95 5-325 60-2 16-4 5:20 15:7 C. 5:30 5:40 5:50 16:3 5:60 5:70 16:2 5:80 16-3 5:50 59-7 16-1 5:20 15-4 C. 5:30 5:40 5:50 16-0 5:60 5:70 16-0 5:80 16:05 5:50 59-4 15:9 5:30 15-2 C. 5:35 5:40 5:45 5:50 5:55 15-6 5:60 15-75 5:45 IB 3 83 bono DWADN Ca baba NS ©] ONANTEHWDAAO ~ Volume of im- mersed Portion (cub. centims.) Vis 1:5471 0771 +3971 2-0213 1-5471 +1117 +3925 20513 1:5471 +1120 1221 1:7812 1:5471 +1017 +3103 1:9591 1-5471 -1002 +2845 1:9318 1-5471 0983 1719 ” 18173 117 Density of Sample. (Density of Distilled Water at 4° C.=1.) Reduced| Reduced to to 15°:56C. t. Ob- served at t’. 45, 45 15°56, 45, 102608 | 1:02541| 1:02770 1-02509| 1:02565| 1:02787 102509 | 1:02566| 1-02788 1:02506 | 1-02524| 1-:02750 1:02521) 1:02533 | 1:02759 1-02558 | 1:02563 1-02788 | | | 118 DR W. S. BRUCE, MR A, KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAYIT Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density o ; ae Depth : 2 i Volu Date (E.) Position (L.) rs Dp Temperature Temperature ume 4 in Fathoms. ae fime ar during Weights hea of i Ss 2 Collection of Experiment. | added ead= | merse Ios ies Dz. d. Sample. Portion 3s] & ay al a] | Colour of Water,}] Time n | of Fed. t Current, and noe -e t’. ~ o f t ate, = | |Month.|Day. Hour] Lat. | Long. | 5¢*| the ee eed = 2\2 Posie Hee of | of of |. of =e tion Eo the the the the Paes L. |) ee a, | Air. | Water. Air. |Sample. | | 1903. ae ae | oF.| °F, oF, | °C. 303 279) Mar. | 14 /19.30; .. x .. | 2532 | 25-6 | 31-0 Large Buchanan} 11.45 (58:8 17-1 525 fal '—0-56C.| Richard water-} 15/3/03 14:9 C. 5°30 | bottle 5:35 Thermometer No. 5:40 102,509 17-0 5:45 5:50 5:55 5:60 16-9 5:65 é 17:0 5:45 303 |279B) Mar, | 14 | 19.40 ate oe «« | 2547, |Same| as 279s | Sounding - tube} 16.0 [57:5 17-5 5:30 bottom with long | 15/3/03 |14-2 C. 5:35 nozzle 5:40 5-45 17:2 5:50 5:55 *| 5-60 5°65 17-0 5:25 17-25 5-45 304 | 280} Mar. | 15 | 4.0 ve me -. | Surface] 29-2] 28-9 |Steamingthrough} 14.55 [58-3 15-7 5-20 —170C. pancake ice,} 15/3/03 |14-6 C. 5:30 some in sampl 5:40 F: - 5-50 15:6 5-60 5:70 5:80 5-90 15:6 6-00 15:65 | 5-60 1-02495| 305 | 281) Mar. | 15 | 8.0 ole oe -» | Surface] 30-2 | 28-9 | Steaming through} 15.30 {57-8 17-6 4-70 if —1:70C, pancake ice, 15/3/03 |14-3 C. 4-80 i] some in samp! 4:90 | Reeve’s French 5:00 t | blue 17:3 5:10 dq 5:20 | 17-2 5:30 | 17-4 | 5-00 1.02235} 306 | 282} Mar. | 15 | 12.0 | 6411S. |4212 W.| .. | Surface] 29-8 | 29-1 Steaming through} 16.30 {55-2 14-6 5-20 | --1:6C. pancake ice, 12:9 C. 5:30 spicules in | 5-40 sample 5:50 Reeve’s French 14:6 5-60 blue 5-70 14-5 5:80 14:55 5:50 307 | 283) Mar. | 15 | 16.0 oe oe .. | Surface] 29-6 | 29-1 Taken whilst} 10.30 (61-1 17:2 5-20 | |—1-6 C. sounding,loose} 16/3/03 '16-2 C. 5:30 ice near, none | 5-40 in sample | 5:50 / Reeve’s French 17:0 5:60 blue 5:70 | | 5:80 5:90 16:8 6-00 | ’ 17-0 5-60 2:0964 | 1-02484\ 1 ) | | ) SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. Data Relating to the Collection of the Sample. , sae Depth ate} o. Position (L. : Temperature ) ( ) in Fathoms. attime of ollection of D. d. Sample. from Fh Vk | } ae which | T- ts .|Hour.} Lat. Long. at the = Posi pample of of tion bie i the the L. | Tecte a Air. | Water | Ged Cw oR, oR, | 4)| 17.30 1900 | 29-0 | 31-31 | —0-39 C. | i” 14] 17.43 p * 2400 | 29-4 | 31-03 —0-54C. | 24.0 n Surface} 28-3 29-0 —17C. 8.45 Surface] 30:0] 29-9 —1:2C. 9.45 ; 2050 | 29-8 | 31-23 —0-42C . | 9.50: 2533 | 29-8 | 31-06 | | —0-53 C, | | | | 119 Data Relating to the Determination of the Density of the Sample. | Temperature _ Volume Density of Sample. during Weights Read oe Lee (Density of Distilled Experiment. antes ae Pochon Water at 4° C.=1.) : £ of Colour of Water,}| Time Hydro- oo Current, and and Te t’. meter Eas Hyste: Ob- |Reduced|Reduced Remarks, Date. (grams).| ™°? (abs served | to to ——- « , °, 8. centims.) att’. |15°-56C. t. of of tole the | the | Air. |Sample.|— w. R. V. a>, |49ise56,| «St. | as SG || | Large Buchanan-} 11.0 60:8 16:8 | 5-30 6:0 | Richard water-} 16/3/03 |16:0 C. 5:35 11:6 bottle 5:40 16:8 Thermometer No. 5:45 21-9 102,509 16-7 5:50 27-1 | 5:55 33:5 5:60 38-7 15471 5:65 43:8 +1042 16-6 5-70 49-0 2497 16-7 5:50 27-6 1:9010 | 1:02539 | 1:02565| 1-02792 Large Buchanan-+} 11.30 58-8 16:3 5:30 4:2 Richard water-] 16/3/03 |14-9 C. 5:40 15:0 bottle 5:50 25:8 Thermometer No. 5:60 36-0 102,504 16-0 5:70 47-6 5:80 57:7. 5:90 68-0 1:5471 6-00 78:0 -1008 16-0 6-10 87:5 4215 16-15 5:70 46-6 2-0694 | 1:02554) 102567 | 102795 12.15 {60-7 17-4 5:20 8-0 16/3/03 |15-9 C. 5:30 18-5 5:40 29-0 5:50 40:0 17-2 5:60 50:8 5:70 60:5 5:80 69-8 1-5471 5:90 79:8 1067 16:8 6-00 89:8 4487 17-1 5-60 49-6 2:1025 | 1:02481| 1-02515 | 1-92743 | Stopped for} 11.35 51-0 11-1 5:40 7-6 sounding in] 17/3/03 |10-6 5-50 18-5 open pool 5:60 29-5 Reeve’s French 5-70 40-0 blue 11-2 5:80 50:6 5:90 61:0 6-00 71:8 1-5471 6:10 82:5 0702 11-4 6-20 93-0 4568 ; 11-25 5-80 50-5 2-0741 | 1:02606| 1-02518| 1-02745 Large Buchanan 7.5 |65-:0 16-6 5:30 5:5 Richard water-} 18/3/03 |18-3 C. 5:35 10:3 | bottle 5:40 16-0 Thermometer No. 5-45 21:3 102,504 16:5 5:50 26-2 Tested by 5:55 31:5 W.S. B. 5-60 35:8 15471 5-65 41-2 1033 16-5 5-70 46:5 2352 16-55 5:50 26-0 1:8856 | 1:02547 | 1-02570| 1-02796 Large Buchanan-} 7.50-8.15 65-2 16-5 5-30 3-2 Richard water-{ 18/3/03 |18-4 C.| | 5:35 9-0 bottle | | 5-40 15-0 Thermometer No. | 5-45 20-0 102,509 16-5 5:50 25-5 1-5471 Tested by 5-55 81:5 +1033 W.S. B. | 166 | 560 | 366 1818 eee ES eee Soll | 16-55 5:45 20:1 1-8322 | 1:02550 | 1-02573)| 1-02798 120 DR W. 8. BRUCE, MR A, KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAV Data Relating to the Collection of the Sample. Date (E.) Position (L.) Temperature at time of Salta |e ae a ave es a Collection of 2 |e Sample. @|§ Colour of Water, n s T t Cees f f Month. Day.) Hour.| Lat. Long. —= rae = = of of Bi 3 the the Z2|G4 Air. | Water. 1903. rare ue Pe WBe alk 309 289 Mar. | 16 | 9.55 Same] as 289 | Sounding-tube 309 | 290} Mar. | 16 | 20.0 29-0} 28-9 Steaming through —d-7 (Gi), ypancake cel; some in sampl 310 | 291) Mar. | 17 | 8.0 Surface] 29-9 | 29-7 Outside edge of —1:3 C. pack Reeve’s French blue (light) 311 | 292| Mar. | 17 | 12.0 | 63 8S. | 42 30 W. 25:8 | 28-9 | Steaming through| —1:7C. pancake ice ; some in sample Reeve’s French blue (light) 312 | 293) Mar. | 17 | 16.0 22:0 | 29-5 Just stopped —1:39 C. for sounding outside edge 312 | 294) Mar. | 17 | 16.55 21-0 | 31:3 Buchanan-Rich- —0:39 C, ard water- bottle Thermometer No. 102,504 312 |295a| Mar. | 17 | 17.0 21:0 | 33-44 Large Buchanan- 0-80 C. Richard water- bottle Thermometer No. 102,509 Data Relating to the Determination of the Density of th Time and Date. 8.30 18/3/03 15.50 17/3/03 16.20 9.30 18/3/03 10.30 18/3/03 11.0 18/3/03 11.30 18/3/03 Whee it 2 CPESCMN a: Spi] CHC SG OC9 ORSS X| NSOwWBONS whe 1-9936 1-9679 20205 Temperature during Weights Experiment. | added to Hydro ye t’. meter (grams) of of the the Air. |Sample.| w. cab °C) 62-2 15:6 5:30 16-8 C. 5:35 5:40 5:45 5-50 15-6 5:55 15-6 5-425 52:8 15:7 5:05 11:6 C. 5:15 5:25 5:35 15:3 5°45 5:55 15:3 5:65 15:5 5:35 55:8 17-9 5:20 3-2 C. 5-30 5:40 5:50 17-4 5:60 5-70 17-2 5-80 17-55 5-50 57:0 17:5 5:10 13:9 C. 5:20 5:30 5:40 17-1 5:50 5:60 16:9 5-70 17-2 5:40 60:0 16-9 5:15 15-6 C. 5:25 5:35 5:45 16-8 5-55 5:65 16:8 5:75 16-85 | 5-45 60-9 16-1 5:30 16-1 C. 5-40 5:50 5:60 15-8 5:70 5-80 5:90 15:6 6:00 15°85 5:65 56-6 15-4 5-40 13:7 C, 5-45 5:50 5:55 15-1 5-60 5-65 5:70 5:75 14:8 5-80 1550 60 17-9308 mersed (cub. Vv. 2-0438 | 1-02458) 15471 1078 +3392 1-02 432 2} 1-5471 *1051 -3157 102474 1-471 | . 0989 . +3745 1-5471 0942 +2895 SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. Data Relating to the Collection of the Sainple. Data Relating to the Determination of the Density of the Sample. 121 i | | +4 Depth Volum : Position (L.) : Temperature Temperature ne Density of Sample. in Fathoms. | “4s time of caning Weights ofim- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) D. d. Sample. me ing Portion ee Colour of Water, } Time Hydro oe oH ~ of |. from T t Current, and ’} and TVs ise qeter ie: oe Ob-_ |Reduced) Reduced lone Sea eee : j Remarks. Date. (grams). | ™°?* (oub. served Z fot. to : A =e — eae att’. |15°-56C. t. Posi- eample of of of of aye) tion} Goj. | the the the the L. | lected. Air. | Water. Air. |Sample.| — w. R. V. Se, |4Siss6.| 4S, | | Sa aa on: 7 C: 1956 | Same} as 2954 | Sounding-tube 14.0 [54:3 14-0 5°35 4-2 ‘ 18/3/03 |12-4 C. 5:40 10-0 5:45 15-0 5:50 20-0 5:55 25:6 15471 5:60 30:6 -0861 13-6 5°65 35°8 +1827 13:8 5:50 20:2 1-8159 | 1:02587 | 1-02548)| 1-02775 100 | 21-0 | 30-62 Large Buchanan} 14.30 [55-3 14:8 5:30 1-0 —0:79C.) Richard-water} 18/3/03 |12-9 C.| - 5:40 11-6 bottle 5:50 22-5 Thermometer No. 5:60 33-0 102,513 14-7 5:70 44-0 5:80 53-4 5:90 63-5 1-5471 6-00 74:2 0917 14-6 6-10 85-0 +3899 14:7 570 45-1 2:0287 | 1:02577 | 1:02558 | 1:02786 500 | 20-9 | 32-73 Large Buchanan-} 15.0 = [65-4 15-9 5:30 2:0 0:40 C.| Richard water-] 18/3/03 |13-0 C 5:40 12: bottle 5:50 23-2 Thermometer No. 5:60 33-8 102,504 15-4 5:70 44-8 5:80 54:0 5:90 63-5 15471 x 6-00 | 74:8 | -0967 15-1 6-10 84:8 3953 15-5 5:70 435-7 20391 | 1:02571)| 1:02570 | 1-02797 1000 | 20-9 | 32:95 4 Large Buchanan-} 15.50 /55-0 15-2 5:30 0-5 0:53 C.| Richard water-} 18/3/03 |12-8 C. 5-40 11-0 bottle 5-50 22:0 Thermometer No. 5-60 32-8 102,509 15-0 5:70 43-6 5:80 54:0 5:90 64:5 15471 6-00 75:0 0936 14:8 6-10 84-6 3899 15-0 5:70 43-1 2:0306 | 1:02576 | 1-02564| 1-02792 1758 | 29-8 | 31-02 Large Buchanan} 10.25 |54-7 13-0 5-40 4-0 —0-55C.| Richard water-} 19/3/03 |12-6 C. 3:50 15-0 bottle 5:60 26:0 Thermometer No. 5-70 37-0 102,504 13-2 5:80 48-0 5:90 58-5 6-00 69-5 15471 6-10 79:8 0824 13-4 6-20 90-0 4297 13-2 5°80 47-5 2-0592 | 1:02614| 1:02565 | 1-02792 1775 | Same} as 2994 | Sounding-tube 12.20 |56-4 13:3 5:35 0:0 - 13-6 C. 5°40 5-2 5:45 10:8 5:50 16-7 5:55 21:8 1-5471 5:60 26:8 0827 13-2 5-65 32:0 1465 13-25 5°50 16-2 1:7763 | 1:02609 | 1-02560 1-02787 122 Date (E.) Month. |Day.| Hour. g| 3 sae Hes s|& D n ‘oso e| 3 2 E|& 3 | 3 Z2/|a 1903. 313 | 300 Mar. 313 | 301 Mar. 313 313 314 315 302 303 304 305 Mar. Mar. Mar. Mar. 18 18 18 18 19 19 11.10 11.20 11.25 12.0 8.0 12.0 Position (L.) Lat. ee Data Relating to the Collection of the Sample. Depth eminent é perature Temperature in Fathoms. | “4: time of ae Collection of Experiment. D. Sample. a Colour of Water, | Time at from t Current, and and tie t’. Sea ee Remarks, Date. atuleeeeen as Posi bits & of of of of tion| oy | the | _ the the | the L. lected Air. | Water Air. |Sample. fee an alt Ge 500 | 29-5 | 32-54 Large Buchanan+ 11.12 |57-0 13-0 0-30 C. Richard water-+} 19/3/03 |13-9 C. bottle Thermometer No. 102,509 13-4 13-6 13°38 1000 | 29-5 | 31-68 Large Buchanan-j 6.30-6.52|60-0 14-7 —0:18 C. Richard water-] 20/3/03 |15-6 C. bottle Thermometer No. 102,513 14:9 Teste by W.S. B. 60-0 | 15-0 15:6 C. 14:85 1500 | 29-5 | 31-23 Large Buchanan-]7.7-7.30 | .. 14-4 —0-42 C. Richard water-| 20/3/03 bottle Thermometer No, 102,504 14-4 Tested by W.S. B. lk 14:7 14:55 Surface] 29:3 | 29-6 In open pool, | 7.35-7.55|60:3 14:5 —1:30C. trawling 20/3/03 |15-7 C. Ice spicules in} - sample Reeve’s French 14:6 blue Tested by W.S. B. 14:8 14:65 Surface} 26-0 | 29-0 Afewice spicules} 10.0 |48-4 8:8 —1:70C. in sample 24/3/03 | 9-1 C. Reeve’s French blue 9-2 9-0 312 | 26:2 | 31-81 Large Buchanan-+} 10.15 |53-1 9-4 Bettom —0-11C. Richard water-} 24/3/03 |11-7 C. at 327 bottle fms. _ Thermometer No. 102,504 9-7 | 10:0 | | 9:7 | QA) ABAOOAanPew &©f NE OUMBADQUME co eS QAnakbwowmnm B®] DARIAaAH AHO Crorver cron eoouuaaa alananaa al aangananaan al agaaganag al anngangnag a}! eaogaunagan a NEOOSOABaQN A oooooo°coo Soy sz a Sis DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAY] Data Relating to the Determination of the Density of th Volume of im- mersed Portion (cub. centims.) Vis Data Relating to the Collection of the Sample. Temperature at time of Collection of Sample. t. 27-0 | 29-12 —1-6C, —1:6C. 29-4 —1-4C. 1D SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. Data Relating to the Determination of the Density of the Sample. Temperature Volume Density of Sample. during of im- (Density of Distilled Experiment. Read- | mersed Water at 4° C.=1.) : ing Boren Colour of Water, | Time of 0 7 a Current, and and i Hydro- | Hydro-) q},. Reduced | Remarks. Date. Eng eats served to ee centims.), att’. |15°-56C.) the Sample. R. Vie 49, | 4S ine-5e. ms In open pool} 10.45 9:8 5:50 21-1 outside edge of} 24/3/03 5-60 32-2 pack 510 | 43:2 | 1-5471 Reeve’s French 5:80 54:0 0624 10-2 5:90 65-0 +3899 10-0 5:70 43:1 1:9994 | 1-02593 | 1:02485 11.0 9-5 5:50 6:5 24/3/03 5:60 Viel 5:70 28:7 Thermometer No. 5:80 39-9 102,513 10-0 5:90 50:8 6-00 61:2 6-10 71:9 1-5471 . 82-2 0615 10-2 89-9 4514 9-85 49-9 2:0600 | 1:02669 | 1:02557 11.30 9-8 6-1 24/3/03 16-9 27:9 Thermometer No. 38-8 102,504 10:2 49-9 60-7 71-2 1:5471 82:0 0636 10-6 92:3 -4478 10-2 49-5 2:0585 | 1:02670| 1:02564 Large Buchanan-| 12.0 10-4 6:0 Richard water-] 24/3/03 Wyott bottle 28-2 Thermometer No. 39-4 102,513 10-7 50:8 61:4 71:7 1-5471 82-0 0661 10:8 92-8 +4514 10-6 49-9 2-0646 | 1:02611| 1-:02512 14,15 10-0 5-1 24/3/03 16-5 27-1 Thermometer No. 38:6 102,504 10-2 49-6 60-1 70:6 1-5471 79-8 -0640 10:5 90-2 4396 10-25 48-6 2:0507 | 102647 | 1:02542 Reeve’s French | 13.45 9-6 22-2 blue 24/3/03 33-0 43-9 1:5471 54:9 0599 9-6 65-6 +3971 9:6 43-9 20041 | 1-02591) 1:02475 Outside edge of} 14.0 9-8 34:5 : pack 24/3/03 45-0 Ice spicules in 50°7 15471 sample 66-0 -0612 Reeve’s French 9-8 76:5 +5030 blue | 55-6 1:02530 | 1:02417 124 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA TT Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the §; : sys Denti | Date (E.) Position (L.) : Temperature | Temperature 2 rE pad in Fathoms. at fime of during Weights = 3 ata! Collection of Experiment. | added S|] ee] D. d. Sample. = 5 : _ | Colour of Water, } Time am | a of ee t Current, and and EG “ | Month. |Day.| Hour Lat, Long. eae the | ——— gee Teor PASE o oe 2; Posi- Sample of of of = =) tion i the the the R= L. Jeohed: Air. | Water. Air, 1908. | ie ht! ells aig as a 320 | 313} Mar. | 22 | 12.0 | Lewthwajte Chan- Surface] 24-7 | 30-0 Reeve’s French | 14.40 {53-2 10-2 5:50 i nel, South) Orkneys. | —1:1C. blue 24/3/03 |11-8 C. a 5:80 10-4 5:90 10-3 5:70 321 | 314) Mar. | 23 | 10.15 Surface} 23:0 | 29-8 Reeve’s French | 15.0 [54:2 10-2 5:50 —1-:2C, blue 24/3/03 |12-3 C. 5-60 5:70 5:80 10-5 5:90 10-35 5:70 322 |315| Mar. | 25 | 9.0 Surface} 31:0] 29:5 Open water, no} 10.15 |56-2 10:7 5°30 —14C.| ice anywhere} 2/4/03 |13-4C. 5:40 near 5:50 Reeve’s French 5:60 blue (pale) 11-0 5:70 : 5:80 5:90 6-00 11-2 6-10 10:95 | 5-70 323 | 316| Mar. | 25 | 12.0 | Abreast | Murray Surface] 30-5 | 29-8 Reeve’s French | 10.40 [57-3 10-6 5:30 Island|s, South —1:2C blue 2/4/03 |14-:0 C. 5:40 Orknelys. 5:50 5:60 11-1 5:70 5:80 5:90 6:00 10-9 6:10 10:75 570 325 |317| June | 11 | 14.0 Bay A, |South 23-2| 28-9 | Sample taken] 10.30 [48-8 75 5:50 Orkneys. /-170C.| from freshly | 14/6/03 | 9-3 C. 5:60 dug hole in 5:70 bay ice 5:80 78 5-90 6-00 6-10 6-20 19 6:30 7-7 5:90 | 325 |318a) June | 11 Upper surface of a piece of ice 23 | 23-6 0-3 0. 11.35 {51-4 8-9 0:90 inches thick taken from a hole 14/6/03 |10-8 C. 1:00 in ice in Bay A. 1:10 Temperature taken, after some of 1:20 the ice had been melted, with 9-1 1:30 thermometer belonging to No. 1-40 14 hydrometer. 1-50 1-60 9:3 1-70 9-1 1:30 26:3 37-0 47-9 58-7 69-5 47-9 25-0 35-7 46-8 57-1 67-6 46-4 Volume of im- mersed Portion 1:5471 -0646 4197 20314 20584 4288 47-4 20327 1 De { (De: W: 1-028: A i | ne ne || ne Data Relating to the Collection of the Sample. exe Depth ; Position (L.) ean HOTnE sporupesaiute ae an an ee Collection of D. d. Sample. f - ou of Wynter, tom urrent, an at which | T. a Remarks. Lat. Long. at the Posi-|S@™Ple) of | of tion nie the the L. Teered Air. | Water. °o , ° , °o ne ° (On | Same as 3184, second | At 14.30 hrs.) 0-2 Taken in Jabora- melting. tory (more ice Temperature taken with than water in thermometer belonging sample) to No. 25 hydrometer. | At 14.4 | Tested by W.S. B. at 10.5-10.25 on 15th June 1903. Middle part of the same piece of ice as sample 3184 taken from. First melting. Temperature taken with thermometer be- longing to No. 14 hydrometer. Same as 319A, second | At 14.3 melting. Temperature taken with thermometer belonging to No. 25 hydrometer. | At 14.4 Tested by W.S. B. at 11.15-11.34 on the 15th June 1903. Bottom surface of the same piece of ice as samples 318 and 319 taken from. First melting. Temperature taken with thermometer be- longing to No. 14 hydrometer. At 16.1 At 19.0 At 16.1 At 19.0 5 brs. 0-0 Taken outside (more ice than water in sample) 5 hrs. 0-0 Taken in labora- tory hrs. 0-2 Do. do. Tested by W.S. B. 23-6 | —0-2 0 hrs.| —0-2 Taken in labora- tory (more ice than water in sample) 5 hrs.| —0:3 Taken outside (more ice than water in sample) 0 hrs.| —0:2 Taken in labora- tory hrs, —0-1 Do. do. Tested by W.S. B. 23-6 | —0-4 TD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. Volume of im- mersed Portion (cub. centims.)) Data Relating to the Determination of the Density of the Sample. Density of Sample. (Density of Distilled Water at 4° C.=1.) 125 Ob- |Reduced Reduced served to to att’. |15°-56C. t. Wa sSty, 4S 157-56, 4S, 1-5471 0568 4306 20345 1-5471 0566 -2280 18317 15471 0474 2967 1:8912 1-5471 0643 +3609 1:97.23 1:5471 0596 +3953 2-0020 1-004096|1-003366 1-004103)1-00337: 1-:00696 | 1:00610 100074) 0-99996 100076 | 0:99997 100785) 1-00680 So 126 Date (E.) Number of Station Number of Sample. Month. Day. Hour. Data Relating to the Collection of the Sample. 325 (3208 325 |3184|) June | 325 3183) June 325 |3194| June 325 3198) June 325 3204 June June 325 '320B : / 11 11 11 11 11 11 11 Depth Position (L.) io Wathen. D. d. Lat. ° , ° , Same as 320A, second melting. Temperature taken with thermo- meter belonging to No, 25 hydrometer. Tested by W.S. B. at 11.50-12.7 on the 15th June 1903. Upper surface of a piece of ice 23 inches thick taken from a hole in ice in Scotia Bay. Temperature taken, after some of the ice had been melted, with thermometer belonging to No. 14 hydrometer. Same as 318a, second melting. Temperature taken with thermo- meter belonging to No. 25 hydrometer. Middle part of the same piece of ice as sample 3184 taken from, First melting. Temperature taken with thermometer be- longing to No. 14 hydrometer. Same as 3194, second melting. Temperature taken with thermo- meter belonging to No. 25 hydrometer. iece 319 Bottom surface of the same of ice as samples 318 an taken from, First melting. Temperature taken with thermometer be- longing to No. 14 hydrometer. Same as 3204, second melting. Temperature taken with thermo- meter belonging to No. 25 hydrometer. Temperature at time of Collection of Sample. At t of of the the Air. | Water. ee POH ate —0:25 23:6 0:3 0:3 23:6 | —0-2 —0-2 23-6 | —0-4 —0:25 Colour of Water, Current, and Remarks, Very little ice in sample Data Relating to the Determination of the Density of the DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT Weights added to | Hydro- meter (grams). Ww. | Temperature during Experiment. Time and Te t’. Date. of of _ the the | Air. |Sample. oh, OC 11.50 [56-0 9-6 15/6/03 |13+3 C. 56:3 | 10:6 13:5 C. 10-1 «» (06:0 9-6 13:3 C. 10-4 10-0 Tested |39-0 41 again at | 3-9 C, 11.20 30/11/03 42 ; 4:15 Tested |39-2 4-0 again at} 4:0 C. 11.45 30/11/03 4-1 4:05 Tested |39-8 4:0 again at| 4:3 C. 11.55_ 30/11/03 4-1 4:05 Tested |39-8 4-2 again at | 4:3 C 12.5 30/11/03 4:3 4:25 Tested |389:9 |. 4-2 again at | 44C, 13.20 30/11/03 4-4 4:3 Tested - |40-1 4-2 again at | 4:5 C 13.40 30/11/03 4-4 4-3 1-80 1-85 1:65 1-00 1-10 1-20 1:30 1-40 1-20 0:80 0:90 1-00 1:10 1-20 1:00 2-10 2:15 2-20 2-25 2:30 2-20 1-50 1-60 1-70 1:80 1-90 1:70 2-00 2-10 2:20 2:30 2-40 2:20 1-50 1-60 1-70 1:80 1-90 1:70 Volume of im- Read- | mersed ing | Portion of of Hydro- | Hydro- meter. | meter (cub. centims.) Re i, 454 2:8 13:8 24:0 35-2 47-2 58:8 71:0 15471 82-0 0630 93:0 4297 47-5 2:03898 | 1:00455 2:8 9-5 13:8 17-4 24:0 30:3 35:2 1-5471 43-2 0624 AT-2 *2243 24:8 1:8338 14:0 24:5 36:0 1:5471 47-2 +0259 58:0 +3248 85-9 1:8978 11:0 22:2 33°5 1:5471 44-0 0253 55-0 -2994 33-1 1:8718 9-2 14:5 19:8 1-5471 24-8 0253 30:4 +1782 19:7 1:7506 18-0 28:5 39-5 15471 50-0 0265 61-0 +3564 39-4 1-9300 12:5 24-0 35-0 15471 46-0 0268 57-0 3157 34:9 1:8896 9-5 20:0 30:5 1-6471 41-2 0268 52:2 2777 30:7 | 18516 SD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 127 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ti Depth ae Volum : jE.) Position (L.) : Temperature Temperature OOO Density of Sample. in Fathoms. | ~) time of during Weights of im- (Density of Distilled Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. da Sampie. Gy ing Ronen Tale ese Colour of Water, | Time Hydro- or e aos from lee Current, and ‘ and Ts te meter Bee Be Ob-_ |Reduced|Reduced Sea | Which 5; ; Remarks. Date. (grams). a served to to Tat. | Long. |” | the : == Supa att Sea. te Posi- Becile of of of of ° poate) i tion rote the the the the L. Tegtede Air. | Water. Air. |Sample.| w. 1Ne Vie 49. |4Sises6.| 4St. Bia? omy? ons cabs Bane as Water frjom hummiockin| Scotia] .. As AD 9.0 |89:8 4:0 0-70 7-0 Bay. 30/11/03 | 4-3 C. 0:80 | 17-5 0:90 28:5 1-5471 1-00 39-0 0253 , 4-1 1-10 49-8 +2569 4:05 0:90 28-4 1:8293 | 1:00049 | 0-99951 12.0 | 5943S. |4810W.| .. | Surface] 34:0] 32-7 Reeve’s French | 10.15 |59-0 13-6 5-50 24:0 0-4. blue 12/12/03 |15-0 C. 5-60 35-0 5-10 45:5 15471 F 5:80 57:0 0855 13-8 5:90 67-5 4143 13-7 5:70 45°8 2:0469 | 1:02566 | 1:02527 | 1-02747 18.0 | 5923S. |49 8W.| .. | Surface] 34:3) 33-0 Pale green 14.0 |59-2 14:0 5:30 7-5 0-6 C. 12/12/03 |15-1 ©. 5-40 | 17-8 5:50 28-5 1:5471 5:60 39-0 +0880 14-2 5-70 50-5 +2596 14-1 5:50 28:7 1:8947 | 1-02542) 1:02510 1-027 26 12.0 | 5828S. |5156 W.| .. | Surface] 31:8} 33-3 Pale blue 14.20 59-8 14-2 5:50 30-5 0-7 C. 12/12/08 |15-4 C. 5:60 | 42-0 5:70 53-0 15471 5-80 63:5 0892 14:4 5:90 74:5 4767 14-3 5:70 527 2-1130 | 1-02529) 1-02501| 1-02717 18.0 | 58 OS. |5310W.!| .. | Surface| 33-6] 35-0 Light blue 18.15 |62-0 15-3 5:30 11:5 176. 13/12/03 |16-7 C. 5-40 | 23-0 5:50 34:5 1:5471 5-60 45-5 -0961 15:5 5:70 55°8 +3085 15-4 5-50 34-1 1:9517 | 1-02510 | 1:02507 | 1-02717 i) 9.0 | 5710S. | 5535 W.| .. | Surface] 30-9} 33-0 Reeve’s French | 16.30 /|62:0 15-4 5:50 35-0 0-6 C. blue 13/12/03 |16-7 C. 5:60 | 45:0 ' 5:70 56-0 1:5471 5-80 67-5 0964 15:5 5:90 78-0 +5093 15-45 5:70 56-3 2:1528 | 1:02507| 1-02505| 1-02720 re 18.0 | 56548. |56 24 W.| .. | Surface] 30:8] 35-2 Reeve’s French | 18.50 |62-2 15-6 5:50 34-8 1-8C. blue 13/12/03 |26-8 C. 5:60 | 45-5 5-70 56:0 15471 5-80 67-0 +0980 15-8 5:90 78-0 6093 15-7 5:70 56:3 21544 | 1-02506| 1-:02509| 1:02717 10.0 |55 5S. |5725 W.| .. | Surface! 29:0| 42-2 Dull blue 14.40 |62-0 15:8 5:50 35:0 o70. 14/12/03 |16-7 C. 5:60 45-5 5-70 57:0 1-5471 5:80 67-0 -0989 15:9 5:90 77-2 +6093 15-85 5:70 56-3 2-1553 | 102506) 1-01513| 1-02683 Surface] 35-6 | 42-0 Reeve’s French | 15.0 |62-0 15-8 5:50 33-2 5:6 C. blue 14/12/03 |16-7 C. 5-60 45-0 5-70 55:2 1-5471 5:80 66-2 0989 15-9 5:90 7T-0 -3002 15-85 5:70 55-3 2-1462 | 1:02510| 1-02517 | 1-02688 . SOC. EDIN., VOL. LI, PART I (NO. 4). 19° 128 DR W. 8. BRUCE, MR A, KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GR. Data Relating to the Collection of the Sample. ae Depth Date (E.) Position (L.) in Fath a 3 = : D. d. 8 2 from al . of | which © | 2 |Month.'Day,|Hour.| Lat. Long. the 8 o at S iT Soe Posi) Pause EIE tion | Was Slee 1 col- ol lng * | lected. 1903. ex. ie 346 | 330, Dec. 1 | 19.0 | 5425S. | 57 32 W. Surface 348 | 331! Dec. 2 | 12.0 | 5211S. | 5755 W. Surface 350 | 332} Dec. | 12 | 12.0 | 44 8S. |5730 W.] .. | Surface 352 | 333] Dec. | 13 | 12.0 | 42 30S. | 59 18 W. Surface 353 | 334} Dec. | 14 | 12.0 | 4032S. | 58 33 W. Surface 354 | 335} Dec. | 15 | 12.0 | 38 24S. | 57 42 w. Surface 355 | 336| Dec. | 16 | 12.0 | Off Cape |Corientes. Surface | ) Mogotes |light10-20) miles} distant, Temperature at time of Collection of Sample. Colour of Water, T t Current, and ; . Remarks, of of the the Air. | Water. Saeelll lal 33°8 | 41:8 Reeve’s French 540 blue 40:2} 44-9 Pale green 7-20. 51:3 | 49-3 Green : 9:6 C, 53:0} 50-0 Pale greenish 10-0 C. blue 56:0 | 54-8 Pale pea green 12:7 C. 62:3 | 61:8 Pale sage green 16-6 C. : 63-0 | 62-0 Pale green 16-7 C. Data Relating to the Determination of the Density o Time and Date. 15.20 14/12/03 15.40 14/12/03 9.45 15/12/03 10.0 15/12/03 10.20 15/12/03 14.0 16.30 Temperature during Weights Experiment. | added a e to Hydro- T% t’ meter ‘ (grams) of of the the Air, |Sample.| w. oa a (6) 162-2 16-0 5:50 16-8 C. 5-60 5:70 5:80 16-1 5:90 16-05 5:70 62-2 16:2 5:50 16-8 C. 5-60 : 5:70 5:80 16:3 5:90 16:25 5:70 1-2 17:8 5:30 16:2 C. 5-40 5-50 5-60 18-1 5:70 17:95 5:50 2-0 17-9 5:30 16:7 C. 5-40 5:50 5-60 18-1 5:70 18-0 5:50 62-0 18-0 5:30 16:7 C. 5-40 5:50 5-60 18-2 5:70 18:1 5-50 69-0 18-4 5:10 20-6 C. 5:20 5:30 5-40 18-6 5:50 5:60 5:70 5:80 18-8 5-90 18:6 5:50 168-0 19-0 5-10 20-0 C, 5:20 | 5:30 5:40 19-2 5:50 5-60 5:70 5-80 19-4 5:90 19-2 5-50 34-0 45:5 56-2 66-5 77-0 55:8 40-0 50-2 61-0 72-0 82-5 61-1 19-5 30:8 41:8 52:5 63-0 41-5 26-0 37-2 48-5 59-5 70-0 48-2 27-5 A] aIaAcne || BPLoe co | S@Ow SMBAQMNEWH | ROMY SOM 3 Nonnoodnodrs a “AQ Cre tb BOWE SSO NOADUNDOO | cr | 00 RO) Pore mae Of the Volume of im- mersed Portion 15471 -1001 +5048 2:1520 | 1-02507 15471 -1014 5527 2.2012 | 1-02480| 15471 +1161 4677 2-1309 +4740 2:1409 | 1-0240 ) SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 129 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ee Depth Volume - Position (L.) F Temperature Temperature Density of Sample. in Fathoms. | “3+ time of during —_| Weights ofim- | (Density of Distilled ag ae Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. d. Sample. kG ing Honea as, Colour of Water,} Time | ~ Hydro- | -, of aan oF from T t Current, and 2 and KA 1, meter ee Hy om Ob- |Reduced| Reduced Sea | Which : : Remarks, Date. (grams). 7 (cub served to to Long. | ‘at . Oe nox f we ane | abt’. |15°-56C.| t. Posi- | are | of of of of ms = tion se the the the the L. lected. Air. | Water Air. Sample. Ww. R. V. 494, 49 15°-56, 4 Si ° , ° ’ ° i ° iE. ° ihe ° CG. 12.0 | Off Mogo tes light. Surface] 66-4) 61-5 Dark sage green} 14.0 |72-0 20:6 5:00 11-0 a 16-4.) ~ 21/12/03 |22-2 C. 5:10 22-2 5:20 33-0 5:30 44-2 20-9 5°40 55:0 [| 5:50 66-0 . 5:60 76-5 | 1:5471 % 5-70 | 87-2 | +1298 21-0 5:80 97-5 “4948 ’ 20-8 5-40 54-7 2-1717 | 1-02331| 1-02459 | 1-:02440 12.0 | Off Mendjana Point Surface] 70:0 | 67-2 Dirty sage green} 14.30 |72-5 21:0 4-20 9-5 (12 miles) bear- 19-6 C. 21/12/03 |22-5 C. 4:25 14:0 ing W. 4:30 20:0 4:35 25:5 21-2 4-40, 31-5 4-45 36:5 4:50 42-5 1-5471 4:55 47-5 +1320 21:3 4-60 53-0 2813 21-15 4-40 31-1 1-9604 | 1-01901} 1:02035 | 1:01943 36 57S. [55 45 W. Surface) 73-1 |} 73-0 Pale green 14.0 |62-4 17-1 4:30 140 22-8 © 25/1/04 |16-9 C. 4-40 25-0 4:50 36:8 1-5471 4-60 47-0 +1067 17-1 4-70 57:5 +3266 17-1 4-50 36-1 1-9804 | 1:01944| 1-01978| 1-01800 | 39 245. [55 2 W. Surface] 63-4 | 58-2 Pale blue 14.20 |62-2 16:9 5-30 19-0 14:6 C 25/1/04 |16-8 C. 5:40 30:5 5:50 41:5 1-5471 5:60 52-5 +1055 16-9 5-70 63-0 +3736 16-9 5:50 41-3 2-0262 | 1:02468) 1-02498| 1:02519 4059S. |55 4 W. Surface) 58:0} 61:6 Light blue 14.35 |62-1 16-7 5:20 11-0 16-4 0, 16:7 C. 5:30 21:0 5-40 32-5 1-5471 5:50 43-8 -1045 16-8 5:60 O45 2949 by 16-75 5-40 32-6 1:9465 | 1:02458| 1-02485 | 1:02466 43 33S. |55 7 W. Surface} 54-7 | 60-0 Blue 14.0 61-8 15:8 5:20 7-5 15-56 C, 28/1/04 |16-6 C. 5-30 18-2 5:40 29-0 5-50 40-5 16:0 5:60 51-0 1-5471 5:70 61-5 0992 16-0 5-80 71-8 -3609 ’ 15-9 5-50 39-9 2:0072 | 102479 | 1:02487 | 1:02487 45 31S. | 55 21 W. Surface 55-5} 56-1 Grey blue 14.20 (62-0 15-1 5-50 31-5 13-4 C 16:7 C, 5:55 36:5 5:60 42:3 5:65 48:0 15:3 5-70 53-5 5-75 59-0 5:80 64:0 1:5471 5:85 69-2 0955 15:5 5:90 74:8 +4812 15-3 5:70 53-2 2-1288 | 1:02523) 1:02518| 1:02563 130 Number of Station. 365 366 368 369 370 372 373 Number of Sample. | | 344 345 346 347 348 350 351 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAYVIT Date (E.) Month. |Day.} Hour. 1904. | Jan. Jan. Feb. Feb. Feb. Feb. Feb. Feb. 29 30 10 10 11 11 12 12 12.0 12.0 20.0 12.0 20.0 12.0 24.0 Position (L.) Lat. Long. 47 478, [56 8 W. 50 358. | 5758 W. 53 225. |56 5 W. 64 35S. | 55 23 W. 55 47 S. |5419 W. 5618S. | 53-20 W. 57 47S. | 51 40 W. 58 53S. | 50 40 W. Depth in Fathoms. D. d. lected. Surface Surface Surface Surface 109 Surface Surface Surface Surface Data Relating to the Collection of the Sample. Temperature at time of Collection of Sample. _| Colour of Water, T. j i Current, and Remarks. of of the the Air. | Water. oR. aN 55:1} 54:2 Grey blue 12:3 C. 51-2 | 50-2 Green 10-1 C. 50-1 | 46-9 Blue 8-3C 45:0 | 44-9 Too dark to see 7-2C 44-1} 39-8 Reeve’s French 4:3(C, blue 43:9 | 41-8 Reeve’s French 5-40. blue Reeve’s French blue 39-0 Je) se GSO a 35:8 | 35. vi 209 fo) Data Relating to the Determination of the Density of th Temperature during Experiment. Time and HN £7, Date. | of of the the Air. |Sample. Sane aC 13.35 (55-5 12:9 80/1/04 |13-1 C 13-0 12:95 13.50 |56-0 12:0 13:3 Cy 12-3 12:15 9.15 |49-3 6-4 17/2/04 | 9-6C. 7-0 6-7 9.80 {51-2 6:8 17/2/04 |10-7 C. 7-0 7-6 7-13 10.0 {53:6 7-4 17/2/04 |12-0 C. 7-8 7:6 10.20 [54-7 8:3 17/2/04 |12:6 C. 8-7 8:5 10.40 [55-0 8-6 17/2/04 |12-8 C. 9-1 8-85 11.0 [55:3 8-7 17/2/04 |12-9 C. “91 89 / of im- Read- | mersed ing | Portion of Hydro- meter R. v. 24-0 29-5 q 345 | 1-5471 40:0 | -0808 45-0 | +3130 34-6 | 1-9409 | 1-02571 5-0 10-5 16-2 | 1-5471 21:0 | -0758 27-0 | +1438 4 ——— | 15-9 | 1-7667 | 1-02559] 6-0 17-5 | 28-0 | 1-5471 38-8 | -0418 | 50-1 | +2542 | 28-1 | 1-8431 | 1-02681|h 13 a 19-0 29-9 | 40:2 51-0 62-1 72-8 | 1-5471 83-1 | -0445 92:3 | -4604 50-9 | 20520 12-0 23-0 34:6 | 15471 45-0 | -0474 55-8 | +3085 34-1 | 1-9030 13-3 24-0 35-0 | 1-5471 453 | +0530 56-2 | -3148 34-8 | 1-9149 15-5 26-5 37-2 | 1-5471 48-2 | -0552 59-2 | -3374 37-3 | 1-9397 10-3 21-0 31-5 | 1-5471 42-5 | +0555 53-5 | +2877 31-8 | 1-8908 y Volume SD SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 131 Data Relating to the Collection of the Sample. i, op Depth ates. Position (L.) : Temperature ) ( in Fathoms. a hee of Collection of D d Sample. i om 3 polaue of Yai f % urrent, an ank which | T. ee Remarks. f.| Hour.; Lat. Long at NS) Posi seople of of tion shed the the L. ieoted Air. | Water. i Sats ort Oe Sakis | J) 12.0 | 5956S. | 49 30 W. Surface] 33-7 | 36-0 Reeve’s French 2-2 C. blue 11.40 Scotija Bay Surface] 31-9 see 4 Dull blue Norr sy D. W. Witton. : 23rd Feb. 1904. ll nly I’ve always left the thermometer in the ii the hydrometer whilst testing sample: m Hw onwards I read the t’, then take the moleter out, and after sample has been bed put thermometer in again for a second dingf temperature. i} 12.0 | 6128S. 4155 W. Surface| 30:0] 33-3 | Reeve’s French 0-7 C. blue | i m/ 18.0 | 6151S. |41 6W. Surface} 29-5 | 33-0 Reeve’s French 0-6 C. blue 24] 6.0 | 6230S. |39 8 W. Surface) 29:5 | 32-1 Reeve’s French : | 0-1C. blue ‘ i | } _ t ° | 24, 12.0 | 6249S. |3812 W.| .. | Surface| 29-1] 31-6 Reeve’s French —0:2C blue Two bergs) to wind|ward jand about |three milejs off. | Data Relating to the Determination of the Density of the Sample. Time and Date. 11.15 17/2/04 14.15 23/2/04 13.50 24/2/04 14,15 24/2/04 14.40 24/2/04 13.45 25/2/04 Temperature Volume Density of Sample. during Weights of im- (Density of Distilled Experiment. | added Read- | mersed Water at 4° C.=1.) to ing Portion Hydro- of of 1 tie meter Hydro: Hyer Ob- |Reduced|Reduced (grams) Z served to to => (cub. | att’. |15°56C.| t centims,) : i . of of the the Air. Sample. w. R. V. 4Sy, 49 15°-56, Se, ae °C. do°2 9-2 5:50 13-8 12:9 C. 5:60 24:8 5:70 35-5 1-547 1 5:80 46:5 0580 9-4 5:90 57-0 3211 9-3 5:70 35:5 1:9262 | 1:02634| 1-:02514)| 1-02720 59:0 11:7 5:50 17:8 15:0 C. 5:05 23-0 5°60 28-0 5:65 34:0 11-9 5-70 40:5 5:75 46-0 5:80 50:5 1:5471 5:85 56-2 0743 12-1 5:90 62-0 -3600 11:9 5:70 39-8 1:9814 | 1:02603 | 1:02528 | 102743 57:3 13-4 5:50 28:0 14:1C 5-55 33-5 5:60 39-0 5:65 44:0 5:70 49:5 5:75 55:0 5:80 60:5 1:5471 5:85 65:5 0849 13-8 5:90 71-0 4487 13-6 5:70 49-6 2:0807 | 1:02547 | 1:02506 | 1:02721 60:0 16-2 5:20 7-0 15-6 C 5:30 17-5 5:40 28:2 5:50 39:0 5:60 49-5 5-70 60:0 5-80 70-0 15471 5:90 80:0 +1014 16:3 6-00 90:5 +4442 16-25 5:60 49-1 2:0927 | 1:02486| 1:02502) 1:02715 60:0 |. 16-2 5:20 6-0 15-6 C. 5:25 11:0 5:30 16:8 5:40 27-0 5:50 38:0 5:60 49:0 bd 5:70 59-5 15471 5°75 65:0 +1011 16-2 5:80 69-8 +3437 16-2 5:50 38-0 1:9919 | 1:02487 | 1:02502)| 1:02715 56-2 16-1 5:20 6:0 13-4 C, 5:25 11:5 5:30 17:0 5:40 27-2 5:50 38-0 5:60 48-8 5:70 59-8 1-5471 5:75 65:0 0995 15-8 5:80 70-0 +3447 15:95 5°50 38-1 1:9913 | 1:02488| 1:02497 | 1:02714 132 Data Relating to the Collection of the Sample. DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT Data Relating to the Determination of the Density « = 1 ‘ ae Depth ‘ : Date (E.) Position (L.) 2 piiRathome. ery era te he —| Collection of & e D. d. Sample. =| 5 rm n rom =a s See which | T. t. | s | = |Month. Day.) Hour. Lat Long “ak the |—— | Sa BS Sample = E Bosi-|~* of of | s/s tion} Go | the the eae L. leaked Air. | Water. 1904. | a ear | °F, enh 381 | 358} Feb. | 24 | 18.0 | 6313S. | 37 33 W. Surface) 30-0 ae a Two or} three berjgs in |the j distance. 382 | 359) Feb. | 25 | 6.0 | 64 58S. |3610W. Surface, 29-3 | 30-8 : —0:7 C. Two or) three berjgs in |the distance. 383 | 360; Feb. | 25 | 12.0 | 64298. | 35 29 W. Surface 29-8 | 31-0 a —0-6 A few bejrgs far away. 384 | 361; Feb. | 25 | 18.0 -| 64388. | 35 13 W. Surface) 29-8 a @ A few bel|rgs far awlay. | 385 | 362) Feb. | 26 | 6.0 | 6535S. | 33 50 W. Surface) 28-8 zee é —0:8 C. Two icjebergs. : 387 | 363) Feb. | 26 | 12.0 | 6559S. |33 6 W. | 2625| Surface] 28-8 ee a Ice abou|t1 mile off) ; loosje pieces i quite near) ship. Colour of Water, Current, and Remarks. Reeve’s French blue Reeve’s French — blue Reeve’s French blue Reeve’s French blue (light) Reeve’s French blue Reeve’s French blue (light) Toupee Voltts | eee Weights Read- mene o Experiment. | added ne Postiaa W; Time of and IN t’. Date. of of the the Air. |Sample.} — w. R. Vi 4Sy, ot, eG. 14.5 [57-0 15-4 5:30 15:8 25/2/04 |13-9 C. 5:35 | 208 5:40 26-0 5:50 36-2 5:60 AT-0 5:70 BY G5) 5:80 68-0 1:5471 5°85 73:0 0955 15-2 5:90 78:2 +4243 2 153 5:60 76-9 20669 | 1:02500 14.385 57-2 15:0 5:30 13-5 14:0 C. 5:35 19-0 5:40 24-2 5:00 34:8 5:60 45-0 5-10 56-0 5:80 66:3 1:5471 5:85 715 +0930 14:8 5:90 76:5 +4089 14-9 5:60 45-2 2:0490 | 1-02: 13.35 |54-:2 155 5:30 16-0 26/2/04 |12-3 C. 5:35 21-0 5:40 26-0 5:50 36:6 5:60 47-0 5:70 By Gs) 5:80 68-0 1:5471 5:85 73-0 0977 15:8 5:90 78-0 +4251 15-65 5:60 47-0 2:0699 | 1-02 14.0 58°8 16-0 5:30 12:8 26/2/04 |13-2 C. 5:35 18-0 5:40 23-0 5:50 33:5 5:60 44-0 5:70 54-2 5-80 65:5 1:5471 5:85 70:5 0989 15-7 5:90 76-0 +3998 15-85 5:60 44-2 2:0458 | 1-02: 14.20 (57-0 14-6 5:30 8:5 13-9 C. 5:35 13:5 5:40 18-0 5-50 28-2 5-60 38:5 5:70 49-0 5:80 60-0 1:5471 5°85 65-0 0911 14-6 5:90 70:2 +3528 : 14:6 5:60 39:0 1-9910 14.45 |57-2 13:9 5:30 12-0 14:0 (0. 5:35 17-2 5:40 22:6 5:50 33° 5:60 44:5 5:70 55:0 5:80 65:5 1:5471 5:85 69:0 0867 13-9 5:90 74:8 +3962 13-9 5:60 43-8 2:0300 | 1-:02521 AB) SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 133 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. 5 oe Depth ; | Volume i ate). Position (L.) F Temperature Temperature Ml Density of Sample. | ) : \ in Fathoms. at orne of during Weights of im- (Density of Distilled {fo ——— | Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) | D. d. Sample. ie Ee pono ] Colour of Water, | Time | Hydro- tees 2 | i of | from T ; Current, and >| and age t, meter yee Bg atee Ob-_ |Reduced) Reduced \ : Sea | Which : : Remarks. Date. (grams). ta (cub served to to .|Hour.| Lat Long. at the Se ; Shan ano). att’. |15°-56C. 7 Posi-/Sample] os x Be ee centims.) tion Neng the | the the | the oa Cay L. | J, oat Air. | Water. Air, |Sample.| w. R. V. aSv, }4Sises6.| 4St, 7 | o¢ or oF, oF | ie lop. 0G. | 4 J = 2|| 7.0 | First piecje of loose jice | Surface} .. xe Bt 13.35 58-7 14-4 0-70 75 met wi|th this sea|son. 27/2/04 |14-8 C. 0:80 18-0 | 0-90 28-5 1:00 39-5 1:10 50:8 1:20 61:0 1:30 72:0 1-5471 1-40 83-0 -0899 14-4 1-50 93-5 4559 14-4 1-10 50-4 | .2-0929 | 1-00014 | 0-99995 7.0 |Same piecje ofice; mjore | Surface] .. ob Ap 13.55 |58-4 13:3 0-70 8-5 i ape than samjple 27/2/04 |14-7 C. 0-75 14-0 i 364a, 0:80 19-8 0:90 30:5 | 1:00 | 42-0 | 1:10 52°8 . 1:20 63-8 | 1-5471 1-25 69-0 0836 13-4 1-00 41-7 2:0079 | 100006 | 0-99974 18.0 | 66 6S. |3245 W.| .. | Surface] 28:8} 30-7 Reeve’s French | 14.15 {59-1 15:3 5-20 4-2 ; —0:7C. blue 27/2/04 |15-1 C. 5:25 9:3 Ship near ice. 5:30 14:5 : 5-40 25:2 5:50 36-2 5-60 47-0 5:70 57:5 1-5471 5:75 62-8 0952 15-2 5:80 67:8 +3266 15-25 5:50 36-1 1-:9689 | 1-02500 | 1:02494| 1-02715 | ; 135 1:30 74:5 *3772 6.0 | 6630S. | 3150 W.| .. | Surface} 26-7| 29-3 Reeve’s French 14.45 |59-0 14:3 5:30 125 : —1-5C. blue 15:0 C, 5:35 17:8 Tn among} loose pac\k ice. 5-40 22:8 : 5:50 33-5 5:60 44-5 5:70 55:2 5:80 66-0 1-5471 5:85 71:0 0892 14:3 5:90 76-0 -4016 14-3 5:60 44-4 2-0379 | 1:02517 | 1:02490| 1:02713 12.0 | 6626S. |3125 W.| .. | Surface] 28-4| 29-5 Reeve’s French 13.40 |54:0 13-0 5:30 10:0 —1:40C. blue 28/2/04 |12:2 C. 5:35 15:0 In among} loose pac|k and| young 5-40 20:0 ‘ ice. 5-50 31-0 : 5:60 A415 5:70 52°5 5:80 63-0 1-5471 : 5:85 68-5 -0817 13-2 5:90 74-0 *3772 13-1 5:60 41:7 2:0060 | 1:02535 | 1:02484) 1-02707 27) 16.40 | 66 14S. | 31 18 W. | 2630] 2625 | 28-2| 29-70*| Buchanan-Rich-| 15.30 |50-9 12:5 | 5-40 10:5 i 10-5 —1-3C. ard water-} 28/2/04 |10-5 C, ee 2) 15:8 Read on bottle 5-50 21-0 bridge | Thermometer No. 5-60 31-2 102,513 5:70 41:8 : 5:80 52:8 the t}rmometer reversed satisfactorily as far as the frame was 5:90 63-2 1:5471 the jlumn of mercury was not broken when the instrument | 5-95 68-2 0774 ‘(fompntil it was made to do.so by giving it asharp tap. The | 12:3 6-00 73-2 | -3799 all)pbability approximates to that of the surface waters. Cle Pe ee | 12: 5:70 42:0 | 2:0044 | 1-02590| 1:02526| 1-02751 134 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAVIT Data Relating to the Collection of the Sample. ae Depth Date (E.) Position (L.) oar Temperature in Fathoms. at time ms a ee 4 ollection of 8 = | | Dd. d. Sample. = =a ~ S nln from p | » |Month.!Day.! Hour.| Lat. Long the ra) ro) r = at Ss 1 | = = Posi ras Simos of 53/3 tion) go. | the the aR L. lected, | Ait- | Water. 1904, | nes Sy aH make 391 | 369} Feb. | 27 | 17.50 | 66148. | 3118 W.| 2630} 2000 | 28:0} 32:62 kept 0-3. C, Read on bridge 391 | 370| Feb. | 27 | 17.45 | 6614S. | 3118 W. | 2630) 1500 | 28:0] 31:58 kept : —0-2C. Read on bridge 391 | 371] Feb. | 27 | 18.0 | 6614S. |3118 W. Surface] 28-0} 30-2 A —1-0 In pacjk ice. 391 | 372| Feb. | 27 | 16.40] 6614S. | 3118 W.| 2630) .. | 28-2 | See 368. Crack in |small jar, |water| all ran out, {could not} test jit. 391 | 373| Feb. | 27 | 18.40} 6614S. | 3118 W. | 2630] 1000 | 28:9} 31-90 kept —0-1C. Read on bridge | 391 | 374| Feb. | 27 | 18.30 | 6614S. | 3118 W.| 2630) 500 | 28:9] 32-42 kept 0-2C. | Read on / bridge | Colour of Water, Current, and Remarks. Buchanan-Rich- ard water- bottle Thermometer No. 102,513 Buchanan-Rich- ard water- bottle Thermometer No. 102,512 Reeve’s French blue Buchanan’s sounding-tube Buchanan-Rich- ard water- bottle Thermometer No. 102,513 Buchanan-Rich- ard water- bottle Thermometer No. 102,512 Data Relating to the Determination of the Density of th Temperature during Experiment. Time and nig tis Date of On || the the Air. |Sample. oT °G, 15.50 51-7 11-6 28/2/04 |10-9 C 11:5 11:55 16.15 {54-0 11-2 28/2/04 |12-2 C. 11:6 11-4 14.0 {55-8 12-6 28/2/04 |13-2 C. 12:6 12:6 16.40 54-0 11:2 28/2/04 |12-2 C. 11-1 11-15 17.35 54-2 10:3 28/2/04 |12-3 C. 10:3 10°83 Weights added Ww. | Srna Sessaessas aAgooo°o°onoe 2 i) ARERR OEE 0 DAD Oh co co AoOoo0codnro a] a Si} o i=} }| 1S) COOOBAOUNES DPAAAMAA NAN SRSSSssss a a S SOU OUCHIEOX SOSOrnAAHQq Uc ASSSOOSAS alo coO}H KS) Volume of im- Read- | mersed ing Portion of Hydro- meter, AD) SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN, 135 Dita Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. z iti Depth Volume ; afie|5.) Position (L.) : _ | Temperature Temperature , Density of Sample. . in Fathoms: | “a5 fire of during Weights of im- (Density of Distilled i> =a ali Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) ; D. d. Sample. iG ing Pemen ___| Colour of Water, | Time Hydro- g ; e ! of | from T t. | Current, and and Wb ties meter ee: Hy Go Ob-_ |Reduced/ Reduced Sea which ; i Remarks. Date. (grams). pec een served to to Hour. Lat. Long ae viet | == (cub att’. |15°-56C. fy Posi-/SamPle} of of of of gents, : | tion ea the the the the | L. | jected. | Ait. | Water. Air. |Sample.| ww. Re Vi aS¢. | 4S ise-s6. | 4S. ‘ { * 4 i} || a TAMAS on els | | | 6.0 | 66 9S. | 2927 W.| .. | Surface] 27-1 | 29-7 Reeve’s French | 14.20 |55-9 11-7 5:30 5-5 —1:3C. blue 13:3 C. 5°35 10-5 Among 1 oose pack. | 5-40 16-0 5:50 26-0 5-60 37-0 5-70 47-5 } 5:80 | 57-6 | 1-5471 5-85 62:8 0733 11:8 5-90 68-2 *3329 11:75 5:60 36-8 1:9533 | 1:02564) 1-02487 | 1-02710 12.0 | 6621S, |2830W.| .. | Surface] 29-9} 29-8 Reeve’s French | 14.45 |55-0 11-2 5-40 15:5 : —1-2C. blue 12-8 C, 5-45 20:6 Among | oose piecesjof ice, 5:50 26:0 5:60 36:8 5-70 47-2 5:80 57-5 ; 5:90 | 682 | 1-5471 | 6-00 18:5 0699 | 11-2 6:10 88:5 -4414 , 11:2 5-7166| 48-8 2:0584 | 1:02569 | 1:02482| 1:02704 18.0 | 6643S. | 2755 W.| .. | Surface] 29-2] 29-8 Reeve’s French | 13.45 |59-3 11:8 5-30 6-5 . —1-20C. blue 29/2/04 |15-2 C. 5:35 11-0 Ice about one or) two | miles 5:40 16:3 ] away. 5:50 27:2 5:60 38-0 5-70 48-5 5:80 59-0 1-5471 5:85 65-0 0749 12-2 5-90 69-5 +3428 12-0 5-60 37-9 1-:9648 | 1:02558 | 1-02486| 1 02708 19.35 | 66 43'S. | 27 55 W.| 2685] 2680 | 29:0| 29-0 Buchanan-Rich-] 14.10 {57-0 12-0 5:50 13-2 —17C.| ard water-| 29/2/04 |13-9 C. 5:55 18-5 Read on bottle 5-60 23:6 bridge | Thermometer No. 5:70 34:0 102,513 5:80 44:8 Again mercury 5:90 56:0 column did 6-00 66-5 1:5471 not break. 6:05 72-0 0752 See No, 368 12-1 6-10 16:3 4071 12:05 | 5-80 45-0 2-0294 | 1:02631| 1:02558| 1-02787 zi) 19.30 | 66 43 S. | 27 55 W.| 2685] 2185 | 29-0] 32-42 | Buchanan-Rich-} 14.40 |55-0 11-7 5-40 2-6 0-2C. ard water- | 29/2/04 |12-8 C. 1| 5°45 7-8 Read on bottle 5:50 12-6 bridge | Thermometer No. 5:60 23:5 102,512 5:70 34-0 5-80 45-0 5:90 55°8 1-5471 5-95 61:0 0733 11:8 6-00 66-6 -3103 11:75 5:70 384-3 1:9307 | 1:02632)| 1-:02554| 1-02776 6.0 | 6740S. |2724W.| .. | Surface! 27-0] 29-6 Reeve’s French 15.25 |55-0 13:5 5:30 10-2 £)) BT eee —1:3C, blue 12:8 C.| 5:35 16:0 No ice in) sight. | 5-40 21-5 | 5:50 32:0 5:60 43-0 5-70 54:0 5:80 64:5 1-5471 5°85 69-6 0852 | 138 | 590 | 75:0 | -3881 13-65 | 5:60 42-9 2-0204 | 1.02527 | 1-02487 | 102710 RO. SOC. EDIN., VOL. LI, PART I (NO. 4). 20 136 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of th DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA) Time and Date. ‘Volume of im- mersed Portion Wa Date (E.) Position (L.) Temperature { at time of Seas ara a aes Collection of io i a. : Sample. la] & _ | Colour of Water, |n | which | t Sue and | 2 | 2 |Month.|Day.|Hour.| Lat. | Long. ——e et } o a/2 of of E|§ the | the 5 Air. | Water. | a lected. } | 1904. | De on oral eacun 397 | 381} Feb. | 29 | 12.0 | 68 8S. |2710W. Surface] 28-8 | 29-2 Reeve’s F ench kept! —1-6C. blue In open} sea four | pack edge. 398 | 382) Feb. | 29 | 18.0 | 6825S. |2710W. Surface] 29-2} 30-0 Reeve’s French —1-10C. blue A few bejrgs in sigh|t. 398 | 382] Feb. | 29 | 18.0 | 6825S. |2710 w. Surface] 29:2} 30:0 Reeve’s French —1-1C. blue 399 | 383!) Mar. 1 6.0 | 68 38S. |25 Ow. Surface] 26-6 | 29-2 Reeve’s French —1-6C. blue Bay ice |forming in | 399 | 383] Mar. 1 6.0 | 68 38S. |}25 OW. 26:6 | 29-2 Reeve’s French —160C. blue | 400 | 384} Mar, 1 | 12.0 | 68 43S. | 2415 w. 29:0] 31-0 Reeve’s French } —0-6 C0, blue (light) | Ship in ‘loose ice. 9.15 2/3/04 9.45 2/3/04 Tested 10.15 3/3/04 $6333 5:60 56999 i 5:60 5:90 5:60 5:40 6:00 5:70 5:40 5:85 5:30 5:90 5-40 5:30 5:35 5:40 5:50 5:60 5-70 5:80 5:85 5:90 1-5471 0583 +3139 19193 15471 0739 +3691 1-901 1-02543 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. go) SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 137 1.) Position (L.) a eee Temperature Temperature Volume Density of Sample. P lame at during Weights| peaq as 1m | Density of Distilled i a ollection o i “ °C =I D. fF eens Experiment. aac’ ing Portion Water at 4 C.=1.) ; Colour of Water, Time Hydro- of ; of ak ae from. z : Current, and and 1, t’. meter ete: ae Ob- |Reduced/Reduced Seana : ; Remarks. Date. (grams).| ™&¥e meter | served to to Hour.| Lat. Long. at |. the —= (cub. att’. |15°-56C t. Posi- eve of of of of centims.) tion Gate the the the the L. Teated. Air. | Water. Air, |Sample. w. R. V. a5v. |4Sisese| «Se. a oie” Oink 20, 2485 268 18.0 | 69 48. | 2355. W. Surface] 28-9 | 30-2 Hooker’s green 10.45 |50-0 13-4 5:30 11:6 —1-00C. (dull) 3/3/04 |10-0 C. 5:35 16-5 Only a fe|w distant |bergs. 5:40 22:0 5:50 32-0 5:60 42-8 5:70 53-0 5:80 63-0 15471 5:85 68-5 0821 12-9 5:90 73-6 +3854 13-15 5:60 42-6 2:0146 | 1:02530) 1:02480 | 1:02704 6.0 | 7028S. | 23 38 W. Surface} 30-6 | 30-2 Dark green 11.5 = |50-0 14-7 5:20 3-6 —100 3/3/04 |10-0 C. 5:30 14-0 No jice. 5:40 24-3 5:50 35:0 5:60 45-5 5-70 56:0 ~ 5:80 65:5 1:5471 5:90 75:8 0902 14-2 6-00 85-9 +4080 14-45 5:60 45-1 2:0453 | 1:02513)| 1:02489.| 1-02713 12.0 | 71 48S. | 2310 Ww. Surface] 29-6 | 30:8 Reeve’s French 11.35 |52-0 15-4 5-20 4-5 —0:7 C. blue 3/3/04 |11-1 5-25 9:8 As berg about 500 jyards} away 5:30 15:0 on port} quarter, |samplje taken 5-40 25-8 from thie lee side. 5:50 36:5 5:60 47-5 5-70 58-0 1-5471 5:75 63-5 0964 15-5 5:80 68-6 3311 15:45 5:50 36-6 1:9746 | 1-02497 | 1:02495 | 1-02716 1| 18.0 | 7130S. | 22 25 WwW. Surface] 28-4 | 31-0 Light Antwerp | 12.15 [58-0 14-6 5:30 10-2 —0-6C blue 3/3/04 |14-4 C. 5:35 16-5 Only one} berg in silght ; jsample 5-40 20:8 taken; from the] lee sjide. 5:50 31-6 ; 5:60 42-0 5-70 52-5 5:80 63-0 1-5471 5:85. 68-5 0917 14:8 5:90 73:8 +3808 14:7 5°60 42-1 20196 | 1:02527 | 1-02509 | 1:02731 7.0 | 7212S. | 18 40 w. Surface] 25-7 | 30-5 Blue 12.45 |58-0 12:9 5:30 6-2 —0-8C. 14:40 5:35 11-2 Two distant bergs} only, 5:40 16:5 5:50 27-0 5:60 37-6 5-70 48-0 5:80 58-2 1-547 1 5:85 63-6 0805 12-9 5:90 69-0 3392 12:9 5:60 387-5 1:9668 | 1:02557 | 1:02502)| 1:02725 10.45 | 7218S. |17 59} W.) 1131) 1131 | 25-2) 32-00 | Buchanan-Rich-] 15.40 [58-1 13-5 5:40 6-2 0:0C. ard water- 14-5 C. 5-45 11:5 Read on bottle 5:50 16-6 bridge | Thermometer No. 5:60 27-8 102,512 5:70 39-0 5:80 49-6 5:90 60-5 1-5471 5:95 65-3 0849 13-7 6:00 69-8 +3483 : 13-6 5-70 3825 1:9803 | 1:02604 | 1:02563| 1-02786 138 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GR. Number of Station. ~ a 406 406 406 406 406 Number of Sample. 391 kept 392 kept 393 kept 394 395 kept 396 Data Relating to the Collection of the Sample. Date (E.) 1904. Mar. Mar. Mar. Mar. Mar. Mar. Month. Day. Hour. Data Relating to the Determination of the Density of th Z, Depth Position (L.) : Temperature Temperature in Fathorhs, at time of during : Collection of Experiment. Dd. qd. Sample. as Colour of Water, | Time of cin T t Curent, and and Wes t’. f i 6 Date. Lat. Long ig CHEN | is aa ale Bios l. Posi ical of of of of tion atl the the the the L. lected: Air. | Water Air. |Sample Vote one oR oF, uns me: 11.45 | 7218S. |17 594 W.| 1131] 1000 | 25-2] 32-4 Buchanan-Rich-} 16.10 (58-0 12-7 0-2C ard water- 14:4 C. Readon}| bottie bridge | Thermometer No. 102,504 13-0 12-85 11.45 | 7218S. |17 594 W.| 1131} 750 | 25-2} 82:50 | Buchanan-Rich-]| 14.20 [53-8 13:2 0-30.) ard water-] 4/3/04 |12-1C. Read on bottle bridge | Thermometer No. 102,512 12-9 13:05 12.30 | 7218S. |17 594 W.|1131} 500 | 26-2] 32-92 | Buchanan-Rich- | 15.0 [52-0 12-6 0-5C.| ard water-}| 4/3/04 |11-1C. Read on bottle bridge | Thermometer No. 102,512 12:5 12-55 12.30 | 7218S. |17 594 W.| 1131] 600 | 26-2] 29-7 | Thermometer No.} 13.45 [54-0 14:3 —1:3C.) 102,504 "4/3/04 |12:2 C. Probably} thermometer djid not Read on reverse} till near} surfalce. bridge 14-1 : 14:2 12.55 | 7218S. |17 591 W./1131] 300 | 26-2] 31-08 | BuchanansRich-] 15.35 [53-0 17-0 —0:5 C, ard water- | 4/3/04 |11-7C. Read on bottle bridge | Thermometer No. 102,512 16-6 | 16-8 12.55 | 7218S. |17 593 W.| 1131} 400 | 26-2} 29-2 Buchanan-Rich-| 16.5 59-0 17-0 —1:6C.| ard water-} 4/3/04 |15-0C. Probably, thermomleter djid not Read on bottle reverse) till near | surfa/ce. bridge | Thermometer No, 102,504 17-0 17:0 5-90 5:70 5-40 5-95 5:70 5:40 5:70 5-40 5:55 5:60 5:50 5:35 5:65 65611 5-30 5-85 5-75 65833 64-6 69-8 60-0 41:9 centims.) Volume of im- mersed Portion of Hydro- meter (cub. 1-5471 +0802 +3247 1-9520 1-02620| 14 1-5471 0814 | +3329 1:9614 | 10261 | 1-5471 0886 +2985 1-9342 | 10251 1-5471 1048 +3401 1:9920 1-5471 +1061 +3790 2.0322 | 1:02 AN SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. (| 12.0 Data Relating to the Collection of the Saimple. Data Relating to the Determination of the Density of the Sample. 139 oat Depth , Volume i j Position (L. : Temperature Temperature ; Density of Sample. ce) in Fathoms. | “44 ae of duving Weights of im- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) Dz d. Sample. to ing Boxzon ais ___| Colour of Water, | Time Hydro- 2 My of | from T t Current, and and Ts tre meter Bate Bye: Ob- |Reduced/Reduced Sea which 3 3 Remarks. Date. (grams) ‘ b served to to Lat. | Long. |“ |_ the rie lecttimgs)| abt. |15°56C.) +t. Posi- eenple of of of of [era aie tion en the the | the the { E L. lected Air. | Water. Air. |Sample. w. R. Vv. aS. | 4Sis-53.| 454 oe (Sas Pe Mallee cas Ch 7218S. |17 59} W.'1131| 100 | 26-2| 28-93 | Buchanan-Rich-] 16.35 60-6 16-6 5:30 11-0 | —1:70. ard water-} 4/3/04 |15-9C 5°35 15-5 Read on bottle 5-40 20:0 bridge | Thermometer No, 5:00 30:8 102,512 5:60 41-0 5:70 52:0 5:80 62-5 1:5471 5°85 68:2 1026 16:3 5:90 72-0 3745 16-45 5:60 41-4 2:0242 | 1:02524) 1:02544| 1-02767 7218S. |17 594 W.| 1131} Surface} 26-2 | 30-9 Reeve’s French 9.5 |62-0 11-6 5:40 10:5 —0:6C. blue 5/3/04 |11-1 C. 5:45 15:8 Distant |berg or gilacier| about 5:50 21-0 twenty miiles. 5:60 32-0 5-70 42-8 5:80 535 5:90 64:2 1-5471 5:95 69:5 0727 - 11-7 6:00 75:0 +3863 11:65 5:70 42-7 20061 | 1:02590) 1-02511| 1-02733 72 228. |18 30 W. Surface] 28-2} 30-9 Reeve’s French 9.50 [54:8 12:2 5:30 3-0 L —0-6 C. blue 5/3/04 |12-7 C. 5:35 8-0 D.R. Off} Coats Lajnd. 5-40 13-6 No ice in} sight, wejather| misty 5-50 24:5 and snjowing. 5:60 35°53 3 5:70 46-0 5:80 57:0 | 1:3471 5:85 62:2 0768 12-4 5:90 67-8 +3193 12:3 5:60 35-3 1:9432 | 1:02570| 1:02504| 1:02726 7228S. | 1855 W. Surface] 26-6 | 30-0 Indigo blue 10.35 |55-8 115 5:30 3-8 —110. 5/3/04 {13-2 C. 5:35 8:8 Tce aboujt four miles ajway to [ 5-40 13-8 port ; a) small berg about one 5:50 24:8 mile ajway to stiarboalrd. 5-60 35:6 5:70 46:5 5:80 57-0 | 1:5471 5:85 62-0 0727 11:8 5-90 67-5 3211 | 11-65 5:60 35-5 1-9409 | 1-02571) 1-02492| 1-02716 72318. |19 0W.| Surface| 26-8 | 28-2 Indigo blue 10.55 55-6 10-4 5:30 2:2 ? —1-6 C. 13-1 C. 5:35 2 Off Coats Land. | 5:40 12-4 Pack ice |about half a mille away. 5:50 23-2 | 5:60 34-0 | 5:70 | 45-0 | 5:80 58°6 15471 | 5:85 61-0 0665 | 10:9 5:90 66-0 +3085 | 10:65 5:60 34-1 1-9221 | 1-02582| 102486 | 1:02712 73 30S. | 21 28 W. | Surface] 24-8 | 28-9 Indigo blue 9.20 |50°8 8-8 5:10 3°5 a | = E7C, 11/3/04 |10-4 C. 5:20 | 14-2 Steaming along edge of pack, 5°30 25-0 about six miles distant from 5-40 35°6 Coats Land—bay ice forming. 5:50 46:8 Sample about three-quarters 5:60 57:0 full of it. 5:70 67-7 1:5471 5:80 175 0558 9-1 5:90 88-0 4170 8:95 | 5:50 46-1 2-0199 | 1-02472 | 102347 | 1-02568 140 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GR, Data Relating to the Collection of the Sample. Date (E.) Depth Temperature | bergs. Position (L.) : / in Fathoms. Ab hia oh é Z Collection of 8 e | D. d. Sample. 3S | FI ee | | Colourior Water, n | n of | from T Current, and s/s Sea | Which : t. Remarks. «| & |Month.|Day.| Hour.| Lat. Long. at the. | 2/2 Posi- Sample | lof of 5 g tion New the the (a | L. | yactea.| Aiz- | Water. | 1904. its Sree Wises [sea | sea 410 | 403, Mar. | 4.40 | 73 30S. | 2128 W.| 159 | 154 23-0} 28-90 | Buchanan-Rich- kept : eae —1:7C. ard water- Read on bottle bridge | Thermometer No. 102,512 411 404) Mar. 6.0 | 73258. |23 45 W.| .. | Surface 18-5 oe “ Indigo Bay ice.) Oldice hialf a |mile off. i About |a tablespojonful| of ice particl|es in sample. 412 | 405) Mar. 12.0 | 73118. | 23 53 W. Surface 25:0] 28-9 Indigo kept 170. Steaming} through |pan- cake icje; pack a/bout half ja mile ajway. Sample} about a qjuarter full of jice particlijes. 412 | 406) Mar. 15.0 | 73 958. | 2353 W.| 1950} 1030 | 20:0] 32-8 Buchanan-Rich- kept 0-40, ard water- Read on bottle bridge | Thermometer No. 102,512 412 407) Mar. 15.0 | 73 98. |23 53 W.|1950*| 1535 | 20:0} 31-6 Buchanan-Rich- —0-2 C. ard water- Read on bottle bridge | Thermometer No. 102,504 } * Weights did not come off when bottom was reached, and, on hauling up, lost about 1800 fathoms of wire, two Buchanan-Richard water-bottles, and | Buchanan’s sounding-tube. 413 | 408 Mar. | 6.0 | 72 28, | 23 40 Ww. Surface 200) 203 Blue | —15C. No ice nejar, only dilstant Data Relating to the Determination of the Density of th Time and Date. 9.55 11/3/04 9.50 15/3/04 10.20 15/3/04 10,50 15/3/04 11.15 15/3/04 11.40 Temperature during Experiment. Wwe t” of of the the Air. |Sample oaks ea€ 54:8 9:3 12:7 C 9:8 9:55 58-0 13-4 14:4 C. 13:8 13-6 60-2 14-4 15-7 C. 14-2 14:3 58-2 ‘13-0 14-6 C. 13-1 13-05 57-5 12:5 14-2 C, 12-7 12-6 58-0 11:6 14:4 C. 12-0 11:8 39-1 | 1-9604 | 1-02615| 41-8 2-0101 37-4 1:9746 | 1: 39-5 1:9858 36-7 1-9577 40-2 | 1-9843 | 1:0260: Volume} of im- ( Read- | mersed W ing | Portion ; of of Hydro- | Hydro- Ob- meter. | meter i. (cub. a t’. centims.) *”®* | R MWe 45 6:6 12-0 17-6 28-2 39-2 50:0 60:6 1-5471 66:0 0596 71-4 +3537 9-5 14-6 20:0 31:3 42-0 62-7 63-3 1-5471 69-0 0849 13-5 +3781 102532 6:5 11:5 16:8 27-0 37-5 48-0 58-5 1-5471 63:0. 0892 68-0 +3383 6:0 11-0 16:5 27-2 37-5 48-3 59-0 1-5471 69:8 0814 80:5 +3573 1-02610, 4:5 10:0 15-0 26-0 37-0 47-2 58-0 15471 64:0 0786 69-0 +3320 1-02617 8-6 14:0 19:0 29:8 40:2 51:0 61-5 1:5471 66:0 0736 715 +3636 \Nj SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 141 Data Relating to the Collection of the Sample. . Data Relating to the Determination of the Density of the Sample. re Depth | Volume : tee Position (L. : Temperature Temperature ByaOs Density of Sample. ) ey) in Fathoms. | “a5 tcae of during Weights of im- (Menai of Distilled «7 |), | Collection of Experiment. | added | Read- | mersed “Water at 4° C.=1.) Dz. d. Sample. te ne bomen : Colour of Water,} Time | | | Hydro- eee e ee le) ae ee of from T t Current, and and 1, te estar Hoge ee Ob- |Reduced|Reduced Sea | Which] *- ; Remarks. Date. (gtams),| “Set | me a served | to to ieee tat. |. Long. |“ | the |——|——— centins,)| abt’. |IB°56C) t. Posi- gonple of of of of ge tion} Go. | the the the the L. | jected. | Ait: | Water. Air, |Sample.) w. R. Vv. ace, |4Sic-56,|| 4S “ibaa Bene” EKG: | 71 50S. | 23 30 W.| 2102) 2092 | 16-0) 31:5 Buchanan-Rich-} 10.0 (61-0 149 | 5:30 2-0 —9:3C, ard water- | 16/3/04 |16-1 C. | 5:35 6:5 Read ou bottle | 5:40 11-6 bridge | Thermometer No. 5:50 22:2 102,502 5-60 33-2 5-70 44-0 5:80 54:8 15471 5-85 59:8 0924 14-7 5-90 65-0 +3003 14:8 5:60 33-2 1:9398 | 1:02572) 1:02556| 1-02777 7150S. |23 30 W.| .. | Surface} 16-0) 29-1 Bright indigo 10.40 (60-0 14-2 5:30 6-7 —1:6C. 16/3/04 |15-6 C. 5:35 11:9 No ice nejar, only dijstant 5:40 17-2 bergs. 5:50 28-0 5:60 38-6 5:70 49-2 5:80 59-5 | 1:5471 5°85 65-0 0892 14-4 5:90 70-2 +3483 14:3 5:60 88-5 1:9846 | 1:02547| 1:02520| 1°02745 7128S. | 2232 W.| .. | Surface] 19-7} 29-0 Indigo 11.5 [58-8 11-9 5:30 2:2 —17C.) .- 14:9 C. 5:35 7-4 No ice in|sight, only} dis- 5-40 12:8 tan|t bergs. 5:50 23-9 5:60 34-7 5-70 45-2 5:80 56:0 | 1:5471 5:85 61-2 0761 12:5 5:90 66-8 +3121 12-2 5:60 34-5 1:9353 | 1-02574| 1-02506| 1-02732 71 28S. | 22 32 W. | 2338) 1338 | 19-7] 31:8 Buchanan-Rich-} 11.30 |60-0 12:3 5-40 2:8 —0-1C.| ard water- 15:6 C. 5-45 8-0 Readon| _ bottle 5-50 13-5 bridge | Thermometer No. 5-60 24:6 102,505 5-70 35:5 5:80 46:5 5:90 57-2 | 1-5471 5-95 62-6 0783 12-8 6-00 68-0 +3202 12-55 5:70 35-4 1:9456 | 1:02624| 1-02561| 1-02784 71 28S. | 22 32 W.|2338| 2328 | 19-7] 31-5 Buchanan-Rich-} 12.0 |54:7 12-4 5-40 4-4 —0-3C.| ard water- 12:6 C. 5-45 9-8 Read on bottle 5:50 14-2 bridge | Thermometer No. 5:60 25-0 102,502 5:70 35-5 5:80 46-0 5:90 56-5 1-5471 5:95 62-0 0768 12-2 6-00 67-0 3220 12:3 6-70 35-6 1:9459 | 1:02623)| 1-02555| 1-02779 71 22S. |1815 W.| .. | Surface] 29-8} 29-6 Reeve’s French 9.35 [58-1 10:6 5-40 9:0 , —1-3C. blue 14:5 C. 5:45 14-2 Only dist\ant bergs | in 5-50 19-8 sight. 5:60 | 30-6 5:70 41-5 5:80 52-4 5:90 63-0 1:5471 5-95 67-9 0674 11-0 6-00 73-0 +3736 10:8 5:70 | 41-3 71-9881 | 1:02600 1-02504° 1-02732 142 Data Relating to the Collection of the Sample. Date (E.) Position (L.) a js i i oot |] SEs = |& mn | © | 2 | Month. Day.) Hour.| Lat. Long. 5 | 2 Ase Ss = | Z 1904. CV mF) o ¢ 416 | 415) Mar. | 17 | 8.15 | 712258. | 1815 W. kept ! 416 | 416) Mar. | 17 | 8.15 | 7122S. | 1815 W. kept . 416 | 417| Mar. | 17 | 7.50 | 7122S. | 1815 W. kept 416 | 418) Mar. | 17 | 8.0 | 7122S. |18 15 W. kept 417 | 419 Mar. | 18 | 8.0 | 7122S. | 16 34 W. / No ice in) sight, 417 420) Mar. | 18 | 8.0 16 34 W. kept el 71228. | Depth in Fathoms. Toswen tire Collection of D. qd. Sample. car Colour of Water, f : Current, and Sea which | T- t. Remarks. a: the Posi-;S2™Ple] of | of tion ue the the L. leeteal Air. | Water. | Minh|| Shank 2370} 200 |19-8| 33-0 Buchanan-Rich- 0-70. ard water- Read on bottle bridge | Thermometer No. - 102,505 2370] 300 | 19-8) 33-0 Buchanan-Rich- 0-6 C. ard water- Read on bottle bridge | Thermometer No. 102,502 2370| 1870 | 19:8) 32-1 Buchanan-Rich- 0-1C. ard water- Read on bottle bridge | Thermometer No. 102,505 2370) 2360 | 19-8} 31-5 Buchanan-Rich- —0°3C. ard water- Read on bottle bridge | Thermometer No. 102,502 Surface] 28:5 | 29-9 Reeve’s French —1:2C, blue : : Buchanan-Rich- 1410) 100 | 28-5 as é ard Besse ‘ ottle on A Thermometer No. 8 102,505 Time and Date. 10.0 10.25 11.0 11.30 10.0 10.30 Temperature during Experiment. Le t’. of of the the Air, |Sample cok “(Ch 53:8 CB. 112-1 C. 9-8 9:75 50-2 9-7 10-1 C, 9-5 9:6 47-0 9:5 8:3 C. 9:3 9-4 51-2 | 10-0 10-7 C. 10-0 10-0 51:0 10-6 10-6 C. 10-8 10-7 52-0 10:7 11:1C 10-5 10-6 HOOD OADM enooooonre BQSSLARSA & S| onocoooonos & © CHAO Co SPIED ED CHEE ENEMIES] (SP op GAINES) hall CAE SME EER SIEN SSoSuNS Sas DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GR Data Relating to the Determination of the Density Volume of im- mersed Portion (cub. centims.) Ws 1-5471 . 0608 -3501 1-9580 | 1-02672) 1-5471 | 0599 +3474 1-9544 1:5471 * 1-0268, 1-5471 0624 3591 1-9686 | 1-02666 * 15471 0668 +3347 1 9486 1-0262: 1-5471 0661 4016 ~ 9.0148 | 1-0264l SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCHAN. 143 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. sys Depth | : Position (L.) a or Temperature Temperature | Volume Density of Sample. f in Fathoms, alee of ' during Weights} poag oe (Density of Distilled ollection o i ~ onirore cig D. Fh aeentet xperiment. oie ee ee Water at 4° C.=1.) Colour of Water, Time | Hydro- H ve | H ot i : of Lies ie. Current, and and nes tis meter neg a Ob- |Reduced) Reduced Sea | Whic : : Remarks, Date. |(grams).| ™°?e? er | served | to to Lat. Long. | “34 |, the oo ee | (cub. | att’, |15°56C,| . Posi- ue of of of of centims.), tion Ais the the the the L. | jocteg, | Ait. | Water. Air, |Sample.| — w. R. Vv. rsa sheen || piss | ee nee eB | ee ek malts ees 8.10 | 71 22S. |16 34 W.| 1410) 500 | 28-5) 33-0 Buchanan-Rich-} 11.0 50:8 10-9 5-50 10-0 0-6 C. ard water- 10-4 C. 5:55 14:5 Read on bottle 5-60 20-7 bridge | Thermometer No. 5-70 30:9 102,502 5:80 | 41:6 5-90 52:0 6:00 62:6 1:5471 6-05 68-0 -0683 ‘11-0 6-10 73-2 +3754 10-95 5:80 41-5 1:9908 | 1:02653 | 1:02560 | 1-02781 18) 7.45 | 71 22S. | 16 34 W.| 1410) 410 | 285] 33-0 Buchanan-Rich-} 11.20 [53-2 13-2 5-50 17-0 } 0-6 C. ard water- 11-8 C. 5-40 5:0 Read on bottle 5-45 10-2 bridge | Thermometer No. 5-60 26-2 102,505 5:70 | 36-9 ; 5:80 47-2 ; 5:90 57-8 - ; ‘ 6-00 68-5 1-5471 6-05 73-5 -0817 13-0 6:10 79:0 | +3808 13-1 5-75 42-1 2:0096 | 102615 | 1-:02563| 1-02782 18 @y.50 | 71 22S. | 1634 W./ 1410] 1400 | 28-5} 31-9 Buchanan-Rich-} 11.45 {53-5 15-4 5-40 13-0 | : —0-1C, ard water- : 11:9 C. §:45 18-0 Read on bottle 5-50 23-5 bridge | Thermometer No. 5:60 34-0 | 102,502 5:70 44-7 5:80 55:5 5:90. 66-0 6:00 76-0 1-5471 6-05 81-0 “0948 15-0 6-10 86-0 4505 15:2 O75 49-8 2-0924 | 1:02568 | 1-:02560| 1-02780 19 }}.0 | 7132S. | 17 15 W. | 1221) Surface] 27-2 | 29-8 Reeve’s French 10.0 37:0 8:3 5-50 8-5 ae =P) blue 2-8 C, 5-55 13-5 No ice in| sight, only two) distant 5:60 19-5 bergs. 5-70 30-0 5:80 40:3 5:90 51-0 = 6-00 61-8 15471 6:05 67-0 0512 8-1 6:10 72:2 +3655 8-2 5°80 40-4 1-:9638 | 1:02668)| 1:02530 | 1-02758 190-0) 7132S. | 1715 W.| 1221) 721 | 27-2) 392-5 Buchanan-Rich-} 10.20 /43-2 1:9 5-50 2-0 0-3 C. ard water- 6-2. 5:55 7-0 Read on bottle 5:60 12-0 bridge | Thermometer No. 5:70 22-6 102,505 5:80 | 33:5 5:90 44-2, 6-00 55:0 1:5471 6-05 60-0 -0487 T- 6:10 65:2 +3030 7-8 5:80 33°5 1-8988 | 1:02705)| 1:02559 | 1-02782 OC. EDIN., VOL. LI, PART I (NO. 4). 21 144 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA Data Relating to the Collection of the Sample. Data Relating to the Determination of the Dens aap Depth Date (E.) Position (L.) rea Temperature Temperature in Pathoms: at time of during Weights i ; te ie Collection of Experiment. | added 3. Bs | a Sample. to | 8 = 7 ________| Colour of Water, | Time Hydro- | a)@ | of Le , t Current, and and iy, we meter he es Month. |Day.| Hour.| Lat. Long. ue the Ee. pete. pate | | 2 2 Posi Sample of of of of esas 4 tion tile the | the the | the 2/4 L. | lected. Air. | Water. Air. |Sample.| w. 1904. F Sed bast Sane oh oT °C. | 418 | 426] Mar. | 19 8.5 | 71328. | 17 15 W.| 1221) 1221 | 27-2) 31-9 Buchanan-Rich- | 10.45 |48-7 TA 5:50 kept —0-1 C6. ard water- 9-3 C. 5:55 | Readon| _ bottle 5:60 } bridge | Thermometer No. 5-70 102,502 5:80 5:90 6-00 6-05 8-1 6-10 7-9 5-80 419 | 427| Mar. | 20 | 18.0 | 7043S. | 17 35 W. | 2513] Surface] 26-0, 30-4 Antwerp blue 13.30 55-0 13-7 5:30 —0:9 C 21/3/04 |12-8 C. 5°35 No ice in|sight, only, three} small 5-40 bergs | about six | miles| off. 5:50 5:60 5-70 5:80 5:85 13-4 5-90 13-55 5:60 420 | 428] Mar. | 21 8.0 | 69 33 S. | 15 19 W. | 2620] Surface] 27-2 | 31-0 Antwerp blue 13.50 |54-4 12-7 5:30 —0-6 0. 12-4. 5:35 No ice an|ywhere in |sight. 5:40 5:50 5:60 5-70 5:80 5:85 12:7 5:90 12:7 5:60 420 | 429] Mar. | 21 |c.8.0 | 6933S. | 15 19 W./ 2620] 1615 | 27-2] 31:3 Buchanan-Rich-}| 14.15 |55-0 12-7 5-40 kept —0-4C.) ard water- 12-8 C. 5-45 Read on bottle 5:50 bridge | Thermometer No. 5:60 102,505 5-70 5:80 } 5-90 5:95 12-7 6-00 : 12-7 5:70 | 420 430} Mar. | 21 |¢.8.0 | 69 33S. | 1519 W.| 2620] 2615 | 27-2] 31:5 Buchanan-Rich-} 13.5 |53-5 9:7 5:50 kept —0-3C.) ard water- | 22/3/04 |17-9 C. 5:55 Read on bottle 5-60 | bridge | Thermometer No. 5:70 102,502 5-80 5:90 6-00 6-05 9-7 6-10 { 9:7 5:80 | 421} 431] Mar. | 22 | 12.0 | 6832S. |1052 W. Surface] 30-0} 31:8 Cobalt blue 13.40 |52-0 9-0 5-40 —0-1C. I1:1 ¢, 5:45 No ice an ywhere in |sight. 5:50 | 5:60 5-70 | 5-80 | 5-90 . 5:95 9-5 6-00 : | ——_ hel 9:25 | 570 | Volume of im- mersed Portion 1-471 0493 2977 1-8941 | 1 15471 0846 3157 19474 15471 +3238 1-9501 | 1-5471 0605 +3510 19586 1-5471 0577 +3112 19160 10 14,40 2) 14.40 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. NE SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 145 ona Depth Pa tRGe | Vol i Position (L.) : Temperature Temperature apie Density of Sample. , | in Fathoms. | “45 time of during —_| Weights ofim- | (Density of Distilled l nes | Collectionol Experiment. | addeqd | Bead- | mersed Water at 4° C.=1.) . D. d. Sample. to | ig | Portion = , fe SS f of |—— | Colour of Water,} Time Hydro- oie | of from T t Current, and and ae ins meter | ee a Ob-_ |Reduced) Reduced | Sea which E : Remarks. Date. (grams). | eee Te = served ; to to Long. | “at E ae =n = ae | esntians.) att’. |15°-56C.| t. Posi- aaa? ©) of of | of of leat SF ee tion Col the the the the Wp lected Air. | Water. Air. |Sample w. R. V. aS. \4 Si5°-56.| 45. oe ira Se PS Ct | 10 52 W. | 2487 50 | 32-2} 304 | Buchanan-Rich-} 14.50 [54-0 12-6 5-40 8-7 | | —0:9C.,; ard water- 12-2 C 5-45 13-9 | Read on bottle 5-50 19-2 | bridge | Thermometer No. 5:60 29-5 102,502 5:70 40-0 5°80 50:6 } 5:90 61-2 1-5471 ’ 5-95 66:5 0786 12-6 | 600 | 71-5 | -3627 | 12-6 3:70 40-1 1-9884 | 1:02600 | 1:02540 | 1-02764 10 52 W.| 2487} 100 | 32:2} 33:5 Buchanan-Rich-} 15.20 |55-0 12-0 5-40 2-1 -0-8C.| ard water- 12-8 C. 5-45 7-2 | i) Readon}| _ bottle 5:50 12-5 ~ bridge | Thermometer No. 5:60 23-0 102,505 5-70 33-2 : 5:80 43-8 5:90 54:3 1-5471 5:95 59°8 0746 11-9 6-00 65-0 +3021 11:95 570 35-4 1:9238 | 1:02636| 1:02561)| 1:02779 68 32:8. | 1052 W. | 2487} 200 | 32:2} 33-5 Buchanan-Rich-} 15.45 |50-0 1-7 5:50 11:2 O-8C.) ard water- 10:0 ©. Byy) 16-2 Read on bottle 5:60 21-0 |} bridge | Thermometer No. 5-70 31-2 102,502 5:80 | 42-0 | 5-90 52-6 6-00 63-0 1-5471 6:05 68-0 0718 11:3 6:10 73:5 +3808 11-5 5:80 42-1 1:9997 | 1:02648)| 1-02566)| 1:02784 68 32'S. | 1052 W.| 2487) 300 | 32-2] 33-2 Buchanan-Rich-} 16.10 {52-0 15-0 5:50 21-0 0-7 €:) ard water- 11-1 C. 5-40 9-8 Read on bottle 5:45 15-0 bridge | Thermometer No. 5-60 31-0 102,505 5:70 41:5 5:80 51:8 5:90 62-0 1-5471 : 595 | 67:2 | -0914 14:3 6-00 72-0 3736 : 14:65 3:70 41-3 2-0121 | 1:02586 | 1-02567 | 1:02783 | 68 32S. | 1052 W.| 2487} 300 | 32-2] 33-2 Buchanan-Rich-} Tested |50-0 9-8 5:50 7-6 a 07C.) ard water-| again |10-0 C. 5-55 12-7 Read on bottle 9.30 5:60 17-4 bridge | Thermometer No.} 23/3/04 5-70 28-0 102,505 5-80 38-5 5:90 49-2 6-00 60-0 1-5471 | 6-05 65-3 0615 | 99 | 610 | 70:0 | -3500 | | 9-85 5:80 38-7 1-9586 | 1:02671| 1-02558)| 1-02779 | 68 32S. | 1052 W.| 2487; 400 | 32-2] 33-1 Buchanan-Rich-} 10.0 [51-0 10-4 5:50 8-2 | 0-6C.| ard water-| 23/3/04 |10-6 C. 5-55 13-2 Read on bottle 5-60 18-5 bridge | Thermometer No. 5:70 29:0 102,502 5-80 39-9 5-90 50:5 6-00 61-2 1-5471 6-05 66-5 -0649 10-4 6-10 71:8 3609 10-4 5-80. | 39-9 1:9729 | 1:02663 | 1:02560\ 1-02781 | 146 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. Number of Station. Number oz Sample. 421 | 437 kept 422 | 438 Month. |Day 1904. Mar. kept 422 | 439 kept 422 | 440 422 | 441 Mar. Mar. Mar. 422 Mar. = Depth . Date (E. Position (L.) + Temperature (E.) in Fathoms. Pi aa ai Collection of D d Sample. ee ___| Colour of Met, of co} Current, an : which | T. t. R k .| Hour. Lat. Long. pe A the |—— SES Posi Sariple of of tion hs the the | L. lected: Air. | Water of Cle 2 oR oR 22 | 16.20 | 68 32S. 1052 W.| 2487, 2480 | 29:0; 31:5 Buchanan-Rich- | —0:3C ard water- | Readon| __ bottle bridge Thermometer No. 102,502 23 | 8.0 | 6832S. | 12 49 W. | 2660| Surface] 30-0] 31-1 ‘ Indigo —0:5 No ice jin sight. 23 | 18.40 | 68 32S. |12 49 W.| 2660! 500 | 30-0] 32-7 Buchanan-Rich- | 0-4C.| ard water- Whilst heaving sample from 1000 Read on bottle fathoms, thermometer No. bridge | Thermometer No, 102,502 wire snapped just 102,505 before reaching surface, thus losing thermometer, Buch- anan-Richard water-bottle,and weights. 23 | 16.15 | 68 32S. | 12 49 W:| 2660; 600 | 30:0} 32:5 Buchanan-Rich- 0-30.) ard water- Read on bottle bridge | Thermometer No. 102,502 23 | 17.50 | 68 32 S. | 12 49 W.| 2660} 1500 | 30-7| 31:5 Buchanan-Rich- —0-3C. ard water- Read on bottle _ bridge | Thermometer No. 102,505 23 | 18.0 | 68 32S. | 12 49 W.| 2660} 2000 | 30:7| 31-4 Buchanan-Rich- —0-3C.| _ ard water- Read on bottle bridge | Thermometer No. 102,502 Data Relating to the Collection of the Sample. Mar. WILTON ON THE TEMPERATURES, SPECIFIC GRA\ Data Relating to the Determination of the Densi Time and Date. Temperature during Experiment. t’. Weights added Volume of im- Read- | mersed ing | Portion of of Hydro- | Hydro- meter. meter (cub. 10.35 23/3/04 10.50 14.20 24/3/04 15.0 24/3/04 15.20 24/3/04 15.40 24/3/04 10:8 11:05 8-2 i RE no 8-3 8:25 13-1 57:5 14:2 C. 13:2 1315 54:5 12-4 12:5 C. 12-4 12-4 14-2 14-1 14-15 15-0 PAAANM AMMA FOOD UMOOADUN oOnnooocoone 6:00 5:70 5:50 . 5:80 5:90 5:95 6-00 5:70 5:40 4-45 5:50 5:60 8-70 5:80 5:90 5:95 6-00 5:70 5-40 5:45 5:50 5-60 5:70 5:80 5:90 5:95 6-00 5:70 i rag centims.) Ve Data Relating to the Collection of the Sample. SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 147 Data Relating to the Determination of the Density of the Sample. | a Depth ate.) Position (L.) : Temperature Temperature in Fathoms. ain (hua during Collection of Experiment. D. d. Sample. See Vi Sol Colour of Water,}| Time an ore T. t Current, and and es Wo Ja} Hour.| Lat. Long. pe (UO | be ks Doe a Aly | Posi- me of of- of of tion col. | the the the the L. ieeted! Air. | Water. Air. |Sample : | ° , he , op | aoe | oF 226s | 11.10 68 41S. 1336 W. Cz. | 1000 | 30:8, 31-82 | Nansen-Petters-} 16.5 58-0 13-6 2500 —0:1C.| son water- 14:4 C. Fi | Read on bottle || | bridge | Thermometer No. iT 18,725 | 13-7 | 13-65 | | 2418.0 | 68 41S. | 1236 W.| Cir.| 700 | 30-0] 32-40 | Nansen-Petters-] 13.50 [59:5 12-9 | 2500 0-20.) son water- | 26/3/04 |14-3 C | Read on bottle | bridge | Thermometer No. 18,725 12:8 i oa 12:85 74) 18.25 | 68 41S. |12 36 W. Cir.| 800 | 30-0] 32-4 Nansen-Petters-} 14.15 [54-0 12-3 2500 0-2C.) son water- | 26/3/04 |12-2 C. Read on bottle | bridge | Thermometer No. 18,725 | — | 2 12-4 44) 17.10 | 68 41 S. [1236 W. Cir.} 900 | 30-0] 32-1 Nansen-Petters-{ 14.40 [54-0 126 2500 0-10.) son water-]| 26/3/04 |12-2 C. | Read on bottle | bridge | Thermometer No. | 18,725 } ) | 12:5 | | 1255 wo} 8.0 | 68268. |1111 W. .. | Surface] 30-2] 31-3 Antwerp blue 14.15 |45-0 i167 : (noon pojsition) —0:4C 7-2C | No icje. E | | | 11-2 j | 11-45 6.0 | 6748S. |1152 W.)| .. | Surface] 30-3| 30-9 Blue 9.30 [51:3 8-8 —0-6C. é 7 C. | Three| distant bergs. Boe 9:3 9:05 oan Density of ar Weights ears Saat (Density of Distille added me aaa Water at 4° C.=1.) eae oe of meter Pee ate Ob- |Reduced| Reduced (grams). MShere (ex served a to centims.)| atl’. [I5°560) t. Ww. R. Vv. aS¢, | 4Sise56.| 4St, 5:40 8-0 045 13-5 5:50 19-0 5:60 29-2 5:70 39-8 5:80 50°5 5-90 60:8 1:5471 5:95 66:0 0852 6-00 71-5 +3600 5-70 39-8 1:9923 | 1:02597 | 1:02557 | 1:02780 5:40 5:0 5:45 10-0 5:50 15-0 5:60 26:5 5-70 37-2 5:80 47-2 5:90 58:0 | 12-5471 5:95 63-0 0802 6-00 68:0 +3320 5-70 36-7 1:9593 | 1-02616| 1:02559 | 1-02781 5-40 3-0 5:45 8-0 5:50 13-0 5:60 24:5 5-70 36:0 5:80 46-0 5:90 57-5 1-5471 5:95 62-5 0774 6-00 68-0 +3202 5:70 35-4 1:9447 | 1-02624| 1:02558 | 1:02780 5:50 16-0 5:40 4-5 5:45 9-0 5:60 26-0 5:70 36-0 5:80 46-2 5:90 57-2 1-5471 5:95 62:0 0783 6-00 67-0 +3257 5:70 36-0 1:9511 | 1-02621| 1.02558 | 1-02780 5:35 4-0 5-40 9-0 5:45 14:5 5:50 19-8 5:55 25-0 5:60 30:8 5:65 36-0 1-5471 5-70 41-2 0714 5:75 45-5 2271 5-55 25-1 1:8456 | 1:02597 | 1-02514 | 1:02739 5-40 6-0 5:45 11-0 5-50 16-7 5:60 27-0 5:70 38-0 5:80 48:5 : 5:90 59:0 1:5471 5°95 64:5 0565 6:00 69:5 +3419 5:70 37:8 11-9455 | 1:02624) 1:02499 | 1-02724 148 Data Relating to the Collection of the Sample. | ; Depth Data Relating to the Determination of the Density of th DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAY Date (E.) Position (L.) | ; Temperature Volume | i Fathoms. aieimelor Went of im- inh os a 7 as lee wl eel Collection of added mersed a jt | 1Ds ae Sample. es Fora 3 E _: ee: Colour of Water,} Time Hydro- ¢ | 2 a . of ee “Gl Me Current, and | and meter Hyd ee : Month. |Day.| Hour. Lat. Long. he the |——|—— AU BENE eras): _ (cub. 3 5 | | Posi-|S@™Ple| og of con Big tion| 8 | the | _ the a : aso L. lacked: Air. | Water. w. Va | | 1904. o 7 o 2 jae Rae ee 428 | 449; Mar. 27 | 6.30 | 67 9S. |1130 W. Surface 30-5 | 30:5 Reeve’s French 9.50 5:40 } | | | —0-8C blue 30/3/04 5:30 5:35 5:50 | 5:60 5-10 5:80 1-5471 5:85 -0861 5:90 +3881 | | 5:60 20213 | 428 | 450| Mar. | 27 66 41S. | 1117 W.| 2715| Surface 30-0 | 31-0 Reeve’s French 10.30 5:30 —0-6C. blue 30/3/04 5:35 Only onje berg in |sight,) about 5-40 four miles} off. 5:50 5:60 5-70 5:80 1-5471 5°85 0733 5:90 38175 5-60 | 1-9379 429 |451| Mar. | 28 66138. |11 22 W. Surface] 28:0 | 30:8 Reeve’s French | 11.30 5:30 —0-7 C. blue 80/3/04 5:35 Several |distant bejrgs. 5-40 5:50 5:60 5-70 5:80 1-5471 5:85 0777 5:90 +3329 5:60 1:9577 430 | 452| Mar. | 28 65 568 8. | 11 24 W. Surface] 28:0 | 31-2 Reeve’s French 12.0 5:30 kept —0-4C. blue 30/3/04 5°35 Only distjant bergs |in sigjht. 5:40 5:50 5:60 5:10 5:80 15471 5:85 0792 5:90 +3338 5:60 1:9601 | 431 | 453} Mar. | 29 63 548. | 10 42 W. Surface] 29:0} 31-2 Reeve’s French 13.10 5:40 —0-4C, blue 30/3/04 5-45 No ice in| sight. 5:50 5:60 5-10 5:80 5:90 1-5471 5:95 -0649 | 6:00 +8410 | 5-70 1-9530 1-0 | 4314 454| Mar. | 30 62 48S. |12 21 W. Surface 28:9 | 30-9 Reeve’s French 13.30 (51: 5-40 | —0-6C. blue 10: 5°45 Only dist ant bergs. 5:50 5-60 . 5-70 5:80 5:90 1-5471 | : 5:95 0652 6:00 +3492 5:70 1.9615 | 1-02 ) at AN SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 149 - Data Relating to the Collection of the Sample. 18.0 17.30 3017.40 30 |16.0 ———————— Position (L.) Lat. Long. 61 25 S. | 12 47 W. Only distjant bergs. 6121S. |13 2 W. 61 21S. | 13 2 W. 61 21S. |13 2 W. 61 21S. |13 2 W. 6121S. |13 2 W. Depth in Fathoms. D. d. from oi | which at the Posi- Sample tion’|) vs L col- * | lected. | Surface 30 2764 50 2764 80 2764 90 2764| 100 Temperature at time of Collection of Sample. t. of the Water. Colour of Water, Current, and Remarks. 30:2 | 30:75 —0:7C. Read on bridge 30:0} 28-92 —1-7 C. Read on bridge 30-0} 29-60 —1:3C. Read on bridge 30:2} 30-80 —0:7 C. Read on bridge 30-2 | 32-01 0-0 C, Read on bridge Reeve’s French blue Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- 3 son water- bottle Thermometer No. e] Data Relating to the Determination of the Density of the Sample. Time and Date. 9.55 31/3/04 9.40 8/4/04 10.10 8/4/04 10.35 8/4/04 11.0 8/4/04 13.10 8/4/04 Temperature Volume Density of Sample. during Weights of im- (Density of Distilled Experiment. | added Read- | mersed Water at 4° C.=1.) ae ing pen Hydro- g 1 tie meter poh nee ae Ob- |Reduced| Reduced (grams) j Sub served to to centims,)| 2%. |I5°56C) - t. of of | a: the the Air, |Sample.} — w. R. Vi 4S¢, | 455-56] 454, 2B. AG5 54:0 12:3 5:30 1-0 12:2 C. 5:35 6-0 5:40 12-0 5:50 22-0 5:60 33-0 5:10 43-5 5:80 53:5 15471 5°85 59-0 0768 12-3 5:90 64:3 *2958 12:3 5:60 82:7 19197 | 1:02583)| 1-02517| 1-02737 49-0 9-7 5:40 3:5 9-40. 5:45 8:8 5:50 14:0 5:60 24:5 5-70 35:5 5:80 46-0 5:90 56:5 1:5471 5:95 62-0 -0608 49-9 9-8 6:00 68-0 -3202 9-9 C. 25 9-75 5:70 35-4 1:9281 | 1:02633 | 1:02519)| 1-02745 50:3 9-1 5:50 9-5 10:2 C. 5:55 15:0 5°60 20:0 5-70 30:5 5:80 41-0 5:90 52-0 6:00 62-0 1-5471 6-05 67:5 0577 50:2 9-4 6-10 72:5 +3718 10:1 C. ———— 9:25 5:80 41-1 1:9766 | 1-02661| 1:02538)| 1-02770 49-9 8-9 5-50 8-0 9:9 C. 5:55 13- 5:60 18-8 5:70 29:5 5:80 40-2 5:90 50:8 6:00 61-2 15471 6:05 66-0 0562 47-0 9-1 6:10 71-0 3609 8-3 C. —|—— 9-0 5:80 39:9 1:9642 | 1:02668 | 1:02541| 1-02772 47-0 7-7 5:50 5:0 8-3, 5:50 10:0 5:60 15:0 5-70 25:5 5:80 36:5 5:96 AT-2 6:00 58-0 1-5471 6-05 63-0 -0484 en 6 78 6-10 68-0 +3302 7:75 5°80 86-5 1:9257 | 1:02690 | 1:02544) 1-02772 | 48:8 13-4 5-40 7-5 | 9-3 C. 5:45 12:5 5:50 18-0 5:60 28-0 | 5-70 38:8 5:80 49-8 5:90 60:0 1-5471 5-95 65-0 -0817 50:0 12:8 6-00 70:0 +3510 10:0 C. — 13-1 5:70 38-8 19798 | 1:02604) 1-02552\ 1:02775 | 150 Data Relating to the Collection of the Sample. 7 On h Date (F.) Position (L.) ba ae Serene yal ES = rd ae : Collection of g§\2 / D. a. Sample. a ma aA |D from ae 3 e143 | | ot | which | 7 t » |g Month. Day.| Hour.| Lat. Long. Ai iste | |[-=—=—||= 2 = | Posi- See of of 5 5 tion} gol. | the the a ee | De fected Air. | Water. = =¥ gee A lees. ll Fs a 1904. 2s es ite Ca 4324) 461) Mar. | 30 | 16.50) 6121S. |13 2 W.| 2764) 200 | 30:0} 32:33 0-2C Read on bridge 432a| 462| Mar. | 30 | 16.30] 6121S. |13 2 W.|2764| 300 | 30-0 ae 0-2C. Read on bridge 432a| 463| Mar. | 30 |17.0 | 6121S. |13 2 W./2764) 400 | 30:0] 32-21 0-1 C. Read on bridge 433 |464| Mar. | 31 | 6.0 | 6050S. |1230 W.| .- | Surface] 30-2} 31-1 —0-5 C, Distant bjergs only. 434 | 465] Mar. | 31 |12.0 | 6037S. |1216W.| .. | Surface] 29-5 | 31-2 —0-4C. No ice |visible for] two | miles ; (weather mis ty). | 486 | 466) Apr. | 2 |18.0 | 5810S. |1150W.| .. | Surface) 31-0] 33-0 | 0-6 C. | No ice in} sight. | / } | Colour of Water, Current, and Remarks, Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- son water- bottle Thermometer No. 18,725 Nansen-Petters- son water- bottle Thermometer No. 18,725 Reeve’s French blue at 7.0 Reeve’s French blue Reeve’s French blue Time and Date. 13.30 8/4/04. 14.0 8/4/04 14.45 8/4/04 10.15 9.50 4/4/04 16,30 3/4/04 Temperature during Weights Experiment. | added to Hydro- Mle t’. meter (grams) of of the the Air. |Sample.! w. pel: 2 (Gh, 50-2 12:8 5-40 10-1 C. 5:45 5-50 5:60 5-70 ; 5:80 5:90 5-95 50:4 12-5 6-00 10-2 C. 12-65 5:70 50-2 11-7 5-40 10:1 C. 5:45 5:50 5:60 5:70 5:80 5:90 5:95 50:3 116 6-00 10-2 C. ———_ 11:65 5:70 49-2 10-6 5-50 9-6 C. 5:05 5:60 5-70 5:80 5:90 6:00 6:05 48-0 10:5 6:10 ~ 8&9 C. — 10°55 5:80 55:5 9-4 5-40 13-1 C. 5:45 5:50 5-60 5-70 5-80 5-90 5-95 10-0 6-00 9:7 5:70 59-6 14:6 5:30 15:3 C. 5:35 5-40 5-50 5-60 5:70 5:80 5:85 59-0 14-7 5:90 15:0 C. 14-65 5:60 61:5 18-1 5:50 16-4 C, 5-60 5:65 5-70 5:20 5:30 5:40 5:10 61-0 17:9 5-15 16-1 C,,—-——— 18-0 5:40 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC | Volume of im- Read- | mersed ing Portion of of Hydvro- | Hydro- meter, meter (cub. centims.) R. Vv. 5-0 9:8 14:8 25-0 35:8 46:0 56:5 1:5471 61:8 0789 67:0 +3229 35:7 1:9489 2-5 7:8 * 12:3 23-0 33-0 43-6 54-0 1-5471 59-0 0727 64-8 +3012 83-3 1:9210 9-8 15:3 20-0 30:3 40-5 50:8 61:5 1:5471 66:3 0658 72-0 3682 40-7, 1-9811 5:0 10:0 15-3 26-0 36:8 AT-5 58:5 1:5471 63-5 0605 69-0 +3329 36-8 1:9405 | 10 7-0 12-2 17:5 28-0 39°5 50:2 61:0 1-5471 66:0 -| 0914 71:5 +3546 39-2 1:9931 | 1-0. 51-0 61:5 66-0 70:5 18:0. 28-0 38-0 15471 8-0 +1123 10-0 +3528 39-0 | 20122 | 102421 NBSALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN, 151 Data Relating to the Collection of the Sample. of Depth Position (L.) P Temperature : in Fathoms. cat time of ad —_ 7, | Collection o D. d. | Sampie. to 4 _| Colour of Water, of from P Current, and Se. which Remarks. rifour.} Lat, Long ae the Posi pee of of tion oi the the L. Tenteal Air, | Water. | ° 4 ° ‘7 ° F, ° F, 6.0 | 5716S. | 1037 W. Surface] 31-6 on é Blue Two dist/ant bergs. ; 18.0 | 5623S. |10 OW.| .. | Surface} 31-0 a8 4 Turquoise Two distjant bergs jin sigiht. #) 8.0|55 88. | 950 W./ 2270 32:2} 316 | Buchanan-Rich- | | —0-2C. ard water- The water-bottle, which was bottle fixed about 3 fathoms from the Thermometer No. sounding-tube, came up with 102,505 its sides pressed in; some water, with a little mud, was in the bottle; the sounding-tube, how- ever, was lost. 12.0 | 55 8S. |10 OW. | Surface| 33-8 ae Turquoise A distant) berg in si/ght. ; 12.0 | 5233S. | 947 W. Surface] 36-2 35-0 Cobalt 170, 12.0 | 51 7S 931 W. Surface} 36:0 | 35-3 Reeve’s French 1-8C.| blue (light) Data Relating to the Determination of the Density of the Sample. big eh eres Density of Sample. uring Weights = (Density of Distilled Experiment, | ‘added | Read: | mersed| ‘Water at 4° C.=1.) io ing yous Time Hydro- and Me t’. meter a Hite: Ob-_ |Reduced|Reduced Date. (grams). gua (ens served to to a ¥ centims.) att’. |15°-56C. t. the the Air. |Sample.| w. Ih. Vv. 4Sy, | 4Sises6.| 4S, ak SiC 10.10 49-0 13-3 5:30 11:5 4/4/04 | 9-40. 5-35 17-0 5-40 22:5 5-50 32-5 5:60 43-0 5-70 54-0 5:80 64-8 1:5471 5:85 70-2 ‘0811 oe 6 12-7 5-90 75:0 38926 “2C. ————— 13:0 5-60 43-4 20208 | 1:02526| 1-02473| 1:02688 10.35 |47-4 12-0 5:30 5 4/4/04 | 8-6C. 5-35 12:5 5-40 175 5-50 28:8 5:60 39-5 5-65 44-5 5-70 50-0 | 1:5471 5-75 55°5 0730 46-2 6 11-4 5:80 60:5 +3175 7-9 C, | 11:7 55611} 35:1 | 1:9376 | 1:02552| 1-02474| 1-02688 Sample lost djuring galje on the |night of 1/8tb Aprijl 1904. 9.15 [50-7 9-0 5-40 15 8/4/04. |10-4 C. 5-45 12-5 5-50 19-0 5:60 29-0 5:70 39-8 5:80 50-5 5-90 60-8 1-5471 5:95 66-0 0565 e s ‘ 9-1 6-00 72-0 +3591 9:05 5-70 39-7 1:9627 | 1-02614| 1-02489| 102705 15.8 |47:3 9-8 5-40 13-0 8-50. 5°45 18-0 5-50 23-0 5-60 33-5 5:70 43-5 5:80 54-0 5:90 64:5 1:5471 5:95 69-5 0605 46-5 7 9-6 6-00 75-0 3962 8-10. SS 9-7 5:70 43-8 2:0038 | 1:02591| 1-02478) 1-02688 13.55 |53-0 11-0 5:50 27-5 15/4/04 |11-7 C. 5:30 7-0 5:35 12-2 5-40 17-2 5-60 38-0 5:70 48-0 5:80 59-0 1:5471 5°85 64-0 0696 52:8 11:3 5:90 69:5 38437 11:6 C,} =| 11:15 | 5-60 38-0 | 1-9604 | 1:02560| 1.02472) 1-02681 22 152 Number of Station. 448 449 450 451 | 453 454 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC @ Data Relating to the Collection of the Sample. Data Relating to the Determination of the Densi Date (E.) Position (L.) ‘ ee a Temperature Temperature Volume f RY UO ssi eros during Weights of im- = Collection of Experiment. | added Read- | mersed ay 1by d. Sample, vo ing Pore g Colour of Water Time Hydro- ° e a ! of Boe T t Current, and : and ie tite meter ie ee ee Month. |Day.! Hour.! Lat. Long. pe the |——|/———— Pemaris. Date, Jas Merams). (oe | { ce } a | Posi-|9@mple} ot | of of |- of | ma =) tion Nia the the the the va L. leetaae Air. | Water. Air. |Sample.| w. Re VE 1904. ee apes met OS ous cary, 2 (Ch 473| Apr. | 10 | 12.0 | 4925S. | 921 W.| .. | Surface] 38:5} 39-0 Cobalt 14.0 51-9 oui 5:40 11-0 | 3-9 C. 16/4/04 |11-1 C. 5:45 16-0 5:50 21:0 5°60 32:5 5-70 43-7 5:80 54:2 ; 5:90 64-2 | 1-5471 5-95 69-5 0618 53:0 10-1 6:00 74:5 +3890 11:7 C. S| 9-9 5:70 43-0 19979 474) Apr. | 11 | 12.0 | 4853S. | 925 W.| .. | Surface] 38-7] 39-4 Dark blue 14.30 |53-0 10-0 5-40 11:3 410, 16/4/04 |11-7 C. 5-45 | 17-0 . 5:50 22:0 5-60 33:5 5:70 44:5 5:80 55:0 5-90 66-0 1:5471 5:95 71:0 0640 53:8 10:5 6:00 76:0 +3980 12:1. ——— 10-25 5-70 44:0 20091 | 475| Apr. | 12 | 13.0| 48 OS. | 950 W.| .. | Surface} 38-8} 40-0 Reeve’s French | 15.0 (54:0 10:3 5-40 14:0 D.R. 4:40, blue (light) 16/4/04 |12-2 C. 5:45 19-0 5:50 24:0 5:60 35-0 5:70 45-5 5:80 56-3 5:90 66-5 1-5471 5-95 72:0 0661 54:5 10:9 6-00 175 +4116 12-5 C. SS 10:6 5:70 45-5 2-0248 | 476) Apr. | 13 | 13.0/ 48 6S. |10 5W.| .. | Surface 35-8) 40:8 Reeve’s French 9.15 (51-2 10-6 5:30 1:0 4-9 C. blue 20/4/04 |10-7 C. 5:35 6:0 5-40 11-0 4 550 | 22-0 5-60 33-0 5-70 44-0 5:80 55:0 1:5471 5:85 60:0 0665 51:5 10-7 5:90 65:2 +2985 10-8 C. —————|— ————S| 10:65 5:60 33-0 1:9121 477; Apr. | 14 | 13.0 | 4635S. |1010W.| .. | Surface] 39-2} 42-9 Cobalt blue 9.55 |52-2 10-7 5:30 05 6-1C. (light) 20/4/04 |11-2 C. 5-35 55 5:40 10:5 5:45 16-0 5:50 21:0 5:55 26:0 5:60 32:0 | 1:5471 | 5:65 38-0 0677 52-7 11-0 5-10 42- +1927 11:5 C. ————_]—-——_ ai 10°85 5:50 21:3 1-8075 | 1: 478) Apr. | 15 | 13.0 | 4554S. |10 4W.| .. | Surface] 48-9] 44-2 Cobalt blue 10.20 |52-9 10:8 5-40 10-5 6-8 C. 20/4/04 |11-6 C. 5:45 16-0 5:50 21:8 5:60 32:5 5-70 43-0 5-80 54:0 5:90 65-0 15471 ) 5:95 70:0 0686 | 53-4 | 11-2 6-00 75-0 +3899 | 11-9 C oe : 11-0 5:70 43-1 2:0056 | 1-02: SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 153 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ont Depth Volume i Position (L.) = Temperature Temperature : Density of Sample. ' in Fathoms. | “of, wane of during Weights ofim- (Density of Distilled Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) D d. Sample. me ing Honion Sa _| Colour of Water, | Time Hydro- of ¢ of | from t Current, and and a tite meter ee ee Ob-_ |Reduced|Reduced Sea | Which : Remarks. Date. (grams) ‘ b. | Served to to ay.|four.| Lat. Long. | “4 |_, the — | ee: | at t’. 115°56C.| +. Posi eample of of Ay of centims. ‘ tion va the the the the L. | jected. | Ait: | Water Air. Sample. w. R. Vv. 4S¢, | 4S1s-56,| 4S¢. | oo. 2 Ch a2 oF, nat oR, ren | 13.0 | 45 25S. | 1019 W. Surface] 46:0 | 45-2 Reeve’s French 10.50 (53-5 10:9 5:40 | 10:5 | 7-3 C. blue 20/4/04 |11-9 C. 5:45 15:5 5:50 20:5 5:60 32:0 5:70 42:5 5:80 53-0 5:90 64:0 1-5471 5:95 69:8 0693 53:8 11:3 6:00 75:0 +3845 12:1, ———_ 11-1 5:70 42-5 2:0009 | 1:02592) 1-02503| 1:02653 3.0 | 4430S. | 943 W.| .. | Surface] 44:8) 47-8 Cobalt 9.25 |51-7 11-4 5-40 9-2 8-8C 27/4/04 |11-1 C. 5:45 14:5 5:50 20-0 5:60 31:0 5:70 41:0 5:80 52:0 5:90 62:5 1:5471 5:95 68:5 0714 52-4. 11-5 6-00 73-0 +3736 11-3 C. —_—_— ———— 11-45 | 5-70 | 41-3 | 1-9921 | 1-02597| 1-02514) 1-02642 18 $12.0 | 48 41S. | 830 W. Surface] 55:0} 52:0 Cobalt 9.50 {53-0 11:5 5:40 8-0 D.R. 11-:1C 27/4/04 |11-7 C. 5-45 | 13-0 5-50 18:8 5:60 29:0 5:70 39:8 5:80 50:0 5:90 61-0 1:5471 5:95 66:5 0721 53-2 11:6 6:00 71:3 +3590 11-8 C. —_—_ — 11:55 5:70 39-7 1-9782 | 1:02605) 1:02523| 1:02613 425758. | 813 W Surface) 45-3] 50:8 Cobalt 9.0 56-8 12:6 5:30 15 Das 10:4 28/4/04 |13-8 C. 5:35 6-7 5-40 12:5 5-50 23-5 5:60 35-0 5:70 45:8 5:80 56:5 1:5471 5°85 61-5 :0802 56-9 13-1 5:90 66-5 +3112 13-8 C. —_—_——_ — eel 12°85 5:60 34-4 19385 | 1:02573| 1:02518 | 1-:02619 41 30S. | 955 W.| 1998] Surface] 49-5 | 55-0 Reeve’s French 9.25 [57-2 13-2 5-40 10:0 D.R. 12-8 C. blue 28/4/04 |14-0 C. : 5:45 15-2 5:50 21-0 5:60 31-2 5:70 42-5 5:80 53-0 5:90 63-5 1-5471 5:95 69-0 0833 57-2 13-5 6-00 74:0 +3817 14:0 C. — SSS] 13-35 5:70 42-2 2-0121 | 1-02586 | 1-02540 | 1:02597 40 20S. | 9 563 W. Surface 56:5 | 55-2 Reeve’s French 9.50 57-2 13-1 5:40 10-0 12:9 C blue 28/4/04 |14-0 C. 5:45 15:0 Gough Is/land. 5-50 21-0 5:60 32-0 A 5-70 42-8 5:80 53-5 5:90 64:0 1-5471 5:95 69°5 0827 57-5 13-4 6-00 74:0 +3835 14:2. a 2-4 | 2-0133 | 1:02585)| 1:02537 | 1:02592 154 Data Relating to the Collection of the Sample. Te Depth Date (E.) Position (L.) . Temperature in Fathoms. “at time of : - "= nll mae) Colleetion/of & = D. d. Sample. @|& Colour of Water, m | 2 of ent T. t Current, and & Month. |Day.| Hour.| Lat. Long. Sea i || Hemerks: 2 = Posi- Sens of of 5 = tion eT the the Z2la L. | jected. | Ait: | Water. | 1904. ed a | Sanat eae 462 | 485) Apr. | 24 | 12.0 | 3958S. | 836 W.| 1807| Surface) 51-4) 55-2 Reeve’s French ; 12:9 C. blue 463 | 486) Apr. | 25 | 12.0 | 4022S. | 545 W. Surface] 52-0] 53-0 Reeve’s French : 11:7 C. blue 464 |487| Apr. | 26 | 12.0 | 4115S. | 238 W. Surface) 51-1] 51:9 Reeve’s French 11-10. blue 466 | 488; Apr. | 27 | 12.0 | 4033S. | 0 7E. Surface] 50-2} 51-4 Reeve’s French 10-8 C, blue 467 | 489) Apr. | 28 | 12.0 | 40 8S. | 150E. | 2645] Surface] 54-8] 52-4 Reeve’s French 11:3 C. blue 468 |490| Apr. | 29 | 12.0 | 3948S. | 233 E. Surface] 53:0] 53-8 3 Light cobalt 12-1 C. Data Relating to the Determination of the Dens Time and Date. DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA) 10.10 28/4/04 10.30 28/4/04 10.55 28/4/04 11.15 28/4/04 12.10 15.10 30/4/04 Tee uring Weights Experiment. Cad Read- wo |g Hydro : ie fie pater Hydro: (grams) z of of the the Air. |Sample.| w. Ra Hal te SC: 57-4 13-1 5:40 85 14:1 C. 5:45 13-7 5:50 19-5 5:60 30:0 5-70 41:0 5:80 52:5 5:90 63-0 5:95 68:8 58:2 13-5 6:00 74-0 14:6 C. 13-3 5:70 41-2 58-2 13-1 5-40 11:0 14-6 C. 5:45 17-0 5:50 22:0 5:60 33-0 5:70 44-0 5:80 55-0 5:90 65°5 5:95 70-5 58-5 13-5 6-00 75:5 14:7 C. 13-3 5:70 43-7 58:0 13:2 5:40 12:5 14:4 C, 5:45 17-8 5:50 24-0 5:60 34-0 5-70 44-8 5:80 56:0 5:90 66-0 5-95 71:5 57-5 13-5 6-00 76-5 14:2 C. — 13-35 5-70 44-8 57:5 13:2 5:40 14:5 14:2 C. 5:45 19:8 5-50 25°5 5-60 36:5 5-70 47:0 5:80 58:0 5:90 69-0 5:95 74:0 58:0 13°5 6-00 79-0 14:4 C, 13°35 5:70 47-0 57-6 12-4 5:40 12-0 14:2 C. 5:45 17-0 5:50 22:5 5:60 34-0 5:70 45:0 5:80 56:0 5:90 66-5 5:95 71:5 57-9 12-9 6:00 77-0 14-4 C, 12:65 5:70 44-6 58-2 14-4 5:50 27:5 14:6 C. 5:30 6:0 5:35 11:8 5-40 17-0 5:60 380 5:70 49-0 5:80 60-5 5:85 66-0 58-5 14-7 5:90 71: 14-7 C, - 14-55 5-60 38:5 Volume ofim- | mersed | — Portion | 1-5471 +0830 | +3727 2:0028 | ; 15471 0830 «3953 2-0254 | 1-025' 15471 0789 4035 2:0295 SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 155 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. | sys Depth : Vol : te@l) Position (L.) : Pp Temperature Temperature ashen Density of Sample. | in Fathoms. | “a¢ time of auvine Weights ofim- | (Density of Distilled reer = =anluegeen |) Collection of Experiment. | added Read- | mersed Water at 4° C.=1.) Dz. d. Sample. i ing | Portion ; : f of Colour of Water,| Time Hydro- a of | from T t Current, and and 1: tie meter yates Hy oe Ob-_ |Reduced|Reduced Sea | Which ‘ ; Remarks. Date. (grams).| eT | Meter | served to to DayHour.} Lat. Long. aE the = ey att’. |15°56C. fy Posi- ee of of of of ane, tion| a) | the | _ the the | the L. feated Air. | Water. Air. |Sample.) w. -R. V. a95t, | 4Sises6.] 4S¢, a4 Qt pelle ells 2. SG 12.0 | 39275. | 550 E. | 2900| Surface] 54:2 | 55-5 Light cobalt 15.30 58-5 13-8 5:40 12-5 13-1 14:7 C. 5:45 17:5 5:50 23:2 5:60 33:8 5:70 45-0 5-80 55:5 5:90 66:0 | 1-5471 5:95 71:0 -0867 58-2 14:0 6-00 76:0 +4026 ; 146 C. —— 13-9 5:70 44-5 2-0364 | 1:02572)| 1-02537 | 1:02588 12.0 | 39 25S. | 10 25 E. | 2625| Surface] 58-0] 55-5 Reeve’s French 15.0 (68-7 19-8 5:50 47-0 ; 13-1 C blue (dark) 4/5/04 |20-3 C. 5:20 16:0 5:25 21-0 5:30 26:0 5-40 37-0 5-60 58-0 5°70 69-5 1-5471 5°75 75-0 +1242 69-0 20-0 5:80 80-0 4315 . 20-6 0. —— 19-9 5:50 47-7 2:1028 | 1:02425| 1:02529| 1-02580 2712.0 | 38 6S. | 1432 E. Surface] 58-0] 64:9 Antwerp blue 15.20 |69-0 20-1 5:20 1:0 18:3 C 4/5/04 |20-6 C. 5:25 6:5 | . 5:30 115 5-40 22:0 5:50 33:5 5:60 44-5 5-70 55-0 1-5471 5°75 60:0 -1260 69-5 20:3 5°80 65-5 +3012 20-8 C. SS 20-2 5-50 83+3 1:9743 | 1:02497 | 1:02610 | 1-02546 12.0 | 3537S. |15 3E. Surface) 62-4] 64-7 Reeve’s French] 15.45 |69-3 20:0 5:30 11:3 18-2C.| blue and Ant-| 4/5/04 |20-7 535 | 168 werp. 5°40 22-5 5-50 33-0 5-60 44:0 5-70 55:0 5-80 65-5 1:5471 5-85 70-5 *1257 : 69:3 20:3 5-90 76:0 3962 H 20-7 C. oS 20-15 5-60 43-8 2:0690 | 1:02499 | 1-02611| 1-02549 4 12.0 | 3458S. |17 OF. Surface] 64-1 | 64:7 Light turquoise | 16.0 (68-8 20:0 5:30 10:5 : 18-2 C 20:4 C. 5:35 16-0 ; 5°40 21:8 : 5-50 | 32:0 ; 5-60 43-0 | 5:70 | 54:2 5:80 65-0 15471 5:85 70:0 *1254 68-4 20:2 5-90 75-0 +3899 20-2 C, a 7 20-1. 5:60 43-1 2:0624 | 1:02503| 1:02614| 1-:02552 5 |12.0 | Off Cape |Peninsula. Surface] 60-1] 61:3 Dull dark green | 16.0 (62:3 17-1 5:30 2-8 16-3 C 16:8 C. 5°35 8-0 5:40 13-5 5-50 24-6 5-60 35:3 5-70 46:5 5:80 56°8 1:5471 | 5:85 62-0 °1070 . 62-4 17:2 5:90 67-7 +3184 16-9 C, = 17-15 5:60 35-2 1:9725 | 102553 | 1:02590 | 1-02572 156 Number of Station. 480 484 485 487 489 492 DR W. 8. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRAY Data Relating to the Collection of the Sample. Date (E.) Position (L.) en Ea ele foie tena ie art he Collection of e | D . Sample. |g ar ie i) from bare oe which t | = |Month. Day.) Hour.| Lat. Long. the. |= 5 at. | Sample a Posi-|"oeP | of of 5 tion ae the the Zi L. | jected, | Alt: | Water. 1904. oe We | euFl sig 497, May | 18 | 13.0 | Off Dassen Island. | .. | Surface) 60-4) 59-5 | 15:3 C. 498| May | 22 | 12.0 | 3138S. |1515 5. Surface] 62:7 | 62-4 . 16-9 C, 499| May | 23 | 12.0 | 30 5S. |1235 5. Surface] 63:0] 64-4 s 18-0 500| May | 24 | 12.0 | 2819S. | 956 HE. Surface] 64-0} 65-0 18-3 C 501; May | 25 | 12.0 | 2612S. | 7 5E. Surface] 64-4} 65-3 18:5 C, 502} May | 26 | 12.0 | 2414S. | 432 E. Surface} 65-0 ee 19-1 Data Relating to the Determination of the Density of eee | vou uring Weights Im Experiment. a ea mere to aa = f of - Colour of Water, Time Hydro- rk Current, and , and A t. meter Hy une Hy ange Ob Remarks. Date. (grams).| tC? (core re Ge at t’, ee of centims.) the the Air. |Sample.| w. R. V. Sage green 9.0 |73-0 22:3 5:20 17-8 30/5/04 |22-8 C 5:25 23-1 5:30 28-2 5:40 39:0 560 | 612 5:60 : 5-70 71-0 15471 5:75 76:3 +1396 iaG 22-45 5:80 81:5 4505 22:375| 5:50 49-8 2:1372 Pale turquoise 9.25 |72-4 22-4 5:20 12-5 30/5/04 |22-4 C 5:25 17-6 5:30 22:0 5:40 32:9 ae ane 5:60 5:0 5:70 66:0 | 1-5471 5:75 71:0 +1401 Le 22-5 5:80 76-5 +3998 ._ :\22:9'C. ———— 22-45 5:50 44-2 2:0870 | 1: Light cobalt 9.45 |73°8 22-4 5:20 12-0 30/5/04 |23-2 C 5:25 17-0 5:30 22-0 5:40 32:3 5:50 43:3 5:60 54:0 5:70 64-5 1:5471 5:75 70:0 +1404 a 22:6 5:80 75:5 +3926 23-1 C. —————SS 22:5 5:50 43-4 2-0801 | 1 Light Reeve’s 10.5 = |73-9 22-45 5-20 10:0 French blue 30/5/04 |23-3 C 5:25 15:0 5:30 20-0 5:40 31-0 5:50 42:0 5:60 53:0 , 5:70 63-5 1:5471 5:75 69-0 +1406 Le 22:6 5:80 74:5 +3799 22:9 C. —_———— 22:625| 5:50 42-0 2-0676 Reeve’s French | 10.25 {73-7 22-4 5:20 9-8 blue 30/5/04 |23-2 C. 5:25 15-0 5:30 20:0 5:40 31-2 5:60 ' 5:70 64:0 | 1-5471 5:75 69-2 +1404 he 22:6 5:80 75-0 +3817 23-3 C, ———_ 22:5 5:50 42-2 2:0692 | 1: Reeve’s French 10.45 |73-9 22°45 5:20 8-0 blue 80/5/04 |23-3 C. 5:25 13:5 5:30 19-0 5:40 30:0 500 | 520 . 0 . 5-70 63:0 | 1:5471 5:75 68-0 +1407 ee 22:65 5-80 73-5 +3700 23:2 C, —_ | 22-65 | 650 40-9 | 2-0578 Data Relating to the Collection of the Sample. SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 157 Data Relating to the Determination of the Density of the Sample. re Depth aa : Volume | F Position (L.) : Temperature Temperature Pye Density of Sample. joe Fathoms. | “34 bane of during Weights of im- (Density of Distilled Ea Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. Sample. i ve onsen Colour of Water. Time Hydro- e by of | from T t Current, and ’} and AN tis meter etal yan Ob-_ |Reduced/Reduced Sea | Which : Remarks, Date. (grams) b served to to our.} Lat., | Long. | ,'|_ the == ; ee =| abt’. |15°56C.| ¢. Posi- Sample of of of of entims.) tion| "eS | the | _ the the | the L. lected: Air. | Water. Air. |Sample, w. R. V. aS, |4Sis-56.| 4S, | oy Sip? cake Sues oF. Or 27} 12.0 | 2223S. | 1425. Surface] 67-1 | 68:0 Reeve’s French | 11.5 |74-1 22:5 5:20 6:0 20-0 C. blue 30/5/04 |23-4 C. 5:25 11:0 5:30 16-0 5:40 27-0 5-50 38-0 5:60 48-8 5:70 59-7 1-5471 5:75 65:0 +1410 13-4 22-7 5:80 70:0 3428 23-0 C. —_—— 22-6 5:50 37-9 2:0309 | 1:02466 | 1-02648)| 1:02538 e 12.0 | 19 45S. | 035 W.| .. | Surface] 68:0} 69-0 Reeve’s French }| 11.25 |73:3 22:5 5:20 1:5 20-6 C blue 30/5/04 |22-9 C. 5-25 7-0 5:30 12:0 5-40 23-0 5:50 34:0 | 5°60 45-0 5:70 56-0 1:5471 5°15 61:5 +1410 73-4 22-7 5:80 67-0 +3085 22:0 C. ———_ 22:6 5-50 34-1 1-9966 | 1-02485 | 102667 | 1-02542 12.0 | 1725S, | 234 Ww. Surface| 72:0 | 70-7 Reeve’s French } 11.45 |73-5 22-45 5:30 10-0 21:5 C. blue 30/5/04 |23-1 C. 5:35 15-0 5:40 20:0 5:50 31-0 5:60 42:0 5-70 53-0 5-80 63-5 1-5471 5:85 69-0 +1407 73:9 22-65 5:90 74:0 38790 23:3 C. — —_—_—_—— 22:55 5:60 41-9 2-0668 | 1:02500 | 1-02680 | 1-02530 307/|2.0 | 1554S. | 459 W. Surface] 71:3 72-2 Reeve’s French } 13.25 |78-9 25:6 5-20 11-0 22:3 ©. blue 6/6/04 |26-1C. 5-25 17-0 5:30 22:0 5:40 32:5 5-50 43-0 5-60 54:0 5:70 65-0 1-5471 5:75 70:2 “1601 79:5 25-7 5:80 75:2 3917 26-4 C. —— 25°65 5:50 43-3 2:0989 | 1-02427 | 1:02697 | 1:02525 | 2.0 | 1431S.) 7 5W. Surface] 72:2 | 74.2 Reeve’s French 13.45 |79-8 25:7 5-10 0-5 23-4 C. blue 6/6/04 |26-6 C. 5-15 57 5:20 11-0 5:30 22:0 5:40 33-0 5:50 45:0 5:60 55:0 5°65 60:8 1:5471 5-70 65:5 1613 80:0 26-0 5°75 71-0 *3347 26-7 C. —————S = 25°85 5-435 37-0 2:0431 | 1:02422) 1:02698| 1-02494 158 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC G Data Relating to the Collection of the Sainple. and Depth Date (E. Position (L. : Temperature (E.) (Li) in Fathoms. cat time of : lec) ae a a ec imeem aman) collection of 5 Z, D. dq. Sample. 3|& | = Colour of Water, n|n t of | from T t Current, and si3 | Sea | Which | *- : Remarks. = | s |Month. Day.) Hour.| Lat. Long. oe Woe} | SN Sample a/|a Posi ae of of E E fren |e the the col- + A|lG L. lected: Air. | Water. } _ 1904. SF a oun ae 502 508 June 4} 12.0 | 1240S. | 9 6W. Surface] 73-0} 73-3 Reeve’s French | 22-9 blue | | 503 | 509} June | 5 | 12.0 | 1046S. |1112 W.| .. | Surface] 75-0 ie é Light cobalt 505 | 510) June 6 | 12.0] 8518S. |13 2W. Surface] 77-5 a P Cobalt 508 | 511} June | 10 | 12.0 | 7386S. | 1433 W. Surface] 77:9 | 78-2 Light Reeve’s 25:7 C.| French blue 509 | 512} June | 11 | 12.0 | 4478S. |15 47 W. Surface| 78-4 eo 5 Dull cobalt 510/513! June | 12 | 12.0| 2468, |1724 W. Surface) 77-5 | 76-5 Turquoise and 24:7 C cobalt ; | | | Data Relating to the Determination of the Densi Henapenarede vo uring Weicht: * Experiment. ie Be mene to , f Time Hydro- oF and is te meter Bee Date. (grams). : x at 3 centims.) the the Air, |Sample.| w. R. We SaR. aCe 14.0 {80-2 26-0 5:10 1-0 6/6/04 |26-8 C. 5-15 7-0 5:20 12:0 5:30 23-0 5-40 34:5 5:50 45:0 5:60 56:0 | 1:5471 5:65 62:0 +1629 80:2 26:2 5-70 67-0 +3094 26-8 C ———_ 26-1 5:40 34:2 14.30 |80-0 26:1 5:10 T5 6/6/04 |26:7 C 5:15 12:5 5-20 18-0 5:30 29:0 5:40 40:0 5:50 50:0 5:60 : 80-2 26:1 26-8 C. 26-1 14.45 |80-4 26:1 26:9 C. 80-2 26:3 26:8 C. 26:2 6.40 |76:8 25:3 12/6/04 |24-9 C. 17:3 25:3 25-2 25-3 6.50 |77-4 25:3 12/6/04 |25-2 C. 17-5 25-4 25-3 C. 25:35 12.5 79:0 25-25 26-1 C. 79-0 25°45 26-1 C. 25-35 a } : AND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 159 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. ate (E.) Position (L.) . ? ae Temperature Temperature Volume Density of Sample. yn Fathomns- | at timeof during Weights of im- (Density of Distilled MM, o.oo OD =| Collectionof Experiment. | added | Read- | mersed Water at 4° C.=1.) 10), d. Sample. ag ing Portion ae Colour of Water, | Time Hydro- | yy oie H Oe of from T i Current, and and as ti, meter ee ai et an Ob-_ |Reduced/Reduced Sea which : : Remarks. Date. (grams). cA b served to to mi. | Day.) Hour.| Lat. Long. | ‘5, |_ the = aaa (cub. | att’. l15°56C.| t. . |Sample centims.) Posi- of of of of ae =| exces ees tion} WeP | the | the the | the L. Fecal! Air. | Water. | Air, |Sample.) w. 13, Vv. aS, |4Sis-s6,| «St, 4 : : = | | ) eae gine Stall ae OTK LGE ul | 13 |12.0 | 0158S. |1832 W.| .. | Surface) 76:9} 76-0 Pale turquoise 6.45 |78-1 25:5 5:10 9-0 24-4 C. 14/6/04 |25-6 C. 5-15 15:0 5-20 20-0 5:30 31-0 5-40 41-5 5:50 53-0 5-60 64-0 1-5471 5-65 69-0 1594 78-6 25:6 5:70 74-0 +3781 25°9 C. | | 25:55 5:40 41:8 20846 | 1:02380| 1:02648| 1-02414 ij | 14/ 6.0 | 156N./1921 W.| .. | Surface] 77-3] 80-1 Antwerp blue 7.0 |78-6 26-25 5:00 15-2 26-7 C. 25-9 C 5:05 20:8 5:10 26-0 5:20 36:8 5:30 47-0 5:40 58-0 5:50 68-8 15471 5:55 74-0 1638 78-5 26-25 5:60 79-0 4279 25:8 C. 26-25 5:30 47-3 2-1388 | 1-02295| 1-02581)| 1:02281 mj | 14} 7.55] 156 N.|19 21 W.| Cir. | 1500 | 77-8 a Nansen-Petters-} 15.30 |79-2 26-2 5:00 17-5 2700 son water- 26:2 C. 5:05 22-5 bottle 5:25 44:0 15471 Thermomleter did njot recjord. Thermometer No, - 5:45 64-9 -1636 18,729 79-6 26-25 5:50 70:0 +3962 26:4 C. 26:225| 5:25 43-8 2-1069 | 1:02286| 1-02571 uj | 14 | 7.45) 156 N.|19 21 W.| Cir. | 1000 | 77-8] 39-9 Buchanan-Rich-} 15.25 /|79-3 26-4 5:00 15-1 2700 440. ard water- 26-3 C. 5:05 20-5 Read on bottle 5-10 26-0 bridge | Thermometer No. 5-20 36:8 102,506 5:30 | 48-2 5:40 59-0 5-50 69-0 1-5471 5:50 74-0 +1647 79:3 26-4 5-60 78-0 -4288 26-3 C. S| 26-4 5:30 47-4 2-1406 | 1:02294| 1-02586| 1-02770 | 14} 910! 156N.|1921 W.| Cir.| 500 | 78-2] 40-6 Buchanan-Rich-} 15.0 /79-:3 26-4 5-00 20:8 2700 48C.| ard water- 26-3 C. 5-05 26:5 Read on bottle 5:10 32-5 bridge | Thermometer No. 5:20 43-0 102,505 5:30 53:0 5:40 64-0 5:50 74:5 1-5471 0:00 79-5 -1647 79-3 26-4 5-60 84-0 +4803 26-3 C. — Satara 26-4 5°30 d31 2-1921 | 1-02265 | 1-:02556| 1-:02736 . ea ij | 14 | 8.30) 156 N.|1921 W.| Cir.| 400 | 77-8 ot Buchanan-Rich-j 14.30 |79-2 26-6 5:10 32-0 2700 ard —water- 26-2 C. 5-20 43-0 bottle 5:30 | 53-9 1-5471 Thermometer No. 5-40 64-0 -1660 102,506 79-2 26-6 5:50 74:2 -4840 26-2 C. 26:6 3:30 | 53-5 2:1971 | 1:02262) 1:02559 4 160 DR W. 8S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GR Data Relating to the Collection of the Sample. Temperature 5 sae Depth Date (F.) Position (L.) 2 RR one iia ; . Collection of 5 = D d. Sample. 3 | Colour of Water, n | nD of oe T t amen ere = lo : W. : R ks, = = | Month. \Day.| Hour.| Lat. Long. A ae ees o 5 = Posi Se of of z 5 tion Men the the . a |G L. Tena Air. | Water. 1904. ) ay Po, | oF | oR 513 519s June | 14 | 9.6 156 N. | 19 21 W. oe 400 78:2 ar Bucane ee 0 | 5:3 C. ar water- Read on bottle bridge | Thermometer No. 102,506 513 |520} June | 14 | 9.80} 156 N.|19 21 W. ae 300 | 78:2 oi [upebenan tics x ar water- bottle Thermometer No. 102,505 513 | 521) June | 14 | 9.26) 156 N./19 21 W. Sie 200 | 78-2 aS te Buphanan, Bich ) 9-30. ar water- bottle : Thermometer No. 102,506 | 513 | 522} June | 14 | 10.0 156 N. | 19 21 W. va 100 | 78-2 ae Bucimnar eh 12- ar water- bottle Thermometer No. 102,505 513 |523| June | 14 | 9.41} 156N./19 21 W. a 50 | 78:2 ae Buchanan-Rich- 14-2 ard water- bottle Thermometer No. 102,506 513 | 523 June | 14 9.41; 156N. | 19 21 W.| oF 50 | 782) 57-5 F Buchanan-Rich- | 27 14:2, ard water- bottle Thermometer No. 102,506 \ Time and Date. 14.45 14.15 14.0 13.40 10.45 Tested again, 13.5 Terapenaiuze Volune ae Weights | Read- mersed Experiment. paiee ing Por tion Oo) 0 1’, ry yee Hydro- | Hydro- i (grams) meter. | meter ey § s (cub. of et a centims.) the the Air. Sample.) w. R. V. rake XG. 79-2 26-4. 5-00 19-5 26-2 C. 5:05 25-0 5:10 30-5 5:20 41-0 5:30 52-0 5:40 63-0. 5:50 73:8 1-5471 5:55 78-5 +1647 79-2 26-4 5:60 83-9 4695 26-2 C. ee 26-4 5:30 51-9 2-1813 79-0 27-0 5:00 23-8 26-1 C. 5-05 29-2 5:10 34:0 5:20 45-0 5-30 55:5 5:40 65-5 - 5:50 16-4 | 1:5471 5-55 81-5 +1682 79-5 26-9 5:60 85-8 +4993 26-4 C. ee 26:95 5:30 55-2 2-2146 79-2 26-1 5:00 15:5 : 26:2 C. 5-05 21-0 5:10 26:5 5-20 37-0 5:30 47-6 5-40 58-2 5:50 68-5 1:5471 5:55 73-5 +1629 79-2 26-1 5-60 78-0 -4279 26-2 C. Se 2 26-1 0:30 47-3 2-1379 79-9 26-1 5-00 11-2 26-6 C. 5:05 16:8 5:10 22-0 5-20 33-0 5:30 44-0 5-40 54-9 5:50 65-7 1:5471 5-55 70:8 +1632 79-2 26-2 5:60 76-0 +3962 26:2 C. 25 26-15 5:30 43-8 21065 79:6 23-4 5:10 5-5 26-4 C. 5-15 10-9 5:20 16:3 5:30 27-0 5-40 38-0 5:50 49-5 5-60 60:5 1-5471 5-65 65:8 +1479 79-4 24-0 5:70 71-0 +3465 26:3 C. | 23-7 5:40 38-3 2-0415 79:0 25:8 5-00 6-5 26-1 C, 5:05 11:5 5:10 17-0 5-20 27-5 5:30 38-6 5-40 49-2 5:50 60:5 1-5471 5:55 66-0 +1613 | 79-2 25-9 5-60 71-0 +3492 | 26:2 C — _——— rl 25-85 | 5-30 | 38-6 | 20576 AND SALINITIES OF THE WEDDELL‘ SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 14 | 14 ie |14 le 15 9.55 12.0 18.0 24.0 6.0 Position (L.) Lat. Long. (we oe 4 156 N. | 19 21 W. 19 26 W. 2 32 N. | 19 32 W. 3 48 N. | 19 36 W. 3 30 N. | 1951 W. 3 30 N. 19 51 W. Data Relating to the Collection of the Sample. 161 Data Relating to the Determination of the Density of the Sample. | Depth Temperature Tem | Volume i : perature f Density of Sample. | in Fathoms. | “st'time of during Weights of im- (Density of Distilled Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. d Sample. +5 ing | Portion —— — P f of 7 ore! Colour of Water, Time Hydro- OF | , | of from T t Current, and and Ms tie meter aye BOE By Cae Ob- |Reduced Reduced | Sea which 3 : Remarks. Date. (erams)s |e les ‘7 served to ton ee at |g tbe, Ae j| abt’. |15°560. t. Posi-|"9 mE '| of of of of | wa tion Asie the the the the | L, lected Air, | Water. Air. (Sample W. h. WS 49t, |4Susse.| 4S¢. } oon, halo ok. OG Cir. 20 | 78:2) 80:5 Buchanan-Rich-} 10.25 |78-9 26-8 5:00 16-2 2700 ; 26:9C.| ard water- 126-1 C. 5°05 21:8 bottle 5:10 26-5 Thermometer No. 5:15 32-0 102,506 5:25 42-5 5:30 A4T-4 5:35 53:2 1:5471 5:40 58-2 -1671 79:3 26-75 5:45 63-8 +3636 26-3 C. 26:775| 5:2277| 40-2 20778 | 1:02290| 1:02593 | 1:02286 Surface] 79:0} 80:8 Reeve’s French 13.20 |79-2 26-7 5:00 17:5 27-1 C blue 26-2 C. 5:05 22:6 2 5:10 27-7 5-20 38-4 5:30 49-0 5:40 60-0 5:50 71-0 1:5471 5:50 76:0 -1665 79-2 26-65 5-60 81-0 4451 < 26-2 C. ——-- 26:675| 45:30 49-2 2:1587 | 1:02285 | 1-02585 | 1-02272 Surface} 78-9 | 80-9 Reeve’s French 6.30 |80-0 26:8 5:00 18-8 27-2 C. blue 15/6/04 |26-7 C. 5-05 | 24-0 5-10 29-8 5:20 39:9 5:30 50:0 5:40 60-6 5:50 71-0 1:5471 5:55 76-0 +1672 80-0 26-8 | 5-60 81:0 4532 26:7 C. = 26:8 5°30 50-1 2-1675 | 1-02279) 1-02583 | 1:02267 Surface} 79:0} 80:9 Too dark 6.50 |80-0 26-7 5:00 18-0 27-2 C. 15/6/04 |26-7 C. 5-05 23-0 5:10 28-5 ) 5:20 39-2 5:30 50-0 5:40 60:5 5:50 71:0 1-5471 . 5:05 76-0 +1669 80-0 26:8 5-60 81-0 +4497 26:7 C. | ——— —-—-—— 26-75 0:30 49-7 2:1637 | 1:02281| 1:02583| 1-02268 2526} Surface] 78-8 | 80-2 Reeve’s French 7.20 |80-0 26-6 5:00 16:3 26:8 C. blue 26:7 C. 5:05 21:8 5710 26:8 5:20 37:5 5-30 48-8 5-40 59-0 5:50 69:8 1-5471 5:55 75-0 -1660 80-0 26-6 5:60 (8) +4369 4 26-7 C. ——__ ———__ 26-6 5:30 48-3 2:1500 | 1:02289)| 1:02587 | 1-:02283 2526| 500 78:8 | 40-0 Nansen-Petters-} 14.50 |78-8 25:7 5:00 18-0 4:40 son water- 26-0 C. 5:05 23-0 Read on bottle 5:10 29-0 bridge | Thermometer No, 5:20 39°8 18,727 5:30 | - 51-0 5:40 61-0 5-50 72:0 1:5471 5:55 77-0 1607 78-8 25:8 5:60 81- +4550 26:0 C. SSS 25:75 | 5:30 50°3 2:1628 | 1:02282| 1:02553| 1-02736 162 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC Gi Data Relating to the Collection of the Sample. Depth Data Relating to the Determination of the Densit Date (E.) Position (L.) : Temperature in Fathoms. atime of : ; ollection of 8 = D. d. Sample. 3|§ Colour of Water,} Time ell ee of ae T oh Caen aud and ow .: R 5 H) e “ Month. Day.| Hour.| Lat. Long. = the, || Map Bore o 2 = Posi seu of of gis tion} WS | the | _ the a|a L. echo Air. | Water. 1904. ° ¢ ° , Oat ine 517 | 5380; June | 15 | 7.80) 330 N.| 1951 W.| 2526) 400 | 77:0} 4092 | Nansen-Petters-] 14.35 5-00.) son water- Read on bottle bridge | Thermometer No. 18,727 517 | 531] June | 15] 8.5 3 20 N. | 19 51 W.| 2526} 300 | 77-0| 43-0 Nansen-Petters-} 14.15 61C son water- Read on bottle bridge | Thermometer No. 18,727 517 | 532) June | 15] 7.25) 3280 N. |] 1951 W.| 2526} 200 | 77-0} 50-2 Buchanan-Rich-} 14.55 10-1 C. ard water- Read on bottle bridge | Thermometer No. 102,505 517 | 533} June | 15] 8.0 3 30 N. | 19 51 W.| 2526} 100 | 77-:0| 56:9 Buchanan-Rich-} 13.40 : 13-80.) ard water- Read on bottle bridge | Thermometer No, 102,506 517 | 5384); June | 15] 7.55] 330N.} 1951 W. | 2526 50 | 77:0} 74:6 Buchanan-Rich- | 11.50 23:7 C. ard water- .| Read on bottle bridge | Thermometer No. . 102,505 518 | 535| June | 15 | 12.0 350 N. | 1958 W. Surface 76:8 | 80-1 Reeve’s French 12.10 26:7 C blue } | ) | Temperature during Experiment. t’. 25-4 25-3 25-1 25:3 25-2 25:6 25:8 25-7 25:8 25-9 26-85 26-1 26-1 26-1 26:6 Weights added ReneS R. 16-0 21:8 26:5 37-3 48-0 58-2 69:8 75:0 80:0 eras Quwnyee wl] aadacn QD Ne Ieoien VS =o NAOAHAWS Ww] ASSOAS 5:5 eS BAW 84:8 54-6 | 22067 | 10 Volume of im- mersed | Portion Hydro- meter centims.) 2:1401 | 1: 2.0675 | 1-02: of (cub. ‘ah Mis 1-5471 +1579 4351 1-5471 -1613 | Per’ ‘| vs 15471 | — +1629 | 12433 | 1-9533 | 1-023 Fie 1-6471 | 1657 | 4939 | ND SALINITIES OF THE WEDDELL SEA Data Relating to the Collection of the Sample. _| Colour of Water, Current, and Remarks. Data Relating to the Determination of the Density of the Sample. AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 163 Aan Depth , te (E. Position (L. : Temperature (E.) (L.) in Fathoms. aii ites OF = Collection of D. d. Sample. from ae which : t. hiDay.. Hour.| Lat. Long. |~ 34 thee Pp Sample f osi al ) of tion hes the the L. Pan | Air, | Water. 4, a7? io we op. | oT, | ie} 15 ) 8.30} 330 N./1951 W. 2026 | 60 | 77:0) 59-9 15-5 C. Read on bridge 6/15 | 8.25] 330 N./1951 W.| 2526 20 | 77:0) 80-6 27-0 C, Read on bridge e |15 | 16.0 414N./2011 W. Surface] 78-1 | 80-6 27:0 C0, e |15 | 20.0 423 N.|2010 W. Surface] 78-0 | 80-6 27-0 e 115 | 24.0 445 N. | 2015 W. Surface] 77:6 | 80-1 , 26-7 C, e 6 | 6.0 5 47 N. | 20 28 W. Surface] 77:2 | 80:3 26-8 C. Buchanan-Rich- ard water- bottle Thermometer No. 102,506 Buchanan-Rich- ard water- bottle Thermometer No. 102,505 Reeve’s French blue Too dark Too dark Antwerp blue 11.35 7.0 16/6/04 7.15 16/6/04 7.50 | Temperature _ Volume Density of Sample. during Weights of im- (Density of Distilled Experiment. | added | Read- | mersed Water at 4° C.=1.) iG ing Portion “| Hydro of of 4 1, us meter Eile Bye Ob- |Reduced/ Reduced (grams) Taeeel (cub, served to to : centims.) att’. |15°-56C. t. of of a aes the the Air, |Sample.| — w. R. i, 4aSy, |4Sis-56.| 4S5¢, (SIRE KO} 78:2 25:8 5:00 5:0 25:7 C. 5:05 11-0 5:10 16-5 5-20 27-0 5:30 37-6 5-40 48-0 5:50 59:5 1-5471 5:50 64:8 *1613 78:8 25:9 5:60 70-0 +3410 26:0 C. ——— 25:85 5:30 37-7 2:0494 | 1:02345 | 1:02620| 1:02621 78-0 26-2 5:00 11:5 25-6 C, 5:05 17-0 5:10 22:0 5:20 32:5 5:30 43-2 5-40 53°8 5:50 64-8 1-5471 5:55 70-0 +1635 78:2 26-2 5:60 75:0 3917 25-7 C. ——— 26-2 5:30 43-3 2:1023 | 1:02316| 1-02601| 1:02291 79:8 26-4 5:00 24-5 26:6 C. 5:05 30-0 5:10 35:5 5:20 46:0 5:30 56:8 5:40 67:8 5:00 17:8 15471 5°55 82-0 +1650 80-0 26:5 5:60 87-2 +5102 26:7 C. — ———— 26:45 5:30 56-4 2-2223 | 1:02248 | 1-02540 | 1:02231 80:0 26-3 5:00 26-5 26:7 C, 5:05 32:0 5:10 37-0 5-20 48:0 5:30 59-0 5-40 69-0 5:50 78:5 1:5471 5:55 83-0 +1644 80:0 26-4 5:60 88:3 5238 26:7 C. ————_ 26-35 5°30 org 2-2353 | 1:02241 | 1-02530 | 1-02221 80-0 26:3 5:00 29:2 26-7 C. 5-05 35°0 5:10 40-5 5-20 51:0 5:30 62:0 5:40 72-0 5°50 81:5 1:5471 5:55 86-5 +1644 80-1 26-4 5:60 91:7 3018 26-7 C, a 26:35 5:30 61-0 2:2633 | 1:02225| 1:02513| 1:02214 80-0 26-6 5:00 24-5 26-7 C. 5:05 29:5 5:10 35:0 5-20 45-0 5:30 55:2 5:40 66-0 5°50 17-2 1:5471 5-55 82-5 1663 80:0 26:7 5:60 87:8 +6057 26:7 C. a 26-65 5:30 559 2-2191 | 1:02250| 1:02549 | 1:02246 164 DR W. S. BRUCE, MR A. KING, AND MR D. W. Date (E.) Position (L.) | § = D. s|§ Se | nin of 5 | Month. 'Day.) Hour.| Lat. Long. Hee |} a|8 Posi 5 5 tion 2i4 L. [ | 1904. | ° f ° (! | 521 | 542; June | 16 | 8.50] 547 N. | 2028 W.| Cir. 1800 | | 521 |543| June | 16 | 8.53] 5 47 N. | 20 28 W.| Cir. i | ; 1800 | | 521 | 544! June | 16] 9.0 5 47 N. | 20 28 W. | Cir. 1800 *Therjmometer |did not re cord. 521 |545| June | 16 | 8.20] 547 N. | 20 28 W.| Cir. | 1800 : 521 | 546} June | 16 | 7.45] 547 N. | 2028 W.| Cir. 1800 521 | 547) June | 16 | 7.0 5 47 N. | 20 28 W. | Cir. 1800 } | 521 |548| June | 16 | 7.20] 5 47 N. | 20 28 W.| Cir. ) 1800 Depth in Fathoms. d. from which lected. 20 50 80 100 200 290 300 Data Relating to the Collection of the Sample. Temperature at time of Collection of Sample. 1 t.- WILTON ON _| Colour of Water, Current, and Remarks. of of the the Aix. | Water. aoe ie 78:3 | 73:6 Buchanan-Rich- 23-1 C.| ard water- Read on bottle bridge | Thermometer No. 102,505 | |78:3| Gd-4 Buchanan-Rich- 18:6 C. ard water- Read on bottle bridge | Thermometer No. 102,506 78:3 PA Nansen-Petters- son water- bottle Thermometer No, 18,727 78:3) 656-1 Buchanan-Rich- 13:4 C, ard water- Read on bottle bridge | Thermometer No. 102,505 78:3 | 49-8 Buchanan-Rich- 9:9 C. ard water- Read on bottle bridge | Thermometer No. 18:3) 465 8:10 Read on bridge 78:3) 45:5 7-5 C. Read on bridge 102,506 Buchanan-Rich- ard water- bottle Thermometer No. 102,505 Nansen-Petters- son water- bottle Therm meter No. ? THE TEMPERATURES, SPECIFIC GRAVE Time and Date. 10.45 13.30 11.20 11.30 11.45 13.15 Data Relating to the Determination of the Densit; Temperature vos iets) Weights Read- meee Experiment. are vials Poxtiae =< ~| Hydro- | ,, °F one |e | aioe | Baio | Ryd —- (grams) ; (cub. of Af centims.) the the Air, |Sample.| w. Te V. Be IG: 81-4 27-2 5:00 15:0 27-5 C. 5-05 20:0 5:10 25:0 5-20 35:8 5:30 47-0 5-40 57:5 5:50 68-2 15471 5°55 74:0 1700 82-0 27:3 5:60 78-0 4224 27-8 C. 27-25 5:30 46:7 2:1395 29: 80-2 26-6 5-00 11-0 26-8 C. 5-05 16-0 5-10 21-5 5-20 32:0 5:30 43-0 5-40 54:0 5:50 65:0 | 1-5471 5:55 70-0 1663 80-5 26:7 5:60 75:5 +3898 26-9 C. - 26:65 5-30 43-1 2:1032 82-2 26-9 5:30 47:5 27-9 C. 5:35 53:0 5-40 58:3 15471 5:45 63:5 +1682 82-0 27-0 5:50 69-3 +6274 27-8 26:95 | 5-40 58:3 2:2427 82-0 26-1 5-20 33-0 27-8 C. 5:25 38-2 5:30 43-5 15471 5°35 49-0 +1635 82-0 26-3 5-40 54:3 +3944 27-8 C. 262 5°30 43-6 21050 | 1 81:8 26°5 5:20 39-5 27-7 C. 5:25 44-5 5:30 49-5 15471 | 5:35 54:8 +1657 81-9 26-6 5:40 59-9 4487 27-7 C. - 26:55 | 45-30 49-6 21615 81-9 26-8 5-20 42-5 27-7 C. 5:25 47:5 5:80 52-7 15471 5°35 58:5 1676 81-2 26-9 5:40 63:5 4785 27:3 C. - 26-85 5-30 62-9 | 21931 |\1 80-1 25:9 5°30. 48:8 26:7 C. 5:35 54:0 5:40 59:0 | 1-547] 5-45 64:5 -1619 80-2 26:0 5-50 69-8 5355 26-8 C. — 25:95 | 5-40 69:2 | 22445 ND SALINITIES OF THE WEDDELL SEA AND OF THE NORTH AND SOUTH ATLANTIC OCEAN. 165 Data Relating to the Collection of the Sample. Data Relating to the Determination of the Density of the Sample. Suis Depth jm | Vol } i » (B.) Position (L.) : Temperature Temperature otume | Density of Sample. ec iar time of during Weights ofim- | (Density of Distilled Collection of Experiment. | added | Read- | mersed Water at 4° C.=1.) D. d. Sample. He ing | Portion | Dee es | Colour of Water, | Time | Hydro- of of of eed T t Current, and and Huge ti meter Hees HyGre: Ob- |Reduced/ Reduced i Sears lb eas : : Remarks. Date. ; (grams). 2c: served to to th. Jay.| Hour.| Lat. Long. at the i a (cub. att’. |15°56C.| +. Tage Sample of of oe y centims.) tion] Wi | the | _ the the | the L. | jected. aa } Water. Air, Sample. w. R. V. | 4St, |4Sases6| 494, ‘on fF rl Fat oR, oF, | 0 6} | 8.30! 547-N. | 2028 W.| Cir.| 400 | 78:3) 43-4 Buchanan-Rich- 7.0 |79:5 26-4 5:00 20:6 | : 1800 6:30.) ard water-}| 17/6/04 |26-4C. 5:05 26-0 = Read on bottle 5:10 31:0 bridge | Thermometer No. 5-20 41-2 102,506 5:30 52:0 5-40 62-0 5:50 74:0 1:5471 5-55 79-0 1647 719:5~ | 26-4 5:60 83-5 4713 |26-4 C. ———— 26-4 5:30 52:1 2-1831 | 1:02270 1-02561| 1-02724 8.35 | 547 N. | 2028 W.| Cir.| 490 | 78-3| 41-2 Nansen-Petters- 7.15 |79:3 26-4 5-00 21-6 1800 5-10.) son ~ water-} 17/6/04 |26-3 C. 5-05 27-0 J Readon} _ bottle 5:10 32-8 bridge | Thermometer No. 5:20 43-2 : 5:30 | 54:0 5-40 65-0 5:50 75:5 1-5471 5:55 80-5 -1647 19-7 26-4 5-60 85-0 -4867 26:5 C. SS 26-4 5:30 53-8 2-1985 | 1-02262) 1:02553 | 1:02730 12.0 6 2N.|2033 W.| .. | Surface} 80:0] 80:8 Antwerp blue 13.45 80-6 27-0 5:20 44:0 27-1 C. 27-0 C. 5-25 49-0 5:30 54:5 1-5471 5:35 60-0 +1685 80:7 27-0 5:40 65:0 4930 27-1 C. Se - 27-0 5:30 54:5 2-2086 | 1:02256 | 1:02565| 1:02253 240 | 7 5N.|21 2W.| .. | Surface] 77-2 | 80-4 Too dark 7.30 |79-7 26-4 5-20 44-0 26-9 C. 17/6/04 |26-5 C. 5:25 | 49-2 . 5:30 54-9 1-5471 5:35 60-1 +1650 | 79:8 26:5 5-40 65-2 4948 26-6 C. —— —_——— 26-45 5-30 64-7 2-2069 | 1:02257 | 1:02549 | 1:02244 |12.0 | 725 N.|2139 W.| .. | Surface| 73-6] 80-0 Dark turquoise 9.0 |75-6 23-9 5:20 41-5 DBs 26-7 C. 28/6/04 |24-2 C. 5°25 47-0 5-30 52:0 | 1-5471 5°35. 57-5 +1498 75:8 24-1 5:40 62:5 4718 24-3 C. ——-—— | ——— 24-0 5:30 b2-1 2:1682 | 1:02279 | 1-:02497 | 1-02198 12.0 | 946 N./2134W.| .. | Surface] 76:7! 80-1 Bright Reeve’s 9.15 |75:8 24-0 5-20 17:5 | 26-7.) French blue 28/6/04 |24-3 C 5:25 22-8 5:30 28-0 1-5471 5:35 33-5 *1501 \75:8 24-1 5-40 38-5 +2542 24-3 C,| ———_|—___ — aa 24-05 5:30 28-1 1-:9514 | 1:02£00| 1-:02622) 1-02520 | 12.0 | 1132 N.|2030W.)| .. | Surfacel 76:9 78-2 Pale Antwerp 9.30 76-0 24:0 5-20 13-3 25-7 C. blue 28/6/04 |24-4 C.| 5:25 18-8 | 5:30 24:8 1:5471 | 5:35 | 31-0 | -1504 \76-0 24-2 5:40 36:0 | -:2243 \24-4 C, == 24-1 5:30 24:8 1-9218 | 1:02417 | 1-02641| 1-02370 166 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE TEMPERATURES, SPECIFIC GRA} Data Relating to the Collection of the Sample. | Depth Data Relating to the Determination of the Density 7 Date (E. Position (L. es ._ | Temperature Temperature Volume | (E.) (L.) in Fathoms. at time oii qunnd Weights eae of 4 x i= = =0ni mcm |) Gollection of Experiment. i ead- | merse' § -z D. d. Sample. saree oe ine Pout a | & Colour of Water, }| Time Hydro- or 0 a a | a of | from T t Current, and and 4 bee Fetes Hydro- | Hydro- | ¢ S| Sea | Which) = : Remarks. Date (grams).| Meter. | meter » | « |Month.|Day.! Hour.) Lat Long. ae the : - (cub. gE Posi- Sample of of of of cone 5 z tion ae the the the the ; 1A L. fected Air. | Water. Air. |Sample.| w. R. Mo | | | | | [ 1904. [pares ead ball lle Baa 21, 2G} 528 | 556] June | 20 | 12.0 | 13- 7 N.} 21 47 W. Surface 75:4) 74:3 | Antwerp blue 9.45 |76-1 24-1 5:20 16-0 23:5 C.| 28/6/04 |24:5 C. 5:25 21-0 | | 5:30 26-0 1:5471 | 5:35 31:0 +1507 | | 76:2 24-2 5-40 36-2 +2352 24:6 C. ———| 24-15 5:30 26-0 1:9330 | 1-0241 529 | 557| June | 21 | 12.0 | 14 27 N. | 23 30 W. Surface] 73:9 | 74:6 Reeve’s French 10.0 |76-2 24-1 5:30 21:5 . 23-7 C, blue 28/6/04 |24-6 C. 5:35 26:8 5-40 82-5 | 1-5471 | 5:45 38-0 +1507 76:0 24-2 5:50 44:0 +2949 24-4 C. — | 24-15 5:40 382-6 | 19927 530 | 558] June | 22 | 12.0 | 1525 N.|2520W.| .. | Surface] 73-0] 173-8 Pale Reeve’s 10.15 |76-1 24-1 5:50 42:5 1 23-2C.| French blue 28/6/04 |24-5 C. 5:55 47-5 1 5-60 52:5 1:5471 5:65 58-0 +1507 | — 76-2 24-2 5:10 |. 63-0 4767 | 24-6 C. - ae 24-15 5:60 §2-7 21745 531 | 559| June | 23 | 12.0 | 1655 N. | 26 22 W. Surface] 71:8 | 72:8 Reeve’s French 10.30 |76:3 24-1 5-30 20-0 22:7 C. blue 28/6/04 |24-6 C. 5:35 25:0 Hi 5:40 30:0 1:5471 | 5-45 35:0 +1507 | — 76:3 24-2 5:50 40:0 *2714 | 24-6 C. - 24:15 5:40 30-0 1:9692 | 532 | 560} June | 24 | 12.0 | 18 43 N. | 27 46 W. Surface| 72:3 | 72:8 Reeve’s French 10.40 |76:3 24-1 5:30 18-7 22:7 C. blue 28/6/04 |24-6 C. 5:35 24-0 ; 5:40 29:5 1-471 | 5-45 84:8 | +1610 | 76:3 24:3 5:50 40:0 +2660 | 24-6 C. —— 24-2 5-40 29-4 1-9641 533 |561| June | 25 | 12.0 | 2019 N. | 2910 W. Surface| 73:6 | 73-5 Reeve’s French 10.55 |76°3 24-1 5:30 13:5 23-1 C. blue 28/6/04 |24-6 C. 5:35 18-8 5-40 24:5 15471 5-45 30-0 +1507 76:3 24-2 5:50 35:2 +2207 | 24:6 C. 24-15 5:40 24-4 1:9185 | 1 534 | 562) June | 26 | 12.0 | 22 44N./3035 W.| .. | Surface] 73-8| 74-4 Reeve’s French 11.10 |76-4 24-2 5:50 30:0 / 23-6 C. blue 28/6/04 |24-7 C. 5:55 | 35:3 | 4 5:60 41-0 | 15477 | / | 5-65 | 47-0 | 1616) 76:6 24-4 5:70 52:5 3727 24:8 C. = 24-3 5:60 41-2 20714 535 | 563| June | 27 | 12.0 | 25 2N.|3151 W. Surface] 73-5 | 75-0 Cobalt 16.0 |'75:8 24:5 5:30 6-0 . 23:9 C.| 24-3 C. 5:35 11:5 | 5-40 17-0 | | 5-50 27-5 4 5:60 38-5 | 5:70 49-5 73°35. | 75°35 | -75:2.\- 66:6: | 70:6. 74 7 ee Horizontal circumference, . 544 520 — 545 re 517 598 | 542) 525 <7 Frontal longitudinal are, 134 | 129 137 | 128): 132)\" F256 _ 28 Parietal ss ij 127 ty | 123 134 = 125 125 136 137 | Occipital - 5 | 116 131 a: 123 140 123 bated | Total _,, i a 362 | 391 380 | 390 395 | 376 Vertical transverse arc, 313 | 298 | 313 307 |, 309%) “298 7) 200 Basal transverse diameter,. 127 117 ye oe Vertical transverse circum- ference, . 440 | 415 56 He oe Length of foramen magnum, Phe 35 34 34 40 of Basi-nasal length, | 100 109 99 98 97 Basi-alveolar length, . : 101 94 92 89 Gnathie Index, 92°7 95 93°8 | 918 Total longitudinal circum- ference, . fs 497 — 5384 513 — 533 ee _ Interzygomatic breadth, $31 Ne V7, eee 132°| 134-1 139°), 13s Intermalar . LOT i hd | 124 La moe 121 |} 120 Nasio-mental length, . oe ; 125.) 126 : Nasio-mental complete facial 4 Index, . (ree eg OF 3900 ace Nasio-alveolar length, 69 67 70 72 62 Mazillo-facial Index, a i ae bO7.| O2:2eolcoa Fore Nasal height, 55 45 51 53 |) 50 55 49 Nasal width, 21 24 23 24 24 24 25 Nasal Index, 3S'2 | 583 45°1 ASG, \ Aa'O. \r4O26 od a Orbital width, 41 38 4% 4] Bae 435 43 Orbital height, . 39 28 33 33 af 32 31 Orbital Index, OST Na horn 786 SOD | ie 74: 72° Palato-maxillary length, Pell 53 54 | 55:5 | 47-5 Palato-maxillary breadth, mu 60 62 | 65:5 60 | Palato-maxillary Index, : 1132 114°8 |117- | 126° Mandible, symphysialheight, | 2866) 3ils5 178 PRINCIPAL SIR WILLIAM TURNER ON TABLE I. Neolithic Period. Antiquities. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 179 The roof was formed of massive flags, superjacent to which three so-called cists 14 to 2 feet in breadth and length were found, constructed of undressed flat stones. Each contained numerous pieces of broken pottery, which had doubtless been ciner- ary urns, and quantities of calcined bones recognised as human, some of which were imbedded in hard vitrified slag. The cists therefore were cremation cists, at 9 4 4 Sy wg) oe Se 9 Feet mI wth oa Fic. 7,—Plan of chambered cairn with passage, Rousay. secondary in time to the cremation remains found in the passage leading to the central compartment. The bones in the chamber, in the passage and in the cremation cists were too fragmentary to allow of reconstruction, and the form of the skull was not ascer- tained. The cairn without doubt was of the neolithic period. The excavations conducted by Professor T. H. Bryce on megalithic inter- ments in the island of Arran have enabled him to describe their structure, their grave goods, and the remains of the skeletons which they contained.* They formed a class of sepulchral cairns constructed of great slabs of stone divided by partitions into three or four separate compartments, none of which opened into a passage * “Cairns of Arran,” Proc. Soc. Antig. Scot. , part i, 1902, part ii, 1903, part iii, 1909, vols, xxxvi, xxxvii, xliii ; The Book of Arran, Sepulchral Remains, 1910. TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 5). 26 180 * PRINCIPAL SIR WILLIAM TURNER ON, directed to the surface. Flint and smooth stone implements, rude urn-like pottery, rounded at the bottom, with simple ornamentation, were obtained, but no trace of bronze. Four skulls were procured at Torlin and Clachaig, which have been described by Professor Brycx, also a fragment of a fifth (Table I). They ranged in length from 186 to 201 mm.; in breadth from 134 to 140 mm. ; in height from 132 to 136 mm. The cephalic index was from 66°6 to 75'2, the mean being 71°3; the vertical index from 65°6 to 73°1, the mean being 68'1 metriocephalic; the breadth in each case was more than the height. In general form the crania were long in relation to the width, dolichocephalic, parieto-occipital slope moderate ; occi- pital squama somewhat bulging; glabella and supraciliaries moderate; face high, narrow ; jaw orthognathous; nose narrow: orbits low in relation to their breadth. The cubic capacity in two skulls was 1480 and 1560 respectively. Cairns enclosing — segmented megalithic interments were also examined by Bryce in Kintyre, Islay, and by GREENWELL and Map.eron, near Crinan, Argyllshire, but no skeletons were obtained for description. 7 Professor Brycr, from researches. in the island of Bute,* exposed chambered cairns like those in Arran, which contained burnt and unburnt burials: the bones of the latter were too fragmentary to be measured; also broken pottery. One chambered cairn had been built on the site of an older kitchen midding, whilst in one a beaker urn of the bronze age was found. Another example of a chambered cairn at Kewing Hill in the parish of Firth + in the Orkneys has been described by Mr M. Cuartuson. It had a central compartment with five smaller cell-like chambers opening off it, as well as an entrance passage. The walls were dry-built on the beehive principle. On the floor of the central chamber was a deposit of an unctuous appearance, the skulls of dogs, jaws and teeth of an ox, tibia of horse, bones of birds. Five human skulls with limb bones were found in the compartment, also two skulls in the cells. In the entrance passage were portions of human long bones which showed evidence of cremation. No industrial relics were found in the chamber. ; The human remains sent to me consisted of portions of five skulls and three thigh bones. They were from the condition of the sutures advanced in life, at least three being males, I attempted reconstruction of the crania, but they were so much broken that in only one could | obtain a numerical cephalic index. The calvaria was 188 mm. long, breadth 142 mm., which gave 75°5 as the cephalic index; the vertex was neither ridged nor highly arched, the post-parietal region was obliquely flattened, the occipital squama was plano-convex, the forehead retreated a little, the glabella and supraciliary ridges projected. The skull, though with a cephalic index fractionally higher than the numerical limit, was essentially dolichocephalic, Another calvaria without the occiput was elongated and narrow; in a third the parieto-occipital region was steep and the squama was faintly convex. * Proc. Soc, Antag. Scot., vol. xxxviii, 1908, + Idem, vol, xxxvi, 1902, THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 181 Mr Aurxr. O. Curve has recently described * a cairn near Sunningdale, Suther- landshire, which showed indications of four horns. It enclosed a chamber 8 feet 9 inches long and 7 feet broad, formed of seven large upright slabs arranged to form a domed roof. A passage ran from one end of the chamber for about 12 feet towards the strface of the cairn. The chamber contained decayed human bones, a flint seraper and a fragment of pottery neolithic in character. Cranology. Tain I. Well-preserved specimens of the skulls from neolithic interments are unfortu- nately few in number, but from those specified in Table I the following general characters may be stated. The crania varied in maximum length: five ranged from 196 to 201 mm, five from 183 to 187 mm.; the mean length of the series was 191°8, which is a high average when compared with modern male Scottish skulls.| The breadth ranged from 134 to 144 mm. and the mean was 139°8, much below the mean of the modern specimens. ‘The height ranged from 128 to 189 mm. and the mean was 133-2, almost the same as in the modern skulls. The horizontal circumference ranged from 517 to 545 mm. and the mean was 528°7, almost equal to the modern male skulls. The cubic capacity taken in four male crania gave a mean 1480 c.c. The cephalic (length-breadth) index was obtained in ten skulls, the mean was 72°8: seven were less than 75, numerically dolichocephalic; three were from 75:2 to 76, or approximately meso-dolichocephalic ; not one was brachycephalic or meso- brachycephalic. The vertical (length-height) index computed in six skulls ranged from 65°7 to 73'1, with the mean 69°3; in each it was less than the cephalic, and no cranium was high, hypsicephalic. In general configuration the cranium was an elongated ovoid, moderately arched but not ridged on the vertex, the side walls flattened, the parieto-occipital slope not steep, the occipital point behind the inion; the forehead somewhat retreating, the . glabella and supraciliary ridges well marked, the nasion a little depressed, the characters therefore are, like the proportions, dolichocephalic. For the facial measurements the proportion of the complete length to the breadth could be taken in only two specimens; the nasio-mental index had a mean 92°3, high-faced or leptoprosopic ; in four specimens the maxillo-facial index worked out at from 45:9 to 52:2, the mean being 50:1, also leptoprosopic. The gnathic index computed by Flower’s method was in each specimen orthognathous. The nasal index was not platyrhine, but either leptorhine or mesorhine. The orbital index, as is customary, varied, ranging from 72 to 95°1; only one was rounded, mega- * Proc. Soc. Anteg. Scot., vol. xliv, 1910, The Reports of the Royal Commission on Historical Monuments in the Counties have recently added examples of chambered cairns in Sutherland and Caithness, also long cairns in Galloway, and notably one in Berwickshire at Byrecleugh, Longformacus. + The references in the text to modern Scottish skulls apply to those described in Part I, vol. xl, of this Memoir. 182 PRINCIPAL SIR WILLIAM TURNER ON seme, 95'1 ; the others were wide in relation to the height, microseme, and the mean was 78°9. The palato-maxillary index was either mesuranic or brachyuranic, the mean of four specimens was 117°7. The characters of the neolithic skulls may be summarised as follows: crania elongated, the occiput not flattened but protruding, the vertex not high; the face long; the upper jaw almost vertical, orthognathous; the nose relatively narrow, nostrils not wide; the orbits with the breadth wide in relation to the height ; the palate short and relatively wide; the capacity of the cranium approximated to that of modern Europeans. The-interments above described as neolithic were limited to particular localities in Scotland, as the chambered long cairns in Argyll, Arran, Bute, Galloway and Nairn, the horned and chambered cairns in Caithness, Sutherland and the Orkneys, and the tumulus at Newbattle. There can be no doubt that a dolichocephalic race lived in neolithic times. he preservation of their remains throughout the centuries was due to the substantial modes of interment, which justify the inference that the builders practised well-designed constructional methods during at least the later stage of the neolithic period. The labour obviously required for their construction points to the tombs being those of the chiefs of the tribes with their families. It should not be inferred that this race, like these cairns, was limited to small areas in- Scotland, and it is probable that the burials of the people generally were frailer in construction and had consequently disappeared. Instruments and weapons made of polished stone and worked flints characteristic of the neolithic period have been found in numerous localities but not associated with interments, though some of these doubtless had been in use in the succeeding bronze age. BronzE AGE—Suort Cists; Cremation. URNs. In Scotland two modes of burial were practised in the bronze period, Inhumation . and Cremation. Inhumed dead bodies were interred in characteristic stone graves, known as short cists, which were concealed in cairns or tumuli, or were placed only a few feet below the surface of the ground. These graves were less massive in construction than the chambered cairns of the neolithic age. Archxology. In general character a short cist was built of a single, or at times of two flat slabs of undressed stone on each side and of a smaller slab at each end; the stones rested on their lower edge, and supported on the upper edge a more massive slab as a cover for the cist, with occasionally a second smaller slab superimposed. The floor of the cist might be the natural rock, or clay, or gravel, seldom a layer of flat stones. In internal dimensions the cist averaged 3 to about 4 feet in length, about 2 feet in breadth and the same in height. The covering slab was longer and overlapped the THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND... 183 ends and sides of the cist. Owing to its short length the body of an adult could not be laid in the extended position but had to be placed on one of its sides, with the hip and knee joints bent, as well as the elbows, so that the hands were brought close to the face. Although objects made of bronze have not been found in the majority of the short cists which have been opened, their similarity in size and construction, the bent position of the skeletons and the presence of definite types of urn pottery have led them to be regarded as of the same period, whether bronze was present or not. Frequently only a single cist had been exposed in a particular locality, sometimes two or three; very seldom were more than five placed in close proximity to each other. Almost every county in Scotland has given to archeologists examples of this type of burial, and it is evident that the people or race which practised it was not limited to a particular area, but were distributed throughout northern Britain. In many localities the practice of Cremation was associated with Inhumation ; though in others cremated bodies were found without being contained in or near short cists. In the great majority of cases the ashes were deposited in a special type of urn, known as a Cinerary urn. These urns have been found either singly, or in small groups of three or four, or in such numbers as to indicate a cemetery of some size. They have been exposed as independent burials in sand or gravel, or have been enclosed in a cairn’or tumulus. A hole had apparently been made in the ground and the urn lodged in it, either upright, resting on its base, or in- verted, with the mouth downwards, on a flat stone. Sometimes the urn had been in contact with the surrounding sand or gravel, but at others loose stones had been arranged around it as a protection. I have now analysed the notices of burials of the bronze-age period recorded in Scotland in no less than 197 localities, which comprised 475 distinct interments.* Two hundred and thirty-two were inhumed in short cists; in 240, cremation had been practised and the ashes placed in a cinerary urn; in 3 cases the record was not precise. Sometimes both kinds of interment were found in the same tumulus, cairn, or cemetery, though at others they were in distinct localities. Ina few eases, Cavers, Pentlands, Pomona, incinerated bones, not enclosed in an urn, were lying along with a skeleton in a short cist. It is obvious, therefore, that two modes of interment were practised in the same district and frequently at the same period of time. It has, however, been observed that cremated interments in cinerary urns prevailed when bodies had been buried in stone circles. On the other hand, in the Orkneys, inhumation in short cists was the rule in prehistoric graves. In 318 cases, urns of rough unglazed clay pottery were said to be associated with the interments ; 206 of these were cinerary urns. Ina few examples a heap of bone ashes and charcoal, obviously indicating an interment, was found not included in an urn or cist. In 112 cases urns either of the bowl-shaped food type or of the * Ir my lecture “On Early Man in Scotland” (op. cit.), I analysed the characters of about 400 interments. 184 PRINCIPAL SIR WILLIAM TURNER ON taller beaker form were found. In a number the presence of urns is not recorded. It has been noted that the beaker urn especially prevailed in the cists exposed in the north-east of Scotland, though not exclusively in Aberdeen and Banff. incense cup,” have been found — in Scotland in cinerary urns. They are from 2 to 3 inches high and about 3 inches ce Occasional examples of a small urn, known as an wide. Archeologists have associated them with the burning of incense, which assumes that the bronze-age people possessed fragrant gums and resins. One discovered in 1857 at North Queensferry, about the size of a teacup, contained calcined human bones; another obtained by Dr James Macponatp of Ayr was occupied by the bone ashes of a child of five or six years. Presumably they were cinerary urns for the reception of the ashes of infants and young children. Objects of bronze, usually thin blades or pins, were obtained in connection with 87 of these interments, some in short cists, others mingled with ashes in- cinerary urns. Flint, in the form of arrow-heads and flakes and implements of stone, horn and bone, were found in 45 burials, mostly in short cists. Ornaments, usually of jet, with occasionally beads of amber or vitreous paste, were obtained in 16 burials. Eleven objects of gold, as armlets and rings, were collected from cinerary urns and short cists. . An interesting description of a Cemetery of cinerary urns, eight in number, was given in 1866 by Mr Anprew JerRVISE;* they were exposed at West- wood, Newport, Fife, arranged in a circle around a central urn.* They were inverted and contained incinerated bones, whilst each of two of the larger urns had a smaller urn inserted in it. On the adjoining property of Tayfield a group of three urns was exposed in 1882. One of the largest bronze-age cemeteries was discovered at Law Park, near St Andrews, in 1859.+ About twenty cinerary urns were obtained, and two bronze blades were found amongst the burnt bones. At Alloa twenty-two cinerary urns were also found, and among the group was a short cist which contained an unburnt skeleton with two penannular gold armlets. In the low ground on the banks of the Esk between Inveresk and Musselburgh numerous bronze-age interments have been exposed. The Rev. Gzorcr Lown, B.D. described } a sandpit in the Kirk Park, below Inveresk Church, as. yielding nineteen cinerary urns, one of which, in addition to calcined bones, contained a piece of flint resembling an arrow-head: in two the bones were stained green as if from contact with bronze. ‘The urns were described and figured by Dr J. AnpERson. Subsequently Professor ARTHUR THOMSON exposed in the same pit an urn which had been protected by stones placed around it, but not forming a built cist. The urn was 10 inches * Proc. Soc. Antig. Scot., vol. vi, p. 388, 1868 ; idem, vol. xvii, p. 272, 1883. + JoserH ANDERSON, Scotland in Pagan Times, p. 36, “ Alloa,” p, 62, 1886 ; Proc. Soc. Antiq. Scot., vol. x. D. Hay Fiemine, LL.D., has described, P. S. Antig. S., vol. xli, 1907, the cinerary urns of 1859, also two additional short cists in Law Park, St Andrews, which contained fragments of two beaker urns and a jet necklace. + Proc, Soc, Antiq. Scot., vol. xxviii, 1894, THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 185 high, 9 inches broad at the mouth, and 4? at the foot. It was ornamented by two groups of incised horizontal lines separated by a zigzag pattern, the lower group being on a raised collar (fig. 8), and it bore some resemblance to fig. 3 of those found by Mr Lows. The urn contained calcined bones and bits of charcoal. Mr Lowe referred to stone coffins having also been found in the Kirk Park. I have been told that they were short cists formed of slabs of sandstone, each with a large cover, about 4 feet in length. They contained human skeletons. At Belfield, on the opposite bank of the Esk, five short cists imbedded in sand were exposed in 1896, during building operations. I measured them along with my assistant Mr James Stimpson. One was 4 feet 4 inches long internally, 2 feet 1 inch broad at one end, 14 foot at the other; its covering slab was 5 feet 8 inches long, 2 feet 6 inches broad, and 7 inches thick. The smallest cist was 2 feet 5 inches long, 1 foot 6 inches broad ; its covering slab was 2 feet 11 inches long, © 2 feet broad, and 5 inches thick. One cover was waterworn and marked by circular depressions like those on the cover of the cist found in Leith (p. 187). The largest cist contained a skeleton in the bent position, the head of which was in the narrow end of the cist, whilst a bowl-shaped food urn, 44 inches high, 5? inches across the mouth,* occupied the broader end (fig. 10). The smallest cist contained the broken skeleton of a child. In a third, the workmen exposed calcined bones, specimens of which I saw; the fourth had a human skull with fragments of other bones; in the fifth, in addition to parts of a skeleton, were a hammer of smooth stone ; two pieces of flint, one, arrow-pointed, smooth on one side, chipped on the other, was 1°7 inch long and ‘8 of an inch at the widest part a little in front of the convex hinder border (fig. 16); the other flit was smaller and without definite shape. At the Kirk Park patches of black earth containing charcoal were found situated external to the cinerary urns, which indicated cremated burials not contained in urns. ‘There can be no question that the Kirk Park and Belfield interments in cinerary urns and short cists were a cemetery formed by people by whom both modes of burial were practised, and that, apparently, cotemporaneously. Without doubt they were a community of fishermen who pur- sued their avocation, as at the present day, in the waters of the adjacent Firth. Recently Mr M‘Letian Mayn has called attention} to a cremation cemetery cairn of sixteen urns at Stevenston, Ayrshire. They contained calcined bones, and in one traces of thin gold-leaf and bead-like objects of vitreous paste were found. I do not propose to consider in detail the archeological features of the bronze- age burials recorded in the important works specified on p. 173 and in the forty- eight volumes of the Proceedings of the Society of Antiquaries of Scotland. But as my notebook contains an account of several interments not previously * Proc. Soc. Antig. Scot., vol. xxii, 1898. + Idem, vol. xl, 1906. Also, vol. xxxix, 1905, at Langside, Glasgow. t The Report on Historical Monuments, Berwick, notifies a group of six cinerary urns at Coldingham and twelve short cists at Ayton, 186 PRINCIPAL SIR WILLIAM TURNER ON published, I may state briefly their characters; the description of those found at Belfield and Kirk Park in Mid Lothian has already been given on p. 185. Leith.—In February 1884, when excavating ground in Merrilees Close, the town authorities exposed a short cist about 6 feet below the surface, lying in sand and gravel not far from the sea beach. The cist, with its contents, was presented to the University, and has been re-erected by me in the vestibule of the Anatomical Museum (fig. 11). The cover of the cist was a massive stone slab, 4 feet 6 inches in length, 2 feet 9 inches broad at one end, and 2 feet 5 inches at the other; it varied in thickness from 6 to 11 inches; its under surface, from its appearance, had been detached from the parent rock, but the upper surface was generally smooth, though hollowed in places, as if from trituration with sand and gravel. The stone which formed the right side of the cist was 4 feet 3 inches long by 2 feet 3 inches at its ereatest breadth, though it narrowed at the ends; it was 34 inches thick; one surface showed the natural cleavage of rock, but the other and two of the edges were smooth and waterworn. The stone of the lett side was 4 feet 1 inch long, 1 foot 8 inches broad, and had a maximum thickness of 9 inches. One end stone had a vertical diameter of 1 foot 11 inches, a transverse of 1 foot 4 inches, and was 5 inches thick: that at the opposite end measured 2 feet 2 inches by 1 foot 9 inches, and was 5 inches thick. A gap, owing to one end stone being defective, was filled by a smooth oblong boulder 13? inches long by 94 broad. The flattened waterworn walls of the cist had apparently been the surface of the rock exposed on the sea beach, and had been detached by the builders in the plane of cleavage (fig. 11). | When the massive cover was removed and the sand taken away two human skeletons were exposed, the limb bones of: which were so friable that, as a rule, they were not preserved. The skulls, though defective, are described on p. 201. One, larger than the other, was at the north-west end of the cist, but the position of the smaller skull was not noted. The interior of the cist was a little over 8 feet long and about 2 feet high, and bodies could not have been buried in the extended position. A bowl-shaped urn was found near the larger skull. It was 6 inches high, 64 inches wide at the mouth externally, and 4? inches inside measurement, whilst — the diameter at the foot was 84 inches. The outer surface was decorated with twelve rows of obliquely impressed zig-zags or chevrons arranged around the urn. ‘Two grooves encircled the urn immediately below and parallel to the rim. At the lower of these grooves four knob-like projections (lugs) had been modelled, one of _ which was imperfect. ‘The rim of the urn was bevelled on its inner aspect and was marked by the chevron pattern (fig. 12). Cousland, Mid Lothian.—Short cists were exposed in 1886 on the ja of Cous- land, Cranston, on Lord Stair’s estate. Mr ANprEw Gow, the factor, kindly gave me the following particulars:—They were 2 to 3 feet below the surface of the ground, and were between 3 and 4 feet long, 2 feet broad and 20 inches deep. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 187 Fic. 12.—Bowl-shaped urn, Leith. aia Fic. TRANS, ROY. SOC. EDIN., VOL. LI, PART I (NO. 5). 18.—Bowl-shaped urn, Bridgeness, 27 188 PRINCIPAL SIR WILLIAM TURNER ON The walls, which rested on their edges on a bed of limestone, had the customary — single rough slab on each side and at each end. The cover also was a single flat massive stone. The peculiarity of this interment was the presence in more than one of the cists of a slab in the middle, extending from end to end, which divided Side and end. Interior. Fias, 14, 15.—Cousland short cist. F the cavity into two parallel compartments (fig. 15) each of which contained a skeleton;* in the undivided cists only a single skeleton was present. The sand and earth in the cist were riddled without fragments of an urn or any implements being discovered. Pentland Hills.—I am indebted to the late Mr Jonn Hunperson for an account of a short cist from a cairn on the East Cairn Hill. The cairn formed a roundish Fic. 16.—Flint, Belfield. Fic. 17.—Flint, Pentland cairn. mound, in the centre of which the cist was exposed resting on the sandstone rock. The sides and ends were each formed by a single sandstone slab, and the cover was a massive unwrought slab. In its internal measurement the cist was 3 feet 8 inches long, 2 feet broad and 1 foot 3 inches deep. It contained earth with which numerous fragments of calcined human bones were mingled. ‘Three barbed flint arrow-heads were also found, one of which Mr Hernperson presented to the * In Proc. Soc. Anti. Scot., vol. xxxii, 1898, is an account by Mr Arex. Hurcuxson of a bronze-age cist found on a hill at West Mains, Auchterhouse, Forfar. It contained a bronze dagger, and two collections of calcined bones, and had a separate interment in a smaller compartment. Mr Hurcuuson also referred to a short cist in the Barnhill burial mound (Proc. Soc. Antigq. Scot., vol. xi) which was longitudinally divided into two parallel compartments, resembling therefore the Cousland grave, THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND 189 Anatomical Museum. It was 32 mm. long and 18 mm. in greatest breadth; the edges were saw-like, point sharp, surfaces finely chipped (fig. 17). No trace of an urn was seen. Birsley Quarry, Tranent, E. Lothian.—In 1884 the workmen employed by Mr Joxun Witson exposed a short cist in the earth covering the sandstone rock. The cist had been built in a layer of sand and weathered sandstone. It consisted of the customary four slabs for the sides and ends, and of a larger cover slab, all formed of sandstone. In internal measurements it was 3 feet 6 inches long, 2 feet broad and 2 feet 3 inches high. The floor consisted of the solid rock. The cist contained a skeleton with the hips and knees bent, and with the face looking towards the east; the bones of some of the fingers were situated at the back of the head, as if the elbows had been bent and the hands raised to the sides of the head. Neither urns nor other grave goods were present. Previously to the discovery of this cist a cinerary urn had been obtained in the earth covering the same quarry, and had been recorded by the Rev. Jonn SrrutHERs,* who stated that a short cist had been exposed within three or four yards of the urn. Morrison's Haven, Preston Grange.—A short cist was exposed in May 1887 in the sand near the mouth of the harbour. Mr Luke, manager of the adjoining colliery, told me that it was built of rough stone ‘slabs, was covered with rough flat stones, and was also said to be paved. ‘The internal dimensions were length about 34 feet, breadth 2 feet, depth 20 inches. It contained a human skeleton in the bent position, but the skull, detached from the spine, was found between the legs (p. 204). No urn nor implement was obtained from the cist. Bridgeness, W. Lothian.—In 1896 Mr H. M. Capett of Grange informed me that a “heavy stone coffin” had been exposed a few yards from the probable end of the Wall of Antoninus at Bridgeness-on-Forth. Owing to absence from home at the time he could not give its dimensions, but he. had ascertained that the skeleton which it contained had been bent around an urn which was preserved. The “coffin” was undoubtedly a short cist. The urn was a good example of the bowl-shaped form (fig. 13). It was 5°38 inches high, 6°9 inches wide at the outer lip of the mouth and 4:9 inches at the inner lip, 2°5 inches in diameter at the foot. The rim was bevelled on the inner surface and marked by short oblique impressions. A broad groove encircled the urn below the outer lip, and the outer surface was marked with closely arranged horizontal rows of pits about the size of small shot, separated by lines, some of which were plain, others minutely pitted. The skull is described on p. 206. Mr Capex subsequently obtained a similar urn from a short cist exposed in 1905 at Cowdenhill near Bo'ness, in a sandbed of the 25-foot beach, about 10 feet above high-water mark. t * Proc. Soc. Ant. Scot., vol. xiv, 1880. In vol. iv of the same Proceedings, a note is made of a large cinerary urn having been got in trenching a field at Tranent. + Idem, vol. xl, 1906. 190 PRINCIPAL SIR WILLIAM TURNER ON Binns, W. Lothian.—In 1875 Sir Ropert DaALzELL wrote me a note on the disclosure close to the House of Binns of a short cist, 2 feet 11 inches long, the ends and one side of which were formed each of a single slab of stone, whilst the other side consisted of the rock itself. The floor of the cist was without any slab. Torphichen.—lIn a sandhill near this town Professor Duns uncovered five cists, one of which from its dimensions was a short cist and contained a human skeleton.* Threipland Farm, Elgin.—In 1869 Mr G. ALLAN, in opening a hillock on his farm, exposed a short cist. The cover stone was about 6 feet long, 4 feet broad and 1 foot thick; both it and the side and end stones were rough slabs of sandstone. The internal dimensions of the cist were 4 feet 2 inches long, 2 feet 1 inch broad and 2 feet 1 inch deep, and the floor was a sandy clay. ‘The cist was said to lie west of south by east of north, and the head was at the south-west. In addition to the human skeleton the cist contained an oval piece of flint, shaped like a spear- head, flat and thin at the edges, 24 inches long, 13 inch broad, which was seen about the middle of the right of the skeleton. The flint was presented to the Elgin Museum. + Duns, Berwickshire.—In excavating gravel near the railway station in 1863 a short cist was exposed, 4 feet long, 14 feet broad and 20 inches deep, which contained an urn of the drinking-cup type and the skeleton of an adult, the brachy- cephalic skull of which I described at the time,? and now on p. 193. In 1897 a cist which was said to be about 3 feet long, wider in the middle than at the ends, and 2 feet deep, was discovered on the farm of Chapel, Duns. The cover consisted of two massive slabs of sandstone, and the sides had smaller stones, as well as the customary large slabs: the floor was the hard subsoil. No urn or implement was found in the cist. A skull, much injured, and portions of other bones were sent to me, but their imperfect condition did not permit a satisfactory description. Roxburghshire.—At Kelso, in 1864, a short cist was exposed in the Knowes, adjoining the Abbey.§ It contained an urn of baked clay which corresponded in size and form to the largest urn, a beaker from Lesmurdie. No bones were procured from the cist. . Cavers.—In the same county a short cist was discovered in 1896 ina tumulus in Belvedere Wood, on the estate of Mrs Parmer Dovenas. It was carefully described and figured by Dr Davin Curisrison.|| The space for the cist had been excavated in the solid rock, and one side and end were each formed of a single slab, whilst their opposites consisted of smaller slabs: the bottom was composed of small stones imbedded in clay. The cover consisted of two whinstone slabs, the larger of which was 7 feet 2 inches long and 18 inches thick. The cist contained an unburnt * P.S.A.S., vol. xii, p. 405, 1878. + See zdem, vol. xxii, p. 341. t Idem, vol. v. § See my description in Proc. Soc. Antig. Scot., vol. vi, p. 18, || Jdem, vol, xxxi, 1897. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 191 skeleton in the bent position and a heap of fragments of incinerated human bones, a flint scraper and arrow-head, a bone pin and perforated disc, but no urn. I had the opportunity of examining the skull (p. 206). In the following year a small cemetery of three cinerary urns, exposed by a land slip, was found on the bank of a stream at Chesters, near Jedburgh.* They were inverted, and contained calcined bones, and in one was also a piece of flint. I repro- duce a drawing, kindly made by Mr M. G. Craic, of one of the urns (fig. 9, p. 187) which was decorated, but the other two, larger in size, had no ornamentation, except that one was marked on the lip. In 1902 a short cist was exposed on a farm near Denholm in the same county. It measured 40 inches by 25 inches, and contained a much decayed skeleton, but no other relic. Ross-shire—Fyrish, Evanton.—In June 1865 a note on a short cist on the farm of Fyrish was communicated to the Society of Antiquaries.t The cist con- tained a beaker urn, a perforated greenish-coloured stone, and a skeleton. The skull is described on p. 198. | Strathpeffer—Dr Fortrscur Fox gave me an account of three short cists, found in 1896 in a mound of sand and gravel at Croch Fionne. They were formed of slabs of a clay slate, and each had a large cover stone. Two were about 34 feet long ; the third was somewhat smaller. Each contained a human skeleton, but neither urn nor implement was detected. In one the bones, judging from the dentition, had belonged to a young person, but they and the other skeletons were fragmentary. Near the centre of the mound a cinerary urn was exposed outside the cists. It was in a black mass containing charcoal, and a similar substance filled the urn, but bones were not seen either in the urn or around it. It was 6 inches high, 74 inches across the mouth, and 5} inches at the foot, and it was marked immediately below the lip with oblique intersecting lines. Dr Fox also called my attention to a description by- the Rev. W. Warson } of two short cists, each of which contained a skeleton in the bent position, | at Aonach, Drummond, Kiltearn, in the same county. In one cist a_bowl- shaped urn full of a dark substance was found in front of the face, also a piece of bronze the size of a darning needle. Dr Fox gave me the opportunity of examining the human remains, unfortunately very imperfect. One skull was that of a man advanced in life; the glabella and supraciliaries were moderate, the bridge of the nose was well formed and projecting, hut the cranium was too imperfect to admit of being measured. Rosemarkie, Ross-shvre-—Karly in this century a short cist was exposed in the manse grounds. It was formed of rough whinstone slabs with a massive cover, and its long diameter was north and south. It contained a skeleton, the head of which was at the north. At the opposite end was an urn, 6 inches high, 6? inches in * Proc. Soc. Antig. Scot., vol. xxxi, p. 199. + Idem, vol. vi, p. 288, 1868 and p. 266. t Idem, vol. xxiii, p. 138, 1889. 192 PRINCIPAL SIR WILLIAM TURNER ON. circumference at the lip, and 3 inches at the foot. It was of dark brown pottery, marked with rings of herring-bone ornament. No other grave goods were observed. Salen, Loch Sunart, Argyllshire.—In 1897 Mr J. Fraser of the Ordnance Survey wrote and told me that he had exposed in a cairn of loose stones a cavity, about 34 feet long, 2 feet broad; and about 24 feet deep, the walls of which were formed like those of the short cists which I had described in my lecture “On Early Man in Scotland.” He noted that the cover consisted of a large flat stone, on the top of which a much smaller slab had been laid. Imperfect remains of a bent skeleton were seen in the cist, but no weapons or other relics. | Cranology. Tapius II, III, IV, V. It is to Sir DanreL Wison that we owe the definite recognition of the presence in Scotland of a prehistoric race having the brachycephalic type of skull, which he regarded as belonging to a race later in time than the primitive dolichocephalic or kumbecephalic people. One example which he gave as characteristic was obtained, along with urns, in a cist exposed during the demolition, in 18338, of the Old Town Steeple of Montrose. He figured the skull and described it* as square and compact in form, broad and short, well balanced, and with a good frontal development. The length was 180 mm., the parietal breadth 157 mm., and the cephalic index was 87°2. Another characteristic example was a skull found in a cist under a tumulus at Ratho ; alongside it stood a small rude clay urn, which contained several bronze rings, so that its association with the bronze age was established. This skull was 177 mm. long, 153 mm. broad, and the cephalic index was 86°4. WILSON also referred to three skulls discovered near Cockenzie, East Lothian, in a group of rude cists of “ primitive circumscribed dimensions.” Two are preserved in the National Museum of Antiquities, and he figured another as a characteristic example. Of these three skulls, two are undoubtedly dolichocephalic, whilst the cephalic index of the third, 78, placed it high in the mesaticephalic group. As the cists in which they were got were doubtless short cists, I have included two, which I have measured, in my list (Table V) from short cists, and not along with the neolithic (kumbecephalic) skulls (Table I), with which Witson, owing to the dolichocephalic form and propor- tions of two specimens, had associated them. He considered that the kumbecephalic and brachycephalic skulls belonged to the early native races of Scotland, which pre- ceded the intrusion of the Celts into Britain. Although he noted several skulls from Argyllshire and the Hebrides as affording a fair average criterion of the Celtic type, he does not definitely specify their characters, except that the parietal and vertical diameters of the cranium were nearly equal, whilst in the brachycephalic the parietal is greater than the vertical and in the kumbecephalic the opposite is the case. Drs Davis and THURNAM described and figured in their Crania Britannica four * Archxology and Prehistoric Annals of Scotland, Edinburgh, 1851, p. 170. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 193 skulls from short cists in Scotland. In one of four short cists at Lesmurdie, Banff, in 1851, a bent skeleton, a drinking-cup (beaker) urn, and some flints were found. The skull, that of an adult male, was broad, with the parieto-occipital region flattened, and a cephalic index of 86. The vertex was flattened, glabella and supraciliary ridges distinct, nose long and narrow, orbit rounded, lower jaw massive. A second male was obtained in the same year at Juniper Green, Edinburgh, in a similar interment ; the skull had the cephalic index 83°6. In one of three short cists exposed near Kinaldie, Aberdeenshire, in 1855, a bent skeleton and an urn of the beaker or drinking- cup type was seen ; the cranium was steep behind and the vertex flattened, but owing to one side being injured the breadth and cephalic index could not be taken. A short cist in a barrow at Newbigging, in Pomona, Orkney, contained two skeletons and a heap of burnt bones. One of the skulls, 71 inches (180 mm.) long and 5:7 inches (145 mm.) broad, gave a cephalic index 80°6. Three of these skulls from bronze-age interments were markedly brachycephalic. Duns.—In 1863 I examined* a cranium, from the short cist near Duns (p. 190), 6°5 inches (165 mm.) long and 5:4 inches (137 mm.) broad, cephalic index 83, in Fie. 18,—Duns. which the vertex was flattened, the parieto-occipital region truncated, and the fore- head well developed. I pointed out that, like the skulls described by Sir Dante Witson and Barnarp Davis, it was distinctly brachycephalic (fig. 18). ; Fyrish, Ross-shire.—I described, in 1865, a brachycephalic skull from Fyrish,+ obtained in a short cist (p. 191). It was from a male in the decline of life. The eranial vault was moderately arched, the post-parietal region was flattened; but as the occipital squama was somewhat bulging, the back of the skull was not truncated. The glabella and supraciliaries projected; the nasion was depressed, the nasal bridge projected, the hones were narrow, the nasal index was leptorhine, 44:2, the upper orbital border was thick, and the index was mesoseme, 87:2; the upper jaw was orthognathous. The cephalic index was 82°3, the vertical index 71. The cranial breadth was much greater than the height. The cranial capacity was 1555 c.c. * Proc. Soc. Antig. Scot., vol. v, p. 279, 1865, + Idem, vol. vi, pp. 233, 266, 1868. 194 PRINCIPAL SIR WILLIAM TURNER ON TABLE ihe Short Cists. so Za: . ie 28 ret) Ove: | ale Aberdeen : ib | 2 Mull: Ardachy. ee ae 5s z= Broomend, | az a RE) we | es @s nyerurie, ce | # E.U,A.M. A e A as F Collection number, . -| Ay | Bl] °C. © |Metopic.| zum.) ... | ET = a, | E.T. | E.T. | E:T. | Age, . ' ; ; - | Juv..| Juv.| Juv. | Ad. | Ad. | Ad. -| Ad, oe Advd. Advd. Ae A Sex, « P i f | MD ee) Sa Me. Be) MM ey M. M. | 3 Cubic capacity, . : - | 1250} 1390] 1350)... Rit |ieat C8 7 L000 1448) 1555 | Glabello- -occipital length, - | 178)" 171) “2744 180.) 164. | 180. |. 176.) 190) 1-183 182| 186 Basi-bregmatic height, . L254), 130) 125) 228) 123" a4 |, eal) AOS 139 134} 132 | Vertical Index, : | L253) SOO! TE8) 717 | 75 ISS | WG 75a \e7O a) ae Galas | Stephanic diameter, . j 110; 104) 106) ... Dak W344) |. bash e Bet a 114} 111 _ Greatest parieto-squamous | _ breadth, ee eal, US6)), DAO! WSR os | 1859, are: eee eon ee ee nee OP men gs Cephalic Index, 4 : | 78:6:| SHO} 75:9). | 82:3 | 87-2 \orachal 75-99 ee | 76:4) 823 Horizontal circumference, -| 401) 492) 495) 4. |» <.. 1.584 | 2. 5d WepIBse Silaleaee Frontal longitudinal arc, . Py 263) 2a) De | Le 118 ie eae 128} 131 Panetal) * eee |, kes) Se Sr aia orl Sia) ee 130°} °127 | 127 193, 182 Occipital 6 ast te TTS) DOT Ve Uo) SLL 4) LOL) oe 1) SOG oe ataiz — 105 } = 125 Total » +| 340) 361] 369) 360] 335/ .. | 354! ... | ...| 361] 38a Vertical transverse arc, .| 282 305 300| ... e: a ee oie cs 303| 324 Lengthof foramen magnum, 388; 35) 32, 34] 337 ... Seer cate eee 38| 32 Basi-nasal length, . ; OOF 9289 1004), S854) oa | Oa, ves 104 100 Basi-alveolar length, . ; 96, 88 SOs. dus ee ne a ee 95 90 Gnathic Index, . «| OO Ge nO 2a ia. S nos sie e Baca mureae wipien/ aot 90 Interzygomatic breadth, i) SLO) Cate eh ama ee |e sh [once gmore ie 5a Intermalar F 107.) LOGI TOT), 115 aac 2 ot alo Bic LA ae Nasio-mental length, . : 98:1) MOB 9) hal 23.) 7 LOS Lat) Seen ee a LUD) OR | Nasio-mental facial Index, = | 82S SB OO see NER 2. ee os A ne fe, me | Nasio-alveolar length, ; DO) O84) DS |) cen Osa wy ans et 0! EG | Mazillo-facial Index, . . | 49:6) 48 | 47 | verde as Gite al nc eee ee Nasal height, . 7 : ADs Veto eee laiiak Dileep. Thee oP are) ge / Nasal width, . : ; 22 21 DONE DAS ets. on ae ee nT eee 23 | heme _ Nasal Index, : . | 92°4) -48°8'| 48:8) 47:7 eae ee “fie 1s ws | JO" |) 44:2) am - Orbital width, : : : 39 36 Sala Hwee ba; DEAT a 4 3 4] 39 Orbital height, . ¢ 3 33 31 30| 35 age sor Spot sae ile oe 34 Orbital Index, . 3 «| OO) 8057 Cort 8725: | > oS. Bi Se | ae we | 805 | 87-2 Palato- -maxillary length, . OL ABN aa G0: ences A eee Re 51 | + 50 Palato-maxillary breadth, 59| 56 SOM! tees 61 58 Palato-mawillary Index, <1) PISO NITGO Ne 272 1196 116 | Mandible, symphysialheight, 281, “2B 692), eB2e) 223 30) «30 THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 195 Broomend, Inverurie.—In 1867 Mr Hay CuHatmers and Mr r. 8. Davipson described * three short cists found at Broomend, Aberdeenshire, in one ot » ich two bent skeletons, urns, and flint flakes were found, while in another a remarkable lamp made of tanned hide was hanging from one side of a beaker urn. I measured the skulls at the time: one, A, was an adult male, the face of which was broken, but the length, breadth, and height of the cranium gave a cephalic index 78°9 and a vertical index 73°7. The skull was meso-brachycephalic. B was from a younger male, as the basi-cranial joint was not closed; the vertex was flattened, and the occipital region sloped backwards ; the left parietal was so much injured that the breadth, and in con- sequence the cephalic index, could not be definitely stated, though it had not the same brachycephalic character as A. The vertical index was 72'1. The glabella and supra- ciliaries were strong. The upper jaw was orthognathous, the nasion was depressed, the orbit was microseme, the chin was protuberant (Table IJ). Tealing, Forfarshire.—In 1870 Mr James NetsuH read an account of a short cist containing a male bent skeleton, the skull of which is in the National Museum. I examined it along with Dr J. A. Smirn.{ The cranium was elongated oval, side walls somewhat flattened, occipital squama almost vertical, sutures obliterated; length 182 mm., parietal breadth 139 mm., height 134 mm.; glabella and supraciliaries fairly marked; forehead somewhat retreating; nasion a little depressed, nasals narrow, bridge projecting a little; lower jaw well formed, angle almost rectangular, chin strongly everted. The cephalic index was 76°4, the vertical 73°6. The cranium was approximately dolichocephalic, and its capacity was 1448 c.c. The nasal index, 50, was mesorhine, the orbital index, 80°5, was microseme, the upper jaw was ortho- enathic. Two cinerary urns, one of which was inverted, containing calcined bones, were between the cover stone of the cist and the surface of the ground (Table II). At Ninewells, Invergowrie, in the same county, a short cist was opened, which contained a skeleton and bowl-shaped urn.t The skull was probably that of a woman ; the forehead was smooth and vertical, the vertex was flattened, and the parieto-occipital slope was moderate. The cranium was 175 mm. long, but otherwise so imperfect that the breadth and height could not be measured. Dunrobin, Sutherland.—In March 1880 a short cist was exposed in a bed of gravel in Dunrobin Park. The Rev. Dr Joass described it§ as containing a female skeleton with the knees bent. An urn of the drinking-cup type was beside the head and a necklace of shale beads was at the feet. The skull was 178 mm. long and 146 mm. in parieto-squamous breadth, which gave a cephalic index 82. The basi-bregmatic height was 132 mm. and the vertical index was 74°7. The nasal index, 51°1, was mesorhine, and the orbital index, 90, was megaseme. * Proc. Soc. Antig. Scot., vol. vii, pp. 110, 115, 116, 1870. + Idem, vol. viii, p. 383, 1871. { Idem, vol. v, p. 81, 1865. § Reliquary, 1903. Also described and figured by Professor Bryce in Proc, Soc. Ant. Scot., xxxix, p. 428, 1905, TRANS. ROY. SOC. EDIN., VOL, LI, PART I (NO. 5). ‘ 28 196 PRINCIPAL SIR WILLIAM TURNER ON Windy Mains, Humbie, HE. Lothian.—Mr Rosert Forman described * in 1857 two short cists in a mound of sand. Lach contained a drinking-cup (beaker) urn, and in one was a male skeleton advanced in years. The skull was imperfect, but the parieto-occipital slope was abrupt, the nasion was depressed, and the nasal bones projected. I measured the length of the cranium, which was 173 mm., the basi-bregmatic height was 138 mm., and the vertical index 79°8. The breadth could not be taken. The orbital index was 80°5. The total longitudinal are was 349 mm., and the basi-nasal length 106 mm. Silver Moor, Carstairs.—A short cist was exposed in 1847 which contained an adult male skeleton, the imperfect skull of which is in the National Museum of Antiquities (E.T. 13). Dr Ranxrne, who presented it, noted special flattening and backward elongation of the occipital condyls.t The face was wanting and the right parieto-squamous region was imperfect. The cranium was 176 mm. long and 134 mm. in basi-bregmatic height, the vertical index was 76'1, the parieto-squamous breadth could not be taken, but as the minimum frontal was 106 mm. broad and the asterionic 118 mm., it is not unlikely that the proportion of greatest breadth to length would have placed it in the brachycephalic group. The supraciliaries were moderate. Arran.—In this island short cists havé been found in several localities. In 1861 Dr James Bryce explored} a group of short cists within a stone circle at Tormore, Mauchrie, in which bowl-shaped urns and flint arrow-heads were found, along with a bronze pin. From one cist a skull was obtained which was measured by Professor ALLEN THomson. Though imperfect, its length was given as 7 inches (175 mm.) and its parietal breadth as 5:7 inches (143 mm.), which would give a cephalic index 81°4, i.e. brachycephalic. At Knochan Kelly a short cist was opened by Dr J. Jamrgson,§ which contained a bowl-shaped urn and the skeleton of a youth aged about ten. Though the skull was injured, Professor CLELAND regarded it as brachycephalous, like those of the short barrows; he gave the greatest length as 6°6 inches (168 mm.) and the parieto- squamous breadth as 6°25 inches (159 mm.), which would give 94°6 as the cephalic index: the vertical height was 5 inches (127 mm.) and the corresponding index 75°6. Professor 'T. H. Bryce has investigated a number of short cists in Arran which were lodged under cairns. He described the characters of the cists, of urns of the bowl or food-vessel type, the presence in some cists of flakes of flint, and in one of a bronze dagger and a fillet of gold. The human remains unfortunately were so much injured that they could not be precisely described. He referred, however, to the Tormore skull quoted above, figuring the norma verticalis, and to the skull obtained at Knochan Kelly, both of which were brachycephalic. * Proc. Soc. Ant. Scot., vol. iii, p. 50, 1862. Skull in National Museum of Antiquities. + Idem, vol. xi, p. 464, 1876. Theskull which I measured is preserved in the National Museum of Antiquities. t Idem, vol. iv, p. 499, 1863. § Idem, vol. xx, p. 170, 1886. See also Professor T. H. Brycr, “Cairns of Arran,” Proc. Soc. Ant, Scot., vol. xxxvi, p. 74, 1902, THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. Tasie III. Short Cists. 197 Davis AND THURNAM. BRYCE. —~Fae SF . Ayr- Fee =.3 ostes S 2 wg el 2 Ts shite: lee | ss | BS |SBe) ss |eee| 8's | 84 | Largs ae | Ba | eo lose) BE Ess} a2 | SE age | Sig | oe z Bo) 48 | <4 as 3 A | Collection number, ellie Bis lee emule | E.T. | E.T. va So I cece (BU. | Age, . ; , : . | Aged.| Ad. Aged.| Ad..| Juv. | Ad. | Ad. | Ad. | Sex, . , : cilen Ml, “iba eaam! M. M. M. F, M. Cubic capacity, . : libel ve ele ae nde alee or |r LDLO) |. Glabello-occipital length, NG6hie too LeOnmaigemianteD | 168): 176) 178 | 175 Basi-bregmatic Briel, fe LSD Het | 130 i 127 | 138 | 132 | 138 Vertical Index, 70'4 | 68: 73°6 US ON ISA TAT | 788 Stephanic. diameter, nies an is Ne se Greatest parieto-squamous | | breadth, ee. | GOs teemiememinemerisza;| 159 | Jot | 146) 148 Cephalic Index, . ; . | 86° | brachy. | 80°6 | 82: Bene 946 | 85:8 || SZ: 54:6 | Horizontal circumference,. 546... ; 43 DAE OSs I tee Frontal longitudinal are, . 130) ... 128 128 | 132 Parietal 3 ee i LOA 124 120 | 118 Occipital * io mae eee al) abe 116 113 | 120 Total _,, - ie ae STONE 368 361 370 . Vertical transverse arc, Bad shite 309 314 | 325 | | Length of foramen magnum, SOR onnse al 33 35 38 | Basi-nasal length, oh Od 96 95 1OL jf $99 Basi-alveolar length, . 96 | 100 | 89 91 95 | Gnathic Index, . le gos 1042 93°7 90: 96:9 | | Interzygomatic breadth, fe = ee LAGS |) aah || | Intermalar 114 | Sia alhiell } Nasio-mental length, . : se 127) 05" Nasio-mental complete Index, |... Aa ee LOT, jets | Nasio-alveolar length, : 65 65 71 WO 08 Mazillo-facial Index, Re Bea Ne ie Peary” a. | Nasal height, . : 50 49 | 56 49 45 | | Nasal width, . , : 23 23 | 90 25 24 | | Nasal Index, : no hee 4" 469 oF oZme) “62: | Orbital width, . : ! 43 42 | 40 37 39 | Gmitalheight,. . .| 33). 29 | 32 32) 35 | Orbital Index, ; 2 5765718 69: | 75: 86:5 | 97: | Palato-maxillary length, . 52 47 55 50 Bi | | Palato-maxillary breadth, . 57 60 58 BT | | | Palato-maailliary Index, 109-6 |127:7 1115°5 | 114: th. | | Mandible,symphysial height,| 29 30. | 33 Ip 34 27 | | | In the island of Bute * Professor Bryce found short cists, either within cairns or tumuli, or near the surface of the ground, without external mark. bodies had been inhumed, in two they had been cremated. The grave goods were scanty ; in three were urns, two of which were bowl-shaped; in two were articles of bronze and jet. In six of these the The human remains were fragmentary, but it was thought that a skull from a cist at Auchantirie had probably a cephalic index above 80. * Proc. Soc. Antig, Scot., 14th Dec. 1904, vol. xxxviii, pp. 52 et seq. 198 PRINCIPAL SIR WILLIAM TURNER ON TABLE IV. Short Cists. Dr Atexr. Low, Aberdeenshire, | Museum, University, Aberdeen. | uel bas] o n | 8 Bm Ss Cr) = " a 8 a | Parkhill p a = Tyrie. z 2 : 3 e = 3 RE 3 | alae les 4. | "Sie ee number, . ot] Ad A BS 3 4 AAG, | Bil0e fie 8 9 iat 12 ge, : ; : ie ees ae ee oe Sex, . : ; va MS) ST Vices) SVM IVI elle Mo | Mi. | Hae Vian _ Cubic capacity, . : = | 1450 | 220 191420.| 1500 | .<...-1 1580. 14605) = 18500) are ee) Glabello-occipital length, -/°180 | 183 | 169 | 188 | 185 | 185 | 172°) 186 | 167) 1725) Sie Basi-bregmatic eteht, .| 135 | 148 | 183 | 146 | 188 | 144) 128} 135) 136 ade 135 | 143 Vertical Index, . «| 70° | 809" 78:7 | 77:7 | 746-\ 778 | 744 72-6 84) os | 4 | Stephanic diameter, : »| 115) 02. | 120) 136 | 120). 124.) 1285)..125 | TOS8s\s 1 NS) sO _ Greatest parieto-squamous breadth, : : .| 153 | 160 | 156) 160) 156 | 150 | 150) 154") 142 | 135 | 156°) See Cephalic Index, . : . | 85° | 87°4 | 92:3 | 85-7 | 84:3.) 81-7 | 87-2 | 82:8 | 85" | 78:5 | SO-TAeeee Horizontal circumference,. | 524] ... | 510 | 540] 528] 548 | 503] 532] 494 | 500) 512] 524 -Frontal longitudinal are, . | 135 | 132 | 128 | 132] 143 | 126] 127 | 130] 123 128) 130] 130 Parietal = x» «| 125) 118)-120] 148 | 140| 140) 110.) 180) 120) 122') 1325) sie Occipital _,, , ft) 1001 4. | 110} 117,) 110} 120.) 116) 129°\8e2 9 St05e| ieee Total _,, » | 360| ....| 359 | 397 | 393 386] 353 | 382] ... | 355 | 375.|08em Vertical transverse arc, 1 awit Ste 300 | 336 | 336 | 310] 304] 330] 306 300 | 324 | 330 Length of foramen magnum, 30. | nes Biel 238 | 390 \) 38a NOSulee ‘ie 4) 365 aa Basi-nasal length, . -| 104) 100} 95); 96 98 | 104) 97) -.. | 100) .2 >) 1OOt aii Basi-alveolar length, . ‘ 98 92 87 Sialic. sal 94 GA Fiche OT.) Tens alll 95 Gnathic Index, . «| G42 92° OIG, 906 |"... || GOLA 96 Oa ae ren o7e . 707-~ ie Interzygomatic breadth, . | 42 )- 22) V6 140 |S. | 18% | 1800). S86 |” ones ee Intermalar ~ | DES ee) 4s V0.) oc. | LI eee ee Nasio-mental length, . 2 HOS eae SUOMI TO |. c.0. a ees - | L087...) oe a Nasio-mental facial Index, Erie POSE GM AOrO WN cr al Os ies ce | 19:4 e SS 7g eae Nasio-alveolar length, : G4)" 80 CAE GT! 121 (G8) se Gi are. 12:) a Mazilla-facial Index, . . | 45° | 53:9 | 505) 47°38 | 05/2525 | 52S 1) AG! | ee oa Nasal height, . 3 A Wi 2) 53 48 48 55 53 48 ha 3) US eo 51 49 Nasal width, . . 3 23 25 25 25 28 26 25 ce 24 iy ieee 23 Nasal Index, . : 2 ALO NAL DN OA O2 | SOOT 40 Ako Zale teem aren ... | 509 | 46@ Orbital width, : oy) ee 40); 44] ... 4] LOR ce AON an 41 | 44 Orbital height, . c ; 33 sist 30 32 ue 31 33 ae 31 one 32 34 Orbital Index, . | CO ce ABE TAZ | 5 POTOSI S2 Sees 7/7) Palato-maxillary length, : 51 56 51 Dy de ame 50 DOU. ax Dori ter 58 51- Palato-maxillary breadth, . 55 64 61 BOM. ae 65 GON Fs cae glee GGi1l! aakee 63 | 57 Palato-maxillary Indea, OT AAA NIOGN O76 | a. ISO M24: .. (120° | ... (1086 \110 OR Mandible,symphysialheight,| 30 | ... 28 30 30 enol ae ye Se ee seo 33 34 | 29 | He also regarded the fragment of a skull obtained in a short cist at Mount Stewart as probably brachycephalic, though too imperfect for measurement. It had a hole in the frontal which Dr Munro regarded as due to an operation for trephining. The cist also contained a piece of bronze, a jet necklace, and a bowl- shaped urn. In a later memoir on a skeleton from Acharole, Caithness,* obtained in a short * Proc, Soc. Antig. Scot., 8th May 1905, vol, xxxix, p. 418. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 199 cist which contained a beaker urn, Professor Bryce described and figured a male skull. It was hyperbrachycephalic, cephalic index 85:8; the parieto-occipital region was flattened; the vertex was flattened and the vertical index was 78°2, platy- chameecephalic, wide and low; the face was chameprosopic, 76°7, broad and low; the upper jaw was orthognathous, 90; the nasal index was wide, 52, high in the -mesorhine group; the orbital index, 86°5, was mesoseme; the palato-maxillary index, 114, was mesuranic. The capacity of the cranium was 1425 c.c. North-East of Scotland.—Important researches into short cists and their contents in this part of Scotland have been conducted during several years by Professor R. W. Rem and Dr Atexr. Low, of the University of Aberdeen, and interesting series of objects from the cists have been arranged in the Anthropological Museum in Marischal College of that University. Dr Low has described the cists, the beaker urns, and other contents, and has given in the Proceedings,* 1902-04, a comprehensive table of the measurements of ten skulls, and in subsequent parts, 1904-08, those of additional specimens from Whitestone, Blackhill, and Leslie. In this series of thirteen skulls, eleven males and two females, from short cists, twelve had the cephalic index 80 and upwards, and of these eight were hyper- brachycephalic. In one only, a female, the index was 78°5. The vertical index was less than the cephalic. In the specimens in which the face was measured the complete index was low, chamzprosopic in five, mesoprosopic in two; in nine specimens the jaw was orthognathic, in one mesognathic; the nasal index in one was platyrhine, in three leptorhine, in eight mesorhine; the orbital indices were low, microseme. ‘The cranial capacity in seven males ranged from 1350 to 1580 cc., with the mean 1458 c.c. ; in one female it was 1460 c.c. (Table IV). Ardachy, Bunessan, Mull.—Sir Arraur MircHEtt recorded} the exposure, in 1892, of four short cists, three of which contained bent skeletons. A food-bowl urn was found in each of two of the cists. At the request of Sir Arthur I examined and measured three skulls which had been preserved. They were from young persons from ten to eighteen years, were probably females, and were so well preserved that both cranial and facial measurements were given in the table accom- panying his original memoir. The skulls were well formed, and varied in the proportional length and breadth of the cranium. B was brachycephalic, cephalic index 81°9; A was mesaticephalic, index 78°6, approximating therefore to brachy- cephalic ; C, again, with index 75-9, approached the dolichocephalic, and its outline was a longish oval approaching the pentagonal form. In each skull the height was less than the breadth, and the glabella and supraciliary ridges were not prominent: the forehead was almost vertical, and the frontal and parietal eminences projected. The nasal index was mesorhine; the orbits in C were low, microseme ; * Proc. Anatomical and Anthropological Soc. of University of Aberdeen, 1902-04, 1906-08 ; also Appendix by Dr Low to Professor Brycz’s Memoir in Proc. Soc. Antig. Scot., 8th May 1905, vol. xxxix ; I/lustrated Catalogue, Aberdeen, 1912. + Proc. Soc. Antig. Scot., vol. xxxi, p. 115, 1897. 200 PRINCIPAL SIR WILLIAM TURNER ON in A and B they were more rounded, mesoseme. In all the jaw was orthognathous. The face was low in proportion to the breadth, chameeprosopic, from the immaturity of the dental arcades. The capacity ranged from 1250 to 1390 c.c. (Table II). Largs, Ayrshire.—In 1906 a cist 44 feet long, 24 feet wide, and 2 feet deep was exposed and described by Dr Rosert Munro.* It contained an aged skeleton, the teeth much worn and sutures ossified. The skull was examined by Professor — D. J. CunnincHam, it was brachycephalic with marked parieto-occipital flattening, Fic. 19.—Ardachy, Mull: profile and face. cephalic index 84°6, vertical index 78°8 (T'able III); some broken pottery which the Hon. Jonn ABERCROMBY recognised as a beaker urn was present. I shall now transcribe from my notebook several skulls from short cists, the characters and dimensions of which have not previously been recorded. . Leith.—This cist contained a female skull injured, A, and the calvaria of a male advanced in years, B. A was broadly ovoid, not ridged in the sagittal region and sloping gently to the parietal eminences. As the occipital squama was broken, the parieto-occipital slope was imperfect and moderately steep. The forehead * Proc. Roy. Soc. Edin., vol. xxvi, 1907. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 201 was almost vertical, the frontal was flattened above, the glabella was feeble, the nasion was not depressed. The maximum length could be only approximately stated at 179 mm., but the breadth, 140 mm., was measured at the squamous suture, and the basi-bregmatic height was 130 mm. The cephalic index was meso- brachycephalic, approximately 78°2, the vertical index was 72°6, the breadth being more than the height. The nasal index 52 was mesorhine, the orbit was rounded, its megaseme index 100; the upper jaw was orthognathous (figs. 20, 21, p. 202). The lower jaw had a good angle, the chin projected a little ; the teeth in both jaws were not much worn. ‘The palate-was hyperbrachyuranic (p. 186). Skull B consisted only of the calvaria, which was shorter and wider than A Fie. 22.—Birsley. in the transverse vertical are; cephalic index 88°6, hyperbrachycephalic. The parieto-occipital slope was flattened ; the forehead was ample and flattened in front of the bregma. Cousland, Mid Lothian.—Three imperfect skulls were obtained. A, probably a male, with a metopic frontal and a right parietal. The glabello-lambdal diameter was 180 mm., but the length and breadth could not be measured. The forehead was capacious, the glabella and supraciliaries were feeble. The nasion was not depressed, the nose was narrow, leptorhine; the orbits were moderately rounded, mesoseme ; the palato-maxillary index was brachyuranic. The lower jaw was broken at the angle, the symphysis had a well-marked chin. The teeth were flattened on the crowns. B was an adult male; only the frontal and parietals were preserved. The glabello-lambdal diameter was 173 mm., the greatest parietal he a misinl A AAG A Ae 202 PRINCIPAL SIR WILLIAM TURNER ON Fic, 20.—Leith, Fic, 21.—Leith, Fic, 24, —Bridgeness. Fic, 23.— Bridgeness. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 203 TABLE V. Short Cists. THE LOTHIANS. | | seal Moveres|ce ey laa. Sir D. Winson. 3 |= i 33 ia Sa Bo | ao as Leith. | Cousland. 2 aq |e 5 ogee | 3 S, aa | | ae Sm | ‘3 Kank Park rae E, Lothian: |W. L.:| AS | | |= | =I “| | Cockenzie. |Ratho. iS | | ~Q ~ Anatomical Museum, University, Edinburgh. | | ere A BAY BO. ty | AS | Boy ET. 6 ET. ey eT Meee; ~~ CSC | A, Aged] Ad: | Ad. | Ad. | Ad. | Ad. | Ad. | Ad. |-Ad. | Ad. | Ad. |.Ad. |Aged. Sex,. . Pees eM ee) |) MM | ae | Mw | eM | *| Me hM. | M. Cubic capacity, . ee Oe rm ee GOL ie Wh ae (orl, gayi liceen de aetliiedw |v Glabello-occipital length, - | 179] 166 ab Peaele2y se 180 LTS wie | 18d: |) tne, | 180)) 177 | 173 Basi-bregmatic height, me oUiee eee eee ee ay | 300) 127°) ose. | V2 16+) 149 | 138 rss = | 720) 2. |. | ee |...) 75° 178-3 | 730 173-8 |... | 73'4.|,70r, | 80- | 79°8 Stephanic diameter, . iii Aah IS LOS MiSs LOR A IE oe. 1S |. TOL | 994... | Greatest parieto-squamous | | | breadth, ; : . | 140s.) 147 vf 146 1 148p:) 137) -... | 141 ... | 189 | 135 | 183 | 153 | Cephalic Index, . 4 2 | 78°2.1°88°6 ... \brachy.| 87°3 | 72-9 brachy. 79:2 \brachy. 75°1 | 78° | 73-9 | 86:4 Horizontal cireumference,.| ..| ../ ... ... | 524] 518) ... | 504] ... | 525 | 494 | 504 | 508... Frontal eal Mees) 195) i) OGIO 120.) 180 134, TAO B25) 12541307120) | 119 120 | Parietal a i Ye ee ms 140 134 , 128 140. 130 | 116 | 130 | 128 | 128 | 108 . | 1380 | Occipital _,, ee se eer ae iG ios | 102)... | 176 | 128 |... | 99 : Total _,, eee eee me eri | |. 3881 386 1-876. | 357 | «.. | 364 | 350 . | 349 | Vertical transverse arc, .| 302] ... ne Pemimozet 2981 ea ie299.) cl Bl) 294%). on Length of foramen magnum, B37" | eae oe Bas +3! Bin voDn lperoe) | 1380 mie 30 | 40 38 | Basi-nasal length, ; OSn) ee ns ee: Pere OOn Oe i Olek O5 ie 2 90 |. 99 106 Basi-alveolar length, . : Olt ae, Seal el ie | Gay ena RG | tO essa) GO: Veh. Gnathic Index, . Be GOS | s.. Peal state ee EOT: PR OPGriY Ae Hod y..\f00: Interzygomatic breadth, 5 Re cet ee Me eA se | LAS eel 33 £ Intermalar eee eG | oo |- 22) LO Ze i P10 | 118 Nasio-mental length, . eee elo t.. | . |-112).... | 11917... | 120 Nasio-mental complete “facial | | . Indea, . Bee | P89 |) 90a el) up. Nasio-alveolar length, vo 085) 5 GO ae One CllemeGlige tulln 71, -... Mazillo-facial Index, 23 | ao-2) 53: 153-3 = eens tO. | 5S iii.) ... | 49 AS | 4G, | bo) |) AT 4] Nasal width, . . : Cumin Oi fis 25 ae | 22 Nasal Index, . 2 Pa esis ed 47-2 | oa awe ie 46°8 | te ee se meee |... |!) 89)... fs 406) ee 40) 185 | Orbital height, . : j S8 ls tel 33 | BO ey aka sr ao, & 32 Orbital Index, . Cormeen. | 82) |’ 185: |. P80" 197-4 Palato-maxillary length, : AOI 51 | BTR babe OSiervche 53 es be 52 Palato-maxillary breadth, . CPR eb 90) eae 2a ena ea Heys | _Palato-maxillary Index, . 26S O° See e ee zy. a sce WORE as Be Mandible,symphysialheight,) 30)°...; 29) ..) 31, ... FRA OOS, Aa Poe | | breadth was 146 mm., the index of which would be 84:4; if the occipital bone had : increased the length of the cranium to 182 mm. the index would still have been 80°2: the skull was doubtless brachycephalic. The forehead was capacious, the supraciliaries well marked, and the nasion had been depressed. C was too frag- mentary to be described. TRANS. ROY. SOC. EDIN., VOL, LI, PART I (NO. 5). 29 204 PRINCIPAL SIR WILLIAM TURNER ON Birsley Quarry, Tranent.—The skull was a male; the cranial sutures were in process of obliteration, and the teeth were flattened on the crowns but not decayed. “It was broadly ovoid, vertex flattened, sloping gently to the parietal eminences, the parieto-occipital slope abrupt, occipital squama not bulging. The — cranium was 182 mm. long, 148 mm. broad, cephalic index 81:3, brachycephalic. The forehead was broad, almost vertical, the glabella and supraciliaries well marked, nasion a little depressed, facial bones mostly broken, but lower jaw massive, angle distinct, chin marked, teeth complete (fig. 22, p. 201). | Morrison's Haven, Prestonpans.—Skull male, sutures partially obliterated, teeth flattened from use. Cranium elongated, somewhat pentagonal in outline, not — flattened in sagittal region, sloping markedly to parietal eminences and in parieto- occipital region, occipital squama somewhat behind inion. Forehead retreating, glabella prominent with mesial suture, supraciliaries prominent. Length 188 mm., breadth 137 mm., height 141 mm. Cephalic index 72°9, vertical index 75, breadth less than height, skull dolichocephalic. Nasion depressed, nasal bones slender, moderately projecting, index 51, mesorhine. Upper jaw orthognathous. Upper orbital border thick, index 82, microseme. Small Wormian in lambdoid suture. — Cranial capacity 1450 cc. (fig. 25) (Table V). Belfield and Kirk Park.—In a male skull from the largest Belfield cist the teeth were only slightly worn and the sutures partially ossified. The outline was incomplete owing to the right parieto-temporal being imperfect. The slope of the vault from the sagittal line to the eminences was moderate, the parieto-occipital slope was abrupt, and the occipital squama was not bulging. The cranium was 180 mm. long and 141 mm. in basi-bregmatic diameter, the vertical index was 78°3, but the cephalic could not be computed. The general aspect of the skull was brachycephalic (figs. 26, 27). The forehead somewhat retreated, the glabella and supraciliaries were well marked (Table V). Two skulls from short cists exposed in the Kirk Park ailtastofing have been pre- served. One (A), given to me by Dr James CralciE in 1890, was that of a man in middle life. The outline of the cranium was broadly oval, the vertex was flattened, ' the parietal eminences were distinct, the parieto-occipital region was incomplete, also the glabella. The cranial length was 178 mm., the breadth was 141 mm., cephalic — index 79°2, the basi-bregmatic height 130 mm. ; the vertical index was 73. The upper jaw was orthognathous. The femora showed auieanens of platymery and the tibize of platyknemia. The lower jaw had a well-marked angle and a projecting chin. Of another specimen (B), a female, the right half is in the National Museum of Antiquities (H.T. 64).. In it the vertex was flattened, the parieto-occipital slope was steep, the occipital squama was flattened, the forehead was nearly vertical, the glabella and supraciliaries were moderate. The cranial length was 172 mm., basi-bregmatic height 127 mm., vertical index 73°8. Though the breadth could not be ascertained, the general type of skull was without doubt brachycephalic.. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. = —- ~ Fic. 28.—Cavers, Fic, 26.—Belfield. Fic. 25.—Morrison’s Haven. Fic. 27.—Belfield. 206 PRINCIPAL SIR WILLIAM TURNER ON Bridgeness, W. Lothian.—The skull was that of a man; the sutures of the vault were almost obliterated, and the teeth, though much worn, were not decayed. The cranium was elongated, the vertex, moderately arched from before backwards, sloped gently to the parietal eminences, the side walls were vertical, the parieto- occipital slope was moderately steep. The glabella, supraciliaries, and ectorbitals strongly projected and contributed to the length of the cranium, 185 mm., whilst the ophryo-occipital length was only 180 mm. The greatest breadth was approxi- mately 139 mm.; the cephalic index was 75:1, or meso-dolichocephalic. The strong supraciliaries, glabella, and ectorbitals made a ledge-like. projection above the orbits, and gave character to the fronto-facial region, which must have had during life beetling eyebrows (figs. 23, 24). The nasion was somewhat depressed ; the interzygo- matic breadth was 133 mm., the nasio-mental length 120 mm., the complete facial index was 90°2, high-faced, leptoprosopic. The orbital width was greater than the height, the index was low, microseme (Table V). The lower jaw was massive, angle well marked, and the square chin projected forward. Cavers, Roxburghshire.—The skull a male, teeth perfect, basi-cranial joint not fully ossified. It had been broken during the opening of the cist, but I was able to repair it so as to make some measurements and ascertain its form, which was broadly ovoid, the breadth being marked both in the frontal and parietal regions, while the breadth at the parietal eminences corresponded with that at the squamous sutures, 145 mm. (fig. 28). The vertex was somewhat flattened, the occipital squama was flattened, but the post-parietal slope was not abrupt; the inion and curved lines were distinct. The length of the cranium was 18 mm., the basi-bregmatic height 131 mm., the cephalic index 80, the vertical index 72°3; the skull was character- istically brachycephalic. The forehead was almost vertical and the supraciliary ridges were well marked. The border of the lower jaw was thick and the chin pro- jected forward distinctly (Table II). The femora were also examined; the upper epiphyses were fused with the shaft, but the condylar epiphysis was not so. A third trochanter was present, which was separated by a vertical groove from a well-defined infratrochanteric ridge. The front of the upper third of the shaft was convex and not platymeric. The linea aspera was moderate and the popliteal surface somewhat concave. The inner condylar articular surface was prolonged upwards on the back of the femur. The tibial shaft was imperfect. Thriepland Farm, Elgin.—The cist contained the bent skeleton of a iman in the prime of life. The skull had rested on its right side, which was softened and broken. In its outline the cranium closely corresponded with the skull A from the Leith cist. Its length was 180 mm., the basi-breematic height was 128 mm., and the vertical index was 71°1; the parieto-squamous breadth and cephalic index could not be taken. The frontal was metopic and the minimum frontal diameter was 104 mm. ; glabella and supraciliaries were strong, upper orbital border thick; the frontal was flattened THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 207 above its external process, and the adjoining temporal ridge was strong; the orbits were moderately high ; mesoseme index 87°5. The incisive region of the upper jaw, though injured, was sufficient to show that it was not prognathic. The nasal region high and narrow, index 47:1, leptorhine. Lower jaw well formed, distinct angle, well- formed protruding chin. Hyoids ossified and fused with temporals (Table II). Roseisle, Hlgin.—In a cairn on the hill short stone cists were exposed in 1860 in which human bones, an urn, and jet beads with an ornament were found. One cist was said to be 2 feet 10 inches long by 18 or 20 inches broad and 18 inches deep ; another, 4 feet 2 inches long by 3 feet deep.* I examined an adult female skull from one of these cists; the cranium was almost complete and the: bones were abraded ; parietal eminences distinct; sagittal region not ridged ; parieto-occipital slope moderate ; occipital squama not flattened ; glabella and supraciliaries feeble ; face much injured; lower jaw feeble, angle oblique; teeth a little worn; the length, 164 mm., gave with the breadth, 135 mm., a cephalic index 82°3, and with the height, 123 mm., a vertical index 75 (Table II). A child’s skull was also obtained. It was much deformed, apparently by softening and super- incumbent pressure, so that its normal characters were obscured. The right femur of the adult had a degree of flattening, platymery, in the upper third of the shaft. In Tables IT to V measurements are recorded of forty-nine skulls found in short cists in Scotland. Many were imperfect, and their dimensions were incom- plete. With four exceptions the skulls were those of adults. An attempt has been made to distinguish as far as possible the sex: thirty-seven are presumably males, twelve females. The majority are contained in the Anatomical Museum of the University, and in the National Museum of Antiquities, Edinburgh; but to give a wider field for comparison I have incorporated in the tables Dr ALEXR. Low's measurements of the specimens in the Museum of Marischal College, Aber- deen, those examined by Professor Bryce of Glasgow, and the older examples in the Crana Britannica and Sir D. Witson’s Archeology of Scotland. In general dimensions, proportions and form the skulls from the short cists had the following characters:—Forty-seven crania varied in maximum length: one was 190 mm. ; twenty-two ranged from 180 to 189; eighteen from 170 to 179; six from 164 to 169. Eleven of the skulls below 180 mm. were apparently females. In basi- bregmatic height nine skulls ranged from 140 to 148; twenty-one from 130 to 139; ten from 121 to 128. In parieto-squamous breadth three were 160; fifteen ranged from 150 to 159; thirteen from 140 to 149; ten from 132 to 139. A large pro- portion of the skulls in the three principal diameters had good dimensions, and the horizontal circumference in twenty-five skulls exceeded 500 mm. In these measure- ments the female skulls were, as is the rule, smaller than the male. ‘ In only a few specimens was it possible to take the cubic capacity of the cranium. In thirteen males it ranged from 1250 to 1580; four of which were 1500 and * See descriptions by Cosmo Inwus in Proc, Soc, Antig. Scot., vol. iii, pp. 46, 374, 490, 1862. 208 PRINCIPAL SIR WILLIAM TURNER ON upwards, one was 1350, and the rest were between 1420 and 1450 cc. The mean capacity of the males was 1448 ¢.c. Four skulls were regarded as female, and the mean capacity was 1435 ce. | The cephalic index was computed in forty measured skulls: twenty-nine were 80 and upwards, brachycephalic, and of these fourteen were hyperbrachycephalic, two being upwards of 90. In five others, too imperfect to be measured, the form was brachycephalic. In six skulls the index was between 78 and 80, approaching the brachycephalic numerical standard; three were between 75 and 78, and only two were dolichocephalic. The mean cephalic index of the forty skulls was 82°5; if five of these are not included in the computation, the index in which was below 77°5, 2.e. dolichocephalic or mesodolichocephalic, the mean index is as high as 85 in thirty-five crania. The vertical index was computed in forty measured skulls: it ranged from 68 to 81°4. In nineteen skulls the index was 75 and upwards; in eighteen it was below 75; in only one below 70. The mean index of the forty skulls was 75, 2.e. metrio- cephalic. In the dolichocephalic skull from Morrison’s Haven the vertical index exceeded the cephalic; in all the others the breadth was greater than the height and the crania were platychamecephalic, wide low skulls. In general configuration the cranial vault was flattened at the vertex; the parieto-occipital slope in many cases was abrupt and the back of the skull approached the vertical; the slope from the sagittal line to the parietal eminences was moderate. The frontal bone was smooth in the females, but had moderately projecting glabella and supraciliaries in the males. In the skull from Bridgeness, where the cephalic index approached the dolichocephalic standard, the cranium was more elongated and the glabella and supraciliaries were so prominent as to form a ledge above the nose and orbits. In a very large proportion of the short-cist skulls brachycephalic characters were pronounced, in the general configuration, in the high cephalic index and in the breadth of the cranium being more than the height. Owing to the bones of the face having been injured in many of the skulls, the proportional measurements of the facial region were imperfect. A nasio-mental index which expressed the relative length and breadth of the face was computed in twelve, which included the skulls recorded by Dr Atexr. Low: eight were low-faced, chameeprosopic, below 85; one was high-faced, leptoprosopic, 90°2; and three were intermediate in proportion, mesoprosopic ; the mean of the series, 83°4, was chame- prosopic.* The maxillo-facial index was computed in thirteen skulls: in five it was above 50; in eight from 45 to 49; the length of the upper face was proportionally greater than that of the complete face, and the mean index of the series was 49°7, almost leptoprosopic. The gnathic index (FLowER) was computed in twenty-four skulls, orthognathous * For the signification of these terms I may refer to Part I of this Memoir, vol. x], p. 606 ; and to Jour, of Anat, and Phys., vol. xxxvii, p. 406. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 209 except in three specimens, two of which were mesognathous, one prognathous; the mean of the entire series was 94°5. In the series of nasal indices one skull was platyrhine, nine were leptorhine, fifteen were mesorhine, and the mean, 48°9, was on the confines of the mesoleptorhine groups. The orbital index varied in amount: in three the orbits were open and rounded, megaseme above 89; in seventeen the transverse diameter was wide in relation to the height, microseme, index below 84; six were mesoseme; the mean index of the series was microseme, 81°4.. The palato-maxillary index also had a wide range: three were dolichuranic, long relatively narrow palates, index below 110 ; fourteen were brachyuranic, relatively broad palates, and of these seven were hyper- brachyuranic, index above 120; two were mesuranic; the mean index of the series was 1185, brachyuranic. . Fron?the number of specimens which have been described or referred to in this section, one is justified in summarising the general characters of the skulls or heads of the people of the bronze age as follows :— The crania were round-headed, brachycephalic ; the vertex was relatively flattened and the height of the skull from basion to bregma was definitely less than the breadth ; the back of the head was flattened and in several approached the vertical ; the length of the face was low in relation to the breadth between the zygomata ; the upper jaw was almost vertical, orthognathous, not projecting; the nose was relatively narrow, and not widened transversely at the nostrils; the orbits were variable, though frequently the breadth was great in relation to the height; the palate was frequently short and relatively wide; the capacity of the cranium was * such as to associate it with the general standard of modern Europeans. An absolutely sharp line, as regards period of time, cannot be drawn between the neolithic people and those of the bronze age, who to some extent overlapped in their occupancy of Scotland ; skulls either dolichocephalic or approximating thereto have been occasionally found in short cists. It would seem therefore that some of the skulls found in short cists, described in Sir DanteL Witson as kumbe- cephalic, were of the neolithic race. Flakes and arrow-heads of flint, worked stone, bone and horn, the characteristic substances from which neolithic man found his implements and weapons, have not infrequently been found in these cists. Both peoples practised cremation and inhumation. Both manufactured. a rude pottery and enclosed examples in their graves, though in their characters the urns of the neolithic period, rounded at the bottom, differed from those of the bronze age. Caves and Rock SHELTERS. TasLE VI. In referring the brachycephalic skulls to the bronze age, and the dolichocephalic skulls to the polished stone age, one has been guided by the character of the tombs, the grave goods which they contained, the position of the body and the 210 PRINCIPAL SIR WILLIAM TURNER ON configuration of the skull. We have next to consider interments where the people did not build graves, but availed themselves of the opportunities afforded in their respective localities and deposited their dead in the caves or rock shelters provided by nature. Caves and clefts in rocks have been in use in Scotland both as dwellings and as places of interment, and were noted many years ago by CHALMERS in Caledonia and by Sir D. Witson in Archxologia. Examples had been explored on the face of sandstone rock overhanging inland rivers like the Esk, the Teviot and the Jed, but they have been more frequently recognised in sea cliffs, which in prehistoric days formed in many places the coastline and were directly washed by the waves. When the earlier beaches subsequently became uplifted, the cliffs were removed from the sea by an interval which corresponded to the breadth and height of the newer raised beach or beaches. . In 1847 a cave at Lower Warburton, Montrose, about half a mile from the estuary of the Esk, 15 feet above high-water mark, was examined by ALEXANDER Bryson.* q It contained bones of existing mammals, many of which were split, shells of edible molluscs, coarse pottery and a few fragments of human bones. The floor of the cave consisted of rounded stones, as if from the sea-beach, upon which was placed a thick stratum of dark loam and sea-shells, on the surface of which a layer of mammalian bones occupied the width of the cave. In 1875 Miss C. Mactacan described caves at Wemyss on the Fife Coast,} the walls of which were decorated with a remarkable series of sculpturings. Their characters had previously been embodied by Sir James Simpson in his Archaic Sculpturings,t in which he stated that bones of existing mammals, frequently split’ so as to remove the marrow, sea-shells, perforated stones and implements of deer- horn had been found in the rubbish of the floor of some of the caves. In 1873 the Rev. R. J. Marieron described § the contents of a cave found in 1862 at Duntroon, Argyllshire, about 23 feet above high-water mark and 186 feet distant. from it. The floor was formed of shingle, with sea-shells of several species, the bones of a red deer, ash, charcoal and scrapers of flint. A human skeleton in the sitting posture and bones of at least six persons were exposed amidst the loose stones which occupied the grave. No description of the characters of the bones has, I believe, been recorded. In 1875 and later a cave was explored on the sea- coast at Borness, Kirkeudbrightshire, the floor of which was about 27 feet above high-water mark. Its contents were most carefully described || and found to consist of burnt wood, bones of birds, mammals, sea-shells, numerous bone, stone, bronze and iron implements, fragments of two human skeletons ;. also a part of a cup of * Edin. New Phil. Jour., 1850; Howpren and M‘Batn in Proc. Roy. Phys, Soc. Edin., vol. iii, 1867; Proc. Soc. Antigq. Scot., vol. x, 1875. + Proc Soc. Antig, Scot., vol. xi, 1876. + Idem, vol. vi, 1868. § Idem, vol. x, p. 306, 1875. || Bruce Crarkg, A. J. ‘Corrte, Ry J. Jounson and A. &R. Honr i in Proc. Soc. Antig. Scot., vol. x, pp. 476, 499 1875; vols. xi, p. 305, and xii, p. 669. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 211 Samian ware. Mr JoHN SMITH excavated* a rock shelter near the Ardrossan railway station in which shell-heaps rested on sand and gravel. Remains of land animals were found, along with human bones from apparently two individuals. The bay of Oban is a part of the coast of Scotland which has furnished several examples of caves in which human remains have been found. In the ancient cliff which bounds the 25-30 feet raised beach four caves have been explored. In 1869 Mr Jonn Mackay, whilst quarrying the rock at the north end of the bay, opened into a cave filled with earth in which many bones were found. I visited it in the following year and obtained from Mr Mackay, whose name | associated with the cave, bones of deer, dog, fox, hare, the larger of which had been split for the extraction of the marrow. Birds’ bones, shell-fish, flint chips and implements were also present. But further, portions of ‘two human skeletons were obtained, one that of an adult male, the other that of a youth about eight or nine years old. t In 1877 I received specimens from a cave in proximity to the Oban Gasworks,} which consisted of bones and teeth of pig, goat, ox and red deer, some of which had been split ; also sea-shells, a flint chip and fragments of pottery, which Dr Joszpx ANDERSON regarded as resembling the cinerary urns of the late neolithic and bronze age. Fragments of human bones were also sent, but too much injured to be restored. In 1890 a cave was exposed in the cliff behind the Oban Distillery, situated about 40 feet above the present sea-level. In clearing out the soil and debris cart-loads of sea-shells were removed. Flint and bone implements were also present. A small number of bones of mammals, birds and fish were also obtained, including as many as eight human lower jaws of adults and children, with other bones of the skeleton, though too much injured to be restored. The most noteworthy discovery was that of the MacArthur Cave, explored and carefully described by Dr JoszpH AnpERSON in March 1895.§ -This cave was about 30 feet higher in the cliff than the level of the present beach, and 100 yards distant from it. A bed of gravel or small water-rolled pebbles, about 6 feet in its greatest thickness, extended over the floor of the cave. Intercalated in the upper part of the gravel, but thinning out towards the sides and entrance, was a deposit of sea-shells, the lower shell-bed, which varied in thickness from 5 to 26 inches. On the top of the gravel was an accumulated refuse-heap from 27 to 36 inches thick of the shells of edible molluscs, interspersed amongst which were patches of ashes, wood charcoal, bones of deer, ox, pig, dog, badger, the larger of which, as well as horns of deer, were splintered for the manufacture of implements, many of which were scattered through the shell refuse and the gravel; bones of birds, fish and claws of crabs were also * Ayr and Gal. Arch. Assoc., vol. vii, pp. 62 et seg., quoted by Dr Rozpert Munro. * + Described in Reports of Edinburgh Meeting of British Association, p. 160, 1871 ; the femora, in my “ Challenger ” Report, part xlvii, p. 97, 1886 ; the bones, more fully in my account of the Caves at Oban, Proc. Soc. Antig. Scot., vol. xxix, May 1895. { The Gaswork, Distillery and MacArthur Caves are described in my memoir on the Caves at Oban, cited above. § Proc. Soc. Antig. Scot., vol. xxix, p. 211, 1895 ; also my memoir on the Caves at Oban in the same volume. TRANS. ROY. SOC. EDIN., VOL. Ll, PART I (NO. 5). 30 212 PRINCIPAL SIR WILLIAM TURNER ON TaBLE VI. Caves—Mausoleum. alot toe De ae Sede Oban Caves. ‘ a; Anatomical Museum, U. of E. | Seacliff Matizolennte iy | _ Anatomical Museum, U. of E. MAcARTHUR. MACKAY. B a F es é | | Metopic. 7 | -Metopic. f 2 Collection number, A. 215 B.216 - O17 A. ‘ C. D. | Age, . Ad. | Ad. Child Ad. Ad. | Ad. | Ad. | Ad. NOx, « M M. et INE M. M. M. M, Cubie capacity, . : ee Us 63 LS OS Rae ee Be Glabello-oecipital length, 183 205 171 188 189 175 a Basi-bregmatic height, 139 Te. es 140) 131) 1315) ae Vertical Index, 70° i m™ Toe 69°3 | 74:9 Minimum frontal diameter, 99 102 96 | OA args ae Stephanic diameter, . ie Tor 115 | TSA es ce a Asterionic diameter, . 115 | 103 | 120 | 100) 105 ) Ws Greatest parieto-squamous breadth, ; : 138 144 1334 147s. 137 Cephalic Index, ‘ eal ows). AOL? 77°8 | 78:2 3 Horizontal circumference,. | 520 564 490 541 — Frontal longitudinal are, 127 140 249, | 128 | 120ap. Parietal a3 5 Wes | US \ ara 128 | 126 Occipital me 4 TD | = on 121 106 Totaly a, — r Sia | ay 377 | 352ap. Vertical transverse arc, eh eels 985 319 a Length of foramen magnum, 34a ae 34 38 37 Basi-nasal length, OS ee ha TOO 7)! Bea] Basi-alveolar length, . Gian 105ap. Gnathic Index, 94-2 | 96°3 | Interzygomatic breadth, Lee 135 Intermalar vs, [28 124 Nasio-mental length, . AS Ee 125 Nasio-mental complete “facial indicia a 92°6 Nasio-alveolar length, 67 ee Mazillo-facial Index, . pod a Nasal height, 48 52 Nasal width, 23 25 Nasal Index, ALD A be 48:2 Orbital width, 40) 492 40 Orbital height, . Bo eal 33 Orbital Index, : 82:5 | 73-8 ae 82:5 Palato-maxillary length, DON eres 53 5) Palato-maxillary breadth, 60] . 59 67 _ Palato-mazillary Index, ol Got2| wees 1113 aL Ales | Mandible,symphysialheight,| 30 35 29 BA recognised. On the surface of this refuse was a layer of black earth which had drifted into the cave, containing bones of birds and small mammals. The imple- — ments were formed of stone, flint, bone and horn, the most interesting of them being seven harpoons, or fish spears, made of deer-horn. Two human skulls were found, along with other bones of their skeletons, as well as the lower jaw of a third, C, and = THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 213 ‘ ; ; Fic, 29,—Skulls A and B, MacArthur Cave, Oban. 214 . PRINCIPAL SIR WILLIAM TURNER ON four axis cervical vertebree, so that at least four persons had been entombed in the eave. One skull, A, was found on the surface of the black earth towards the back of the cave; another, B, was embedded in the shell-bed immediately subjacent to the black earth. The skulls from the MacArthur cave were adults. B was a powerful male, A possibly also a male, C possibly a female. In B the general form of the cranium was elongated ovoid, pentagonal in both norma verticalis and occipitalis, broadest at the parietal eminences, the parieto-squamous region vertical, the parieto-occipital slope moderate, cranial sutures in course of closure, cephalic index 70:2, dolichocephalic ; | forehead somewhat retreating; glabella and supraciliaries strong ; nasion depressed ; nasal bones short, moderately projecting; facial bones injured, vertical diameter of orbit low, index microseme, lower jaw massive, angle marked, chin strong. In length, breadth and horizontal circumference the dimensions were large, and in the cubic capacity, although the basi-occipital was broken, the cranial ee could contain 1715 c.c. of water (fig. 29, A and B) (Table VI). In A the cranial sutures were unossified and the dimensions were smaller than in B. The teeth were erupted and not decayed. The general form of the cranium approximated to B, but the slope of the vault from the sagittal line to the parietal eminences was not so steep. The right parieto-squamous region was imperfect, and the greatest breadth, 138 mm., could only be given approximately, which, with the — glabello-occipital length, 183 mm., gave a cephalic index 75°4, in form essentially dolichocephalic ; the vertical index was 76. The forehead was not so retreating as in B and the glabella and supraciliaries were not so strong; the nasion was not so depressed and the nasals not so projecting; the nasal index 47:9 was narrow leptorhine ; the gnathic index 94:2 was orthognathous; the orbital index 82°5 was low, microseme; the palato-maxillary index 113-2 was mesuranic, and the palate was symmetrical. The lower jaw had a forward projecting chin, and a distinct angle. The lower jaw was the only bone of C to be identified ; it was well formed and with a definite chin. The skulls from the Mackay cave were those of a man and a child. That of the adult was much injured; evidently the man had been in the prime of life for the crowns of the teeth were not much worn; the lower jaw was well formed, the supra- ciliaries, the inion and the curved lines were prominent, the palate was highly arched. The child was probably eight; the face was broken away, the cranium was 171 mm. long, 133 mm. broad, the cephalic index was 77°8. The cranium was metopic, small Wormians were in the lambdoid and each pterion had an epipteric. The bones of the shafts of the lower limb showed interesting characters. In the ' Mackay skeleton the tibize were platyknemic and an articular surface for the astragalus was prolonged in front of the lower end of the tibia. In 1886 I pointed out* that the upper third of the anterior surface of the shaft of the femur was transversely * “ Challenger Reports,” Zoology, part xlvii, p. 97, 1886. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 215 widened and flattened, and an external infra-trochanteric ridge gave a defined outer border to the femoral shaft distinct. from and in front of the gluteal ridge. I had also described a similar conformation in the femora of Maoris, Sandwich Islanders, Lapps and Esquimaux. Attention has since been called by Manouvrier to this character, which he has named platymery, in femora from neolithic burials in France and from the Guanche cave-burials in the Canary Islands. In the Mackay tibize the index of platyknemy was in one 63°6, in the other 65°6 ; that of platymery 564 and 58°8 respectively. In the MacArthur cave these characters in the shafts of the tibize and femora were not so accentuated. Tt should be pointed out that in each cave_ considerable quantities of soil and rock debris were present. In those in which skulls were found the entrances had been obstructed; in the Mackay cave by an embankment of earth 8 to 9 feet thick, in which beech trees were growing, and in the MacArthur cave by a talus of earth and stones from the fall of superincumbent rock, which blocked and concealed the entrance. Evidently, therefore, the human remains found in these caves were not from recent occupancy. The presence of bones of at least fifteen individuals of different ages and sex, with the remains of food and cooking, points to the caves having been at some former period places of residence and ultimately of interment. It is to be noted that no mention is made of either bronze or iron implements; stone, flint, horn and bone, were the material provided by nature and rudely fashioned by the people for their purposes. In one instance only, the Gasworks cave, were some fragments of coarse pottery found. The implements were therefore those employed by neolithic man, and it was to this period that the caves were ascribed by Dr ANDERSON and myself in our respective memoirs. As regards the skulls in the MacArthur cave, it has been argued that they were not cotemporaneous with the makers of the implements owing to their superficial position on the floor of the cave, one being on the surface of the black earth, another embedded in it and almost on the top of the subjacent shell-bed. It should, however, be stated that of the boxes containing human bones which reached me for examina- tion, No. 2 was labelled “obtained in the shell-bed below the layer of black earth,” -and No. 3 “bones from the shell-beds and pockets in and under a layer of gravel, situated below No. 2.” There are grounds for thinking that the human bones which were mingled with the food refuse and implements were those of the original occupants. The skulls, again, found more superficially were presumably those of the last occupants, which remained where they died. It does not, however, follow that they were not the descendants of the older dwellers in the cave. The skulls differed so greatly in form and proportion from the rounded brachycephalic heads which are associated with the characteristic tombs and implements of bronze burials, that we cannot associate them with that period. They must therefore belong either to an antecedent or a subsequent period. There is nothing in the great length of the 216 PRINCIPAL SIR WILLIAM TURNER ON skull B, its dolichocephalic proportions, or its large cubic capacity, adverse to its being from the neolithic maker of the implements found in the cave; an argument that its size and capacity harmonised rather with those of people of more modern times does not necessarily confirm it as being a relatively recent introduction into the cave in which it was found, for skulls undoubtedly neolithic have been obtained © with large capacity. Dr ANDERSON in his memoir emphasised the discovery of harpoons of deer-horn in the MacArthur cave as the first examples of the kind seen in. a cave in Scot- land, though similar implements had been found in a refuse mound at Cazsteal nan Gillean in the island of Oronsay. He also referred to a specimen from the neolithic stratum of the Victoria cave, Settle;* and others from Kent's cavern at Torquay associated with implements of palzeolithic type. Since the publication of his memoir a rock shelter at Druimvargie, Oban,{ was cleared of its talus, and a refuse-heap of shells and broken bones, from which waterworn pebbles, borers and chisels of bone, together with portions of two harpoons of deer-horn, were obtained. In an article commenting on the Ohsen: caves M. Bove called attention} to similar harpoons found in caverns in France by himself and by M. Carratuac, also to their discovery by M. Pierre in the cavern of Mas-d’Azil. By these observers harpoons of this type were regarded as occurring in a stratum of, or later than, the age of the reindeer which had been succeeded by the red deer, ve. at a time inter- mediate to the paleolithic and neolithic epochs. M. Bou.e considered therefore that the archeological data of the Oban caves are not to be regarded as neolithic, as was stated by Dr ANDERSON and myself, but, like the discoveries in the Mas-d’Azil, inter- mediate between it and the palzeolithic epoch. In a memoir § published last year, Mr HENDERSON Bishop gave a detailed account of his excavations on a sandhill, Cnoc Sligeach in Oronsay, in which were beds of sea-shells with food refuse and fragments of charcoal. Implements of stone, flint, bone and horn, including portions of harpoons, were figured. He regarded the harpoons as for the most part of characteristic Azilian type, and as evidence of the culture of the time when the sandhill was occupied and the refuse-heap was in process of accumulation by the islanders. He concluded that this pened was die correlated with that of the occupation of the Oban caves. In connection with the assumed relation of the Oban and Srondiy vets with the Azilian stage in the French caves in which harpoons of this type have been dis- covered, the French naturalists have associated them with a modification in the species of the fauna, remains of which have also been found, with the disappearance of the reindeer and the advent of the red deer, a difference which expressed an * Boyp Dawkins, Cave Hunting. + Ropert Munro in Prehistoric scotland, p. 54, 1899. { L Anthropologie, vol. vii, p. 319, 1896. § Proc. Soc. Antig. Scot., vol. xlviii, p. 52, 1914, . THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 217 interval of time and change of temperature. In the Oban caves and the Oronsay refuse-heap there is no corresponding modification in the fauna, the species of which correspond with the neolithic period, and are also characteristic of the present time. No painted or coloured stones were found in either of the Scotch localities such as have been described by M. Pierre* in the Azilian layer in the cavern of that name, which is regarded as immediately preceding the neolithic age. Iron Ace. Tasie VII. In the course of centuries the bronze-age people ceased to be the dominant race in Britain, and were succeeded by successive invasions from the Continent of other races, Celts, Norsemen, Anglo-Saxons, Danes, whilst for a period of about four centuries the Romans had occupied South Britain and for less than a century the more southern part of North Britain. The use of bronze in the manufacture of implements, weapons and personal ornaments gradually disappeared and was replaced by iron. The Celts, the immediate successors of the bronze-age people, occupied doubtless the country along with them, and the use of bronze and iron for a time overlapped. It is generally admitted that the extraction of iron from its ores and its economic applications were known on the Continent before the first invasion of Britain by Celtic tribes took place, and that it was through them that this metal was introduced and gradually displaced the use of bronze. Celts.—It has been shown that the neolithic and bronze-age inhabitants can be recognised by the character of their interments, by the grave goods and the skeletons found in their tombs. - The question therefore naturally arises if the prehistoric Celts in Scotland had a distinctive mode of burial, and if their skeletons could be relied on as possessing definite ethnographical features. Dr JosrpH ANDERSON has shown in his classical Rhind Lectures that abundant specimens of the artistic work of the Celts have been preserved from both Pagan and Christian times. The Celtic art of the pagan period in bronze, iron and even in gold displays originality in design and skill in execution, whilst evidence of constructional architectural ability is shown in the ancient brochs which must undoubtedly be assigned to the Celtic and not to the Norse period of occupancy of Scotland. But in regard to a distinctive type of interment he is emphatic in stating that he knows of no characteristics as distinguishing the burials of the pagan Celts in the iron age in Scotland, so that, through the burning of the bodies of the dead, our archeology is absolutely destitute of recorded data for this purpose. Since the publication of his lectures discoveries have been made which seem to throw some light upon the former obscurity. In 1903} a grave was opened at Moredun near Edinburgh and was described by Mr F.'R. Cotns. A cist was * LT’ Anthropologie, vol, vii, p. 385, 1896. + Proc, Soc. Antig, Scot, vol. xxxviii, 1904. 218 PRINCIPAL SIR WILLIAM TURNER ON TasLeE VIL. Tron Age. CELT, NorsE. 3 ssis 5 Orkney : Se 3 = Gullane. / pieromell ‘a 2 A a A a Collection number, . Mille wee or Biase E.T. 63\E.T. 62|E.T. 20, E.T. | Age, . : A ; --| Ad. | Ad. | Ad. | Ad. -| Ad. | Ad. | Ad, | Ad. Sex, . : ; Sih Pe BEN phe M. M.S) oe tee _ Cubic capacity, . é one ie Hye ot og 1260 | 1350 |’ ... ei. | Glabello-occipital length, Sl Oe sien cat ae 175 | 1857) “189 | ekse 7 Basi-bregmatic height, . 140. ... Merril pet | 105:)) 91205) 4124. aes Vertical Index, . : S| 72:8 a9 717 2049 | 60: 64:9 | 65°6 | 67:2 Stephanic diameter, . |) Oa aire Sas 3 1139) AAS ASE eee Greatest parieto-squamous | breadth, F : . | 144 aes hac Soe 143 | 139 147 145 Cephalic Index, ‘ 4 pV Jom Oona on (0775 STP Ne FOd, \PIES* | wen Horizontal circumference, 5 (qa DOO ul eeere ae ee 504 | 525 | 542 =ta Frontal longitudinal arc, .| 123] ... Baa im I3T a28 a 2s Parietal “ pe Bl 25) ie x. th | 116°; 13838 | 124) - Occipital r Rem Weed its tes es ic Te LOOM S25 Total _,, E> AIS SOC a meeME Rae! ae oy. | | 358 | 371] 364. Vertical transverse arc, .| 320)... sat al eck 293) | S300 SOs a tees Length of foramen magnum, SE ze a Be 33 36 35 | ee Basi-nasal length, . do) LOO) ied ee. Sie | deere 92 93 105 — Basi-alveolar length, . ! a esa hal en 97 94 | 107 | Gnathie Indge, eee i GFN a ee Nia! 105°4 |101°1 | 101-9 Nasio-mental length, . ie lb! awe eal a5 5 sath Nasio-alveolar length, — . ail | Ae aa 71) Orbital width, , ef eet +3 co eee 4] 40 41_ Nasal height, . : 3 eal ee bemeae sic e “e 54 Nasal width, | 22 Nasal Index, | | Ree be 40:7 Orbital height, . | | 31 29 38 Orbital Index, . ee diag oe TION WIZ Sy GOA Mandible,symphysial height, 7 oho ke em | | ecg of ye aa | | exposed which gave evidence of an interment at an early period of the iron age, when bronze-age modes of burial were still practised. The sides, ends and cover of the cist were formed of sandstone slabs from 34 to 5 inches thick. It was an almost regular oblong and was directed practically east to west. The inside measurement was approximately 4 feet in length and 2 feet 3 inches in breadth. The cist con- tained two decayed human skeletons, a young adult and an adolescent about twenty- one ; an iron fibula, an iron circular brooch, and a long iron pin. ‘The cist in general form and construction was not unlike a short cist of the bronze age, and the skeleton was bent on itself; but the presence of iron objects and the absence of urns associated — it with the iron period. It is stated that a portion of cloth with which the body had been wrapped was adherent to the pin of the fibula. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 219 The skull of the adult in the Moredun cist, though much damaged, enabled Professor Bryce to estimate its form and proportions (Table VII). The cranium was ellipsoidal in the norma verticalis, vertex flattened, no sagittal crest, the sides rounded, the parieto-occipital slope gradual, the occipital probole slightly marked, the frontal bone full and rounded, the glabella and supraciliaries not pro- jecting, the upper orbital rim thin, characters which suggested that the skull was that of a woman. The maximum length was 192 mm., the basi-bregmatic height 140 mm., the maximum breadth 144 mm. The cephalic index was 75, the vertical index 72°8. The skull was dolichocephalic and the vertical index, less than the cephalic, was metriocephalic. The face was high leptoprosopic and the jaw orthognathous. In 1908 a cairn and interments weré explored at the Black Rocks, Gullane, Hast Lothian, by Dr Epwarp Ewart and Mr Atexr. O. Curty.* The cairn consisted of whinstone boulders placed on the shore between high-water mark and the line of the 20-foot beach. Six human skeletons were found in the cairn, some of which were in the contracted position, but in no case was one enclosed in a built cist. Near one of the skeletons a small spiral ring of bronze was found, and beneath another was a rust-corroded, fragile, disc-shaped iron knife-dagger, 7 inches long and 2 inch broad. A disc or whorl of sandstone 22 inches in diameter, and per- forated in the centre, was found at one end of the cairn. In the neighbourhood of this cairn were others similar in type, though some were smaller in size and con- tained single burials. In-only one of these did a cist appear to have been noticed. Tt was thought that a number of skeletons had been buried in these cairns. So many imterments in a comparatively limited area point to the locality being an ancient cemetery. Three skulls were procured, which enabled Dr Warerrston to examine and obtain some measurements. In one, an adult male, the cephalic index was 765, the vertical 74°9 ; in a second, an adult female, with a metopic suture, the cephalic index was 7%°3, the vertical 71:7 ; in a third adult the coronal, sagittal and lambdoid sutures were obliterated, the skull had a subscaphocephalic character and was so elongated in relation to the breadth that the cephalic index was only 67°3 and the vertical 64:9 (Table VIT). A femur showed strongly marked platymery; also an articular facet for the tibia above the inner condyl due to great flexion of the knee in the squatting posture. WATERSTON estimated, from the length of the femur and tibia, the stature of the individual at 5 feet 3 inches. The characters of the crania approximated to dolichocephalic and not to brachy- cephalic proportions, and the burial of the bodies in the sand which surrounded the boulders, and not in built cists, together with the presence of the iron knife, points, notwithstanding the bronze ring, to the interment having been made at or later than the transition from the bronze to the iron period. These interments may have been * Proc, Soc, Antig. Scot., vol. xlii, 1908. TRANS. ROY. SOC, EDIN., VOL. LI, PART I (NO, 5). 31 = 220 PRINCIPAL SIR WILLIAM TURNER ON somewhat later than that at Moredun, and they are perhaps to be regarded as Celtic burials. In 1909 Mr James E. Cree described* two caves which he had explored at Archerfield, East Lothian, which had been used as habitations. Bones of domestic animals, shells of edible sea-molluscs, fireplaces and charcoal were found. Also broken pottery, some of which was coarse, others of finer paste and a few frag- ments of Samian ware. Objects of stone, bone, horn and glass fitted to be of use were found. An iron knife corroded, a spear-head and nails, also a bronze pin and small portions of this metal were obtained, but no human remains. Mr CRE associated the relics and the use of the caves with the early iron period, though the Samian ware makes it probable that they were occupied during Romano- British times. + Many years ago a group of graves was exposed at Cramond,} in one of which — an iron key was found along with a skeleton. Its presence might seem to point to the burial as of the iron age, but this grave, like the others in the same group, was — arranged and built on the type of the long stone cists to be described in a later section (pp. 229, 231). A figure of the key, which evidently had been little corroded by rust and had a modern aspect, is given in the Archxologica Scotica. I am inclined to regard these graves as later than the pagan iron age. In 1864 a skull © was presented to me by Dr Lumeair which had been obtained at Largo, in the field in which the standing stones are situated. It was found in a “ coffin built of loose slabs,” and it showed a fracture in which a piece of iron was lodged. ‘This burial | have also associated with the long stone cists (p. 228). Norsemen.—Without taking into consideration the Roman invasion of Scotland, which probably did not exercise permanent influence on the physical characters of the natives, other invasions followed which have left their mark. The most important of these was the establishment by the Norsemen of settlements at various parts of the coast, relics of which have been preserved. An admirable account of some of these burials and of the objects found was given by Dr ANDERSON in the first and second of his Rhind Lectures.§ Sutherland, Caithness, the Orkneys, the Shetlands, Islay, Mull, Tiree, Barra, Sangay and St Kilda provided examples. Some were Viking burials; they contained implements and weapons of iron, and various objects in bronze, also silver ornaments. They were in mounds containing human skeletons and sometimes bones of the horse and dog. ‘The skeletons did not seem to be contained in stone cists, though in some cases planks of wood with iron rivets were present, as if the Viking had been buried in his ship. Subsequent to the publication of these Lectures, additional examples have been * Proc. Soc. Antig. Scot., vol. xliii, p. 243, 1909. + In the section on Ethnography (p, 247) the recent discovery of iron objects by Mr ALExr. O. CURLE in a fortification on Traprain Law, East Lothian, has been included in the text. { Archaxologia Scotica, vol. iii, p. 40, 1831, § Scotland in Pagan Times—The Iron Age, 1883. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 221 recorded. ANDERSON himself described in 1907 objects found in a low mound near the beach on the east of the island of Oronsay:* brooches of bronze with iron pins, beads of amber and serpentine, an iron knife and two human skeletons at full length directed with the feet to the 8.S.E.; also iron rivets, as if the burial had been in a ship. His description included an account of a Viking grave at Kiloran Bay, Colonsay, from notes made by the late Mr Wm. Gattoway. A mound 15 feet by 10 feet, with rough schist slabs, contained an iron sword, spear, axehead, umbo of a shield, with some bronze articles, Anglo-Saxon coins of the ninth century, also a human skeleton in a crouched position, that of a horse, and numerous iron rivets, as if this also had been a ship burial. In 1896, within a gravel mound at Lamlash, Arran, about 170 feet above high- water mark, a Viking burial was exposed, and was subsequently described by Mr J. A. Batrour.t It contained a single-edged iron sword and the iron umbo of a shield of the type of the eighth or ninth century. More recently, in 1909, a cairn at King’s Cross Point, Lamlash, was found to contain a heap of calcined human bones, with charcoal, iron and bronze objects, a bronze coin of the ninth century, also iron rivets, as if the body had been cremated in a boat. The skeleton from Kiloran Bay was said to be that of a powerful man, not of great stature, with a long narrow dolichocephalic skull ; those from Oronsay were from a large man and a woman, both advancing in age. ANDERSON described in his Rhind Lectures on the Vikings, graves and mounds at Westray and at Pierowall in Orkney in which iron implements and weapons, human bones and those of the horse and dog were found. Those from Pierowall were associated with wooden planks and iron rivets, which suggested an interment along with the ship. Inthe National Museum of Antiquities is a male skull (E.T. 63) from Pierowall (Table VII), the vertex of which was flattened; the parieto-occipital slope was relatively steep. ‘The basi-occipital had been fractured transversely, and subsequently united obliquely, which might have affected the measurements in length and height of the cranium and conse- quently the cephalic and vertical indices, which were 81°7 and 60 respectively. Another male skull (E.T. 62), stained brown from peat like E.T. 63, and possibly from the same place, was neither flattened nor keeled on the vertex, and the parieto- occipital slope was not so steep. It was longer than but not so broad as in 63, and the cephalic and vertical indices were 75°1 and 64°9 respectively. It is possible that one of these skulls may have been that referred to by Mr Wm. Renpatt}{ as from the Norse cemetery at Pierowall, Westray, and the other from the group of grave mounds at Pierowall examined by Messrs FarReR and G. PErrig in 1841 and 1855. Opportunities of measuring the skulls of the Norse Vikings have been scanty. * Proc. Soc. Antig. Scot., vol. xxv, p. 482, 1891, and vol. xli, p. 487, 1907. + Idem, vol. xliv, p. 221, 1910; The Book of Arran. { ANDERSON, The Iron Age in Scotland in Pagan Times; also RENDALL in Proc. Soc. Antig. Scot., vol. xiv, p. 85, 1880. ; 222 PRINCIPAL SIR WILLIAM TURNER ON In two of the above specimens one cranium was approximately dolichocephalie, — another brachycephalic, but the proportions in the latter may have been modified from the fracture of the basis cranii. : PREHISTORIC Mausoteum. Tasir VI. In November 1865 I accompanied the late Sir J. Y. Stmpson to Seacliff House, North Berwick, to inspect a structure, containing human skeletons, which had been exposed by the late proprietor Mr J. W. Larpiay, in the course of excavating a deep pit for the construction of an ice-house, in the policy adjacent to the mansion- house. In notes made at the time I stated that the pit was dug in a mound of sand higher than the level of the surrounding ground. In the middle of the mound — the roof of the structure had been brought to view 6 feet below the surface, and in Fre. 30.—Section of Mausoleum, Seacliff. the further process of excavation one of the walls, which I may call the front wall, had been uncovered. This wall had fallen down before our arrival, and the interior of the structure was partially disclosed. The sand above and around was somewhat stratified. The walls of the structure were built of fat flagstones superimposed on each other, without any trace of intermediate mortar. They enclosed a four-sided, low, rectangular chamber longer in one diameter, the transverse, than in the other or antero-posterior direction. The roof was vaulted and formed of flat stones, placed almost horizontally, and— arranged in five rows, which converged across the transverse diameter of the chamber. The top of the wall on each side supported the outer row of the flat stones of the roof, which in their turn supported the second row, and on these again — the stones of the middle row rested (fig. 30). The interior of the chamber exposed by the falling down of the front wall was seen to be oceupied with sand, which had apparently entered through the inter- spaces between the stones. For the most part the sand was loose and easily taken out, but near the floor it was more compact, blackish in colour, and so much inter- — mingled with the skeletons resting on the floor as to interfere with their removal. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 223 In the course of the investigation the bones of six skeletons—four male adults, a young woman and a child—were recognised on the floor of the chamber. In attempting to remove the child’s skull the frontal bone came away in my hand. Three of the skulls, two adults and the child, were close to the back wall of the chamber, whilst two other adult skulls lay close to the front wall. The position of the sixth skull was not noted. The mingling of the bones with the compacted sand, and their softened and fragmentary condition, made it impossible to disengage the separate skeletons. As two of the skulls were near the front wall and three at the back wall, it seemed as if the bodies had been laid in the transverse dia- meter of the chamber side by side in a row, alternately heads and feet. The Fic. 31.—Skull A, Seacliff. conditions under which the skeletons were found did not enable me to say whether the bodies had been buried in the bent or the extended position. The dimensions of the chamber were not taken, but it was sufficient to accommodate the adults and the child lying side by side at full length. My notes do not refer to urns, implements or other relics as found in the chamber. A few days after our visit Mr Larpnay brought to the University a box containing five crania; but the skull of the child, the bones of which had come asunder, had not been preserved. Only one of the adult skulls was sufficiently entire to permit of a fairly complete examination. The other four adults had no vestige of a face, excepting portions of the lower jaw, and the cranium was so much injured that accurate measurements of the dimensions could only be partially obtained. The entire skull, A, Table VI (fig. 31) belonged to an adult man, apparently in 224 PRINCIPAL SIR WILLIAM TURNER ON middle life, for the teeth were much flattened from use, but the sutures, including the frontal, were distinct. In the norma verticalis this skull was broadly ovoid, and in its general character well filled; it was not ridged in the sagittal region. Although the length was 188 mm., the greatest breadth was as much as 147 mm, and 120 mm. at the asterion. Its cephalic index was 78'2, mesobrachycephalic. It was cryptozygous. The glabella and supraciliary ridges moderately projected, the forehead was slightly receding. The vertex was somewhat flattened, the slope outwards and downwards to the eminences was moderate and in the parieto-occipital region it was not abrupt; the occipital squama projected behind the inion. The basi-bregmatic height was less than the maximum breadth, and the vertical index 75 was metriocephalic. The nasion was somewhat depressed ; the nasal bones were moderate in length, concave in the bridge and projected downwards and forwards ; the height of the nose was more than twice the width, and the index was almost leptorhine, 48°2. The transverse diameter of the orbits was greater than the vertical, and the index was microseme, 82°5. The alveolar border of the upper jaw was damaged in the incisor region, but the gnathic index, approximately 96, was obviously orthognathous; the face was moderately high, the index 92°6, lepto- prosopic. The lower jaw was massive, with a strong chin, deep symphysis, well- defined angle, and the teeth were much worn. The palate was deeply arched. The palato-maxillary index 121 was hyperbrachyuranic. No special modifications of the bones were noted, except the metopic suture. The cranial capacity was 1530 c.c. The cranium B, apparently that of a man, was 189 mm. long; but as the left side, which was probably that on which it had rested in the chamber, had been broken away, it was not possible to obtain either the frontal or parieto-squamous breadth. The asterionic diameter was only 100 mm. From the general aspect of the fragment, it is probable that in its proportion of length to breadth, in its gradual slope downwards in the parieto-occipital region, and in the convex occiput, the cranium was dolichocephalic; the basi-bregmatic height was materially below that of skull A, and the vertical index 69°3 was chameecephalic. The cranium C, apparently a man, had also lost a large part of its left side. Its length was 175 mm., but the frontal and parietal breadths could not be taken. The parieto-occipital slope was very like that of B; but as the asterionic diameter was somewhat greater, it, is possible that in its proportion of length and breadth C may have been intermediate between A and B. Its vertical index 74:9 was — metriocephalic. In D, only the frontal and parietal bones had been preserved; the glabello- lambdal length was 165 mm., the parietal breadth was 136 mm., and the index computed from this length was 82°4. In KE, only the right half of the cranium had been preserved; the frontal region had feminine characters, which were also “ met with in one of the broken lower jaws, so that one of the skulls was without doubt that of a woman. Skull E inclined to be dolichocephalic. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 225 Owing to the absence of grave goods it is difficult to fix the time when this mausoleum was built and to associate it with a definite period. The small flagstones with which it was constructed remove it from the neolithic period, as well as from the short cists of the bronze age, each of which was characterised by its massive con- struction ; moreover, there was no evidence of cremation. From the constructive skill displayed in the superposition, without binding mortar, of the flat stones which formed the walls and roof, it had probably been built on the surface of the ground, and, after the bodies had been placed in it, covered by a mound of sand for its protection. The size and shape of the stones reminded one of the flags employed in the construction of the long cists found in many counties north and south of the Forth, but they were laid horizontally, so as to form the walls of a chamber much higher than that of a long cist and destined to contain not a single but several bodies. The roof also, instead of having only a single row of flat stones, approxi- mated to the vault shape, such as has been seen in some chambered cairns ‘and brochs. Whatever the period may have been, the tomb was obviously the mausoleum of a family, for it contained the skeletons of four men, a woman and a child, and there was no evidence of wooden coffins. The antero-posterior diameter of the chamber was apparently sutticient to allow the bodies to be buried as extended. They had evidently been inhumed at the same time, as if the family had died through a common calamity. At Seacliff Mr Larpuay described, in 1870, an ancient building of stone, without mortar, found at the beach on the Ghegan Rock, about 22 feet above high-water mark, adjoining which was a kitchen midding.* A few human bones, those of Bos longifrons, small sheep, goats, small horses, red and roe deer, various rodents, birds, fishes and sea-shells were contained in the refuse-heap; also implements of bone and combs, but none of metal, fragments of pottery, with one vessel of considerable size, a rude quern and numerous round stones which bore evidence of fire. It is dificult to assign a date to this building and its original occupants, though the remains point to a prehistoric habitation. Karlier in date than Mr Larptay’s paper a former proprietor of Seacliff, Mr GrorcE Sxico, had described a cave discovered in 1831{ in a cliff bounding the bay in which the Ghegan Rock is situated. The floor of the cave, about 20 feet above high-water mark, was paved with flat stones. A large stone, 44 feet high, stood on a mound 8 feet 1 inch in height at the middle of the mouth of the cave; the top of the stone was flat and 64 feet in diameter. Mr Sxico regarded it as an altar and the cave as a place of sacrifice, for the floor of the cave was covered with wood ashes mixed with bones of horse, dog, pig, sheep partially calcined, and limpet shells. The skeletons of two young children were found at the base of the so-called altar-stone, which was there probably for defensive purposes. * Proc, Soc. Antig. Scot. vol. viii, p. 372, 1871. + Archxologica Scotica, iv, p. 3538. Lo bo for) PRINCIPAL SIR WILLIAM TURNER ON Lone Cistrs—Buitt Graves—SronE Corrins. ‘Tasie VIII. Archxologists have long been conversant with interments in Scotland in stone cists made about 2 feet longer than the short cists of the bronze age, so as to allow the dead to be buried at full length and not in the bent or contracted posture. The sides, ends and cover stones of the long cists did not consist of single massive slabs, but were built of several thin, undressed flagstones; those of the ends and sides rested by the lower edge on the ground, whilst their upper edge gave support to the horizontal slabs of the cover, which, like those of the sides, were several in number. The lateral borders of the flags were not connected by cement or clay, but were simply apposed, and the floor might either be paved with thin flags or formed by the soil. In many of these interments the cists were directed east to west, and it was frequently noticed that the skull was at the west end. These cists differed from those described in the previous sections by the absence of urns, implements, weapons and ornaments, and they have consequently received little attention from archeologists, though in a few cases objects of iron have been noticed lying near the contained skeleton. Interments of this kind have been frequently exposed near the surface during agricultural operations. Kirkliston, W. Lothvan.—The occasion on which I first saw long-cist burials was in April 1864, when Sir JAMES Simpson invited me to accompany him to the cemetery then being excavated by Mr Roperr Hurcuison of, Carlowrie in the Catstane field, Kirkliston. Fifty-one graves were exposed, arranged in nine rows, with their long axes directed east to west. In the longest row were thirteen graves, in the shortest only one. ‘They were constructed as above described, and the slabs were usually of unhewn freestone, though a few were of a black shale-like stone. Twenty cists were 6 feet or upwards in length, two being 6 feet 9 inches, twenty- seven between 5 and 6 feet, while four were less than 5 feet, the smallest only 2 feet 4 inches. The average depth below the surface of the field was about 15 inches, and about a foot of space existed between the graves in each row. When the covers were removed they were seen to be paved with flat stones and to be occupied by sand and earth in which human bones, usually much decayed, were present; the skeletons had been lying at full length with the heads at the west end. Four imperfect skulls were obtained (Table VIII), and were described by me in an appendix to Mr Huvrcuison’s memoir. In one the cephalic index was 73°83, in another 75°9, in a third 77; the proportions were dolichocephalic or approximating thereto. Subsequent to the publication of the above discovery Mr Hurcuison exposed near the house of Carlowrie a coftin almost similar to those in the Catstane field, except that the floor was not paved; it was 6 feet long, 20 inches broad and 18 inches deep. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 227 Taste VIII. Long Cists. West LorTHIAn. | FIFE. M. AND E. Loru1an.| 33 ; | a { } : a ad Aerie | Q So $ Ee Kirkliston. _ Hopetoun. | Lundin. | 2 Dunbar. a" } E S | | = aa RT. Edinburgh University Anatomical Museum. ; D: | ee ee Omnia Nessa eos leennlmeme neonates (in.. fw. | B. | C, Ad. | Ad. | Ad. | Ad. | Ad. | Ad. | Ad. | Ad, | Ad. | Ad. | Ad, | Ad. |Advd.) Ad. | Ad, Aged.|/Aged. Aged. : , M.? M.? AI ag rena ee ree eel ee ere ieee ellie IVE IES St) Mies Mie i AY | Mi M. eee le a oe. abo isane... I. | ... 11585] ... | 1580 ngth, .|170 180 | 178 | 168 /188 | 178 | 178 | 170 | 191 | 189 | 186 | 177 | 190 | 188 | 176 | 186 | 185 189 ht, . pe lieeaw! | ose wees} wee | 120 1122) 2. | .., | 188) 132) | 129 | 126 | 138 | 125 | 127 | 142 | 120 137 ‘ Colt Sere y skal (ence 63°8 1685 | ... |... |72°3 |69°S \69°4 \71°2 |70- \66°5 |72:2 \76°3 \64:9 | 79-5 a = | 102 | 104 |,117 | .«.. ths dla a ... | 111 | 114 | 115 106 | 112 | 108 | 105 | 109 | 126 113 squamous . | | | ; | 129 | 182 | 137+ ... | 183 | 180 | 130 | 185 | 144 | 142 | 187 | 132 | 188 | 140 | 136 | 147 | 1388 144 a DCO Ore! Nir ws WOT NIB" WTB" AO \T54 N7bT 78-7 46 726 \74°5 W773 179° \72°6 | 76:2 MeMmeese | cas | ses “ii a esis ae ao ... | 5384 | 525 | 524 | 505 | 525 | 534 | 499 | 535 | 524 | 532 ROM | eee ol eas me aan ae m a ... | 183 | 129 | 120 | 121 | 1385 | 120 | 119 129 | i24 139 an eels i's sie ees i ae Sep hes ... | 128 | 138 | 122 | 119 | 128 | 126 | 120 | 122 | 130 129 - oa oul ane wae Ae AiG 5 ean at w. | 212) 118 | 126 | 120 | 123 | 119 | 113-| 124 | 116 120 his elects noc oi Bc talaht ae a ... | 873 | 385 | 368 | 360 | 386 | 365 | 352 | 375 | 390 388 are, Be ll errata eee Soe | cee ae wa ee ... | 307 | 300 | 288 | 286 | 298 | 290 | 285 307 | 298 309 ee eee. Peon leer e| se fe eOnol | 40°! 83\ 133 39 | 33] 38 | 32 32 : | eras aon ae eae nfo oe hee ee LOAN oom LOOM Ion OZ SOS Dial 107 |) OF 107 Bex 6 Falters fin ae ste co as rr ee GBa Oa Sie leesn aol 97.) 96 | 102 | 89) | fi = | Geena Meee oe Re na Nec . |... \89°4 |102°1197* 191-6 |89:2 99°. '99: 95-3 |91°8 | me ce We egies Io fe... | 184 | 182 | 120! 125) 124 |... | 121°) 184 | 121 | 138 ee ee ee ome PON ito iis 110 110 | 115 1114 128 | 109 |* 116 Some Ret het ete de tas a ae ae Be ee eUb dally EUGs I WSS 2b | 108) |) 122) 120)... Beigel |. |... | | es | we |. | (878 [87-2 189°9 [888 98-5 |... -189°2 91° \99-1 gth, ma eneeal Deen bees a eesti ele 60) 165°. 68:|)-71'| 78 | 65 | 72.) 72 : | 52°2 \50°3 |\50°4 57:2 | 537 538°7 |59'5 | DOM AOn Aone 48 Weba b4 1) 48) 55 | 52) 46 De | AG) LS XD |) Sey D3 || P| WL || oye 4S \54:2 \45-7 \41-7 \42°3 4t9 455 46" | 52-2 : 39 | 41 | 39| 38) 38 36 40 39 Soni ne2 || oo!) 380) | ov 34.) 384) 37 1) 83 ee re ee fee ee |e ISS 78" 1846 1947 974 | «. 944 85° |... | 846 | Pe eee re tec | x eo 8 49) | 40) 47) 2. | 55: 56 | 54 | breadth, . | ... aes x: na ate ie le ne tae Gon eOta Ge: || 56) (O58 | i... 60 62° 59 | | y Index, aan te ie eed hate pad ee .. (128°5|120°7\128 51142117" neo |VAOD)S TAORG! ORISA | nysial height, reali kewl tat I nd | Piece ie eiecen rece | 30 | 27%) 82) 34.) 28, 84.) 30 Sutherland : Kintradwell. 117 A cist, only 3 feet 8 inches long, was found at the same time, which was a 1 example of a short cist, and its walls were formed of rude slabs of sandstone. Armston and Melville, Mid Lothian.—In the same year I examined on the estates Arniston and Melville long cists similar in construction. Seven of these were posed at Arniston, whilst draining a field; from one cist I obtained a skull ble VIII). No grave goods were found, and the bodies had been buried in the ‘tended position. At Melville Grange a single long cist was on view when I visited he place, but the factor told me that in previous years others had been observed nis; in character. No grave goods had been found in any of these cists, and in that which I saw the skeleton was too fragile to be removed. ‘TRANS. ROY. SOC. EDIN,, VOL. LI, PART I,(NO. 5). 32 ee ap PS Selkirkshire : Yarrow 127 228 PRINCIPAL SIR WILLIAM TURNER ON Kelso.—In 1864, whilst digging a drain in Butts Lane, near the Abbey, several cists upwards of 5 feet long were exposed, one of which, built of freestone flags, was 6 feet 4 inches long, with a head piece at the west end, 18 inches wide at the shoulders tapering somewhat to the feet, and 12 inches deep. It contained an extended skeleton, the skull being at the west. The graves were parallel to each other and not more than a yard apart. A coarsely woven mort-cloth was wrapped round the skeleton, but unfortunately no bones were preserved.* Selkirkshire.—In 1865 the minister of Yarrow, Dr Jas. RussELL, described a cist about 5 feet 9 inches long, 16 inches broad at the head, and directed east and west. It was made of undressed flat stones, the covering stones of which were only about one foot from the surface. It was found on the farm of Whitehope near the manse, and was about 25 yards from the spot where some years ago eight similar cists had been exposed. The skull was obtained and examined by me along with Dr J. A. Smiru.f It was well formed, dolichocephalic, and with the breadth more than the height (see Table VIII). The capacity of the cranium was 1362 ¢.c. The skeleton was moderate in stature, and the marks on the bones indicated well-developed muscles. Hopetoun.—In 1876 several long stone cists, lying east and west, close together, and from 5 feet 10 inches to 6 feet 1 inch long, were exposed in a sandy mound near the shore. Portions of four adult skeletons were obtained and examined by Dr J. ALEXR. SmitH and myself. Three skulls were dolichocephalic ; one was. essentially brachycephalic, 79°4 (Table VITI). Lundin Innks.—In October 1864 I was present at the exposure of a long cist on Lundin Links, Largo, Fife, a few yards above high-water mark. It was about 3 feet below the surface and was covered by a layer of waterworn pebbles ; its direction was east to west, and it had the characteristic construction of undressed freestone slabs. The cist contained the bones of a child which had been buried in the extended position. Mrs Dunpas had described cists of the same construction discovered on the links in 1858.§ At a meeting of the Field Naturalists’ Society, Largo, the secretary stated that the cists had been arranged in parallel rows, from east to west, at regular distances from each other. The contained skeletons were extended, and usually lay on shells and pebbles which covered a floor paved with flagstones. I obtained several skulls from the graves at Lundin (Table VIII). In one cist a piece of corroded iron about 2 inches long lay across the bridge of the nose. At the same meeting it was stated that a similar burying-place had been exposed on Hallow Hill, St Andrews, and on the skull from one of the graves a piece of corroded iron had been found. Dr Lumeair of Largo also presented’ me in 1864 with a skull obtained some twenty years previously from a “‘coftin built of loose slabs” which had been exposed in the field with the well-known standing stones and immediately adjoining them. He told me that it showed a fracture in the occipital region in which was a * Proc. Soc. Antig. Scot., vol. vi, p. 246, 1868, + Idem, vol. vi, p. 65, 1868. t Idem, vol, xii, p. 65, 1878. § Idem, vol. ili, pp. 68, 76, 159, 183, 1862. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 229 piece of iron that ‘‘ crumbled into rust” on being touched. The presence of iron in these graves and in that from Cramond might be regarded as indicating that they should have been placed in the group belonging to the iron age, but the construc- tion of the graves seems to justify their inclusion amongst the long cists. The skull from Largo had evidently been injured by an iron weapon such as might have been used in a fight in days subsequent to pagan times. ‘Three skulls were dolicho- cephalic, and two were meso-dolichocephalic. Dunbar.—From time to time during heavy storms the sea had washed away portions of the coastline near Dunbar and had disclosed ancient interments. In 1865 several stone cists were partially laid bare on the links at Winterfield Mains, about 34 feet from the surface of the soil. When fully exposed three were seen, one of which measured a little more than 6 feet in length, whilst other two were said to be 54 feet long, about a foot broad, and the same in depth. They had the _ characteristic construction of long cists, and were lying east and west, the head being at the west end. Each contained a skeleton at full length, but no um nor other relic. I obtained three of the skulls. In 1891 another great storm encroached on the cliff at Belhaven, Dunbar, and exposed several cists. A note is recorded by the Rev. Rosert Paut, who stated* that they were about 7 feet above high-water mark, and at a uniform level of about 4 feet from the surface. Nine cists were counted, situated 6 feet from each other. They were placed due east and west and were of the long-cist type; the floor seemed in each to consist of a single slab. In _ 1908 another storm exposed in the same cliff two additional cists which belonged to _ this group. One skull was dolichocephalic, one meso-dolichocephalic, one meso- _ brachycephalic (Table VIII). In addition to the above examples, many of which I personally examined, similar interments have been found elsewhere. Amongst the earliest was a group of twenty-four graves exposed on Lord Rosepery’s estate at Cramond in 1822. Similar coffins were also exposed on the adjoining estate of Craigiehall. They were arranged in parallel rows from 3 to 6 feet apart, and the skeletons were extended : some were of smaller size for children. In one grave at Cramond an iron key was found near the skeleton.f In 1849 a similar key was obtained from a grave at Cockenzie. In 1829 a group of thirty cists was exposed at Old Haacks, Fife,t arranged in two parallel rows, containing bones, but no grave goods were noted. In 1859 a single cist was found at Ardyne, Argyllshire,§ which contained an extended skeleton ; also a wedge-shaped flint implement, about 2 inches long and # inch at its broadest end. In 1862, at Milton, East Lothian,|| three long cists were observed, containing extended bodies with the head at the west end, but without grave goods. In 1864, whilst foundations were being dug in Bonnington Road, Edinburgh, | * Proc. Soc. Antig, Scot., vol. Xxxix, p.350,1905. + Archeologia Scotica, vol. iii, p. 40, 1831 ; referred to supra, p. 220. } Proc. Soc. Antag. Scot., vol. iii, p. 505, 1862. § Idem, vol. ii, p. 251, 1859. \| Jdem, vol. iii, p. 508, 1862. 230 e PRINCIPAL SIR WILLIAM TURNER ON ascertained that four stone coflins of this type were exposed, but the skulls were not preserved. In 1868 Mr Lawson Tair excavated at Kintradwell, Sutherland,* a similar cist 5 feet 3 inches long: the skull was at the west end and the skeleton was extended. The skull was removed and sent to me for examination ; its dimensions are given in Table VIII: the cephalic index was dolichocephalic. No grave goods were found in the cist. Three short cists were found in proximity to the long one. The late Mr Wm. Gattoway informed me that in 1896 he had seen, some miles south of Inveraray, a group of long stone coffins which had been opened and partially destroyed. In the course of the excavation of the Roman Camp at Inchtuthill, Perthshire, the Hon. Jonn ABERCROMBY exposed in a mound of clayey loam a cist 7 feet 6 inches long which contained an extended skeleton. He also described | three long cists with skeletons on the south side of the Gladhouse reservoir. Their floors were not paved. Subsequently a group of twenty-four graves was described by him and Mr MacTrmr PrirRiE at Nunraw, East Lothian §: they had the characteristic construction of long cists, and contained skeletons. In 1906 Mr J. W. Loney recorded, on a small island in the North Esk reservoir, a group of six graves from 5 to 6 feet lone arranged in rows. They contained human remains, but no grave goods. The graves were approximately oriented. In 1901 the Rev. J. Primrose noted the find of over twenty lone stone graves at Uphall, West Lothian.|| They were oriented, arranged in at least two rows, con- structed of flagstone of the customary size and form, but not paved on the floor. In 1905 Mr Axexanpder Hurtcueson described@] two full-leneth stone coffins found on Auchterhouse Hill, Dundee, one of which contained a human skeleton. In the same year Dr Ricuarpson recorded** eight long coffins lying east and west in Stenton parish, Hast Lothian, one of which contained a skeleton but no other relics. In 1909 Mr Wm. Rem noted at Leuchars, Fife,+{ a cemetery of thirty-four long cists, with two skeletons in each cist ; no other relics. At Broughty Ferry }} Mr Hurcurson found two long cists, one of which had only a single slab on each side, which he regarded as the only specimen up to that time recorded in a long cist. : The examples now specified show that about 260 interments in this fort of stone-built cist have been recorded in Scotland. In the Lothians, Roxburgh, Selkirk, Fife, Forfarshire, Argyll, Perth and Sutherland one or more graves have been exposed in the same locality. Klsewhere in the Lothians and in Fife the graves were at times so numerous as to give the spot the aspect of a cemetery. Thus at Kirkliston 51 were counted, at Leuchars 34, at Old Haacks, Fife, 30, at Cramond and Nunraw each 24, at Uphall 20, Dunbar 12, Yarrow 9, Stenton 8, Arniston 7, North * Proc. Soc. Antig. Scot., vol. vii, p. 515, 1870. + Idem, vol. xxxvi, 1902. + Idem, vol. xxxviii, p. 96, 1904. § Idem, vol. xl, p. 60, and p. 328, 1906. || Idem, vol. xxxv, 1901, “I Idem, vol. xxxix, p. 3938, 1905, ** Idem, p. 441. tt Idem, vol. xliii, p. 170, 1909. ti Idem, p. 317, 1909. THE CRANIOLOGY OF THE PEOPLE OF SCOTLAND. 231 Esk 6. In their arrangement in a cemetery they were in rows parallel to each other, and the distance between the graves in each row and between the rows.was such as - to ensure economy in the use of ground, as in a modern cemetery. Except in such _ modifications in size as were needed for the burial of an adult or a child, no material ‘difference in their appearance, as might indicate relative rank or wealth, was observed. It should also be noted that whilst several have been exposed at and near the seashore, others again have been found inland a number of miles from the coast, which disposes of suggestions at one time made, that they were interments of ship- wrecked sailors. Their orderly arrangement in cemeteries and the special graves for chi dren showed that they did not mark an ancient battlefield, but were the burial- places of the people of towns and villages in the district, and had apparently been use for generations. Their uniformity and simplicity of design, the extended y and the absence of grave goods and cremation indicated a community of thought amongst the builders, which gave to them a character of their own, distinct from megalithic graves of the stone age and the short cists and cinerary urns of the onze-using people. The question therefore naturally arises, were they constructed 7a race distinct from and later in time than the people of the stone and bronze s and the pagan Celts?* In regard to this question a careful perusal of the orded examples shows that fragments of rusted iron have been found in a few ves. In one of the long graves at Cramond an iron key was found near the Pat Stolon: Regenerative Capacity of the Stolon : . 259 | Karly Developmental Stages of Dicoryne conybearet . 278 Hydrocaulus and Hydranth . : ; : . 260 | Distinctive Characters of the Species of Dicoryne . 280 Blastostyles . 3 : 5 5 : . 261 | Specific Characters of Dicor yne conybearet (Allman), Development of the Senate : : : . 263 emend. . : ; : : . 282 The Sporosacs of Dicoryne conferta . ‘ 266 | Dicoryne, Allman, char. prend: : : ; . 282 _ Affinities of the Free-swimming Srecuutes of List of Works Quoted . ; ; : . . 283 Dicoryne conybearet . 3 F : : . 269 ! Description of Figures. . : ; : : . 284 INTRODUCTION. Hitherto there has been described only a single species—Dicoryne conferta (Alder)—amongst the Hydrozoa in which the reproductive body is a sporosac which becomes_ciliated and free-swimming. We have found that another species which, ~as we shall show (pp. 269-271), must be named Dicoryne conybearei gives rise to reproductive * bodies of this type, which, however, differ from those of Dicoryne conferta in several important respects. The free-swimming type of sporosac has not been the subject of close observation since 1872; we have therefore made as complete a study of the structure and development of the sporosacs, especially of D. conybearet, as material gathered from several sources would permit. The material of D. conybearei at our disposal consists of colonies on seven gastropod shells,j dredged off the island of Nisida, in the Bay of Naples, in April * The cost of the blocks for text-figs. 2 and 3 and of reproduction of the plates has been defrayed by a grant from the Carnegie Trust for the Universities of Scotland, and we are further indebted to the Trust for a grant for artistic assistance in connection with Plate VI. + The specimens were brought to the Naples Zoological Station, and handed, along with other hydroid material, to Mr T. J. AnpmRson, who was at that time studying the Hydrozoa while holding a Carnegie Scholarship. I was working in the Naples Station at the same time, and, on seeing these colonies and their sporosacs, realised, as did Mr AnpERson, that they presented several new and interesting features. Mr Anpurson left Naples shortly afterwards, and, before he had been able to work further at the material, went to Nairobi to take up the appointment he now holds. On his departure he kindly placed the specimens in my hands. Unfortunately, Mr Anprrson did not make any notes on the colours of the colonies, or on the movements of the sporosacs, so that on these points I have merely impressions derived from glimpses of the specimens in the brief intervals of my own work on other subjects.—[J. H. A.] TRANS. ROY. SOC. EDIN., VOL. LI, PART I (NO. 6). 37 258 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE 1906, and of two slides containing a considerable number of sporosacs of both sexes, from the same source, mounted in formalin. The eruption of Vesuvius of that year had just concluded, and the showers of fine dust had produced great mortality among the marine organisms. To this cause we attribute the fact that, of the seven shells, one bore a dead colony, and five others bore colonies from which the ordinary polyps had almost wholly disappeared. In these five colonies, however, living substance in the form of ccenosare in the stolon was present, and a number of blastostyles—though on two of the colonies there were very few—had managed to survive. The remaining shell bore a colony in good condition, and it is on this that most of our observations have been made. GENERAL DESCRIPTION OF THE COLONIES OF DICORYNE CONYBEAREI.* ; (See Pl. VL.) The colonies are in each case on small, gastropod shells of the genus Nassa, and those which are well established spread over the entire shell, showing no preference for any particular portion, though, as a rule, the individuals in the neighbourhood of the mouth of the shell are more luxuriantly developed. In young examples the stolon can be easily seen forming an irregular meshwork on the surface of the shell, but in older examples the interspaces between the strands of the stolon become filled with diatoms and debris so that a uniform rusty-grey covering is formed, obscuring the stolon and obliterating the finer. marking of the shell. From the stolon, at irregular intervals, arises either a short stalk terminated by a single hydranth, or a longer branched stem bearing two, three, or four. (rarely more) hydranths. The stems are rusty grey in colour, due, as in the case of the stolons, to a covering of foreign bodies (in this case chiefly volcanic dust) adherent to the chitin. The perisare of the stem is irregularly corrugated or wrinkled, and at its distal end widens slightly, forming a small, transversely wrinkled cup, which bears a dense coating of foreign particles. This elementary hydrotheca covers, in the full-grown polyps, only a very small proximal portion of the hydranth body. The hydranths are ovoid or fusiform according to the state of expansion. Hach has a prominent conical hypostome surrounded by a whorl of filiform tentacles. The number of tentacles varies from six in the youngest specimens observed to sixteen in the largest adults. ; The sporosacs are borne on blastostyles which arise either directly from the - stolon or from the stems of polyps. The blastostyle varies in form according to its phase of development. It is at first almost cylindrical or slightly dilated at its distal end, which is armed with numerous nematocysts, but soon becomes swollen about the middle of its length and is henceforward more or less vase-shaped. * For the diagnostic characters of this species see p. 282. FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 259 The fully developed blastostyle has a narrow base of attachment to the stolon or to a hydranth stem, a wide middle region on which the sporosacs are borne, and a narrower, dome-shaped distal region bearing plentiful nematocysts. The blastostyle, which has neither mouth nor tentacles, is enveloped at its base by a simple hydrotheca similar to that of a hydranth. A full-grown blastostyle bears usually from eighteen to twenty-four sporosacs, but occasionally more are present; the largest number observed was thirty-four. All the ‘sporosacs on a blastostyle are of the same sex, and throughout any one colony the sex _ of the blastostyles is uniform. The free-swimming sporosac is provided with a single tentacle * and is ciliated all over. The female sporosac is sub-spherical and bears in every case only a single oocyte, at one side of which a well-developed spadix is present. The male sporosac is usually rather smaller, and more oval in form, and the mass of spermatozoa is arranged round an axial spadix. Further details of the various portions of the colony may now be given. Sroron (Pl. VI, fig. 1; Pl. VIII, figs. 14, 15). The external surface of the stolon or hydrorhiza is moderately smooth, 7.e. without corrugations, and there are no internal chitinous projections of the perisare such as are found supporting the walls in several gymnoblastie and calyptoblastic hydroids, e.g. Podocoryne anechinata, Ritchie; Sertularia heterodonta, Ritchie; Plumularia — lagenifera, var. septifera, Torrey. Branches are produced from the stolon strands - (fig. 1), and for a time their ends are free and rounded, but later they anastomose with neighbouring strands of the stolon, and thus the meshwork becomes complex. The perisare of the stolon is about 6p thick and of a brown colour. The histology _ of the ccenosare does not present any unusual feature. Regenerative Capacity of the Stolon.—A point of considerable biological interest may be noted here. It has already been remarked that on all the colonised shells, _ except one, almost all the larger hydranths are disintegrated and their stems empty of ccenosare. One colony was evidently dead when preserved, but even in the least promising of the others a few very small hydranths, with only six tentacles (fig. 1, H 1), ean be seen growing out directly from the stolon, thus giving evidence that, in spite of the destruction of the more visible portions of the colony, a residue of living cells _ still remains within the stolon ready to start into being a new series of individuals. Probably these young hydranths had developed during the few days the colonies had ~ been living under favourable conditions in an aquarium in the Zoological Station. Microscopic examination of the stolons from these colonies shows that although many parts are empty, other portions contain a plentiful amount of normal ccenosare. _ This living ccenosarc had been protected and shut off from the influence of deleterious * One abnormal sporosac with a bifid tentacle and two others each with two tentacles have been observed _ (see p. 277). 260 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE conditions by the formation of thin transverse partitions of chitin barring the lumen of the stolon at certain points. In several cases barrier after barrier had been successively formed by the ccenosare, which had been forced to retreat stage by stage within the stolon (Pl. VII, fig. 14). Such chitinoid barriers are formed by a plug of coenosare (fig. 15), composed of a circular plate of thick ectoderm and of solid endoderm, similar in general form and structure to the “ Deckenplatte,” which secretes the operculum in many species of calyptoblastic hydroids. It is clear, therefore, that, in addition to serving as a hold-fast for the colony, the stolon performs an important function during time of stress in retaining and protecting from harmful environmental influences sufficient ccenosarc to give rise, when con- ditions have again become favourable for normal existence, to new series of nutritive and reproductive individuals. HyprocauLus anp Hypranti (Pl. VI). The smallest hydranths observed on the stolon are sessile, are about ‘2-3 mm. in length, and have already developed six slender tentacles (fig. 1). The stem of the polyp gradually elongates, and not long after the polyp has attained a length of about a millimetre a branch usually appears, others being formed later. The largest living stems observed are 2°8 mm. long and bear four individuals. There are, how- ever, on two of the colonies dead stems 4-5 mm. long, which bore seven to eight individuals. The branches arise at intervals along the stem and lie at acute angles to the main stem, to which they are more or less parallel. It is rare to find the branches subdivided; this occurs only ina very few of the basal branches of the largest stems. Such branches bore two, and in one case three, hydranths, one of which was small—a blastostyle-bud or a very young hydranth. Examination of hydranths of various ages gives the impression that there are longer and shorter tentacles alternating, and that the longer ones in life pointed distally while the shorter ones were directed more or less horizontally. It is, however, difficult to make certain of this in preserved specimens in which the tentacles exhibit various phases of con- traction. When well extended the longer tentacles slightly exceed 5 mm. in length. Towards the base of the stem or hydrocaulus the perisarc, which as usual is laminate, is 3-4 thick, and of a brownish-yellow colour. Here the covering of extraneous debris is often meagre, in strong contrast to the dense coating present at the distal end of the stem at the base of the hydranth body. In its distal portion the perisare is less definite in outline, its laminzee more feebly marked, and its colour gradually pales until the portion surrounding the base of the hydranth is represented by a thin hyaline substance the outline of which is merely indicated by the presence of adherent foreign particles. The hydrocaulus and hydranth have been examined in several series of longi- tudinal sections. As the two cell-layers do not appear to present special features, we do not describe their histology in detail. The ectoderm of the hydrocaulus lies in FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 261 close contact with the perisare, especially in the distal region. At the level of the distal edge of the hydrotheca the cells of the ectoderm are generally more columnar, elsewhere they are less regular. Among the ectoderm cells of the stem may be noticed a few large nematocysts. The bases of the ectoderm cells are prolonged into very thin contractile processes which run longitudinally. The ectoderm is separated from the endoderm by a well-defined lamella of mesogloea. The endoderm of the stem is a layer of flattened epithelial cells with granular protoplasm and large vesicular nuclei in most of which a large nucleolus is present. Here and there are large oval cells crowded with deeply staining spherules, probably of excretory nature. In the greater part of the hydranth the mesoglceal lamella is more strongly developed than in the stem, but diminishes in thickness towards the mouth and also on entering the tentacles. The ectoderm at the apex of the hypostome contains numerous oval nematocysts. The largest of these are about 6m long and 3p broad, but many others are only about two-thirds this size. The endoderm, which in the stem is thin, becomes in the hydranth a very thick layer. Among the ordinary elongate, vacuolated endoderm cells are many clavate cells packed with spherules of secretion. The wide lumen of the hydranth narrows rapidly proximally, 7.e. on approaching the stem, to a canal about 25-30» in width which is continuous through the similar lumen in the stolon with the cavities of the neighbouring individuals of the colony. The tentacles are solid, the endoderm of each being composed of a single column of discoidal cells with stout walls. Hach cell is much vacuolated, the protoplasm being aggregated into a sinall mass around the central nucleus, from which a few strands radiate. The nematocysts of the tentacles are small and oval, about 4 long and 2 » broad. BuastostyLes (Pl. VI). The blastostyles arise either directly from the stolon or from the hydrocaulus, but most of them are borne on the stolon. Those arising from polyps (fig. 1, B v) are more recently formed than the majority of those on the stolon; possibly they are formed in seasonal succession, first on the stolon and later on the hydrocaulus. A blastostyle arising from a hydrocaulus is at first difficult to distinguish from a polyp-bud of the same size, but the latter soon develops tentacles and a mouth, structures which are not present in blastostyles. The basal portion of each blasto- style is protected by an envelope of perisare the distal region of which is sometimes moderately thick, widened into a shallow cup-shape, and easily seen; but more often the perisare is thin and almost transparent. Invariably, however, the distal margin of the perisare is marked by a coating of extraneous matter. The blastostyles vary in form according to their phases of development, and there are also considerable individual variations. Young blastostyles are, for the most part, nearly cylindrical or slightly dilated at their distal ends, where numerous 262 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE nematocysts are present (fig. 1, B1). In some cases, however, the blastostyle is more clavate, and in others is even more dilated, being almost ovate distally owing to the great development of the ectoderm of the region bearing the nematocysts (fig: 1, B m1). The nematocysts are of two kinds: those at the tip of the blastostyle are large and elongate oval in form (PI. VII, fig. 7), their length being about 15-16 and their breadth 4-5; the nematocysts on the sides of. the distal portion of the blasto- style are smaller and of more rounded oval form, their length and breadth being about 6-7 » and 4 respectively. ; In young blastostyles the endoderm consists of high columnar cells, filled with finely granular protoplasm, so that the ccelenteron is represented by a narrow axial slit, but in mature blastostyles a large ccelenteron is present, though in the distal — portion—corresponding to the hypostome of a hydranth—the cavity is smaller. The greater part of the endoderm of such a mature blastostyle is composed of elongate, — thin-walled cells which are highly vacuolated, but the endoderm of the hypostomal region consists of smaller columnar cells similar to those of the young blastostyle, and their granular protoplasm is but little vacuolated. Among the ordinary endoderm-cells are globular cells—numerous in some blastostyles, more scanty in others—containing masses of granules, probably excretory, which stain deeply with iron-heematoxylin. The ectoderm of the column of the blastostyle is similar to that of a hydranth, and its cells are provided with muscle-processes which run longitudinally, but at the distal end of the blastostyle the ectoderm is thicker and heavily charged with nematocysts of two kinds (see above). The external surface of the ectoderm is covered by a thin cuticle. The mesogloea of the middle region of the blastostyle attains a great develope often reaching a thickness of 6p, and on the endodermal side is raised into many irregular folds. In the hypostomal area the mesoglcea is much thinner, and at the tip of the blastostyle is a film of extreme tenuity. Oocytes about 6 @ in diameter are distinguishable in the ectoderm of the blasto- styles by the time the latter have attained a length of little more than ‘2mm. They are situated in a zone around the blastostyle near the middle of its length. By the time the blastostyle is about ‘5 mm. long the oocytes become so large as to cause ectodermal elevations *—the beginnings of the sporosacs—on the. surface of the blastostyle. About this time the blastostyle generally becomes swollen in its middle region and is henceforward vase-shaped, the sporosacs being borne on this wider portion (fig. 1, Brv). The distal region, which throughout the life of the blastostyle is armed with the two varieties of nematocysts described above, remains narrower and somewhat dome-shaped. The longest blastostyles observed are about "7 mm. in length (see fig. 1, B tv). * Occasionally such elevations are present in smaller blestosnyies in one case when the blastostyle was only ‘25 mm. long (see fig. 1, B11), but this is exceptional. FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 263 DEVELOPMENT OF THE SPOROSACS. A. The Female Sporosacs (Pl. VII). The oocytes are differentiated and grow in the ectoderm of a blastostyle in the positions which the sporosacs will subsequently occupy; there is no migration of oocytes from endoderm to ectoderm, nor does there appear to be any movement of ; oocytes in the ectoderm... The differentiation and early phases of growth of an oocyte precede the formation of the sporosac which is to contain it. Oocytes of apparently very different ages are present in the reproductive zone of a single blasto- style, and there seems to be no definite rule as to the period of growth or size of oocyte governing the first appearance of the sporosac. For instance, small oocytes 7-10 in diameter have been observed within distinct ectodermal swellings, but two others, each about 18, in diameter, lie in ectoderm the surface of which is not elevated. ’ The first stage of formation of the sporosac is due directly to the growth in volume of the oocyte. As the bulk of the latter increases the overlying ectoderm is raised into a dome-shaped mound, and the ectoderm cells surrounding the oocyte become considerably elongated. At this stage (fig. 2) the sporosac-elevation is entirely ectodermal, no changes having taken place in the neighbouring endoderm and mesoglea. } Further increase in the size of the oocyte soon involves the formation of a more _ definite ovoid outgrowth (fig. 3). The ectoderm cells covering the rapidly growing oocyte multiply and form a single layer of moderately regular cells; at the same time the subjacent endoderm increases and the mesoglcea becomes arched outwards and diminishes in thickness to a mere film. The base of the sporosac thus becomes occupied by a mass of endoderm—the spadix—connected with the endoderm of the _ blastostyle by a narrow neck. The cells of the spadix are arranged radially and with considerable regularity, and there is a small axial cavity (frequently filled with granular matter as indicated in the figure) continuous with the ccelenteron of the blastostyle. The base of the oocyte applied to the endoderm lies obliquely with regard to the axis of the gonophore, and this obliquity is still further increased as _ growth proceeds, the spadix eventually being pushed to one side of the oocyte. In the older sporosac, shown in fig. 4, the spadix has a large central cavity and its cells contain brownish granules which apparently increase in number with the age of the sporosac. Some time prior to the stage figured there appears, immediately _ proximal to the sporosac proper, a constriction which is of great importance in the _ subsequent history of the sporosac. For convenience of description we shall refer to the region proximal to the constriction as the sporosac-stalk, to the distal region as the sporosac, and to the connection between the two at the zone of constriction ___as the sporosac-neck. In the sporosac-stalk the ectoderm retains its original character and resembles that of the blastostyle, being composed of relatively large cells with 264 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE correspondingly large oval nuclei. The ectoderm cells of the sporosac are smaller and of a different type; they have undergone rapid division, resulting in the formation of a thin layer of uniform squarish or oblong cells with small, spherical, deeply staining nuclei. Occasional small oval nematocysts are present in the ectoderm of the sporosac and the cells are finally ciliated. A thin continuous cuticle envelops the whole ectoderm both proximal and distal to the constriction. In the sporosac- stalk the mesogloea is of moderate thickness, but at the constriction it rapidly diminishes in amount and in the sporosac is very thin, At an early stage in the differentiation of the neck-region of the -sporosac an outgrowth—consisting of ectoderm and endoderm—appears at the base of the sporosac. This outgrowth undergoes rapid development and becomes a finger-like process—the future tentacle of the sporosac—with ectoderm like that of the sporosac and with solid endoderm in a single column of cells. The tentacle is confined between the ectoderm and cuticle of the sporosac, its tip being directed distally. With the further development of the sporosac the constriction at the neck becomes more and more pronounced. The ectoderm of the sporosac and that of the stalk no longer maintain continuity, with the result that in its final stage the neck ‘consists only of a narrow tube of mesogloea with a slender central core of vacuolated endoderm (fig. 4). During the later phases of growth of the sporosae and oocyte a narrow peripheral zone of the latter becomes differentiated; it is more finely granular than the remainder of the yolky cytoplasm and reacts differently to stain (p. 278). Several of the larger oocytes examined in sections are enveloped by a thin pellicle about 1» in thickness—the vitellme membrane. By the time the oocyte has attained a diameter of about 80-90 the tentacle of the sporosac is fully developed, the neck has been reduced to its minimum, and the sporosac is ready to break away from the blastostyle and become free-swimming. The process of release appears to be purely mechanical, and to be due to the action of currents in the surrounding medium playing upon the relatively large bulk of the sporosac and breaking its slender attachments. At any rate, no histological change could be observed in sporosac-stalks from which the sporosacs had recently escaped. The escape is evidently due simply to the rupture of the loose cuticle in the neighbourhood of the neck and then of the thin neck itself. In a few cases where the cuticle had failed to rupture, the sporosacs have remained attached to their stalks, and their respective oocytes have undergone degeneration. On the sporosac becoming free the loose cuticle which covers it is usually soon discarded. The tentacle, now released from its confined position, becomes extended so as to lie more or less in line with the long axis of the sporosac, in the position shown in fig. 5. The numerous cilia of the general ectoderm and of the tentacle come into play and the sporosac swims, at the same time rotating on its own axis, probably with the tentacle directed anteriorly, though the observations on the FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 265 living specimens in captivity were not sufficient to establish this point with certainty. The subsequent life of the sporosac is probably of short duration. The ectoderm is ruptured (fig. 6), the oocyte escapes, and maturation and fertilisation no doubt soon take place. We have, however, seen three examples in which fertilisation and ego-cleavage had occurred while the egg was still contained in the remains of the sporosac, but these are probably exceptional cases and may have been determined by the abnormal conditions of captivity. All the other developing eggs were found free from the sporosacs. Whether the eggs, when liberated, float or sink to the bottom we are unable to say. _ The blastostyles apparently produce successive series of sporosacs, for on a blasto- style which bears almost fully formed sporosacs are oocytes only about 15m in diameter in small elevations of the ectoderm, forming incipient sporosacs of a new series. B. The Male Sporosacs (Pl. VIII, figs. 8, 9, 10). The material at our disposal for the study of the male sporosacs, while not so complete as for the females, is sufficient to enable us to give an account of the main features of their development. Several blastostyles bearing male sporosacs in their early and middle stages of development were present on three of the colonies; these have been studied entire and in longitudinal serial sections, and about a score of free male sporosacs have also been examined. The male sporosacs are confined to a narrow zone about the middle of the blastostyle. The first indication of their appearance is the differentiation in the base of the ectoderm of one or more cells (usually there seem to be two or three). These are ovoid in form, their protoplasm stains more deeply than that of the neighbouring cells, and their nuclei are larger and of a more vesicular type, with peripheral granules of chromatin and a large nucleolus (fig. 8). The elevation which eventually becomes the sporosac seems to be due to the almost simultaneous activity of both cell-layers. The ordinary ectoderm cells of such an incipient sporosac multiply pari passu with the spermatogonia, and the subjacent endoderm, judging from the special character of the cells, is also in a state of considerable activity, so that the condition seen in fig. 9 soon results. The thick ectoderm of the distal end of the sporosac is composed chiefly of a mass of deeply staining cells—the spermatogonia—on the periphery of which are scattered cells which stain more faintly, these being the ordinary cells which will later form the ectodermic envelope of the sperm mass. The endoderm cells below the spermatogonia are more granular and stain rather more deeply than the cells proximal to them. This endodermal spadix is either solid or has a narrow central lumen continuous with the ccelenteron of the blastostyle. The mesogloea of the sporosac is thin and grades into that of the blastostyle. Such a sporosac, having TRANS, ROY. SOC. EDIN., VOL. LI, PART I (NO. 6). 38 266 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE an axial spadix, is radially symmetrical, a symmetry which it retains until the formation of the tentacle. We have no examples intermediate between that just described and the free sporosac (fig. 10), but no doubt the male sporosac becomes detached from the blastostyle in the same manner as the female. Like the latter, it is clothed with long cilia, as also is the tentacle, which comes to lie in line with the long axis of the sporosac ; and probably the male sporosac, like the female, swims with the tentacle directed anteriorly. In shape the male sporosac is readily distinguishable from the female, since it is elongate oval—the female being sub-spherical,—and is usually a little smaller. The average length (excluding the tentacle) and breadth of twenty male sporosacs are ‘096 mm. and ‘064 mm. respectively, the largest bemg “12 mm. long and ‘07 mm. broad. The tentacle, when extended, is ‘04—'06 mm. long. The average length and breadth of twenty female sporosacs are ‘115 mm. and ‘1 mm. respectively, the largest being “135 mm. long and ‘115 mm. broad, and the tentacle is ‘(05-07 mm. in length when extended. The free male sporosac is of simple structure (fig. 10). Externally it is covered with ectoderm composed of regular ciliated cells.. The endodermal spadix, which has a central cavity, runs axially from one pole to the other, passing at one end into the solid endoderm of the tentacle and at the other abutting against the ectoderm. The cells of the spadix of most free sporosacs exhibit a degenerate appearance. Between the ectoderm and the spadix lies the cylindrical mass of spermatozoa. We have no observations on the mode of escape of the sperms from the sporosac. THE Sporosacs oF DicoRYNE CONFERTA (ALDER). The characters of the reproductive bodies described in the preceding pages clearly place them in close relationship with the free-swimming sporosac of Dicoryne con- ferta described by AttMAN (1861 and 1872). The most marked differences lie in the single tentacle of the Neapolitan examples as contrasted with a pair in D. conferta, and in the single oocyte of the former as against the normal pair* in the latter. ALLMAN was content merely to mention the “defined line” along which the sporosace of D. conferta becomes detached, without indicating its origin or nature, and, owing to lack of mature material, the detailed descriptions of later writers have stopped short of the stage at which the tentacles develop. In the hope of correlating more definitely the developmental stages of these remarkable and closely related sporosacs, we have re-examined their development in D. conferta. The material examined consists of colonies from (a) Kristineberg, Sweden, lent by Professor Hsatmar THbet from the collections of the Stockholm Museum; (0) * Goxrrrn (1907, p. 67) has stated that in young sporosacs of D: conferta there are occasionally three or four oocytes, and sometimes only one ; and HartTLAvus (1897, p. 480) has recorded the finding of free-swimming sporosacs of this species, in plankton collections off Heligoland, which possessed three oocytes and the normal two tentacles. FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 267 Blacksod Bay, Plytnouth, lent by Dr E. J. Aten, F.R.S., from the collections in the Laboratory of the Marine Biological Association, Plymouth; and (c) North Sea, lat. 57° 43’ N., long. 0° 38’ E., in a collection received for identification by one of us from the North Sea Investigation Committee of the Scottish Fishery Board. It is a matter for regret that the specimens from each of these localities bear male sporosacs only. As in the Neapolitan examples these are clustered upon blasto- styles borne upon the hydrorhiza or upon the hydrocaulus, but in D. conferta they are relatively more frequent on the latter. The spermatogonia are first to be distinguished in the ectoderm of the blastostyle on account of the deeper stain absorbed by their nuclei. They do not migrate: their situation already indicates the position of the future sporosacs. In all the blasto- styles of D. conferta which we have sectioned the mesoglcea is of unusual thickness, averaging 6 and reaching even 13p, and this induces a slight modification in the early development of the gonophore. For whereas in the Neapolitan examples of D. conybearei the mesoglea, even at the earliest stage, was bent outwards into the _ sporosac-mound, in D. conferta the abrupt edges of mesoglcea at the junction between blastostyle and gonophore show that a dise of the mesoglea has been almost com- pletely resorbed, the thin film remaining being then pushed outwards by the growing endoderm taking part in the formation of the sporosac. The early development of the sporosac follows the course described for the Neapolitan examples (see pp. 265, 266). There is a large endodermal spadix which either has a narrow central lumen or is solid, a thin film of mesoglcea, and a thick eap of ectoderm consisting of a mass of spermatogonia with more peripherally placed covering cells. The sporosac is invested with a cuticle considerably thicker and “usually developing at an earlier stage than that present in the Neapolitan ones, _ and foreign particles are adherent to it. | The subsequent development is marked by a constriction in the proximal portion of the sporosac. This proceeds to such an extent that finally a neck is formed containing only a narrow cylinder of endoderm invested by a thin film of mesoglcea _ which is separated by a considerable interval from the cuticle still strengthening the connection between the newly developed sporosac-stalk (see p. 263) and the sporosac proper (Pl. VIII, fig. 16). The blastostyle type of ectoderm is retained in the -sporosac-stalk, but in the sporosac a thin ectoderm of almost uniform cells, with relatively large nuclei, is formed by rapid division. With the appearance of two tentacles, the origin of which dates from the beginning of the constriction at the sporosac-neck, the sporosac assumes bilateral in place of radial symmetry. The tentacles originate as a couple of buds which arise opposite each other from the wall of the sporosac immediately distal to the narrow neck. Pushing distalwards they are retained alongside the sporosac by the cuticle. With the maturing of the reproductive products the cells of the spadix undergo regenerative changes, and in most cases the spadix exhibits a distinct central cavity. The cuticle enveloping the Fs 268 DR J. H ASHWORTH AND DR JAMES RITCHIE ON THE sporosac appears to become gradually thicker and of deeper colour, and its coat of debris also increases inamount. The length of a mature male sporosac is ‘17-19 mm., and its maximum diameter ‘1-13 mm. Although we have been unable to follow the development further and to study the actual release of the sporosacs, it is evident that, as in the Neapolitan examples, the cuticle, probably under mechanical tension, becomes ruptured in the neighbourhood of the sporosac-neck, that the neck itself, being unable to bear the whole strain put upon it, breaks, and thus the sporosac is liberated. The tentacles, on being released from confinement, come to lie more or less parallel to the long axis of the body. ALLMAN, in 1861 (p. 169), gave a description of the adult male sporosac, and this he repeated in substance in his monograph (1872, p. 226). He gives no information regarding the development of the sporosac, nor does he mention the peculiar con- striction which ultimately leads to its release; indeed his figure (1871, pl. viii, fig. 5) is defective in showing no trace of a constriction nor of the “defined line” to which he refers as forming the breaking plane proximal to the tentacle bases. It seems to us also that ALLMAN has distorted his description and obscured the issues by impressing on this unique type of sporosac a terminology derived from and depending on medusoid homologies. Thus, as applied to the male sporosac of D. conferta, AutMAN’s “internal sac” or “ endotheca” is no more than the ectoderm, a direct derivative, slightly modified, of the ectoderm of the blastostyle; and the “external sac” or “ectotheca” is simply the cuticle secreted by the ectoderm, homologous and continuous with the cuticle which covers the ectoderm of the blasto- style and of the colony in general. The assumption that the sporosac is a medusoid has led ALLMAN into a further error; for assuming the usual construction of a medu- soid to be followed in this case, he described and figured a distinct space between — the “ectotheca” and the “endotheca” (1871, pl. viii, fig. 5). So far as we have observed, the cuticle, except for such small distortion and displacement as are unavoidable in preservation and mounting, lies close upon the ectoderm, and there is no determinate space between the two as represented in AtuMman’s figure. The tentacle in its growth forces apart the cuticle from the ectoderm, but there is not an already pre-existing space. In his description of the release of the sporosac an obvious slip occurs, for the sporosac could not break through the ‘endotheca,” as — he states, but escapes from the so-called “ ectotheca.” Female sporosacs were not discovered on any of the mature colonies of D. conferta examined by us. In this respect our experience agrees with that of Gorrrs (1897, p. 67), who found that colonies with female sporosacs were entirely unisexual. ALLMAN, however, states that he “ occasionally met with both male and female stems in the same colony” (1872, p. 226). The earliest stages in the development of the female sporosacs have been described by Gorrrem (1897, pp. 67, 68, Taf. vi, figs. 122-127). They closely resemble those of our Neapolitan material, except that in place of the single oocyte of the latter the sporosac of the former includes two. In FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 269 the Neapolitan sporosacs the spadix lies to one side of the enlarged oocyte, in D. conferta the spadix is situated between the two oocytes and, notwithstanding considerable compression in the later stages, retains its original axial position through- out development. Gorrre’s observations on the development of the female sporosac cease at a stage long before the appearance of the tentacles, but ALLMAN has given a moderately full description of the free, tentacled sporosac (1861, 1872). From direct observation we can add nothing to AtLMan’s account, but the close- ness of the developmental and structural relations between the male sporosacs of D. conferta and D. conybearet, together with the similarity in the general lines of development of the earlier stages of the female sporosac in these species, suggest that the adult female gonophores of the two species will present similar structure. We may, therefore, indicate probable additions to and slight emendations in the description given by Anuman. Some of the remarks made on his description of the male gonophore (p. 268) apply also to the female, eg. in regard to the “ectotheca” and “endotheca” and their supposed homology with the layers of a medusoid gonophore. ALLMAN does not mention the formation of a constriction at which the sporosac breaks away from the sporosac stalk, although such a neck is no doubt formed in the female as in the male. Further, he states that “each ovum is invested by a proper membrane, which presents the remarkable and unique character of possessing considerable thickness and being richly set with thread-cell-like bodies” (1872, p. 227). In connection with this statement we draw attention to the fact that the oocyte of D. conybeawrer exhibits a differentiated peripheral zone of cyto- plasm and not infrequently also a thin vitellime membrane, but these are not cellular layers and they do not contain nematocysts. Probably the condition is similar in D. conferta. AFFINITIES OF THE FREE-SWIMMING Sporosacs OF DicoRYNE CONYBEAREL The free-swimming sporosacs from Naples are evidently closely related to those of Dicoryne conferta, and the former should undoubtedly be referred to the genus Dicoryne. The origin and development of the gonads and of their investment agree in the two cases almost to the smallest details, the main point of difference being that the Neapolitan sporosacs possess only one tentacle and the female bears one oocyte, while the sporosacs of D. conferta have two tentacles and the female contains as a rule a pair of oocytes. A close relationship also exists between the sporosacs from Naples and those of Heterocordyle conybearet, Allman, described by ALLMAN (1872, pp. 307, 308; pl. x, figs. 4-7) from material collected in Glengariff Harbour, Co. Cork, and by WEISMANN (1883, p. 84, Taf. xi, figs. 7, 8) from Neapolitan specimens. ALLMAN names the parts of the sporosac of H. conybeare: in accordance with his view of their medusoid homologies, but from WEISMANN’s account it appears that in every respect the 270 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE development of the male and female sporosacs, as far as it was traced, agrees with that of our Neapolitan sporosacs. The female sporosac of H. conybearei bears a single oocyte. According to the accounts of ALLMAN and WEISMANN, the develop- ment of the sporosacs in H. conybearer stops a long way short of the stage reached by our Neapolitan sporosacs, for in the former no free-swimming stage was observed and no trace of tentacle detected, so that the sporosac was considered by both authors to belong to the fixed type. Evidence we have accumulated indicates that the sporosacs of Heterocordyle conybearet examined by ALLMAN and WEISMANN were not fully mature, and that consequently these authors (whose accounts have been followed by later writers) were misled as to the final structure and essential nature of the sporosacs. The identity of the early stages of the sporosacs of H. conybeawrei with those of our material has already been pointed out, but it was desirable to ascertain definitely whether or not the later phases were identical, which could only be determined by examining a well- preserved, mature colony indubitably belonging to H. conybearer, and preferably obtained from some source other than Naples. We were unable to obtain such a colony until this paper was standing in proof, when Mr E. T. Browne kindly lent to us a colony, collected near Plymouth in 1895, and identified by him as H. conybearet *_ an identification with which we fully agree. This colony covers an area of about 9x 4mm. on a portion of a shell, and is probably younger than those from Naples examined by us, since nearly all the stems bear a single hydranth, only a few having two. The characters of the trophosome of Mr Browner’s colony are in agreement with those of our Neapolitan examples, but the largest polyps of his colony are distinctly larger than any of ours. Further, the mature blastostyles of his colony are longer than ours (his longest being about 1°3 mm. and ours ‘7 mm. in length respectively), but have approximately the same width. We do not attach any significance to these differences, as they-may both be due to the exceptional conditions under which the Neapolitan examples passed their last days in the sea t (see p. 258). The blastostyles of Mr Browne’s colony are, with one exception, confined to the stolon, and are for the most part young, only five bearing sporosacs. One blasto- style, which was removed for detailed examination, bears a dozen large sporosacs— evidently its first series—in a cluster slightly distal to its middle. More distal, and extending to the base of the hypostomal area, is a broad zone the ectoderm of which exhibits numerous small elevations (incipient sporosacs) each containing a single oocyte. Had these sporosacs become mature, the blastostyle would have been similar (except in length) to that shown in fig. 1, Biv. The large sporosacs on this blasto- style are about ‘13 mm. long and ‘1 mm. broad; each has a single tentacle and a * Recorded from Plymouth on shells of Nassa (cf. p. 258) and Buccinwm, particularly the former. See Journ, Marine Biol. Assoc., vol. vii (N.S.), p. 190, 1904. + It is possible that the Neapolitan and British specimens belong to distinet geographical sub-species or races, but this can only be determined by examination of a longer series of mature examples. wheel \ FREE-SWIMMING SPOROSACS OF THE HYDROID GENUS DICORYNE. 271 single oocyte, and exhibits a distinct neck or constriction near its base, indicating clearly that the sporosacs would soon have broken loose and become free-swimming. The sporosacs borne upon this specimen of MHeterocordyle conybearet from Plymouth have approximately the same form and size, and present exactly the same morphological features, as our Neapolitan sporosacs; and the trophosomes of the two specimens being also in agreement, there can no longer be any doubt that the specimens are specifically identical. Hence Heterocordyle conybearei, Allman, is really a Dicoryne, and should therefore be known as Dicoryne conybearei (Allman). _ To this species belong the British examples described by AtLMan, and the Neapolitan examples * described by WerIsMANN, and by ourselves in the earlier part of this paper. Tf A Discussion or THE HomoLoGy oF THE Sporosacs oF DICORYNE. Several views have been advanced as to the homology of the sporosac of Dicoryne conferta. The discovery in another species of a closely related sporosac, of which we have been able to trace the complete development, and the re-examination of the male sporosac of D. conferta, have provided new facts on which to base a reconsidera- tion of the homology of the sporosacs. The views hitherto advanced may be grouped into two classes :— _ (1) The medusoid homology. (2) The polyp-homology. To take them in the order of their appearance : 1. The Medusoid Homology. In 1861 (p. 170) AttMaN stated his opinion that the sporosac of D. conferta corre- sponded to a normal medusiform gonophore in which certain parts were much reduced. This theory he retained, and restated in somewhat ereater detail in his classical monograph (1872, p. 227). The significant points in his statement are as follows :—‘“The planoblast [=free sporosac] admits of a very instructive com- parison with an ordinary medusiform gonophore. It is, in fact, a medusa in which the place of the umbrella and its canals is taken by two tentacles, the manubrium of the medusa being represented by the rest of the planoblast. It will be recollected * The species apparently extends over the northern portion of the western Mediterranean, for Mme. Morz- KossowsKa has recorded specimens from the Mediterranean coast of France (Banyuls) and Spain (Blanés), and from the Balearic Islands (Cabrera, Mahon), in Arch. Zool. Hxpér., sér. 4, tome iii, p. 76, 1905. + Owing to the lack of mature material indubitably belonging to H. conybearei for comparison with our Neapolitan specimens, and being faced with the definite and accepted statements of ALLMAN and WrEIsMANN that the sporosacs of this Species did not become free, we considered that we were not fully justified, on the evidence available when our paper was read to the Society, in regarding our Neapolitan examples as identical with H. conybearei, and as the former clearly belonged to the genus Dicoryne we designated them as a new species—D. parthenopeia. (This name, which has appeared only in the brief report of the meeting given in Nature, vol. xcv, p. 552, now lapses.) We suggested, how- ever, that further study of the sporosacs of H. conybearei might reveal their identity with those described by us from Naples. Conclusive evidence now available has shown this to be so, and accordingly the name parthenopeia has been deleted from the proof. 272 DR J. H. ASHWORTH AND DR JAMES RITCHIE ON THE that the two tentacles are turned forward, while the sporosac is still invested by its ectotheca, and that they then hold exactly the place of an umbrella (mesotheca) between endotheca and ectotheca. They are, in fact, the radiating canals of the medusa reduced to two and developed as free tubes, instead of being immersed in the walls of an umbrella. ... The free sporosac of Dicoryne is thus a medusa reduced to the condition of a manubrium and two opposite radiating canals. In order, indeed, to convert it into an ordinary medusa, little more is necessary than to suppose the number of tentacles increased to four by the symmetrical development of two others, their extremities connected by a circular canal, and their sides by a continuous muscular membrane (umbrella) inflected at its fore margin so as to form a velum.” ALLMAN was led to this view by the interpretation he put upon the parts of the sporosac. He regards the external cuticular envelope as homologous with the “ectotheca” of an ordinary medusiform gonophore, and the ectoderm layer in which the oocytes or spermatogonia are differentiated is, in his view, the “ endotheca.” Consequently the tentacles of D. conferta, lying as they do between the supposed and “endotheca,”’ occupy exactly the position of the umbrella or and its canals, in a normal medusoid gonophore. ) ‘ectotheca’ ““ mesotheca’ y) But this view appears to us to be untenable and to have been supported by false homologies. The typical medusoid gonophore exhibits, according to ALLMAN’S terminology (1860, p. 3), an ““ ectotheca,” a “ mesotheca,” and an “ endotheca.” “The ectotheca is a simple extension of the ectoderm of that part of the zoophyte from which the gonophore arises, and it encloses either a sporosac or a medusoid” (loc. cit.). Of the other layers “the inner (endotheca) is the equivalent of the ectoderm layer of the manubrium of the medusa; the middle layer (mesotheca) corresponds ‘to the umbrella, and like it may have a system of canals more or less completely developed in it” (1871, pp. 39, 40). If these terms are to be employed in support of any homology, their application must be uniform. The “ectotheca,” for instance, must be understood in its original significance as applied by ALLMAN in the case of a medusiform gonophore, where it is represented by the external investment—an ectodermal layer which arises, along with the ectoderm of the ex-umbrella, from the ectoderm of the blastostyle. The ‘ ¢ ¢ ‘ectotheca” may secrete on its outer surface a delicate cuticular film, as, for instance, in Bougainvillea. This ectoderm is totally distinct in origin from that covering the manubrium, for the latter arises from the thick mass of ectoderm cells—the ‘“ Glockenkern”—at the distal end of the medusa, in which, by the development of the future umbrella cavity, the manubrium becomes differentiated from the ectoderm of the sub-umbrella. In a fully differentiated medusa (‘‘ phanerocodonic gonophore ”) there are four layers of ectoderm from without inwards, namely, the “ ectotheca,” the ectoderm of the ex-umbrella, that of the sub-umbrella, and that covering the manubrium. a e. * @6 » e ins ° 6 6 «6 e o, a ¢ 2 @ & @:", oe o @ Pad ° 04 Po °*@ * *s* @ eiawiy 3 e e Text-FiG. 4,—Sinus terminalis (s.¢,) of a 21-days embryo and trophoblastic disc (¢.d.) seen from within. b.v., blood-vessels on growing margin of unsplit mesoderm beyond sinus terminalis. x 90. of endoderm which forms the inner boundary of a space occupied by the unsplit — stalk, like the cells lining the stalk, are columnar in form (fig. 50); but within a short distance of the yolk-stalk, right up to the sinus terminalis, the yolk-sac endoderm * In the sheep and pig the mesoderm is soon completely split into somatic and splanchnic layers. The resu t of this splitting is the formation of a free yolk-sac vesicle (text-fig. 13). The yolk-sac is nearly, but never quite 7 a free vesicle in the horse, 7 STUDIES ON THE DEVELOPMENT OF THE HORSE. 303 are connected by protoplasmic bands with the vessels of the unsplit mesoderm (fig. 28). Except within the trophoblastic discs, the endoderm beyond the margin of the unsplit mesoderm also consists of a single layer of polygonal cells (fig. 30), but opposite the discs the endoderm cells proliferate to form tubercles which project into the cavity of the yolk-sac (figs. 29 and 82). The first indication of a trophoblastic disc is a local thickening of the coagulum lying between the trophoblast and endoderm, presumably the result of the increased activity of a group of trophoblastic cells (fig. 30). The first indication of a yolk- sac tubercle is an ingrowth from the thickened coagulum which causes the endoderm to bulge into the cavity of the yolk-sac (figs. 31 and 31a). The endoderm cells in contact with the globular ingrowth from the coagulum stain more deeply and are decidedly more granular than the surrounding cells (fig. 314). It may hence be inferred that the coagulum is attacked by the endoderm cells in its immediate vicinity. As the coagulum expands and projects further into the cavity of the yolk-sac it acquires an almost complete investment of endoderm cells with relatively large nuclei. A section through the centre of a growing tubercle (still connected _ by a short stalk with the subtrophoblastic layer of coagulum) is represented in fig. 32, while fig. 324 represents a section through the edge of the same tubercle. If, as is usually the case, the coagulum forming the kernel of the tubercle reaches a considerable size, the capsule of endodermic cells ruptures, with the result that the coagulum projects freely into the cavity of the yolk-sac (fig. 29). Just as it is im- possible to say to what extent the uterine milk is modified as it passes through the trophoblastic discs to form the coagulum, it is impossible to say how the contents of the tubercles are modified by the endodermic cells forming their capsules. 5. The Mesoderm. At the middle of the third week the mesoderm is in the act of splitting in the region of the mesodermic somites; at the end of the third week, as already mentioned, the splitting of the mesoderm into somatic and splanchnic layers has extended some _ distance beyond the embryo (fig. 34 and text-fig. 12). The inner portion of the space resulting from the splitting becomes the ccelom, the outer forms the exoccelom (text- fig. 12). As fig. 34 and text-fig. 12 indicate, the greater part of the mesoderm at the end of the third week is still unsplit, and lies between the endoderm and trophoblast. This unsplit mesoderm is in contact neither with the trophoblast nor the endoderm ; it occupies a. space between these layers, and supports the vitelline vessels carrying blood to and from the embryo. The blood comes direct from the two aorte by the two _ yitelline arteries (text-fig. 7). The left artery (/.v.), very much larger than the right (figs. 27 and 35), eventually bifurcates and encircles the blastocyst as the sinus terminalis (fig. 34); the left artery, before bifurcating to form the sinus, gives off numerous branches, some of which anastomose with branches from the small right TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 7). 44 304 PROFESSOR J. COSSAR EWART. artery. These branches, together with many others from the sinus terminalis, form an arterial network (fig. 34). From the network small veins proceed which unite ultimately to form the large vitelline veins, by means of which the blood from the yolk-sac is poured into the sinus venosus. The large left and small right vitelline — arteries and the large vitelline veins are seen in fig. 35. The general scheme of the yolk-sae circulation is given in fig. 34, while fig. 22 and text-fig. 5 represent on a | larger scale the network formed by branches proceeding from the sinus terminalis. Trxt-Fic. 5.—Portion of vitelline network near sinus terminalis. x 90. In the case of the rabbit and other mammals in which the lower polar region | of 4 alis appears. In the horse at the end of the third week, though a sinus has been ~ established, it does not, as in the rabbit, mark the limit of vascularisation, for, as fig. 2: 2- and text-fig. 4 indicate, blood-vessels or blood-islands extend some distance beyond © the sinus. But by the end of the fourth week there are neither blood-vessels nor blood-islands beyond the sinus terminalis ; hence the yolk-sae circulation in the hor; has even a different history from that of the pig, in which a temporary sinus i followed by general vascularisation. 5 . STUDIES ON THE DEVELOPMENT OF THE HORSE. 305 6. The Amnion. In Martrn’s embryo the amnion was only represented by indistinct lateral ridges (text-fig. 1, am.) ; in my 21-days embryo the amnion was complete (figs. 33, 34, and text-fig. 12). In the sheep and pig the amnion, well advanced on the fifteenth day, may-be complete on the sixteenth day, z.e. at the stage characterised by from 8 to 10 mesodermic somites. As there are neither head nor tail folds in the horse on the seventeenth day, the amnion is probably only completed on the twenty-first day, ze. at the stage characterised by about twenty somites. It may hence be assumed that the amnion appears relatively later in the horse than in the sheep and pig—the early development of the amnion in even-toed Ungulates is doubtless correlated with the early disappearance of the zona pellucida. Whether in the Hquide the amnion is _ mainly derived from a head fold, as in the chick, or from a tail fold, as in the rabbit, will doubtless be ascertained when a more complete series of embryos is available. But seeing that the head is sharply bent backwards (fig. 33) at the end of the third week, the probability is that in the horse, as in the pig, the head fold contributes most. In a longitudinal section of the 21-days embryo the amnion (in contact at its origin with the allantoic diverticulum) is seen to arch upwards and forwards over the external opening of the spinal cord (fig. 33). Ina transverse section on a level _ with the minute diverticulum (fig. 33, cl.) which seems to represent the cloacal chamber, the crescent-shaped amniotic cavity is seen to lie immediately above the lateral extensions of the allantoic diverticulum (fig. 50, a/.d.). In the semi-diagrammatic drawing of the blastocyst (fig. 33) the amnion is represented as having a large cavity. As a matter of fact, there is very little amniotic fluid at the end of the third week, with the result that in transverse sections the amnion is seen to be only separated by a narrow space from the embryo (fig. 8). The structure and relations of the amnion and the extent of its cavity will be gathered from figs. 36 to 53. It will be observed that in the horse, as in the sheep and pig, the amnion is directly continuous with the somatopleure, and that in some of the sections the mesodermic layer of the amnion is thrown into more or less distinct longitudinal ridges (fig. 45). It will be further observed that owing to the head projecting into the amnion as into a cowl or sae, the facial portion of the embryo is completely enveloped by the amnion (figs. 36 and 37). In front of the vitelline veins (figs. 39 and 40) the amnion again assumes the form “of a sac or hood, with the result that in the most anterior part of the flexed embryo the dorsal portion of the amniotic cavity is no longer separated from the ventral portion. It will be noticed that many of the sections figured suggest that the one and only object of the somatopleure is to provide a water-jacket over the back of the embryo.* * With a water-jacket (the amnion) above and a water-bed (the yolk-sac) underneath, the embryo horse is as well protected from jars and pressure as a chick all but completely surrounded by amniotic fluid, 306 PROFESSOR J. COSSAR EWART. 7. The Allantois. In the rabbit the allantois is represented on the eighth day by a mass of mesoderm cells at the posterior end of the embryo. On the ninth day a diverticulum from the hind-gut extends into this cell mass and soon expands to form the allantoic vesicle. In the pig and sheep a similar diverticulum (which extends into a mass of mesoderm on or about the fifteenth day) gives rise at a comparatively early stage to a long compressed allantoic vesicle (text-fig. 15). If the allantois appeared relatively as soon in the horse as in the sheep, one would expect to find allantoic mesoderm investing the posterior end of Martry’s embryo (fig. 7), but, as already stated, there is no indication of an allantoic diverticulum at the middle of the third week. | In the 21-days embryo the caudal end was surrounded by a large fin-like expansion (fig. 11), mainly composed of mesoderm, into which extended a diverticulum from — the hind end of the intestine (fig. 33). From the series of transverse sections of the 21-days horse embryo it was impossible to make out the exact form and relations of either the allantoic diverticulum, the heart, or the pharynx. These difficulties, together with the difficulty of accounting for the difference between Bonnet’s and Mart1n’s so-called 21-days embryos and the still greater difference between Martin's embryo and my 21-days embryo, ultimately led to a deadlock. Repeated attempts to obtain 14-days horse embryos failed; nevertheless, progress eventually again became possible. With the help of information gained from breeding experiments I was in course of time able to account for the differences between my 21-days embryo and the so-called 21-days embryos of Bonnet and Marrin, and at an opportune moment my colleague Professor Ropinson was good enough to offer to have a model of my 21-days embryo constructed by the wax-plate method. This model was in due time completed in the Anatomical Department of the University of Edinburgh by Mr A. Grsson, M.B.* In the description of the model of the 21-days embryo,t Professor Rosrnson, in the section dealing with the foetal membranes, points out that “the allantoic diverticulum is a narrow-necked sac (fig. 33) so flattened dorso-ventrally that its cavity is reduced to the dimensions of a narrow cleft” (fig. 52). He further states (a) that the diverticulum expands as it passes caudally into the somatic mesoderm beneath the caudal part of the amnion fold, until] it attains a width of 235m; (b) that “the endo- dermal diverticulum is surrounded laterally, ventrally, and caudally with mesoderm (fig. 33), but dorsally it lies in relation with the ectoderm in the region of the caudal end of the amnion”; and (ce) that “‘ the allantoic blood-vessels consist of a number of dilated capillaries which form a coarse network on each side. Hach lateral network * Mr Gipson, before leaving Edinburgh to occupy the Chair of Anatomy in the University of Winnipeg, was good enough to place at my disposal notes and drawings of the model he made under the supervision of — Professor Roprnson. + “Description of a Reconstruction Model of a Horse Embryo Twenty-One Days Old,” Trans. Roy. Soc, Hdin. vol. li, by ARTHUR Rosrnson, M.D., Professor of Anatomy, University of Edinburgh. “0 STUDIES ON THE DEVELOPMENT OF THE HORSE. 307 receives two branches from the caudal end of the dorsal aorta of the same side, and it terminates at the caudal end of the allantoic mass in a terminal transverse sinus from which the umbilical veins take their origin” (text-fig. 7). That. the endodermic diverticulum from the hind-gut (fig. 33, al.d.) is allantoic and not simply a portion of the cloacal chamber is suggested by its narrow neck, its relation to the vascular and capillary plexus of blood-vessels in the allantoic meso- derm, and by its agreement with the diverticulum recognised as the rudiment of the allantois in the sheep and pig. In the chick the allantois, a small diverticulum at the close of the second day, is flask-shaped and half the length of the embryo on the fifth day, while on the ninth day it almost completely surrounds both embryo and yolk-sac. In the sheep and pig (in which the foetal appendages appear relatively sooner than in the horse) the allantois has the form of a small two-horned diverticulum on the sixteenth day and of a large double-horned sac on the eighteenth day (text-fig. 15). Nevertheless, the time of appearance of the allantois in the horse may be said to closely coincide with that in the chick, for, though small at the stage characterised by about twenty-two somites, it surrounds two-thirds of the embryo on the twenty-eighth day and nearly invests the entire yolk-sac on the sixty-third day, 7.e. at the stage in the horse which may be said to correspond with the ninth day in the chick. C. Tue Empryo at THE END ofr THE THIRD WEEK. 1. Size and External Conformation. Martrn’s embryo (fig. 7), which represents the phase reached in the horse at or about the middle of the third week, had a total length of 3°25 mm., and the greatest breadth was 1°3 mm. The 21-days embryo, instead of lying, like Marrin’s, at Tight angles to the long axis of the blastocyst (fig. 4), occupied a position nearly parallel to its long axis (fig. 5). Measured round the curve this embryo had a length, when fresh, of 11 mm., 7.e. it was more than three times the length of Martin’s embryo. The length of the dorsal portion of the trunk seen in fig. 10 was 6°25 mm. The length of the head and the adjacent part of the trunk seen in fig. 11 was 3°25 mm., and the greatest breadth 2 mm. If Marrin’s embryo represents the stage reached in the sheep at the fifteenth day of gestation, the 21-days horse embryo probably represents the stage reached in the sheep at the end of the eighteenth day. In the 21-days embryo horse one especially notices :—(1) the curving downwards and backwards of the head and front portion of the trunk round the inner ends of the vitelline veins (fig. 8); (2) the large pericardial sac lying in the space between the branchial region and the vitelline veins (fig. 8) ; (3) rudiments of the external gill slits and of the gill pouches, between the branchial arches (figs. 8 and 9); (4) the openings into the otic sacs (fig. 11); (5) the absence of a maxillary process from the mandibular arch; (6) the absence of nasal sacs and of 308 PROFESSOR J. COSSAR EWART. limb rudiments ; (7) the presence of a broad fin-like expansion at the caudal end of the embryo (fig. 11); (8) the external opening from the posterior end of the spinal cord (fig. 10); (9) the broad yolk-stalk connecting the embryo with the yolk-sac (fig. 33); and (10) the mouth (fig. 33) between the fronto-nasal process and the first pair of branchial arches (fig. 9). 2. The Nervous System and Sense Organs. At the middle of the third week the nervous system is represented by a medullary groove, shallow at its origin immediately in front of the primitive groove, but deep in the region of the mesodermic somites (fig. 7). This groove communicates by means of a neurenteric canal with the rudiment of the hind-gut. Before the end of the third week the medullary groove is converted into the medullary or neural canal. Though at the end of the third week the neural canal still opens to the exterior at the caudal end (fig. 33), it is closed in front and the cerebral vesicles are already modified to form rudiments of the fore-, mid-, and hind-brain. The fore-brain, separated by a shallow groove externally and a faint ridge internally from the mid-brain, occupies the greater part of the obliquely directed fronto-nasal process (fig. 33). The transverse sections through this process show amongst other things the original cavity of the anterior cerebral vesicle and the primary optic vesicles (fig. 36). The mid-brain, small and indistinctly separated from the hind-brain, has a relatively large oval cavity. In fig. 37 the large cavity of the mid-brain is seen also, the notochord lying between the mid-brain and the diverticulum (Seessel’s pouch), which for a time projects from the roof of the fore-gut immediately within the bucco- pharyngeal membrane. The hind-brain (figs. 38 to 40), long and tubular, is neither distinctly ope from the mid-brain nor the spinal cord. After curving backwards and downwards the hind-brain forms a dorsal curve which probably marks the beginning of the spinal cord. The chief interest of the hind-brain at the end of the third week is its relation to the otic vesicles. These vesicles (fig. 38) lie close to the hind- brain, nearly on a level with the second branchial pouch.* The spinal cord, in the form of an ectodermic tube with a distinct lumen, extends from the hind-brain to the caudal end of the embryo to end in a mass of cells with — which the notochord and the cloacal diverticulum are intimately related. The rela- tion of the spinal cord to the notochord and hind-gut, and the opening of its canal to the exterior, are indicated in fig. 33, while in fig. 52 a section of the cord immedi- ately in front of the external opening is represented. From the figures of transverse sections of the embryo it will be observed that the spinal cord assumes a triangular form in the caudal region (figs. 50 to 52), but a more or less oval form throughout the greater part of the trunk (figs. 41 to 49). * Owing to the slight obliquity of the section, only the external opening of one of the vesicles is seen in the figure. STUDIES ON THE DEVELOPMENT OF THE HORSE. 309 -~ - 3. The Alimentary Canal. At the end of the third week the alimentary canal is closed in front and behind, - but the middle portion is in free communication with the yolk-sac (fig. 33). An ectodermic involution in front of the fore-gut represents the mouth, while an endo- _ dermic outgrowth from the hind-gut represents the allantois (fig. 33). The mouth (stomodzeum) is an irregular deep pit between the naso-frontal process and the first (mandibular) pair of branchial arches (fig. 8). The floor of the pit which separates the embryonic mouth from the pharynx consists of only two layers (fig. 33), a layer of Trext-ric. 6.—Diagram of branchial pouches and branchial arches, I.).p. to 1V.b.p., the four branchial pouches ; I.b.a. to 1V.6.a., the four branchial arches ; I.b.g. and III.b.g., the first and third branchial grooves ; I.a.a., first aortic arch ; II.a.a., second aortic arch ; mo., mouth pit ; Z., possible lung rudiment ; ¢., tuberculum impar. ectoderm and a layer of endoderm, the latter intimately related to the notochord and the mandibular arches. Immediately behind this partition (the bucco-pharyngeal membrane) between the mouth and the pharynx a diverticulum projects upwards towards the notochord (fig. 37), which Professor Rogrnson regards as Seessel’s pouch. =) The Fore-gut.— At the end of the third week the pharyngeal part of the fore-gut is especially interesting because of the presence of branchial or gill pouches and their _ related branchial arches. In the chick there are during development four pairs of pouches and five pairs of arches, and for a time each of the three foremost pouches _ opens to the exterior and thus gives rise to imperfect gill clefts. In man at the end _ of the third week there are four pairs of branchial arches and four pairs of branchial 310 PROFESSOR J. COSSAR EWART. pouches, but only the first two pairs of pouches and the first two pairs of arches (the mandibular and hyoid) are well developed, and even the large first pouch has never — been found opening to the exterior. . In the 21-days horse the branchial pouches and arches generally agree with the corresponding structures in man. The first pouch (figs. 9 and 33 and text-fig. 6) projects outwards and upwards between the first (mandibular) and second (hyoid) | arches until it is only separated from the first branchial groove (fig. 8 and text-fig. 6) by a thin partition (consisting of endoderm and ectoderm) which eventually takes part in forming the tympanic membrane. The second branchial pouch (fig. 9 and © text-fig. 6), though smaller than the first, extends outwards and then upwards towards the otic sac (fig. 38), thus filling up the greater part of the space between the second (hyoid) and third branchial arches. On the way it approaches the second branchial groove (text-fig. 6); but this groove is neither as long nor as deep nor yet as intimately related to its pouch as the groove between the mandibular and — hyoid arches. | “a The third branchial pouch (figs. 9 and 39), decidedly shorter and narrower than — the second, extends outwards and slightly upwards towards the space between the third and fourth branchial arches (text-fig. 6) to end some distance from the shallow — third indication of a branchial groove (fig. 8). In the horse, as in man, there is only — a minute diverticulum (text-fig. 6) representing the fourth branchial pouch of the chick, and only at the most a faint indication of a branchial groove lying behind the rudimentary fourth branchial arch (text-fig. 6). In the floor of the pharynx of man at the end of the third week there is a small mesial tubercle (tuberculum impar) between the mandibular and hyoid arches, a forked elevation (furcula) between the second and third arches, and furrows lead- ing into the branchial pouches; the tubercle is the first indication of a tongue, while the furcula gives rise to the epiglottis. In the 2l-days horse there is a smal] mesial prominence in the floor of the pharynx between the first and second branchial arches (text-fig. 6), and furrows leading to the branchial pouches, but no rudiment of either larynx or epiglottis. It may be mentioned that the fore- gut expands behind the fourth branchial pouch and gives off at each side a — shallow diverticulum (text-fig. 6). These diverticula may represent rudiments of the respiratory system. ; The Mid-gut and the Hind-gut.—The mid-gut, together with the anterior part of the hind-gut, communicates with the yolk-sac by means of a broad yolk-stalk (fig. 33 _ and text-fig. 12). In the region of the yolk-stalk the cavity of the hind-gut has the — form of a narrow channel (fig. 50), but behind the yolk-stalk it assumes the form of a wide flattened tube (fig. 51), the end of which projects upwards into the mass of cells occupying the space between the notochord and the terminal portion of the spinal cord: this dorsal diverticulum (figs. 33 and 52) seems to represent the cloacal chamber. From the ventral aspect of the hind-gut the allantoic diverticulum (fig. 33) STUDIES ON THE DEVELOPMENT OF THE HORSE. 311 extends backwards and upwards into the thick layer of mesoblast surrounding the caudal end of the embryo. 4. The Notochord. From Marriy’s statements it may be assumed that the notochord in the horse agrees in its grigin with that of the mole. As in the mole, the endoderm cells under- lying the medullary canal in front of the primitive streak are enlarged and arranged first to form a groove and then a cord (the chorda dorsalis), the thickened posterior end of which may contain a canal continuous with the neurenteric canal. At the middle of the third week large cells, continuous with ectodermic and endodermic cells at the anterior end of the primitive streak, form.a groove immediately under the medullary groove. At the end of the third week the place of this groove is occupied by the notochord. In the last fifteen sections through the tail end of the embryo there is no indication of a notochord, but in the sixteenth section (fig. 52) the notochord, nearly one-third the size of the spinal cord, is seen to the right of the middle line: like the spinal cord, it has a distinct central canal.* Though eccentric in position near its origin, the notochord soon occupies the middle line and lies immediately below the spinal cord. If the drawings of transverse sections through the trunk (figs. 43 to 52) are referred to, it will be observed that the notochord and its canal vary both in shape and size. The sections through the head and pharynx (figs. 37 to 40) show that the notochord varies in shape and size and is inti- mately connected with the endoderm of the anterior third of the roof of the pharynx (figs. 33 and 39) and with the hind-brain. 5. Heart and Blood-vessels. The embryo when first exposed was seen to be surrounded by a countless number of blood-vessels engaged in carrying blood to and from the yolk-sac (figs. 5, 34 and 35). Of these vessels the most obvious were the left vitelline artery, the sinus terminalis, and the large vitelline veins proceeding to the large bulging heart occupying the space between the pharynx and the middle portion of the trunk (fig. 33). The heart in the horse at the end of the third week, like the heart in a 3-days chick, consists of a sinus venosus, an atrium, a ventricle, and a truncus arteriosus.t In a 3-days chick, as in man about the end of the third week, there are five pairs of aortic arches, but in the 3-weeks horse only the first two pairs of arches (text-fig. 7) have made their appearance. There were neither rudiments of a heart nor yet of blood-vessels in Martin’s embryo; hence it may be assumed that in the horse during the second half of the third week there is (1) the vascularisation of the yolk-sac and the formation of the vitelline veins; (2) the fusion of these veins to form a single tubular * The notochordal canal, unlike the neural canal, is closed, does not open into the amniotic cavity. + Professor Ropinson in his description of the model recognises (1) a sinus venosus, (2) a sinu-atrial canal, (3) a ventricle, (4) an atrio-ventricular canal, (5) a bulbus cordis, and (6) a truncus aorticus. TRANS. ROY. SOC. EDIN. VOL. LI, PART II (NO. 7). 45 312 PROFESSOR J. COSSAR EWART. heart; (3) the elongation and folding of this heart so that it assumed the form of the letter ““S”; and (4) its constriction into four more or less distinct chambers, Trext-FiG. 7.—Diagram of heart and blood-vessels, .v., right vitelline vein; s.v., sinus venosus ; at, atrium; v., ventricle; b., bulbus arteriosus; I.a.a., first aortic arch; II.a.a., second aortic arch; La., left aorta; 7.a., right aorta; a.o., dorsal aorta; J.v., left vitelline artery ; r.v., vight vitelline artery; pl., plexus in allantoic mesoderm; a.c., anterior cardinal vein ; p.c.v., posterior cardinal vein ; c.d., duct of Cuvier; w.v., umbilical vein ; I.b.a. to IV.d.a., four branchial arches ; I.b.p, to [V.b.p., four branchial pouches, The vitelline veins, which play so important a part during the earlier weeks, are seen | in figs. 34, 35, and 43 approaching the embryo. Fig. 43 also shows the lateral portions of the sinus venosus in the pleuro-peritoneal canals. The intimate relation STUDIES ON THE DEVELOPMENT OF THE HORSE. 313 formed from the vitelline veins, receives blood from the allantois and the embryo as well as from the yolk-sac. The allantoic (umbilical) veins (text-fig. 7), one at each side, spring from a transverse sinus in the allantoic mesoderm at the caudal end of the embryo (text-fig. 7). The blood from the embryo reaches the sinus by the short Cuvierian veins formed by the junction of the anterior and posterior cardinals. Hach anterior cardinal (primitive jugular) vein (text-fig. 7) begins near the mandibular arch ; as it runs backwards it approaches the first two visceral pouches and the otic vesicle. Hach posterior cardinal vein (text-fig. 7) runs forwards from the caudal end of the embryo; on the way it is intimately related to either the pronephric duct or the Wolffian body. The blood flows from the sinus (text-fig. 7) by a wide opening into the atrium; the atrium, separated by a constriction from the sinus, opens by a valved aperture into the ventricle—portions of the atrium and ventricle are seen in figs. 38 and 40. While the sinus and atrium occupy the uppermost and left portion of the space between the pharynx and the middle curved portion of the trunk, the ventricle occupies the middle and right portion and lies in contact with the floor of the fore-gut and the yolk-stalk (fig. 33). The truncus (bulbus) arteriosus springs from that part of the ventricle lying in contact with the floor of the pharynx, and runs forwards in contact with the atrium (fig. 39) to reach the middle line and give off two pairs of aortic arches (text-fig. 7), a pair to the mandibular arches and a pair to the hyoid arches.* As text-fig. 7 indicates, the first aortic arch on the left side proceeds backwards in front of the first branchial pouch to meet and unite with the smaller second aortic arch and thus form the left aorta. Ina like manner the first aortic arch on the right side unites with the second arch to form the right aorta. The two dorsal aortz as they proceed backwards lie one at each side of the notochord (fig. 42), and are also more or less intimately related to the pharynx and gill pouches, the spinal cord and otic sacs, the pleuro-peritoneal sinus and the anterior cardinal veins (figs. 38 to 41). On nearing the sinus venosus the two aorte approach each other and completely fuse, but they again separate when a point about 3 mm. from the caudal end is reached (text-fig. 7). As the two aortez proceed backwards through the posterior third of the trunk they are intimately related to the notochord and the mesodermic somites. On reaching the posterior border of the yolk-stalk the distance between the aortz (text- fig. 7) becomes greater, and it continues to increase until they end in the allantoic “mesoderm. Ag the aorte proceed backwards the left gives off the large left vitelline artery (text-fig. 7), which eventually bifurcates to form the sinus terminalis (fig. 34) ; the right gives off the small right vitelline artery (text-fig. 7). Before ending blindly each aorta gives off branches to the hind-gut; some of these branches give rise to a plexus with which the allantoic (umbilical) veins communicate (text-fig. 7). * In only having two pairs of aortic arches the 21-days horse agrees with a human embryo of about 15 days figured by His. Ina figure by His of a human embryo of about 3 weeks all five aortic arches are present. If the age of these human embryos is approximately accurate, the aortic arches appear later in the horse than in man, 314 PROFESSOR J. COSSAR EWART. - D. Earty DEVELOPMENTAL STAGES IN THE HorsE AND SHEEP.* If horses are descended from ancestors allied to Hyracotherium of the London Clay, ample time has elapsed to admit of prenatal as well as postnatal variation in directions different from those followed by even-toed Ungulates. It is hence probable | that almost from the outset horse embryos will differ from sheep embryos at a like — stage in the life-history. , The Ovum.—In the Eutheria the ova, though always small, vary considerably ;— they may be under ‘1 mm. or over ‘3 mm. in diameter. For example, in deer the © TEXT-FIG. 8,—Segmenting ovum of sheep with distinct zona pellucida (2.p.): TEX1-FIG. 9,—Blastodermic vesicle of sheep. ¢., trophoblast ; z.c.m., inner cell mass. Note absence of zona pellucida, (After ASSHETON.) Text-FiG. 10.—Ovum of deer. The inner cell-mass (text-fig. 9) has differentiated to form (1) the embryonic shield (e.s.) continuous with trophoblast (¢7.), and (2) the endoderm (end.). At the corresponding stage in the horse there is still a zona pellucida—the line ex- ternal to the shield and trophoblast represents the zona in the horse. (After KEIBEL.) ovum may only measure ‘(07 mm.; on the other hand, the human ovum varies from ‘22 to '32 mm. In sheep the ovum, exclusive of the zona pellucida, measures ‘15 mm. ; in the horse it probably never exceeds ‘18 mm. In the ripe ovum of the sheep, figured by AssHETON, there is no indication of a perivitelline space; but even in the — unripe ovum shown in fig. 2 from a 6-mm. ovarian follicle of a mare there is a distinct space between the vitellus and the zona pellucida. In the sheep the zona (text-fig. 8, z.p.), attenuated at the end of the first veaia ruptures and disappears about the ninth day of gestation. As the amnion is only completed at the end of the second or beginning of the third week, it follows that * For information about the development of the sheep I am mainly indebted to papers by AssuHEToN and Bonnet, more especially to “The Morphology of the Ungulate Placenta,” Assumron, Phil. Trans. Roy. Soc., 1906 ; “'I'the Segmentation of the Ovum of the Sheep,” AssHEron, Quart. Jour. Med. Sci., 1898 ; “ Beitrige zur Embryologie der Wiederkiuer,” R. Bonnet, Archiv f. Anat. u. Physiol., Anat. Abth., 1889. - STUDIES ON THE DEVELOPMENT OF THE HORSE. 315 for some days sheep embryos come into intimate relation with the lining of the uterus. In the horse the zona pellucida, according to Bonnet, has a thickness of 4 when the blastocyst measures 13 mm. (fig. 3),7.e. at the end of the second week. Further, during the first half of the third week the blastocyst acquires an albumen coat, which at the middle of the third week has a thickness of 3 or 4 mm. (fig. 4, alb.). In the sheep as segmentation proceeds it soon becomes evident that certain cells will give rise to the trophoblast (text-fig. 9, t7.), and that others will form the inner cell-mass (text-fig. 9, 7.c.m.) which is eventually differentiated into endoderm and embryonic shield. In text-fig. 10 the embryonic shield (e.s.), trophoblast (¢.), Trxt-Fic. 11.—Diagram to indicate the stage reached in the horse about the middle of the third week. albn., thick (8-4 mm.) albumen coat; am., amnion fold ; ecd., ectoderm ; end., endoderm ; ms., mesoderm ; pr., primitive groove. Albumen layer is lost before end of third week. and endoderm (end.) of a deer are represented. Like differentiations and phases in all probability occur in the Equide. The Blastocyst.—In the horse the blastocyst, nearly spherical at the end of the second week (fig. 3), is egg-shaped at the middle of the third week (fig. 4) and pear- shaped at the end of the third week (fig. 5). At this stage it has a length of 50 mm., and occupies the lower part of one of the uterine horns (fig. 15). In the sheep soon after the rupture of the zona pellucida the blastocyst loses its spherical outline (text-fig. 13) and soon elongates into a narrow tube which occupies both uterine horns. This tubular blastocyst may have a length of 240 mm. at the end of the twelfth day, and measure 400 mm. (nearly 16 inches) at the end of the third week (text-fig. 16). The Trophoblast.—In the horse up to the middle of the third week the tropho- blast, except in the vicinity of the embryo (text-fig. 11), is in contact with the endoderm, but by the end of the third week, over and immediately beyond the embryo, it is in contact with non-vascular somatic mesoderm (text-fig. 12). Beyond the exoccelom up to the sinus terminalis the trophoblast forms the outer wall of a space 316 PROFESSOR J. COSSAR EWART. (fig. 28) occupied by the unsplit highly vascular mesoderm; beyond the sinus it is in contact with the yolk-sac endoderm (fig. 34). : In the sheep, up to the appearance of the amnion folds, the trophoblast, except — under the embryonic shield, is in contact with a thin imperfect layer of endoderm, — TExtT-FIG, 12.—Drawing to indicate the stage reached in the horse at the end of the third week, after the albumen coat as well as the zona has disappeared, a., amnion, now complete ; ch., chorion, made up of a layer of trophoblast anda layer of somatic mesoderm ; s,c., spinal cord; d.a., dorsal aorta; sp., splanchnopleure ; ¢r., trophoblast ; mes., unsplit vascularised mesoderm ; end., endoderm. — Trxt-FIG. 13,—Drawing to indicate the stage reached in the sheep about the middle of the third week. In the 21-days horse the mesoderm is only split in the vicinity of the embryo. In the sheep at a corresponding stage the splitting of the mesoderm is com- plete ; the inner (splanchnic) layer has united with the endoderm to form a free yolk- sac vesicle (y.s.), the outer with the trophoblast to form the chorion (ch.). am., amnion ; s.c., spinal cord ; d.q., dorsal aorta ; sp., splanchnopleure, (After BonNET. )- As the splitting of the mesoderm proceeds, the exoccelom extends in all directions, — with the result that the trophoblast is completely separated from the endoderm and provided with a lining of non-vascular mesoderm (text-fig. 13). In the horse the trophoblast at the end of the second week seems to consist of simple polygonal nucleated cells. At the middle of the third week the cells are cubical, but by the end of the third week they are long enough to rank as columnar cells. Up to the sinus terminalis there are simple columnar cells (fig. 28); beyond — the sinus, in addition to typical columnar cells, there are very tall columnar cells STUDIES ON THE DEVELOPMENT OF THE HORSE. Bly forming the larger trophoblastic discs (fig. 29) and cells with sac-like processes (fig. 30), some of which surround the discs, while others line the shallow areas seattered over the distal end of the blastocyst. Up to the end of the third week the trophoblast in the horse consists of a single layer of cells, which are either quite free or loosely connected. with the uterine epithelium. In the sheep the trophoblast at the outset consists of a single layer of simple polygonal cells; by the tenth day the cells are wedge-shaped, and by the fourteenth cubical. By the eighteenth day the cells which roughly correspond to the coty- ledonary burrs projecting from the lining of the uterus, have assumed a columnar form.* These columnar cells develop minute ridges which by fitting into slight ie i re 4 ee eae S.C SCO eee grooves in the cells of the burrs may assist in fixing the blastocyst. The Uterus.—In the sheep, when the ovum reaches the uterus “the mucous membrane is that typical of an ancestrous period” ; but as development proceeds there is a general increase in the complexity of the glands, with increased secretion, an invasion of the epithelium by leucocytes, and towards the close of the third week a thickening of the stroma. These changes are accompanied by a destruction of the uterine epithelium over the cotyledonary areas by means of binucleate trophoblastic cells insinuating themselves between the cells of the uterine epithelium to form a layer between the epithelium and the sublying stroma. The nourishment for the embryo, which mainly consists of “uterine milk,” is taken up by the cells of the trophoblast lying between the cotyledonary burrs. In the mare, even at the end of the third week of gestation the mucous membrane is still that typical of the ancestrous period. The glands at the end of the third week may be more active, and leucocytes may be more numerous in the uterine eavity, but of this there is no evidence, neither is there any indication of erosion of the uterine epithelium or of changes in the sublying stroma. f Martin states that the mucous membrane of the uterus from which his embryo was obtained was congested and spotted. There was neither congestion nor spotting _ of the lining of the uterus from which I removed my 21-days embryo. In all pro- bability, in the Equide at the end of the third week the whole of the trophoblast takes part in procuring nourishment for the embryo. Ammnon.—In the sheep the lateral folds are conspicuous on the fourteenth day, and extend well over the embryo on the following day. By the end of the sixteenth day the amnion is complete and connected by a cord to the inner surface of the chorion (text-figs. 13 and 14). Though in the sheep the amnion is well advanced on the fifteenth day, at the stage characterised by two pairs of mesodermic somites (text-fig. 14), there is no indica- * Opposite the cotyledonary burrs AssHETON says the trophoblast perhaps consists of two layers. + The pig closely agrees with sheep during the earlier weeks of gestation. The blastocyst begins to elongate on the eleventh day. As it increases in length it is thrown into transverse folds, with the result that, though apparently of no great length, it may measure when extended over 1000 mm. at the middle of the third week. The uterine epithelium begins to degenerate on the fourteenth day, and is reduced to a thin layer by the eighteenth day. 318 PROFESSOR J. COSSAR EWART. tion of the amnion in the horse until the middle of the third week. Even when four pairs of somites are present the amnion in the horse is only represented by very — slight lateral ridges (text-fig. 1), but by the twenty-first day the amnion is complete. 15. Texr-ric, 14,—A 14-days 22-hours sheep embryo with two mesodermic somites (m.so.), a complete amnion (am.), and an allantoic diverticulum (@/.d.). (After Bonnut.) Texr-Fic, 15,—A 17-days 23-hours sheep embryo, with about as many somites as a 21-days horse. Note the large allantois (a//.) in connection with the hind-gut (h. 4: ), the amnion (am.), and the yolk- catalle (y.s.). (After BonNE?.) — — see eS Se a Trxt-F1G, 16,—Diagram of a 21-days sheep embryo and its membranes. Compare with diagram of a 21-days horse embryo (fig. 34). A 21-days sheep is at about the same stage of development as a 28-days horse. m,, amnion; ch., chorion con- sisting of a trophoblast (¢7.), and somatic mesoderm (s0.m.); ¥.s., yolk-sac ; a/l., allantois. The early rupture of the zona pellucida and the absence of an albumen layer may perhaps account for the amnion appearing at an earlier phase in the sheep than in the horse. 3 Yolk-sac and Mesoderm.—In the sheep the splitting of the mesoderm already ye ie ee ee STUDIES ON THE DEVELOPMENT OF THE HORSE. 319 begun at the end of the twelfth or the beginning of the thirteenth day is completed at or about the end of the sixteenth day. When the yolk-sac endoderm is completely invested by splanchnic mesoderm a typical yolk-sac (¢.e. a free vesicle separated by a distinct space from the chorion) is the result (text-fig. 13). As the blastocyst assumes a tubular form the yolk-sac elongates: by the twenty-first day it may reach a length of over 350 mm. (14 inches) (text-fig. 16). As the mesoderm splits the splanchnic layer undergoes vascularisation, without, however, a sinus terminalis being established. In the horse the splitting of the mesoderm begins on the seventeenth or eighteenth day, but proceeds so slowly that it has only extended a short distance beyond the embryo on the twenty-first day (text-fig. 12). As the unsplit mesoderm extends between the trophoblast and endoderm it undergoes vascularisation, and the left vitelline artery bifurcates at an early stage to form a sinus terminalis (fig. 34).* As the splitting of the mesoderm is never complete in the horse, the most distal part of the yolk-sac endoderm retains throughout feetal life a slight connection with the trophoblast. This implies that the yolk-sac in the horse is never, as in the sheep, a free vesicle. In the horse at the end of the third week tubercles project into the cavity of the yolk-sac (fig. 32), each endodermic tubercle being related to a trophoblastic disc. There are apparently neither trophoblastic discs nor endodermic tubercles in the sheep. Allantois.—In the sheep there is a rudiment of the allantois at the end of the second week (text-fig. 14); a like stage is not reached in the horse until the end of the third week (fig. 33). On the sixteenth day the allantois in the sheep is a two- horned sac 2°7 mm. in width; on the seventeenth day it measures 16 mm. (text- fig. 15); the following day the width is nearly double ; and on the twenty-first day, as text-fig. 16 shows, it has a length of nearly 300 mm. (nearly 12 inches). In the horse at the end of the third week the allantois is still a minute diverti- culum 235 in width ; in the sheep at a corresponding stage of development the width between the horns is over 30 mm. The Embryo.—The youngest sheep embryo figured by Bonner was 3 mm. in length (text-fig. 14). It represents the phase reached 14 days and 22 hours after service, and is characterised by two pairs of mesodermic somites. In Bonnet’s figure (text-fig. 14) the mesodermic somites, primitive streak, neural groove, allantoic rudi- ment, and parts of the amnion are represented. The horse embryo figured by Martin (fig. 7) differs but little from Bonnet’s 14-days 22-hours sheep embryo (text-fig. 14). Marrin’s embryo, which represents the stage reached at the middle of the third week, measured 3°25 mm. in length and had four mesodermic somites. Though in this embryo the primitive and neural grooves * In the pig, in which the yolk-sac is a free vesicle as in the sheep, there is a temporary sinus terminalis followed by general vascularisation ; hence in a sense the arrangement of the yolk-sac vessels in the pig is intermediate between that of the sheep and that of the horse. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 7). 46 320 PROFESSOR J. COSSAR EWART. are distinct, there is no rudiment of an allantois, and the amnion is represented by two. hardly perceptible lateral ridges; as the mesoderm is not split under these ridges, VouRi San HORSE Text-Fic, 17.—Diagram of the endoderm of a 21-days horse embryo. Note the gill pouches in connection with the fore-gut, the wide yolk-stalk leading into the pear-shaped yolk-sac, the cloaca, and the minute allantoic diverticulum projecting from the hind-gut. Allantotis 4 Gill Pouches Yolk -Sac~ J Yolk-Sac SHEEP Trext-Fic. 18.—Diagram of the endoderm of a sheep at about the same stage of development as a 21-days horse. <2 Note the large crescent-shaped allantois and the long tubular yolk-sac which extends into both uterine horns, — one is perhaps hardly justified in regarding them as rudiments of the amnion Martin and Bonnet believed this 3°25 mm. embryo was 21 days old, but I have given — reasons for regarding it as representing the stage reached on the seventeenth eighteenth day of gestation. ee ih 3 STUDIES ON THE DEVELOPMENT OF THE HORSE. 321 During the fifteenth and sixteenth days marked progress is made by sheep embryos. The 16-days 22-hours embryo figured by Bonner has ten mesodermic somites, a long deep neural groove, rudiments of the fore-, mid-, and hind-gut, and two branchial arches; the primitive streak is still distinct, the amnion is complete, and there is a crescent-shaped allantois. In 1840 Hausmann figured a 19-days horse embryo with ten or eleven meso- dermic somites. While generally agreeing with Bonnet’s 16-days 22-hours sheep embryo, HausMann’s 19-days horse embryo (text-fig. 19) has apparently no rudiment of an allantois, and very little progress has been made in the development of either the nervous system, gut, or branchial arches.* In a 17-days sheep embryo there are over twenty somites, the greater part of the neural groove is converted into a canal, and there are indications of otic as well as of TEXT-FIG. 19.—HavsMANN’s figure of a 19-days 45-minutes horse embryo. TExT-FIG, 20.— TExtT-Fig. 21. — » ” 19-days 7-hours a ” ~ 19-days 20-hours . cerebral vesicles; the fore- and hind-gut have assumed a definite form, and the allantois has a width of 16 mm.; further, there are branchial arches, a three-chambered heart, two aorte, and a rudiment of the notochord (text-fig. 15). In a sheep embryo at the beginning of the nineteenth day there are more than twenty mesodermic somites, and the development of the nervous system has advanced so far that the three regions of the brain are easily recognised ; there are distinct optic as well as otic vesicles, a three-chambered heart, a maxillary process from the first visceral arch, and a large allantois in connection with a well-developed hind-gut. The 21-days horse embryo is in some respects intermediate between a 17-days and a 19-days sheep embryo, but it obviously differs in shape, in the less-developed condition of the hind-gut, and especially in the blastocyst, yolk-sac, allantois, and the structure of the trophoblast. For some years it has been assumed that the rate of eee in the horse varies greatly during the first three weeks. This assumption is the result-of Bonner stating in 1891 that a 21-days blastocyst varies from 13 mm. to 35 mm.f Now * Hausmann’s 19-days horse embryo lends strong support to the view that Marrin’s so-called 21-days embryo represents the stage reached at the middle of the third week. 7 “Hier vom 21- Tage schwanken zwischen 1:3 cm.-3°5 cm. Linge,” Bonnet, Grundriss des Hntwickelungsgeschichte, 1891, p. 240. 322 PROFESSOR J. COSSAR EWART. that a 21-days blastocyst 50 mm. in length has been described, presumably Bonner would state that in the horse the blastocyst at the end of the eat week varies from 13 mm. to 50 mm. By studying the cestrous period in the mare and the condition of the ovaries before and after service, I arrived at the conclusion that the difference between my 50-mm. blastocyst and Bonnet’s 13-mm. blastocyst was not due to arrested develop- ment but to a difference in age—a conclusion supported by creme of horse embryos (text-figs. 19-21) published in 1840 by Hausmann. Further, it has been taken for granted that during the earlier weeks the horse closely agrees in its development with other typical Ungulates. It is conceivable that up to the end of segmentation, perhaps up to the formation of the embryonic shield, there may be almost complete agreement between even-toed and odd-toed Ungulates ; but by comparing young horse embryos with sheep embryos at a corre- sponding stage, evidence has been obtained in support of the view that from the time the mesodermic somites make their appearance the two main groups of the modern hoofed mammals follow different routes—the sheep, pig, and other even-toed Ungulates especially differing at the end of-the third week from the horse in the form of the blastocyst, yolk-sac, and allantois, and in the structure and relations of the trophoblast. EK. SuMMARY. During the last fifty years so much progress has been made in clearing up the pedigree of the horse that we are now familiar with every chapter in the ancestral history between the 14-inch four-toed Hyracothervwm of the London Clay and the 60-inch Equus sivalensis of the Indian Pliocene. But while our knowledge of the remote ancestors of the Equideeis well-nigh com- plete, we still know surprisingly little about the more interesting chapters of the life- history of the horses now living under domestication. We are acquainted with the ancestral history because Corr, Marcu, Osporn, Scott, and other American paleontologists made a systematic and exhaustive study of “ fossil” horses; we know little about the more interesting phases of the life-history of the living Equide because embryologists have not yet succeeded in working out the development of either horses, asses, or zebras. In America several expeditions have been sent out in search of material for the study of extinct horses, but until recently only one serious attempt seems to have been made to obtain material for a systematic study of the development of recent horses. Unfortunately, this attempt (made in Hanover by Hausmann some eighty years ago) yielded meagre results. Professor Bonnet of Giessen, famous for his memoirs on the development of domestic animals, when referring in 1889 to previous work on the foetal membranes of the horse, pointed out that about the first stages of STUDIES ON THE DEVELOPMENT OF THE HORSE. 323 development we have nothing to depend on except the statements and drawings of Hausmann, who had a stud including fifty-two mares at his disposal for his researches. These statements, which relate to the development at the end of the third week, owing to the mistaken views of their author about the ovum and its membranes, BoNNET points out, have caused more confusion than enlightenment, while the drawings which illustrate Hausmann’s contributions are almost incomprehensible and worthless. In the home of the thoroughbred racehorse no systematic attempt has been made by embryologists to work out the life-history of the Equide, and sporadic attempts to elucidate special points in the early history have hitherto been well- nigh as fruitless in England as the ambitious attempt made in the ’thirties of last century in Hanover. From first to last the difficulty in England has been lack of material. This was especially true of an attempt made by Hux.ey to find rudiments of the second and fourth digits. When preparing the lecture on fossil horses delivered in 1876 in New York, it occurred to Huxuey that strong evidence in support of the fact of evolution would be forthcoming if it were proved that the modern horse passes through a Hipparion or 3-toed stage during development. Armed with all the available material Sir W1tt1am Fiower could place at his disposal, HuxLEy proceeded to search for rudiments of the phalanges of the second and fourth digits. To Hux.ry’s great disappointment the search was in vain. Huxtey failed, not because the 3-toed stage is absent in the recent horse, but because it appears, comes and goes, much earlier than he expected. Up to the end of the Miocene period all horses had throughout life three complete toes both in front and behind. For this reason it seems to have been assumed that the 3-toed phase, if it actually occurred in the modern horse, would persist for a considerable time and be preceded by a 5-toed phase. As a matter of fact, as I shall show in a subsequent paper, the abbreviation in the development in the Equide is so marked that the 3-toed stage is reached during the fifth week of gestation and practically comes to an end during the seventh week. It seems to have been taken for granted that during the earlier weeks horse embryos develop along the same lines as other Ungulates. I find, however, that from the second week onwards the horse follows a route different from that of all the other mammals hitherto studied. Since 1891 it has been assumed that during the earlier weeks the development of the horse is retarded in some cases, accelerated in others. In the roe deer the development is arrested as soon as the cleavage stage is reached, with the result that little or no progress is made for several months. There is, however, no evidence that at any stage the development is arrested in the horse. The belief that in the horse progress is sometimes retarded is based on a statement by Bonnet, who in 1891 asserted without any reservation that the blastocyst at the end of the third 324 PROFESSOR J. COSSAR EWART. week varies in length from 13 mm. to 35 mm.,?.e. from half an inch to over 1°5 inches. Bonnet arrived at the conclusion that a 21-days blastocyst may vary from 13 mm. to 35 mm. because he received in 1889 a 13-mm. horse blastocyst taken from a mare twenty-one days after the first service, while in 1890 Paut Martin of Zurich obtained a 35-mm. blastocyst also taken twenty-one days after service. I find that at the end of the third week of gestation, 2.e. twenty-one days after service, the blastocyst measures 50 mm.; that Bonnet’s so-called 21-days embryo probably represents the stage reached at the end of the second week; and that Marri’s so-called 21-days embryo represents the stage reached at the middle of the third week of gestation. The age of Bonnet’s and Martin’s embryos having been dealt with, attention is next directed to the condition of the ovaries, oviducts, and uterus at the end of the third week, and to the position of the 50-mm. (21-days) blastocyst in the uterine horn (fig. 14). The trophoblast at the end of the third week is unusually interesting; in several essential points it differs from that of the sheep and pig and other Ungulates hitherto examined. Up to the sinus terminalis it consists of typical columnar cells (fig. 28); beyond the sinus there are (1) groups of very tall columnar cells arranged to form trophoblastic dises (fig. 29) which probably help to fix the blastocyst to the lining of the uterus, (2) groups of columnar cells in the act of elongating to form additional discs (fig. 30), and (3) columnar cells with sac-like processes (fig. 30)— these phagocytic cells occur around the margins of the discs and in the grooves surrounding the discs, but especially in shallow depressions beyond the sinus terminalis which probably lie opposite the openings of the uterine glands and are concerned in taking up the more solid particles of the “ uterine milk.” In the region of the embryo the trophoblast by uniting with the somatic layer of mesoderm forms the outer wall of the exoccelom (text-fig. 12). From the exoccelom to a short distance beyond the sinus terminalis the trophoblast forms the outer wall of a narrow space occupied by highly vascular but still unsplit mesoderm (fig. 28). The distal portion of the trophoblast is intimately related to the yolk-sac endoderm (fig. 34). Sections of the blastocyst afford no evidence that solid particles penetrate the cells of that part of the trophoblast extending between the embryo and the sinus terminalis, but they afford abundant evidence that granules of various sizes enter both the tall columnar cells of the discs and the cells with sac-like processes which form a considerable part of the trophoblast beyond the sinus terminalis (figs. 29 and 30). The sections further demonstrate the presence of a continuous sheet of coagulum between the trophoblast and the yolk-sac endoderm (fig. 30). Between the dises this coagulum (which probably consists of nutritive material derived from the “‘ uterine milk”) resembles a basement membrane, but under the dises it reaches — a considerable thickness and is continuous with similar coagulated material entering into the formation of yolk-sac tubercles (fig. 32). In the vicinity of the embryo the yolk-sac consists of endoderm and a layer of splanchnic mesoderm, but from the exoccelom to the sinus terminalis it is only STUDIES ON THE DEVELOPMENT OF THE HORSE. 325 represented by endoderm (text-fig. 12). Beyond the sinus the yolk-sac endoderm is intimately connected with the trophoblast (fig. 34). Hxcept opposite the discs the yolk-sac endoderm consists of a single layer of cells ; opposite the discs, where it takes part in forming the yolk-sac tubercles, the endoderm may consist of two layers of cells (fig. 32). At the middle of the third week the splitting of the mesoderm is only beginning, and the amnion is only represented by indistinct lateral ridges (text-fig. 11). At the end of the third week the amnion is complete and the mesoderm is split so as to form a ccelom and a small exoccelom (text-fig. 12). The outer (somatic) layer of mesoderm together with a layer of ectoderm forms the amnion; further, by uniting with the trophoblast the somatic mesoderm gives rise to the chorion of the embryonic area (text-fig. 12). The inner (splanchnic) layer of mesoderm by. uniting with endoderm forms the yolk-stalk and the upper free and only complete part of the yolk-sac. The unsplit mesoderm (text-fig. 12) which extends from the exocelom to the sinus terminalis is highly vascular ; it receives blood from the embryo by the vitelline arteries and returns it to the embryo by the vitelline veins (fig. 34). There is no indication of an allantoic diverticulum at the middle of the third week, but at the end of the third week there is a considerable mass of vascularised allantoic mesoderm at the caudal end of the embryo, into which extends from the hind-gut a small allantoic diverticulum (fig. 33 and text-fig. 17). Martin’s embryo, the age of which is almost certainly 17 or 18 days, has four pairs of mesodermic somites (fig. 7). Hausmann’s 19-days embryo (text-fig. 19) has probably ten somites, but the 21-days embryo has over twenty somites (figs. 10 and 11). Marrin’s embryo in shape resembles the sole of a shoe; my 21-days embryo is hook-shaped (figs. 7 and 8). That this strongly bent embryo is older than Martin’s so-called 21-days embryo is indicated by its size, by the phase reached in the develop- ment of the nervous, alimentary, and circulatory systems, and especially by the complete amnion and the allantoic diverticulum. The nervous system is represented by a fore-brain with optic vesicles, by an indistinct mid-brain, by a hind-brain adjacent to which are otic vesicles, and by a spinal cord completely closed except at the caudal end, where it opens by a neuropore into the cavity of the amnion (fig. oa The alimentary system consists (1) of a fore-gut from which branchial pouches project outwards towards shallow branchial furrows (text-fig. 6) lying between branchial arches, (2) of a mid-gut which communicates by a yolk-stalk with a large yolk-sac (fig. 34), and (3) of a short hind-gut from which a cloacal rudiment projects upwards and an allantoic diverticulum extends backwards (fig. 33 and text-fig. 17). The circulatory system is represented by a sinus venosus, an atrium, a ven- tricle, and a bulbus arteriosus; by two pairs of aortic arches, two dorsal aorte, and by vitelline and other arteries, vitelline, cardinal, umbilical, and other veins (text-fig. 7). 326 PROFESSOR J. COSSAR EWART. A notochord (fig. 33) and branchial arches (text-fig. 6) represent a skeleton, and excretory organs are represented by Wolffian bodies and pronephric ducts. In a comparison between horse and sheep embryos it is pointed out that even at the stage characterised by four pairs of mesodermic somites the horse differs from a sheep at the corresponding phase in its life-history, and.that, as development proceeds, the differences, more especially in the foetal appendages, become more pronounced, EXPLANATION OF PLATES. Puate IX. Fig. 1. Ovarian follicle of a mare containing an unripe ovum; this follicle only measured 6 mm, The large follicle in fig. 16, which contained a ripe ovum, measured 37 by 33 mm. Fig. 2. The unripe ovum seen in fig. 1. x 300. This ovum measured ‘11 mm. exclusive of the zona pellucida; the average size of the ripe ovum of the mare, exclusive of the zona pellucida, is probably ‘18mm. In the sheep the ovum is about ‘15 mm. In deer the ovum may be under ‘1 mm.; in Dasyurus it averages ‘24 mm.; while the human ovum varies from ‘22 to ‘32 mm.; hence the ovum in the horse, as in the sheep, is relatively large. Between the ovum and the zona pellucida (z.p.) is the perivitelline space (p.s.) ; the cells external to the zona form the corona radiata (c.7.). The egg of amphioxus only measures ‘12 mm., yet it contains sufficient deutoplasm to provide all the energy required for development up to the gastrula stage. It may hence be assumed that the ovum of the mare contains all the nutrient required for develop- ment up to at least the morula stage. Fig. 3. This figure indicates the actual size of Bonnert’s so-called 21-days blastocyst—it probably represents the phase reached at the end of the second week.. Bonner states that the pase was invested by a zona pellucida 4 thick. Fig, 4. Marrin’s so-called 21-days blastocyst, natural size, invested by a thick (3 to 4 mm.) albumen layer (alb,). This 35-mm. blastocyst probably represents the stage reached at the middle of the third week. After Martin. Fig. 5. The blastocyst, natural size (50 mm.), taken from a Highland mare 21 days and 2 hours after the one and only service on the seventh day of cestrus. Note the difference in the size and position of the embryo from the embryo in fig. 4, the left vitelline artery, the sinus terminalis, and the vitelline veins, Fig. 6. A section through the lower end of the left uterine horn, The blastocyst is kept in contact with the tongue-like processes of the uterine mucosa by hydrostatic pressure. Natural size. Fig. 7. A dorsal view of Marrrn’s so-called 21-days embryo. Note the four pairs of mesodermic somites, the primitive streak, and the primitive and neural grooves. 15. After Marvin. Fig. 8. The 21-days embryo seen in fig. 5. This hook-shaped embryo is three times the size of, and much further advanced in its development than, Martin’s so-called 21-days embryo. Note the right vitelline vein (v.v.) on its way to the relatively large heart (At.), the fronto-nasal process (/.p.), the four branchial arches (b.a.) and branchial grooves, and the depression between the first arch and the fronto- nasal process, which represents the mouth or stomodeum (s.). Only the front part of the amnion (am.) is represented. x 15, Fig. 9. Part of the same embryo with its amniotic investment. Note the swelling near the end of the fronto-nasal process due to the optic vesicle (0.ve.); the mouth between the fronto-nasal process and the first pair of arches (1 6.a.); and the cavities of the three branchial pouches ().p.). x 15. Fig. 10. A dorsal view of the same embryo. Note the mesodermic somites and the neural canal with its opening to the exterior at the caudal end (n.p.). x 15. Fig. 11. The same embryo seen from below, the anterior part surrounded by the amnion (am.). Note the openings of the otic vesicles (0.v.), the great width of the pericardial chamber, the somites, and the fin- like caudal expansion which consists in part of allantoic mesoderm. x 15. STUDIES ON THE DEVELOPMENT OF THE HORSE. 327 PiatE X. Figs. 12 and 13. The right and left ovaries of the Highland mare from which the 21-days embryo was obtained. Natural size. Fig. 14. The uterus, oviducts, and ovaries of the same mare, half natural size, with part of the wall of the body of the uterus and of the left uterine horn removed to expose the 21-days blastocyst. Note the longitudinal folds of the uterine mucous membrane cut by transverse furrows into tongue-like processes, and the opening leading from the right uterine horn. Fig. 15. The left uterine horn, blastocyst, and oviduct, and the ovary sectioned to show ovarian follicles at various stages of growth. Had the blastocyst escaped about the end of the third week the large follicle on the right (0,f.) would have discharged a ripe ovum on or about the day the mare was killed. Had the gestation gone on to full time the size of the large follicles would have been gradually reduced. There is no evidence of erosion of the lining of the uterus by trophoblastic cells, or of epiblastic processes from the trophoblast projecting into the spaces between the cells of the uterine mucosa. Natural size. Puate XI, Fig. 16. Section through the ovary of a mare which died on the second day of cestrus. Note that the outer wall of the large follicle (0,f.) is thinner than in the large follicle in fig. 15. Had this mare survived four or five days longer this follicle would have doubtless discharged a ripe ovum. Natural size. Fig. 17. Section through the ovary of a mare which died three months after giving birth to a foal. Note the corpus luteum (c. /m.) above the circular follicle near the middle of the ovary. Natural size. Fig. 18. The corpus luteum found in the left ovary of the mare from which the 21-days embryo was ' taken. Natural size. The corpus luteum in the mare from which Martin obtained his so-called 21-days embryo was about double the size of a lentil. Fig. 19. The “trumpet” of the oviduct and upper end of the left uterine horn represented in fig. 14- Note the small opening (0.7.) leading from the horn into the oviduct, the numerous lamelle lining and extending over the edge of the trumpet, and the interrupted lamelle lining the expanded part of the oviduct adjacent to the trumpet. Natural size. . Fig. 20. Transverse section through the upper end of the left uterine horn represented in fig, 14. Natural size. Fig. 21. Part of the left uterine horn in contact with the 21-days blastocyst represented in fig. 15. It shows the appearance assumed by the ridges when the horn is pinned out on a flat surface. As the horn _ expands the ridges become less prominent and the spaces between the ridges are gradually increased. Natural size. Fig. 22. Part of the sinus terminalis (s.¢.) taking an irregular course through the unsplit. mesoderm of the 21-days embryo: v.a., vessels proceeding from the sinus towards embryo; 0.v., vessels in mesoderm beyond the sinus ; ¢.d., trophoblastic disc. x 28. Pratt XII. Fig. 23. Section showing the highly congested condition of the uterine mucosa on the second day of cestrus. x 40. Fig. 24. Part of the same more highly magnified, showing congested capillaries, distended glands, and yellow bodies resembling hematoidin crystals. There was no evidence of the escape of blood corpuscles into the cavity of the uterus or of destruction of the uterine epithelium. ~ 250. Fig. 25. Section through a ridge from the uterine horn containing the 21-days embryo. x 14. Fig, 26. Part of the same section more highly magnified (x 100). At the end of the third week of _ gestation the glands are decidedly less active and there is less congestion than is the case during cestrus. Fig. 27. The 21-days embryo, 15 times natural size. The head and front portion of the trunk are bent backwards round the large heart. v.v., the large vitelline veins reaching the sinus venosus from the yolk-sac; r.v.a., small right vitelline artery; /.v.a., large left vitelline artery; the left aorta and the _proximal part of the left vitelline vessel (text-fig. 8) are not represented in the drawing. TRANS ROY. SOC. EDIN., VOL. LI, PART II. (NO. 7). 47 328 PROFESSOR J. COSSAR EWART. Pratt XIII. Fig. 28. Section of the blastocyst immediately beyond the sinus terminalis. ¢7., trophoblast consisting : of columnar cells—note that the cells are in contact with each other along their whole length, have square-cut ends and deeply placed nuclei; md., unsplit mesoderm lying in a space between the trophoblast and the yolk-sac endoderm ; md.e., growing edge of mesoderm, in contact with inner surface of trophoblast; y.e., yolk-sac endoderm, in contact with the trophoblast, immediately beyond the growing edge of the mesoderm; b.v., a blood-vessel beyond the sinus terminalis. Fig. 29. Section through the centre of a trophoblastic disc. The disc consists of very tall cells, devoid, except round the margin, of sac-like processes. Some of the cells contain one or more round bodies derived perhaps from the “uterine milk.” The cavity of the disc is occupied by a deeply stained coagulum con- tinuous with a membrane-like coagulum (cm.) lying between the trophoblast (¢r.) and endoderm (end.). The endoderm between the discs consists of a single layer of cells, but opposite the discs it sometimes consists of masses of cells in contact with a lightly stained material which projects into the cavity of the yolk-sae, Many of the cells surrounding the disc end in sac-like processes (pc.). Fig. 29a. Section through the edge of a disc. In such sections, owing to the presence of many nuclei, the dise seems to have resulted from a heaping up of cells. Many of the endoderm cells (end.) seen in this section differ from the typical endoderm cells in figs. 30 and 31. ‘i Fig: 30. This section represents an early stage in the development of a disc. On the left of the section are nearly typical columnar cells (¢7.) ; then come cells with sac-like processes (pe.), apparently in the act of taking in food particles ; then cells which by proliferating and elongating eventually form the disc; cells with — sac-like processes (pe.) also occur on the right of the elongating cells. In the region of the developing dise the coagulum is thick and irregular; beyond this region it has the appearance of a basement membrane. Further, within the developing disc the endoderm has undergone proliferation and lies in contact with a lightly stained coagulum. Puate XIV. Fig. 31. A minute coagulum lying between the trophoblast and the endoderm. This represents the first stage in the development of a trophoblastic dise and of a yolk-sac tubercle; tr., trophoblast ; end., endo- — derm ; cm., coagulum, x 400. 5 Fig. 31a. The coagulum in fig. 32 seen from the cavity of the yolk-sac. The cells in contact with the — coagulum are more granular and more deeply stained than the surrounding endoderm cells. x 400. Fig. 32. A later phase in the development of a yolk-sac tubercle. The coagulum (cm.), now larger, is invested by highly granular endoderm cells (end.) x 400. Fig. 32a. Section through same tubercle beyond the margin of the coagulum. x 400. Fig. 33. Diagram of longitudinal section of 21-days embryo and its membranes. Note the complete amnion (am.), the small allantoic diverticulum (al.d.) embedded in the allantoic mesoderm (al.m.), the wide yolk-stalk (y.s.), that the mesoderm (md.u.) except in the immediate vicinity of the embryo is still unsplit, — and that the spinal cord opens at its caudal end (c.m,) into the amniotic cavity (am.c.); at., atrium; el., cloaca; d.a., dorsal aorta; ecd., ectoderm; end., endoderm; evo., exocelom; 7.0., fore-brain; h.b., hind- brain; m., ‘longitudinal growth-centre of embryo”; m.b., mid-brain; md., mesoderm; mo., mouth; nc., notochord; ne.v., canal of spinal cord; p.o.v., optic vesicle; s.c., spinal cord; s.v,, sinus venosus; — t., tuberculum impar; v., ventricle; y.sa., yolk-sac; 1, 2, 3, 4, internal branchial slits. Circa 15 times natural size, 21-days embryo. Fig. 34. Semi-diagrammatic drawing of the 21-days blastocyst with part removed to show the embryo. — Note the complete amnion (am.), the yolk-stalk (y.s.) leading into a large yolk-sac, the left vitelline artery (v.a.), the sinus terminalis (s.f.), and the right vitelline vein (v.v.), that the greater part of the mesoderm is still unsplit (md.a.), and that the non-vascularised distal portion of the blastocyst consists only of tropho-— blast (#7.) and endoderm (end.). Beyond the sinus, dots (¢.d.) represent some of the trophoblastic discs. am.c., amniotic cavity ; exo., exoccelom ; md., mesoderm. Circa 3 times natural size. Fig. 35. The 21-days embryo as it appeared when the chorion was removed. Note the large left vitelline artery (v.a.), the large right and left vitelline veins (v.v.), and the rich plexus of vessels in the vicinity of the embryo. bs STUDIES ON THE DEVELOPMENT OF THE HORSE. 329 Puate XV.* Fig. 36. Section through fronto-nasal process of the 21-days embryo. am., amnion; am.c., amniotic cavity ; 7.b., cavity of fore-brain ; p.o.v., primary optic vesicle. Fig. 37. Section through head end of notochord (ne.). m.b,, mid-brain ; f.g. fore-gut ; s.p., Seesse]’s pouch ; sto., stomodeum ; | a.a., first aortic arch. Fig. 38. Section through otic vesicles (0.v.). h.b., hind-brain; d.a., dorsal aorta; j.g., fore-gut, from which notochord projects upwards towards floor of hind-brain; 2 b.p., second branchial pouch ; d.a., dorsal aorte formed by union of first two aortic arches; .at., atrium; 0.co., bulbus cordis ; pe., pericardium. Fig. 39. Section through third branchial pouch (3 b.p.). at., atrium; a.v.c., atrio-ventricular constric- tion ; b.co., bulbus cordis ; d.a., dorsal aorta ; h.b., hind-brain ; pe., pericardium ; v., ventricle. Fig. 40. Section through fourth branchial pouch (4 b.p.). at., atrium; at.v., atrio-ventricular valve ; b.co., bulbus cordis ; d.a., dorsal aorta ; h.0., hind-brain ; v., ventricle. Prate XVI. Fig, 41. Section through anterior part of sinus venosus (s.v.). am., amnion; d.a., dorsal aorta ; f.g., fore-gut; nc., notochord ; pp.c., pleuro-pericardial cavity ; s.c., spinal cord ; v.v., vitelline vein. Fig. 414. Section through middle part of trunk in a line with above. m.g., mid-gut; m.p., muscle plate; p.d., pronephric duct ; u.v., umbilical vein; w.b., Wolffian body. Other abbreviations as before. Fig. 42. Section through trunk and middle portion of sinus venosus (s.v.). d.c., duct of Cuvier ; lu., lung ; v.v., vitelline vein; ¢.v., branch of vitelline vein. Other abbreviations as before. Pratt XVII. Fig. 43. Section through posterior part of sinus venosus (s.v.) and the terminal portions of vitelline veins (v.v.). pe., pericardium. Other abbreviations as before. Fig. 44. Section through middle part of trunk in region of yolk-stalk. sp., splanchnopleure. Other abbreviations as before. Figs, 45 and 46. Sections through posterior third of trunk. Note that the amnion (am.) is folded to form longitudinal ridges. c.v., posterior cardinal vein. Other abbreviations as before. Fig, 47. Section near the point where the dorsal aorta divides in two. Abbreviations as before. Puate XVIII. Figs. 48 and 49. Sections nearer caudal end than fig. 47, showing two aorte (d.a.). Abbreviations as before. ; Fig. 50. Section through hind-gut (h.g.) immediately in front of posterior wall of yolk-stalk. Note the triangular form of the spinal cord (s.c.), the mesodermic somites (m.p.), cardinal veins (c.v.), and the two aortz (d.d.). Fig. 51. Section through the posterior wall of the yolk-stalk (y.st.). al.b., allantoic branches of aorta ; _hg., hind-gut. Other abbreviations as before. Fig. 52. Section through cloaca (c/.) immediately in front of opening of spinal cord into amniotic cavity. The notochord vc., is tubular, and lies at one side of the middle line. Note the wide allantoic diverticulum (al.d.) and the crescent-shaped amniotic cavity (am.c.). Other abbreviations as before. Fig. 58. The tenth section from the caudal end of the embryo. am.,-amnion; am.c., amniotic cavity; al.d., terminal portion of allantoic diverticulum; c.m., caudal neuropore ; s.c., spinal cord; sp., splanchnopleure ; w.v. umbilical vein. * The figures in Plates XV-—XVIII are all magnified 75 diameters. 5 Roy. Soc. Edin. i Vol. LI- Plate IX. Development of the Horse. M‘Farlane & Erskine, Lith. Edin Vol: i= Plate x. Pins. soy. OOC. Edin. Ewart: Development of the Horse. .% a M‘Farlane &Erskine, Lith. Edin Wo Tl Pilatesx, M‘Farlane & Erskine, Lith. Edin Vol. LI. M‘Farlane & Erskine, Lith., Edin. s = eos Og o8- sity er : 5 ad eee i a ae © 9: "50, DEVELOPMENT OF THE Horse.—Puate XII. “ys e “pwFa0d \ ween, Ae S. e fe ee a, SNE EN cor eceoes® i \o on ej =) = es fs % = oe on 7) 2 od <2) ° SS tm 6 by Richard Muir; Fig. 27 by Fiddes Watt. ue ae. [=| a= Cy ica) me ro) ° M = is. Roy. Soc. Edin. Vol. LI— Plate XIII. Ewart: Development of the Horse. M‘Farlane & Erskine Lith Edin ‘49% sepply Aq Ge ‘Bi { quourey] vsndny pues yen” “OD f hq #E-18 ‘S51 “UPA ‘“qavT ‘euryerg 9 oueeT, i %, ws AN \ | 4 t ~Jns. Roy. Soc. Edin. Vou. LI. Proressor J. Cossar Ewart: “Studies on the Development of the Horse.”—Puate XV. Neill & Co., Ltd., Edin Oi, JLIt = / \ Trans. Roy. Soc. Edin. E XVI. LAT a P, Proressor J. Cossar Ewart: “Studies on the Development of the Horse. UIpa “PIT “OD 8 ION GP SIA We “Pp au a a iy, “4 + tie oe ee me) , f doe or ne / VIP ‘old —== babe UF ih Tad UN AUVHOTY Vou. LI. un. dns. Roy. Soc. Ed “Studies on the Development of the Horse.”—Puiate XVII. Cossar Ewart: ES ise] fe) nM mM lca} fe So [2] AY UID “PIT “OD % ION, “Lb Std a a) Gants eet " ran kenese™ Bqu e™ = ae = jae om “ ad UIA GauvHONy cee I iis eee ae as Me Vou. LI. Trans. Roy. Soc. Edin. “‘Studies on the Development of the Horse.”—Puatp XVIII. Proressor J. Cossan Ewart “UIP “pyT ‘09 % TLLe IN Ch Pee ae len Se Dh Sod eae te a Oa , ae ‘Tad HIN GAVHOIY Og “Oly F ae : « _-VitellineVein Viteline Vern--> UmbiliialVen =<. : Wolttian Body Amnin’ Spinal Medulla Dorsal Aorta Text-FiIc. 25.—Oblique section through the embryo along the line 1, fig. 58, Pl. XIX. . Wolffian Body ‘Wuscle Plate 1 1 1 | ' 1 On account of the curvature of the embryo the ventral part of the sinus venosus has a comparatively reduced cranio-caudal extent, as. contrasted with the dorsal part ; and at its caudal end, near the median plane there is a slight evagination which may possibly be the rudiment of the cranial end of the inferior vena cava, but at this stage the evagination in question has no connection with any venous channel in the liver region. The cephalic end of the sinus opens through a short and wide sinu-atrial canal into the caudal part of the atrium. The position of this canal is marked on the exterior, on each side, by a very definite sulcus (fig. 57, Pl. XIX). The tributaries which open into the sinus are the two vitelline veins, the two umbilical veins, and the two ducts of Cuvier. Each vitelline vein enters at the caudal end of the corresponding lateral border of the sinus. The terminations of the umbilical veins are close to those of the vitelline veins, but are situated a little more medially and more dorsally. The short duct of Cuvier, on each side, enters the cranial end of the sinus on its dorsal aspect and near the lateral border (fig. 58, Pl. XIX). 342 PROFESSORS ARTHUR ROBINSON AND A. GIBSON. The Atriwm.—Kven when the shrinkage (figs. 57, 62, Pl. XIX) which the atrium has undergone has been fully taken into account, it is obviously a smaller chamber than the sinus venosus. It lies more towards the left than the right side of the embryo, cranial to the sinus venosus and dorsal to the left portion of the ventricular chamber (figs. 57, 58, 62, Pl. XIX). Its dorsal wall lies close to the ventral wall of the fore-gut, in the region of the ventral ends of the second and third branchial pouches, and, as it bulges towards the fore-gut, it is largely responsible for the elevation which forms the ridge-like caudal boundary of the groove which connects the ventral ends of the second branchial pouches together (figs. 57, 63, Pl. XIX). The ventral wall of the atrium lies in relation with the dorsal wall of the left portion of the ventricle, mto which it opens through a short but distinct atrio-ventricular canal (fig. 62, Pl. XIX). 0074/ Medulla Fold of Vitelline Fein Fudiment of Ti Umbilical Vein’ \. \Wtuscle Plate Dorsal Aorta Spinal Medulla TEXT-FIG. 26.—Oblique section through the embryo along the line 2, fig, 58, Pl. XIX. Woltivan Body ;! To the left of the atrio-ventricular canal the left margin of the atrium projects ventrally, forming the rudiment of the left auricle. The right auricular portion of the atrium which lies medial and caudal to the bulbus cordis is not so un deliminated. The Ventricle-—The ventricle forms the most ventral portion of the heart. It crosses the median plane from left to right, but lies more to the right than to the left side (figs. 55, 56, 62, Pl. XIX). Its ventral wall, on the left side, is deeply invaginated, a condition obviously due to shrinkage, for in the drawings of the fresh specimen the ventricular area is uniformly convex. ‘There is already definite surface indication of the separation of the ventricle into right and left portions (fig. 62, PI, XIE): The bulbus cordis commences at the cranial end of the right-segment of the ventricle and it runs medially and dorsally till it reaches the median plane, where it ~~. .greagie (grr as YK PV eS ee eer Se ee RECONSTRUCTION MODEL OF A HORSE EMBRYO TWENTY-ONE DAYS OLD. 343 terminates in a short truncus aorticus (figs. 58, 62, Pl. XIX). The truncus aorticus lies dorsal to the ventricle, cranial to the atrium, and, at its termination, it is on the right side of the atrio-ventricular canal (fig. 62, Pl. XIX). Truncus Aorticus.—The short truncus aorticus extends from the bulbus cordis to the ventral wall of the fore-gut, where, immediately ventral to the ventral ends of the first branchial pouches, it terminates by dividing into four branches, two on each side. The more cranially situated branches are the ventral roots of the first aortic arches ; the more caudally situated branches are the second aortic arches. The Aortic Arches.—Two aortic arches are present on each side. The ventral root of the first arch runs cranialwards from the truncus aorticus, below the aperture of the first branchial pouch, to the ventral end of the mandibular arch ; there it turns dorsally and becomes the first aortic arch. The first aortic arch runs dorsally, along the’ cranial border of the aperture by which the first branchial pouch communicates with the pharynx (figs. 58, 63, Pl. XIX), and at the level of the dorsal margin of the aperture it turns caudally and becomes the dorsal root of the first arch. The dorsal root of the first arch terminates by joining the dorsal end of the second arch at the level of the otic vesicle and ventral to the anterior cardinal vein (fig. 58, Pl. XIX). The second aortic arch runs directly dorsally, through the second branchial arch, from the truncus aorticus to the caudal end of the dorsal root of the first aortic arch, with which it unites to form the commencement of the dorsal aorta (fig. 58, Pl. XIX). The Dorsal Aorte.—The dorsal aortee run from the dorsal ends of the second aortic arches to the caudal end of the embryo, where they terminate by breaking up into branches which join the capillary plexus of the allantois (figs. 58, 59, 60, 61, Pl. XIX). In the head region each dorsal aorta lies medial to the dorsal extension of the second branchial pouch of the same side, and then along the dorsal surface of the lateral margin of the more caudal part of the pharynx. In the thoracic region it is run along the dorso-lateral border of the gut, dorsal to the pleuro-pericardial canal of the same side (text-figs. 25 and 26, pp. 341, 342). Near the caudal end of the fore-gut, and in the transverse plane in which the vitelline veins join the sinus venosus, the two dorsal aorte fuse together to form a single trunk. This trunk lies in the median plane, dorsal to the yolk-sac and ventral to the notochord, until within a short distance from a point near the cranial end of the cleft which indicates the position of the hind-gut. There it separates again into a right and a left dorsal aorta which run caudally, in the angles between the spinal medulla and the dorsal wall of the hind-gut, to their terminations in the allantoic region (ties; 58; 99,60 6k Pl. XIX). A short distance from its termination the caudal portion of each dorsal aorta gives off large branches to the wall of the yolk-sac, and a number of branches which pass ventrally to the allantoic capillary plexus already described (p. 340). The Anterior Cardinal Veins.—Kach anterior cardinal vein commences near the 344 PROFESSORS ARTHUR ROBINSON AND A. GIBSON. dorsal end of the mandibular arch by the union of a number of radicles. From its commencement, it runs, first caudally and then ventrally, to its termination in the duct of Cuvier of the same side. In the cranial part of its course it lies dorsal to the dorsal root of the first aortic arch and the most cranial part of the dorsal aorta, ventral to the otic vesicle, and medial to the dorsal extensions of the first and second branchial pouches. As it reaches the thorax it turns ventrally and crosses the lateral side of the dorsal aorta a short distance caudal to the fourth branchial pouch (fig. 58, Pl. XIX). The Posterior Cardinal Veins.—Kach posterior cardinal’vein commences in the mesoderm near the caudal end of the embryo, and, as it runs towards the cranial end of the body, it lies at first ventral to the corresponding Wolffian duct, then dorso- lateral to the duct, and next on the ventro-lateral aspect of the more cranially situated portion of the Wolffian body. After leaving the Wolffian body it runs along the dorso-lateral wall of the pleuro-pericardial canal (fig. 58, Pl. XIX), and, shortly before its termination in the duct of Cuvier, it lies dorsal to the more caudal part of the sinus venosus (fig. 58, Pl. XIX). Ducts of Cuwer.—Hach duct of Cuvier is formed by the union of the anterior and posterior cardinal veins of the same side opposite the cranial end of the dorsal wall of the sinus venosus, and a short distance caudal to the fourth branchial pouch (fig. 58, Pl. XIX). It is quite a short vessel, which is compressed from side to side, and which runs ventrally and somewhat caudally round the lateral wall of the pleuro-pericardial canal, to its termination in the sinus venosus (text-figs. 25 and 26, pp. 341, 342). The Vitelline Veins.—The right and left vitelline veins are the largest blood- vessels of the embryo, and each terminates in the corresponding ventro-lateral angle of the caudal portion of the sinus venosus. The orifice of termination of each vein is quite distinct from that of its fellow of the opposite side. The Umbilical Veins.—Each umbilical vein takes origin from the transverse sinus which lies in the allantoic mesoderm at the caudal end of the body of the embryo, and it receives a number of additional tributaries from the capillary plexus in the allantoic mesoderm (figs. 59, 60, 61, Pl. XIX). As it passes towards the heart each vein lies in the most lateral part of the body wall of the corresponding side, and it terminates, in the dorsal wall of the caudal part of the sinus venosus, a little dorsal and medial to the termination of the vitelline vein (text-figs. 25, 26, pp. 341, 342, and fig. 58, Pl. XIX). The Celom.—The greater part of the intra-embryonic portion of the ccelom is a narrow cleft, which communicates freely with the extra-embryonic ccelom at the mar- gins of the embryo and at its caudal end. The more cranially situated portion of the ccelom is enclosed, and it forms a pericardial cavity and two pleuro-pericardial canals. The pericardium surrounds the heart. Its cranial extremity is situated at the level of the second branchial arch. The caudal end lies at the cranial border of the : RECONSTRUCTION MODEL OF A HORSE EMBRYO TWENTY-ONE DAYS OLD. 345 umbilicus, where the cavity of the pericardium is separated from the extra-embryonic ccelom by the mesodermal septum transversum. At its dorsal boundary on each side, at the level of the fourth branchial pouch and the cranial border of the duct of Cuvier, the pericardium communicates with a pleuro-pericardial canal (fig. 58, Pl. XIX). The Plewro-pericardial Canals.—EKach pleuro- eOrcardial canal commences from the dorsal part of the pericardial cavity, at the level of the cranial border of the duct of Cuvier. It runs from the more cranially situated to the more caudally situated end of the bend of the embryo, terminating, in the latter situation, in what will eventually be the peritoneal portion of the ccelom. As it passes between its two terminations it lies medial to the duct of Cuvier; dorsal to the sinus venosus, by which its ventral boundary is invaginated; and ventral to the dorsal aorta and the posterior cardinal vein of the same side (text-figs. 25, 26, pp. 341, 342, and fig. 58, Pl. XIX). ‘DESCRIPTION OF FIGURES ON PLATE XIX. Fig. 54. Dorsal view of the caudal portion of a reconstruction of a horse embryo 11 mm. long. Fig. 55. Dorsal view of the cranial portion of the reconstruction, and ventral view of the most caudal part of the caudal portion, of a horse embryo 11 mm, long. Both portions have been partially dissected. Fig. 56. Left lateral view of a reconstruction of a horse embryo 11 mm. long. A portion of the heart is shown, and the lines of incision for the removal of the ventral wall of the pharynx and the dorso-lateral wall of the left pleuro-pericardial canal are shown. Fig. 57. View of a median section of a reconstruction of a horse embryo 11 mm. long. The heart has not been divided, and its left side is shown. Fig. 58. Right lateral view of a reconstruction of a horse embryo 11 mm. long. The reconstruction has been dissected to show the relations of the branchial pouches and the otic vesicle to the blood-vessels and the heart. Fig. 59. Dorsal view of the caudal end of a reconstruction of a horse embryo 11 mm. long. Dissected to show terminal part of right dorsal aorta. : Fig. 60. Ventral view of the caudal end of a reconstruction of a horse embryo 11 mm. long, showing the blood-vessels. Fig. 61. Dissection of the ventral portion of the caudal end of a reconstruction of a horse embryo 11 mm. long, showing cavities of allantoic diverticulum and hind-gut. . Fig. 62. Ventral view of a reconstruction of the heart of a horse embryo 11 mm. long. Fig. 63. View of the dorsal aspect of the ventral wall of the pharynx of a horse embryo 11 mm. long, showing section of the first three branchial arches and sections of the first and second aortic arches. Norz.—Figs. referred to, other than those on Plate XIX, are in the Plates of Professor Cossar Ewarv’s paper. [Key to LETTERING 346 puAa: 2AA. Al Alb. ALD. ALM, Am. ACY. Ana. APpC. AT, AVC. 1BA. 2BA. 3BA. 4BA, BC. IBC. 2BC. KEY TO LETTERING OF FIGURES ON PLATE XIX. First aortic arch. Second aortic arch. Allantois. Allantoic branches of aorta. Allantoic diverticulum, Allantoic mesoderm, Amnion. Anterior cardinal vein. Anastomoses, between allantoic arteries and umbilical veins. Arrow , in pleuro-pericardial ~ canal. Atrium. Atrio-ventricular canal. 1st branchial arch. 2nd 5 5 3rd ” ” 4th » ” Bulbus cordis. 1st branchial cleft. 2nd branchial cleft. BM. Bucco-pharyngeal membrane. Wage 2BP. 3BP. 4BP, Cl, ON. DA, DC. FB. ist branchial pouch. 2nd ” ” 3rd = an 4th ” » Cloaca. Caudal neuropore, Dorsal aorta. Duct of Cuvier. Fore-brain. . Hind-brain. . Hind-gut. Rudiment of lungs. . Mid-brain. . Notochord. Neural canal. For ZBP, tig. 55, read 2BP. OV. Otic vesicle. Ot. Otic vesicle. PCV. Posterior cardinal vein. PoV. Primary optic vesicle. Ppr. Pleuro-pericardial ridge. Sm. Somatopleure. Sp. Splanchnopleure. SpM. Spinal medulla. SV. Sinus venosus, TA. Truncus aorticus. TI. Tuberculum impar. TS. Terminal sinus. UV. Umbilical vein. V. Veutricle. VV. Vitelline vein. / — Roy. Soc. EDIN. WO, oe GIBSON AND ROBINSON: MODEL OF HorRSE EmMBRYO.—PL. XIX. LL \\ ‘ie ‘ A RITCHIE & SON, EDIN™ IX.—A Contribution to the Study of the Scottish Skull, By Matthew Young, M.D., Senior Demonstrator in Anatomy, University of Glasgow. Communi- ited by Professor THomas H. Bryce. (With Three Plates.) (Read June 21, 1915. MS. received July 27,1915. Issued separately January 31, 1916.) CONTENTS. PAGE PAGE Introduction . i ‘ . 847 (C) The correlation of the West Scottish 1. A consideration of ye mean ae of the . skull as regards many of its dimen- various dimensions and indices of the re- sions and characters, and a com- presentatives of a group of 405 male and parison with that obtaining in 1 other 100 female skulls, and their comparison with ; series of skulls. : : : . 392 (A) The mean values given by TuRNER (1) 3. The application of Kiaarscu’s craniotrigono- for his complete series of Scottish metrical method to a subsidiary group of skulls. : 349 100 male skulls, taken from the larger group (B) The mean values of the ame in his of 405, and the results obtained . : . 405 E collection derived from Renfrew- 4, The application of ScHwase’s — special shire : 358 measurements to the same series of 100 - 2. Examination of the West Srothion aienill Be skulls, a criticism of the method as applied biometric methods é 2 366 to modern skulls, and a discussion on the (A). Comparison of the mean ales with results obtained . x ee 431 those of various other types of skull 5. The various features brought to light and an or skulls of other races. 368 analysis of the values of the characters (B) The variability of the West Beare measurable after the preceding 100 skulls skull as evidenced by the range of were sectioned in the median oe variation of the above values, their plane . : 437 standard deviations and coefficients 6. Conclusions as to the ciiaeeion Bere een ae of variation, and a comparison of its type of Scottish skull and other pee both variability with that of the skulls pre-existent and modern. ; . 449 of other races . : F 5 E. -Bieel The present communication is based principally on the results of a series. of ‘Investigations into the form, dimensions, and characters of the. skulls that comprise - the unique collection which has been located in the Anatomy Department at the University of Glasgow for a number of years, and of which no account has yet been published. We owe the collection and preservation of this large and valuable series of crania—comprising over 700 specimens—to the zeal of Mr W. K. Hurrony, Lecturer in Anatomy, Queen Margaret College, University of Glasgow, to whom I am much indebted for permission to continue the work begun, but given up, by him.* I am very grateful to Professor Bryce for assistance and much encourage- ment during the progress of the research. * Mr Hurron intended to carry out a research, not dissimilar to my own, and to that end obtained a grant from the Carnegie Trust for the purchase of craniological apparatus. Unfortunately, from circumstances over which he had no control, he was unable to continue the work, and I have made use of the instruments obtained by him. The present paper formed part of a Thesis presented for the degree of Doctor of Medicine to the University of Glasgow, - [have gratefully to acknowledge a liberal grant from the Carnegie Trustees in aid of its publication. 3 TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 50 7 - ae 348 DR MATTHEW YOUNG, The skulls were obtained in a Glasgow burial-ground, during the course of excavations preparatory to the erection of a building on the site, but beyond that little is known as to the exact period of time to which the skulls belong. In one of the graves a token was found bearing the date 1840, but such evidence does not carry us far, and probably many of the skulls belong to a period antecedent to that date. The skulls are undoubtedly, in the great majority of cases, Scottish skulls, and the remarkable uniformity of the series, which is evident even to the untrained eye, points to the fact that they, for the most part, belong to one type. Proof of this uniformity will be supplied afterwards when we consider the variability exhibited by the series. No doubt there may be a few “foreign” skulls present in the collection, but the number is certainly small, and in my calculations in this paper I think that in most cases I work with a series sufficiently large to prevent a few aberrant speci- men influencing at all, or at most to a trifling degree, the final results. In order to pursue my investigations in connection with this subject I have found it necessary to subdivide the series into the following subsidiary groups :— (A) A group of 405 presumably male skulls. (B) =,°4 5 ~100demale skulls. (Oy es. 90 skulls of doubtful sex. (ch) een 45 male metopic skulls. (En) eee ete 20 female metopic skulls. PU was eee oe 45 juvenile skulls of various ages. (G) ,, 100 taken from group (A) for the application of Kiaatscn’s and SCHWALBE’S methods of examination, and afterwards for median sagittal section. (Series K in Appendix.) It will be seen that, in arranging these skulls in the above sub-groups, I have endeavoured to arrange them according to their sex. This is not a difficult task when one is confronted with typical male and female specimens presenting well- marked distinguishing features peculiar to the two sexes (Plates I, II, and III), but in a series of the magnitude of the present one, as in other collections, there are many skulls which present no very definite distinguishing sexual features, which le on the border-line between the two sexes as it were, and which might reasonably be placed in either division. These I have placed, for the present, under the heading of “doubtful sex.” The problem of allocating skulls to the proper sex when there are no other skeletal remains to assist one, has presented difficulties to most craniologists who have attempted the task.* The 405 skulls put down in the subsidiary group as males have what are usually regarded as the distinguishing male features pretty well marked, and I have little hesitation in affirming that the great majority of those are the skulls of males, — * TURNER (1 and 2), Barren (3), ArBy, VircHow (4), Kuaatson (5), WARREN (6), Parsons (7), Pearson (8), Ropertson (9), and Bryce (10) all agree that it is sometimes difficult, if not impossible, to ascribe certain skulls to their proper sex. ie Crees a a SE ale ee ee Pe A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 349 although it is quite possible I have ascribed a few to the wrong sex. A similar statement may be made regarding those classed as females. The 90 of doubtful sex I have decided to leave aside in dealing with the sexed skulls, instead of following the method advocated by some craniologists, namely, to ascribe them to the sex to which they showed a preponderance of characters. In the present work I have merely recorded for future use their principal measurements, and have calculated some of the mean values and variations of their dimensions. With such a large collection of material, moreover, I can afford to set aside the * doubtfuls,” as the numbers left to represent the two sexes, 400 and 100 respectively, are sufficiently great to make the probable error reasonably small and practically to negative the influence of the inclusion of a few specimens of the wrong sex on the final results. Besides, 400 is as large a number as is convenient to work with when calculating the correlation coefficients by the most reliable method. In the Appendix I have also recorded the detailed measurements of the metopic and juvenile skulls, but shall not further deal with them in the present publication. 1. (A) Tae Mean Vaturs or Dimensions anp InpIcEs or MALE SKULLS. Table I shows in four parallel columns the mean values of the dimensions of the principal cranial characters given by Sir WiLL1aM Turner (1) for his complete Scottish series and his Renfrewshire group as regards the male sex, and the corresponding values in the two male groups (400 and 100) of the series of Scottish skulls under examination. Table II, which follows, gives in three parallel columns the mean values for TuRNER’S complete female group, his female group derived from Renfrewshire, and a group of 100 female skulls from our collection. The “K” series is the group of 100 presumably male skulls which became separated from the remainder in the following way :— As a preliminary to a complete investigation into the characters of the skulls comprising the collection, | began with a group of 100 skulls taken from the whole series which, having the characters usually ascribed to the male sex fairly well marked, I presumed to be, in most cases, the skulls of males. Unconsciously, in making this selection, but as might have been expected, I have included in the group selected an unusually large proportion of the larger skulls. This influences to a marked extent, as will be seen later, the mean capacity and, to a less extent, the other measurements. This group of 100 may be accepted, notwithstanding, as fairly representative of the relative proportions of the male skulls, although the ‘respective mean values may be a little higher. It is to this group of 100 that I have applied Kuaarscu’s “craniotrigonometrical” method of examination and ScHWALBE’s measurements, and thereafter I had them all bisected in the median sagittal plane to study several features only measurable after this has been done. 350 Comparing the measurements of the present series with TuRNER’s series and beginning with the capacity of the male skulls, we find that, in the large male group: the mean capacity is 1459°5.¢.c., while for TURNER’S male series the mean capacity is In estimating the capacity I have followed carefully the directions of TurNER both as to material, 7.e. shot, and technique, ¢ Reports and in his memoir on the Scottish skull, bi as is well known, the “ personal equation ” capacity of skulls even by similar methods. Comparing the lengths in the two series of Scottish skulls, ¢.e. the gbello- occipital lengths or greatest lengths, we find that there is a difference of about | mm. in the mean values, with the advantage on the. side of the present series, values being 186°6 mm. and 187°52 mm. respectively. For the ‘‘K” series the mean 1478 c.c., a difference of about 20 c.c. furnished in iis “ Challenger” glabello-occipital length is 189°2 mm. Character. Cubic capacity : : Glabello-occipital length . Basi-bregmatic height Vertical index . Maximum breadth . Cephalic index Horizontal circumference . Vert. trans. circumference Basi-nasal length Basi-alveolar length. Gnathic index Total long. circumference Interzygomatic breadth Nasi-alveolar length Maxillo-facial index Nasal height . » width index Orbital width . » height. Peeing ex.. f Palato-alveolar length ; breadth "3 index The mean values of the basi-bregmatic heights are practically equal, viz. 182°4 mm. DR MATTHEW YOUNG. TaBLeE [. Mean Values of Dimensions of Male Skulls. Present Series. 1459°5 44-11 187:52 + :20 132°724°17 70°81 + 09 139°56 +°16 74°36 + 08 526°60 + 47 426°23 100°37 + 13 95:08 + °16 94:68 SB P8} 127:16+°18 70°96 +16 55°81 51°72 +°11 23°47 +06 45:58 4°15 38°67 + ‘05 33°05 +:07 85°51 + °17 52°94 59-59 112°76 TuRNeER’s Series, Renfrew Group. 1478 186°6 1324 70°9 144°3 17°4 531 434 1014 - 96 94:5 513:2 1322 71°6 54°3 53°5 23°1 389 39 34 87:2 55°6 60°9 113 is an important factor in ote 1526 190 133°5 70°3 142°8 75°1 531 429°9 101-4 94:5 93:3 523 127°5 73°5 58'1 52°2 23°3 45°2 39°3 35 89:08 54:5 60°7 “li for TURNER’S series and 132°72 mm. for that under examination. the mean value is about 1 mm. greater, z.e. 133°71 mm. With regard to the greatest breadth or parieto-squamous breadth we find a notable difference in the two series of Scottish skulls; the mean value given in In the “ K”-series “KK” Series, ; 1511-348: 60 18918 + 39 i 133°71 53m 70°7 + ‘20 140°7 + 29 74°41 +°16 52969 428-03 100-96 +°27 — 94:87 + ed | ‘ | = a 51743 12814 72°35 53°32 23°18 39-24 33:76 50°84 57°91 ‘ ‘ . ‘ —— SEES Las SS FS A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 351 TuRNER’S memoir is 149°3 mm. (which in Macponngtt’s paper in Biometrika (iii) is given as 144°3 mm.), while the mean value for the group of 405 is 139°56 mm.— sensibly different from 144°3 mm. For the group of 100—the “K” group—the mean value is 140°7 mm.: still much smaller than TuRNER’s value. Herein lies the most notable difference between the two series of Scottish skulls, TURNER’s and the present one. While there is comparatively little difference between the respective mean lengths and heights, there is a distinct difference between the mean breadths, amounting to 10 mm. if we accept TurNmrR’s figures, or 5 mm. if we accept MacponneELL’s correction. Almost wholly due to this difference in breadth, but minimised to a slight extent by the difference in the greatest lengths, we have the higher value of the cephalic index in TuRNER’s series, namely 77°4, 2.€. mesaticephalic, while the mean cephalic index for the present series is 74°36, t.e. dolichocephalic. For the “K” series the mean value of the cephalic index is 74°4, while for the group of 11 male skulls derived from Renfrewshire and included in TURNER'S series the mean cephalic index is 75°1, distinctly below that for the whole series. The details are given in Table II. TasBLe I1.* ‘TURNER’S Series. West Scottish Series. “K” Series 2 of 100. Class. No. Per cent. No. Per cent. No, and per cent. Dolichocephalic, —75 49 28 232 57:3 58 Mesaticephalic TOTO ae a 45 26 129 31°8 34 ee T7579 45 26 35 3-6 7 Brachycephalic, +80 : : 35 20 ~ 9 2°2 1 Following TurNsR’s method of subdividing the mesaticephalic group into two subsidiary sections, one with the index from 75 to 77°4, the other from 77°5 to 79°9, as those in the lower section approach more closely to the dolichocephalic type and those in the higher section to the brachycephalic type, it was found that, in the “K” series, in the higher division of the mesaticephalic group and the brachycephalic group were contained only 8 per cent. of the skulls examined, the remainder either being dolichocephalic (58 per cent.) or in the lower section of the mesaticephalic group (34 per cent.), whereas in TURNER'S series 46 per cent: were either brachy- cephalic or showed a tendency towards that type, while the remaining 54 per cent. were dolichocephalic (28 per cent.) or approached that type (26 per cent.). | * In making up this and the succeeding tables the measurements recorded in the Appendix were all first reduced to the “arithmetical mean.” In Tables IT and XI, however, in subdividing the mesaticephalic skulls into TuRNER’s two classes, this was not done for the indices between 77 and 78, but all the skulls with indices from 77 to 77:4 were placed in the lower group and those with indices from 77°6 to 78 in the higher group, while of the skulls with indices of exactly 77:5 half were placed in the lower, half in the higher, group. 352 DR MATTHEW YOUNG. In the complete male series of 405 skulls the proportions in the different groups were as follows :— .* Dolichocephalic, 57°3 per cent., or, with a tendency to dolichocephaly, 31°8 per cent. ; in all, 89 per cent. Brachycephalic, 2°2 per cent., or, with a tendency to brachy- — cephaly, 8°6 per cent. ; in all, 10°8 per cent. Although less well marked than in the — “ K” series, this large group shows well the remarkably small proportion of skulls : contained in it which show a brachycephalic tendency. . In the Renfrewshire group in TuRNER’s series, which comprised 21 skulls, of which 11 were male, 8 skulls were dolichocephalic and 7 had an index between 75 and 76°7, which is a much higher proportion of skulls showing a tendency towards dolicho- cephaly than is seen in the generalised results in TURNER’s memoir. They belong — to the same type as the group “ K.” The measurements of the latter furnish additional evidence of the fact, which has been demonstrated by several authorities, including Ripuey (14), that while the cephalic index throughout Scotland—in fact, throughout Britain—is remarkably uniform, it has a lower value over some parts of the west and south-west of the former country than over the remainder, especially the east, where it has a distinctly higher value. I shall return to this subject later. . The conclusion that one would draw from an examination of this male series is that the type of skull prevalent in Glasgow and the surrounding district is the dolichocephalic type, and that it exists almost to the complete exclusion of the — brachycephalic type. There is little, if any, evidence in this series of “that strong brachycephalic strain which pervades the population of Scotland,” according to TURNER (1). On comparing the height indices or vertical indices of the two series there was found to exist a close similarity, between them, as we could expect, knowing that the mean lengths and heights were approximately equal in value. In TURNER’s series of 98 male skulls (part of his group of 150) the mean value of the index was 70’9, in the Renfrewshire group of 11 the mean value is 70°83; in the subsidiary group of “K” of this series the mean value of the index is 70°7, while in the complete male group the mean value is 70°81. | i < TasuE III. / Turner’s Series. “K” Series. a Class. ——— - No. Per cent. Toh ne No. Per cent. | | per cent. Below 70: . - . Chamecephalic : 65 43 40 156 38°5 Between 70°1-75: Metriocephalic : 68 45 56 220 _ 54:3 Above 75'1: Hypsicephalic . ; 17 Lt 4 29 Tl A comparison of the results in the above series shows that the percentages found A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 353 in the various classes are very similar to those found in the subsidiary series “ K,” while about 5 per cent. of those in TURNER’s chamecephalic group and 4 per cent. of those in his hypsicephalic group are transferred in this series to the metriocephalic group. A feature of Scottish skulls, as pointed out by Turner (1), is the fact that the breadth is greater than the height, except in rare instances. In his series of 150 skulls the height was greater than the breadth in two instances only (1°3 per cent.), the height was equal to the breadth in four instances (2°6 per cent.), while the height was less than the breadth in the other specimens, v.e. the Scottish skull is platychamecephalic. In the complete male series now under examination, associated doubtless to a large extent with the smaller value of the maximum breadth, the height was equal to, or greater than, the breadth in 67 specimens, t.e. 16 per cent. compared with 4 per cent. in the former series. In the “ K” subsidiary male series the height was greater than the breadth in 12 per cent. and equal to the breadth in 8 per cent.— in all, 20 per cent. With regard to the circumferences in the two series, we see that while the mean values of the total longitudinal circumferences are practically equal, as regards the horizontal and vertical transverse circumferences, TURNER'S series shows a slight excess in value over the present series. The longitudinal circumference comprises the complete sagittal arc, with its frontal, parietal, and occipital sezments, and the opisthio-nasal length or CLELAND’s base line. With regard to the seoments of the sagittal arc, | quote the following figures from TURNER: and these may be compared with the corresponding measurements in the “ K” subsidiary series in the following table :— TaBLeE [LY. TURNER. “K.” Series. Longest frontal arc. ‘ 148 152 Shortest ,, - : : 3 111 120 Longest parietal arc. : : ; 148 143 Shortest ,, % 3 3 : : 102 112 Longest occipital are. ; 139 141 Shortest __,, fae: ; : ; 94 107 TURNER says that the occipital arc was greater than the frontal arc in 13 cases out of 144, and less than it in 131 cases. In the “ K” series of 100 the occipital are was greater than the frontal in 3 cases and equal to it in 2 cases. In TuRNER’s series of 138 the occipital arc was greater than the parietal in 26 cases, 7.e. less in 112 cases. In the “K” series the occipital are was greater than the parietal in 23 cases and equal to it in 2. In TURNER’s 354 DR MATTHEW YOUNG. group of 158 the parietal are was greater than the frontal in 55 cases and equal to itin 7. Inthe “K” series the parietal are was greater than the frontal in 25 cases 4 and equal to it in 8. As TURNER writes: “It is the rule, therefore, for the parietal and frontal longitudinal ares to exceed the occipital, though exceptions to the rule — occur in recognisable numbers.” And further: “ It is obvious, therefore, that so much | variation occurs in the relative length of the longitudinal ares, they have no appreci-_ able value as race characters in the Scottish skulls, and the variation occurred in both the brachycephalic and dolichocephalic types.” Evidence which may have an influence in modifying the last statement will be supplied later when we come to consider the — correlation that exists between the total are and its various segments, frontal, parietal, and occipital, and between these segments and the glabello-occipital length in the present series of Scottish skulls. As has been noted by several observers, including : Huscuke, WErspacH, and CLELAND (15), the relative proportions of the various segments not only differ in the two sexes, but undergo a change during the course — of development, and evidence in support of this is supplied by the values in the series of juvenile skulls. ) a: With regard to the other measurements in the two series of Scottish skulls, we find that the mean basi-alveolar lengths are approximately equal, and the same relation 7 ‘ship holds with regard to the nasi-alveolar lengths. As a consequence, the gnathie index which is obtained from these measurements is practically the same in the two groups. | ‘a The mean value of the basi-nasal length is only about 1 mm. less in the West t Scottish male series of 405 than it is in TuRNER’s group. Table V shows the skulls Isa classified according to the gnathic index :-— TABLE V. a West Scottish Series. TURNER'S Series. Class: No, Per cent. No. Per cent. Orthognathous (below 98)... : : 308 17:2 72 74°22 Mesognathous (98-103). : : ; 84 21 24 24°74 Prognathous (103 and upwards). ; ; i ILS 1 1:03 The near approach to equality between these proportions in the different classes and those in the series under examination is notable, and confirms TuRNER’S con- clusion that the Scottish skulls are characterised by an almost complete absence of prognathism. Skulls are classified into three groups according to the value of their orbita index as follows. Table VI shows the two series classified according to the value of the index. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 359 Taste VI. | West Scottish Series. Due: Series. Class. No. Per cent. | Per cent. Microseme (below 84) : : ; | 132 33 26°4 Mesoseme (84-89) . ; : ns 149 37°3 26°4 Megaseme (89 and upwards) : : 118 29°5 45°6 It will be observed that in the series of 399 male skulls we find that 132 are microsemic, 149 are mesosemic, and the remainder are megasemic. The percentages in the microsemic and megasemic classes are seen to be approximately equal, while the mesosemic group has a slight excess. In TuRNER’s series, which included both male and female skulls, it is evident that there is a much larger proportion in the megaseme class. There is evidently not the same tendency for the orbit to be high relatively to the breadth in the present group, as is manifestly the case in TURNER'S series. TURNER expresses the opinion that the orbital index is very variable, and that it only possesses “a secondary value as a race character.” We shall refer to the variability of the orbital index later when treating of the variability of the Scottish skull and also to MacpoNNELL’s comments on TURNER'S view. With regard to the nasal index the following table expresses the relationship between the two series :— : Pie vane West Scottish, TURNER’S Series. Class. No. Per cent. No, Per cent. Leptorhine (below 48) _.. : : 271 67°9 93 756 Mesorhine (between 48-53) : : 3 103 25°8 26 21:1 Platyrhine (53 and upwards) . : ‘ 25 6:2 1 3:2 The percentage number in the mesorhine group in the two series of Scottish skulls is not vastly different, as it is only about 4 per cent. The leptorhine group in TURNER'S series contains about 8 per cent. more than the leptorhine group of the West Scottish series of males contains of the whole number; while there is about half the proportion of skulls in TurNngEp’s platyrhine group than obtains in this other series of Scottish skulls. The presence of this relatively small platyrhine group in ce both series is possibly due to the explanation given by TuRNER, 7.e. “accidental, and TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 51 a = 356 DR MATTHEW YOUNG. iad due perhaps to intermixture, through an ancestor of a strain of some race in which a platyrhine nose was an ethnic character.” : —_ Owing to the fact that the mandibles have not been preserved, it is impossible to obtain a complete or nasio-mental facial index for the present series. The maxillo- nasi-alveolar length x Loc ) interzygomatic bread has been calculated, however, for both the male and female groups, and in the former the skulls may be arranged in the following way, adopting TURNER’s modification of facial index or upper facial index of KoLLMann, 2.e. y tT on KoLuMaNy’s mode of arrangement :— 2a TasLe VIII. West Scottish Series. TurRNER’s Series. Class. No. Per cent. No. Per cent. Leptoprosopic (50°1 and perene) : ‘ 380 95-2 72 91 Mesoprosopic (45-50) ‘ F : 18 4:5 7 88 Chameprosopic (below 45) : , ‘ 1 2 0 0 While in Turnur’s series a relatively small proportion has been transfe: from the leptoprosopic to the mesoprosopic group, there is a remarkably el similarity between the proportions contained in the respective groups in the series of skulls. The general type of face in the series under examination is even more notably leptoprosopic than in TURNER'S series. As regards the palato-alveolar or palato-maxillary index, I have adopted classification devised by TuRNER, and the grouping for the 399 male skulls is as follows :— TaBLe IX. West Scottish Series. | Turnur’s Series (82). Class. No. Per cent. No. Per cent. Hyperdolichuranic (below 105) . : 5 55 13°8 1 13°4 Dolichuranic (105-110). : : : 96 24 20 24°4 Mesuraniec (110-115) : : : ; 85 21°3 15 18:3 Brachyuranic (115-120) . : ; 81 20°3 17 20°7 Hyperbrachyuranic (above 120). ; 5 82 20°5 19 23°1 A comparison of the percentages in the corresponding groups requires no comment; the close similarity is obvious. In the series of 400 male skulls the A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 357 palato-alveolar breadth was less than the length in 12 cases only, z.e. 3 per cent., whereas in TURNER'S series this occurred in 3 specimens or 3°6 per cent. There are a few other measurements in this series which we can compare with the corresponding dimensions in TURNER’S series. £.g. Minimum frontal diameter with a mean value of 96°42 mm. Interstephanic diameter “sg . AS ie a Asterionic diameter * a 109°68 ,, Intermalar diameter bs % SO mage. Transverse base i, si Goalies Transverse arc . < SIZ) With regard to some of these measurements in his Scottish series of skulls Turner (1) makes the following statement: “‘ As a general rule the frontal stephanic diameter materially exceeded the minimum frontal; the asterionic diameter, except in one skull, was greater than the minimum frontal, but as a rule it was less than the stephanic, although there were several exceptions.” The results in the present series are in accord with TurRNEr’s results, as will be evident from an analysis of the above mean values. Material for comparison with the present series is supplied by TuRNER’s memoir for other two dimensions of the undivided skull, namely, the length of the base of CLELAND, ?.e. the opisthio-nasal length, that of the sagittal arc, and their relationship to one another. The mean length of the base line in the group of male skulls (22 in number) examined by TuRNER was 134°3 mm., and that of the sagittal are was 376°5 mm. ‘The average base line in the series “‘ K” was 135°3 mm. (for the 405 male skulls it was 134°5 mm.), and the sagittal arc in the same series was 382°3 mm. In the series of 100 the mean relation of the base line to the are was ~ 1 to 2°819. This is slightly greater than the figure given for Scottish skulls by CLELAND; which was 1 to 2°72; the latter also gives 1 to 2°67 for French skulls, 1 to 2°80 for German, and 1 to 2°89 for Irish. In a series of 5 male skulls measured by TuRNER the relation of the base line to the are was as 1 to 2°78. In a larger series of 15 the relation was as 1 to 2°8. In a series of 20 male Australian aboriginals, dolichocephalic in type, the mean base line was 139°8 mm. and that of the are 380°4 mm.; the proportion of base line to are being as 1 to 2°72. We observe that the mean arcs in the dolichocephalic West Scottish series and in the Australian aboriginal group are approximately equal, while the difference in the relative pro- portions of base line to arc is mainly due to the longer base line in the lower type, the Australian, and not so much to the slight degree of mean difference in the arc. It was long ago pointed out by CLELAND (15) in his well-known memoir on the skull that “uncivilised or lower forms are in general characterised by the possession of a relatively longer base line, while the length of the arc is very variable.” 358 DR MATTHEW YOUNG. (B) THe Mean Vaturs or Dimensions anp INDICES OF FEMALE SKULLS. Table X demonstrates the mean values of the principal measurements of the. female skulls in TURNER’s series in apposition to those of the 100 female skulls contained in the collection at present under examination, and a brief review and _ comparison of these values shows on the whole a nearer approach to equality — between the two groups of female skulls than between the two male groups already examined. dividuals. TABLE X. Mean Values of Dimensions of Female Skulls. Character. Present Series. | TuRNER’S Series. Rete roup. Cubic capacity ; 1314-5+6-30 1322 1300-5 Glabello-occipital, length . 177-97+-32 178-7 WC Basi-bregmatic height 125-01+:31 126 127 Vertical index 70:31+:20 70-5 71:3 Maximum breadth . 135-16+:29 138 135-7 Cephalic index 76-03+-19 17-2 76:3 Horizontal circumference 503:61+-71 506 501-6 Vertical transverse circumference 408-10 409-6 406 Basi-nasal length 93-82-+ -20 95:3 95-2 Basi-alveolar length 88-614-31 91 91-3 Gnathic index 94-45 94-8 95:8 Total longitudinal circumference 490-13 488-8 489-9 Interzygomatic breadth . 118-16+-26 121-5 mlekeey Nasi-alveolar length 65-834 -24 67 68-5 Maxillo-facial index 55-71 55:1 56-9 Nasal height . 48-41+-20 49-9 50-6 » width 22:56+:11 22-1 21:3 » Index 46:77+-30 44-4 42-2 Orbital width . 37-24+-09 37-4 36:8 » height. 32:95+-13 33-0 33°3 Pr index:. 88-538-+4:35 84:6 90:4 Palato-alveolar length 49-23 51 52-1 - breadth . 56-08 58:3 59-5 5 index 114-16 109-8 114-6 As regards cubic capacity, the difference of the mean values is only between — 7 and 8 cubic centimetres, with the excess in favour of TuRNER’s group; and this: difference, as already explained, may easily be accounted for by the fact that the cubage in the two series has been estimated in a similar fashion by different in- The glabello-occipital length and the basi-bregmatic height in the two groups differ in their mean values by less than 1 mm., with the advantage in favour of TuRNER’s group. The vertical index, obtained from the above two measurements, is practically equal in the two groups. | As regards maximum breadth, the dimension which was notably different in the two male groups, there is a difference in the two mean values, but it is less than 3 mm., and, as in the male series, TuRNER’s value is the greater. TURNER in his A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 359 memoir gives measurements for 11 female Scottish skulls, derived from Renfrew- shire, which give a mean value for maximum breadth of 135°7 mm.—practically equal to that of the eroup under consideration. Although the number of skulls is so small and there is necessarily a large probable error, it is interesting that there should be such an approach to equality. The cephalic index mean value derived from glabello-occipital length and maximum breadth is about one point less than that of TuRNER’s female group, and nearly equal to that of his Renfrewshire subsidiary group. We see that the female groups are more nearly equal as regards this value that the two male groups. In TURNER'S series the values for the male and female groups were nearly equal; in this series the mean value for the female cephalic index is nearly two pomts higher than that of the male, so that, while the mean male skull is dolichocephalic, the mean female skull is mesaticephalic. With regard to the respective circumferences there is little difference in their mean values. The horizontal and vertical transverse circum- ferences of TURNER'S séries are slightly greater than in the present one, while the total longitudinal circumference in the latter series is greater than in the former one ; but in none of the three did the mean measurement of one series exceed the corresponding measurement in the other by more than 2} mm. The mean values of the basi-nasal and basi-alveolar lengths were slightly lower in the present series than in TurNeEr’s, but the mean gnathic index derived from the above measurements was practically equal in the two groups and almost the same as the approximately equal values for the two male groups. As regards nasi-alveolar length and inter- zygomatic breadth the mean values in the present series are slightly less than in TURNER'S, while the mean maxillo-facial index, derived from the above two measure- ments, is practically the same in the two groups, and almost exactly equal to that in the Scottish male group of 405. From Table X we see that, while the mean nasal widths in the two series are nearly equal, the mean nasal height is slightly greater in TuRNER’s series than in the other, and, consequently, the nasal index is slightly lower in value in TURNER'S series than in the one with which it is being compared—the difference amounting to fully two units; a still greater divarication occurred in the male groups, with the excess in favour of the present series. The mean orbital heights are approximately equal in the two series, as are also the mean orbital widths. In TuRNER’s series the mean orbital index is distinctly less, however (to the extent of four units), than the orbital index in the other group—a reversal of the magnitudes of the mean values of the indices in the male groups. It is interesting to note that the Renfrewshire group of 11 skulls has a mean index still further removed from that of TURNER’s complete series than the present group. The values for palato-alveolar breadth and palato-alveolar length are appreciably smaller for the present series of 100 than for TuRNER’s series. The palato-alveolar index, on the other hand, has a mean value for the 100 distinctly greater than that for Turner’s series, the difference being 360; DR MATTHEW YOUNG. equal to fully four units, although the index derived by dividing the mean breadth by the mean length gives a value approximately equal to that given by the present — series. Of course, such a value is only an approximation to the mean, as it differs from the true mean by a quantity dependent upon the variation and correlation coeflicients of the absolute measurements (PEARSON). It is interesting to note ie the mean value of the index in the Renfrewshire group of 11 females is practi the same as in the present series of 100. @ TURNER gives no figures for females alone in the various groupings according to. the value of the respective indices, but takes the male and female together in arriving g at his results. We must therefore combine the male and female groups to get figure es. comparable with those given by TuRNER. Taste XI. Cephalic Index. West Scottish Series. Turwen’s Series. | Class, F. (100). | M. (405). | M.& F. (505). M. & F. per cent. per cent. per cent. per cent. Dolichocephalic ; 34 36 57:3 53 28 (a) 34 31°8 32°2 ome Mesaticephalic : : : ; 54 40°5 43:1 52 (b) 20 86 10°8 wet Brachycephalic ; ; é ; 10 2°2 eels 20 We observe from the above table the distinctly greater tendency towards brachy- female crania together brings the proportions in the various groups slightly, but not to a marked extent, nearer to those obtaining in TURNER'S series. 7 As regards the vertical index we may examine the following table :— a TaBLe XII. West Scottish Series. TURNER’S Series. : Class. SS ed F, (100). M. (405). | M. & F. (505), M.& FL / per cent. per cent. per cent. per cent. Chamezcephalic ; 4 é 46 38°5 40 43 Metriocephalic ; : 5 : 49 54°3 53°2 45 Hypsicephalic : , ; 5 12 67 eel It will be seen that the proportions in the three groups in the case of the female series are not greatly different from those in the male series. In the female have 74 per cent. more in the chameecephalic group, 5 per cent. less in the metrio- A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. cephalic group, and 2 per cent. less in the hypsicephalic group than in the correspond- ing groups in the male. The grouping of the male and female does not sensibly alter the percentage in the hypsicephalic group compared with the male group. As regards the breadth-height index we have the grouping as follows :— Tarte XIII. 361 West Scottish Series. TURNER'S Series. | Class, F. M. M. & F. M. & F. per cent, per cent. per cent. per cent. Hypsistenocephalic . : : 16°5 14 4 Platychamecephalic : ; ; 96 83°5 85°9 96 In Turnepr’s series of 150 the percentages in the two’ groups were exactly the same as those found in the present female series of 100. By associating the male and female percentages together we find that the numbers in the two groups for male and female together are sensibly altered, owing to there being a relatively high pro- portion of hypsistenocephalic skulls in the male group. Passing now to the indices that are derived from the dimensions of the facial skeleton and beginning with the gnathic index, we find in the female series the following grouping, and a comparison with the other series gives the following table :— TABLE XIV. West Scottish Series. TURNER'S Series. Class. j F. (100). M. (399). |M.& F. (499). M. & F. per cent. per cent. per cent. per cent. Orthognathic . 82 17:2 78°1 74:2 Mesognathic . 15 21 19°8 24°7 Prognathic 3 JE 2°2 1 The proportions in the different classes, taking male and female together, are not very different in the two series of Scottish skulls. Grouping the female skulls into the three classes according to the magnitude of the nasal index, we get the following results :— TABLE XV. West Scottish Series. TURNER’S Series, Class. F, (100). M. (399). |M. & F.(499). M. & EF. per cent, per cent. per cent. per cent. Leptorhine 60 67°9 663 75°6 Mesorhine 34 25°8 27-4. ileal Platyrhine 6 6°2 6-2 3°2 362 Analysing the above table, we find that, comparing the percentages in the male and female groups, those in the platyrhine group are equal, while there is about — 8 per cent. less in the female leptorhine group than in the corresponding male group. — Combining the female and male groups, we get figures comparable with those — supplied by TurNER, and we observe that there is 9°3 per cent. less in the lepto- rhine group in the present collection than in TURNER'S series, and of this 6°3 per cent. is found in the mesorhine group and 3 per cent. in the platyrhine group. — With regard to the orbital index, we have for the female the same three group- ings as for the male :— DR MATTHEW YOUNG. TaBLE XVI. West Scottish Series. TURNER'S Series. Class. F. (100). M. (399). | M. & F.(499). M. & F. | a per cent. per cent. per cent. per cent. Microsemic 18 33 30 27-1 Mesosemic 36 37°3 37 27-1 Megasemic 46 29°5 32°8 45°6 in the mesosemic groups are approximately equal, about 15 per cent. less are found — in the female microsemic group than in the corresponding male group, which is com- — corresponding male group. Comparing the present series, taking male and female — together, with TuRNER’s series, where male and female results are aggregated, we find 3 per cent. less in TURNER’S microsemic group ; 10 per cent. less in his mesosemi¢ group than in the corresponding groups of our series, while 13 per cent. is added to the megasemic group. The tendency for the orbit to be relatively high compared — with its breadth is thus better marked in TuRNER’s series than in the present one. As regards the maxillo-facial index we can analyse the following table :— TasLE XVII. West Scottish Series. F TURNER’S Series. Class. SS F, (100). | M. (399). |M. &F. (499). M. & F. per cent. | per cent. per cent. per cent. Leptoprosopic 95:2 95°6 91°1 Mesoprosopic . 3 4:5 4:2 8'8 Chameprosopic Rs "25 “20 a Kven more markedly than TurNER’s series does the present series, including male — A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 363 and female skulls, show the great uniformity in the relatively long and narrow character of the Scottish face. For the series of 100 female Scottish skulls we can construct the following table, according to the magnitude of the palato-alveolar index, subdividing them into three groups as in the male :— Taste XVIII. West Scottish Series. TURNER’S Series. Class. F. (100). M. (400). | M. & F. (500). M. & F. per cent. per cent. per cent. per cent. Hyperdolichuranic . 4 : 15 13°8 14 13°4 Dolichuranic . 3 : 19 24:0 23 24°4 Mesuranic ; : 18 21:3 20°6 18°3 Brachyuranic . f : : 22 20°3 20°6 20°7 Hyperbrachyuranic . : ’ : 26 20°5 21°6 23°1 What strikes one on surveying the above table is the near approach to equality in the numbers of the different groups contained in the various classes. The hyper- dolichuranic class contains a smaller proportion than is contained in the other classes, but the latter are remarkably similar, only slight variations occurring from group to group. : In the series of 100 female skulls the mean value of the opisthio-nasal length, or CLELAND’s base line, is 126°98 mm., which is considerably shorter than the mean length of the base line of 100 male skulls. Im the female series the mean length of the sagittal are is 363°15 mm., and the mean value for the proportion of base to arch in the 100 skulls is as 1 to 2°859, which proportion of base to arch is slightly less in the female than in the male, where the proportion for 100 is as 1 to 2°819. This larger proportion of the arch to the base in the female skull.is held to be the rule by Wetcker (16) and Ecker (17). I shall return to this subject when considering the variability of the skull. In the 100 female skulls | have measured the lengths of the frontal, parietal, and occipital seements of the total sagittal arc, and find the following mean values :— Frontal segment, mean value : . 126°96 mm. Parietal segment _,, me Peas © = Occipital segment ,. 25 : a lA 76 -,, Total are - A 3 SOLON. In the series of 100 the occipital are is greater than the frontal in 8 per cent. of the cases and equal in 1 per cent.; the occipital is greater than the parietal in 20 per cent. and equal in 4 per cent.; the frontal is greater than the parietal in 70 per cent. and equal in 5 per cent. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 52 364 DR MATTHEW YOUNG. | This relationship may be expressed in the form of a table as follows :— TaBLe XIX. Group. Occip. > Frontal. — Occip. > Parietal. Frontal > Parietal. Female (100) —_. | 8 per cent. (=1 per cent.) | 20 per cent. (= 4 per cent.) | 70 per cent. (=5 per cent.) TURNER : aio *5 is} 5 61 A (=4'4 ,; “K ” series (100). | 3 - (=2 e. ) | 23 - (29s ae oe as (= 3 From an analysis of the above table it is easily seen that, comparing the male — series of 100 with the female series, the numbers expressing the frequency of a definite — relationship of the arcs in length to one another are closely similar, whereas the percentages for the three classes in TuRNER’s series (including both male and female) are not markedly different from the male and female groups of the series under ex- amination. The individual variation in the proportional length of the different — regions is very considerable, as will be shown when we consider the variability of © the segments of the sagittal are; and the above table seems to be confirmatory of — CLELAND’s view that it would be “unsafe to deduce any definite relationship between the variation in relative lengths of the ares and the sex of the skull.” CiELanp (15) also says with regard to the relative values of these segments of the sagittal are that — national variation is slight, while Scu#werz (18) says that “ studies on the proportions — of the different bones of the vault to one another have shown that we possess in these — characteristics a useful means of help for the pursuit of racial research”—a view somewhat different from that of CLELAND. It has been asserted that the proportion which the parietal segment bears to the occipital and frontal segments of the sagittal are varies, not only during development, but also to a marked degree in the two sexes. It is stated that the examination of a large number of female skulls goes to show that there exists a lower proportion of both occipital and frontal ares compared with the parietal in the female skull than in the male. HuscHKE, quoted by CLELAND, states that “in the female, the capacity of — both the frontal and the occipital segment is smaller in proportion to the parietal than — in the male,” while WersBacu speaks of the small forehead of the German female. For 100 male and 100 female skulls of this series we get the following mean — values :— Frontal Are. Parietal Are. Occipital Are. Males. . 133°49 mm. 127-91 mm, 120°78 mm. Females . Me ADGFes Fs 12 Ae rae 11476. oe frees aie frontal are _ 1-949 occipital aYe _ 94 parietal are parietal arc frontal are _ 1045 occipital are _ ‘4 In the female . $$ SE parietal are parietal arc A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 365 IT am quite well aware that the index obtained from the mean values of the absolute measurements is not exactly equal to the mean index, as the latter depends to some extent on the coefficients of correlation and variation of the absolute measurements. The above approximate results, however, for female and male, which are practically equal, would seem to show that there is no marked sexual difference in the proportions obtaining between the parietal and the occipital and frontal arcs. I have instituted, in considerable detail, a comparison between the principal mean values of the various dimensions of the male and female groups in TuRNER’S series and the corresponding groups in the series at present under investigation, because it is only in that important memoir that I can obtain material for comparison. While there are many slight differences in the mean values of the dimensions in the two series—differences by no means negligible, but probably partly due to the “ personal equation,’ —the feature that stands out pre-eminently, and in a sense overshadows the minor differences, is the divergence in value in the greatest parieto-squamous or maximum breadth, which is distinct and definite in the female groups, but exception- ally well marked in the male groups. To this, and not to a difference in the greatest length, is due the difference in the cephalic index. There is a marked tendency to dolichocephaly in the series under investigation, whereas, in the series examined by Turner, there is a distinct tendency towards brachycephaly. The diminution in cubic capacity in the present series compared with TuRNER’S series is undoubtedly a real difference, not entirely due to the personal element but dependent largely upon the smaller maximum breadth. In the following table I have arranged the mean indices according to their classes for both male and female groups :— TaBLE XX. Class. Index. Remarks. M. F. Cephalic index . | Dolichocephalic Mesaticephalic (lower divi-| F. show distinctly greater tend- sion) ency to brachycephaly Height index . . | Metriocephalic Metriocephalic Both sexes practically similar Height-breadth index | Platychameecephalic | Platychamecephalic Male nearer hypsistenocephaly Maxillo-facial index. | Leptoprosopic Leptoprosopic Sexes practically alike Orbital index . . | Mesosemic Mesosemic F. index greater (on border of |, megaseme class) Nasal index . . | Mesorhine Mesorhine F. higher index value Gnathic index . . | Orthognathic Orthognathic Sexes nearly alike. M. slightly higher mean index Palato-alveolar index | Mesuranic Mesuranic F. higher index than M. (See Plates I, II, and III for examples of typical male and female skulls from the series. ) 366 DR- MATTHEW YOUNG. 2. EXAMINATION OF THE ScorrisH SKULL BY Biometric MetuHops. Within recent years much attention has been directed to the application of modern biometric methods to investigations into the characters of various collections of crania, and what is described as the “rational” method of treating craniometric — statistics has been evolved. Two of the pioneers in this movement have been Pearson (19) and Garon (20), who, while not devising this method of treating statistical facts (which is due, I believe, to QUETELET (21) as far back as 1846), have — revived and extended the method devised by the latter. Fawcrrr and Lex (22), according to Ropertson (9), were the first. “to apply modern statistical methods to any series of skull measurements,” though Strepa (24) published “the first scientific — determination of the variabilities of the skull.” Others who have taken a notable | part in this movement are MacponNELL (12), Boas (26), TscHEPOURKOWSKY (27), anda Berry (28). Fawcrrr (22) in her paper on the Naquada crania states “that cramio- — metry cannot in future content itself with either the raw measurements, tables of — mere averages, or graphical exhibition of correlation results, but must adopt the methods of modern statistical investigation, tabulating means, variabilities, corre-— lations, and their probable errors in order to draw safe inferences and make racial comparisons”; and further, that “the correlation of the mean values of the chief 4 craniological characters in 50 or 100 races would be a most valuable investigation, — breaking practically untrodden ground” : while MacDoNNELL (12) says, “I venture to think that the chief aim of craniologists at present should be to table means, standard — deviations, and correlations of further long series of skulls” ; and again, “‘ Only when that collection is far more complete will it be possible to state general conclusions applying to the whole range of craniology, and when such tables are formed for 40 or 50 long series we shall have more light not only on intraracial but on interracial — problems.” It is impossible to estimate at its proper value how much we are indebted _ to PEARSON not only for the numerous and important, in fact, invaluable papers he has — contributed himself on the subject, but for the way he has influenced, controlled, and advised various other investigators, in addition to supplying formule which have in ~ many cases made comparatively easy what seemed formerly interminable calculations. Having made a full and complete comparison of the values of the characters of — the male and female groups of our series of skulls with the values in the correspond-_ ing groups in TURNER’ series, adopting TURNER’s measurements and methods of — examination, which the supporters of the new school are pleased to call the | “empirical” method of treating cranial statistics, although none the less valuable on that account, I shall proceed to apply the modern biometric methods to the present series of Scottish skulls. I feel I cannot do better than follow the scheme adopted by Fawcert in her study of the Naquada crania, and also by MacDONNELL in his study of the Whitchapel English crania and other series of English skulls. It will be necessary in the first place for comparative purposes to construct a table with A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 367 the mean measurements of the present series in apposition with the mean measure- ments of various other series to which these methods have been applied. The measurements and nomenclature made use of by these investigators are principally those of the Frankfurter Verstdndigung, and a very full account of the measurements taken and the precautions necessary to ensure accuracy and exactly comparable figures are given in the two papers above alluded to. If further in- formation is required, that can be obtained in the Correspondenz-Blatt Deutsch. Anthropol. Gesellsch., Bd. xiv, 8. 1. The principle underlying the adoption of the German system as expressed in the above volume is, as pointed out by Fawcsrt, to bring any unrecorded series of skulls into relationship with, and into a suitable form for comparison with, as many measured series as possible. I shall content myself therefore in the present paper with merely enumerating the various measurements, only adding a few words of explanation when the method of obtaining the measure- ment differed from that of the above investigators. It is convenient to be able to refer to certain measurements by a reference letter (C, F, L, etc.), as pointed out by the above authors, and I follow their lead also in that respect. The measurements made in the groups of male and female skulls were as follows :— 1. Capacity (C). In obtaining this measurement I made use of the method advocated by TuRNER, using shot as a medium and exercising due precautions.* I have performed the measurements with great care, so that though the mean may be altered by the personal equation the relative capacities of the various skulls have the proper proportions. 2. FLowsr’s ophryo-occipital length (F). 3. Greatest length from glabella to occiput (L). 4. Horizontal length. This measurement was not taken, as both Macponnext and Fawcett agree that little importance is attached to it when the other two measurements of length are given. 5. Greatest horizontal breadth of skull (B). 6. Least breadth of forehead (Bt). This measurement was taken from one temporal crest to the other across the frontal bone. 7. Height of skull (H). Basi-bregmatic height (405 male and 100 female). Frankfurt method (100 male). . 8. Auricular height (OH). 100 male measured on sagittal outline of skull traced by dioptrograph. 9. Basi-nasal length (LB). 10. Horizontal circumference (U,). Measured directly above the superciliary ridges and round the most projecting part of the occiput (FLOWER). Ditto. Circumference measured over the superciliary ridges as adopted by Turner (U,). 11. Sagittal are (S). From nasion over top of head to opisthion, 12. Cross circumference of transverse are (Q). From upper rim of one auricular passage to the other over the bregma. 13. Complete face height. This measurement could not be taken, as no skulls were preserved with mandibles attached. 14. Upper face height (GH). From nasion to alveolar point. * Both MacponneLL and Fawcerr have devoted prolonged research to the various methods, some of them complicated, of estimating the cubage by using various materials and also by formule. MAaAcDONNELL’s conclusion is that different observers using different methods may have an average difference of 20 c.c., and if they use the same method an average difference of 10 c.c. The “personal equation” is thus not a negligible quantity, and is a factor that will always influence the result. MacpoNnNELL says that we must be content with the degree of agree- ment above described at the present stage. 368 DR MATTHEW YOUNG. 15, Face breadth (GB). From the lower end of one zygomatic-maxillary suture to that of the other. 16. Zygomatic breadth (J), From outermost point of one zygomatic arch to that of the other. 17. Nasal height (NH). 18. Nasal breadth (NB). 19. Breadth of orbit for right and left eyes (O!). 20. Greatest height of orbit for right and left eyes (O7): 21. Length of palate (G,). From point of spina-mentalis-posterior to an imaginary surface tangential to. the inner alveolar surfaces of the middle incisors. 22. Breadth of palate (G,). Between the alveolar walls at the second molars MacponNELL says “that — owing to great unevenness of the alveolar walls I have frequently found this a difficult and unsatisfactory measurement.” 23. Profile length (GL). From the basion to the alveolar point. 24. Profile angle (P). Measured in the case of the series ‘‘K” on the sagittal outline projected by the dioptrograph ; in the case of the female by the same method and by the goniometei. 25. Alveolar angle (A). Between GL and GH.) Determined in 100 male by measuring angles on projec- 26. Nasial angle (N). Between GH and LB. \ tion outline, in female by completing the triangles with 27. Basilar angle (B). Between LB and GL. appropriate length for sides and measuring the angle. 28. Basio-nasal-horizontal angle (180°- 66:86 BG FoLL-8Z "103: F.FE-8 '9> o$L-0P ; 0%: = .$F-1F SI: F.68-ZP q> 209-99 6G: F.61-E9 : ZG: F.9E-F9 N> o68°GL o 8G: F.88-S1, 61: F.G8-Eh v> 068 |.00:38 | olf-F8 |.9F-98 [87-98 |.01:68 Ge: F.60-98 0%: F.Z1-L8 d> 6-FOL | €&-GOL | ZL-F6 ~ 2 "+ Ige. 96-46 | 1-96 |Z&- F 18-76 |9T- F80-96 "TY Ze | SL-6E e¢-0r | O@LE | Z-8E | OEE [PZ F 8L-9e 7 80: FPT-LE 9) 0-9 | ZP-6F 08-6G | 1¢-9F | 89-19 | OGFr [es = 12-87 ; 2. Il: F¥-0G 9 18-TE re ge = Qi = cree 91-e¢ |L0- FG0-E¢ 1°0 6-FE | 6L-FE 19-7 | ZEEE | OLE | ZL-Ee |GI- F 9F-EE | L-FE TL-¢g |LO- *g0-E¢ 1°O 09-GF sees, a "* OT. F 66-ZF a F%-6E |GO- F19-8E wo 6-0F | 9G-6E Iver | 12-8¢ | 81-6 | 06-6 IGT. F 90-¢h | T-6E GT-6¢ |G0- F1¢-8¢ «TO 9-GZ | 62-12 ZI-GZ «| SLES | 16% | O8-FZ |LI- F 6Z-7Z | ZZ 8T-E% |90- FLF-EZ aN Gog |2c-6r | 6-87 | BIG | FTG |06-0G loz: F ZI | F-29 €8-1G |IT- FZL-1g HN E-JET | LE-GZL | €9-GZ | OL-OST | EE-ET | OO-SET |LG- F GO-0ST "a FL-8Z1 |S: F9T-LZ0 ui T-ZOL | £996 Gg-c6 | 97-26 | 97-26 | OL-S6 Gh F 18-06 8:06 |9I: FZI-06 rsa) 8:69 | 00:89 6-29 | 62:89 | 9F1L | 08-02 0g: LT-OL G¢-ZL |9T- *96-0L HW G-gze | G3-90€ | Zz-FOE | OBIE | LE-EZE | OL-6SE |ZL- F $6-L0E a €Z-Z1¢ |9E: FOT-TIE 0) Q-ZLE | ZL-L9E | ZO-eLE | 96-G9E | 0Z-L9E | OT-G9E |I8- F IT-LLE : EL-Z8E |SP- F09-81E "9g G-ZZG | 17-80 | ZO-ITG | O%-8TG | Z8-L1S | Ge-PZG |gg- F Gz-FzG | Z-1E¢ 6E:9G |Lb- FEG-22G Fo} 69-62E |LP- F09-92G ise) F-GOL | 96-ZOL | #£:66 | OL:66 | 09-86 | O&-001 |GZ- F 09-10 | FIOL |L3- * 96-001 |€T- FLE-00T qT €-61L | FO-STL | 69-STL | 98-211 GIL | GL-OZL |G¢: * 6¢-F1T "- leg. F LL-811 e HO ZPeFELIel| FF ELLZEL |G-6E1 | 1e-GEl | IZ-SeT | 89-0€T | F6-OGT | SL-EST [PE F FO-ZEL | FSET [Le F TL-SeT |LT- FSL-E1 H 6:96 | 16-46 90-16 | 12:96 | 61:26 | OL-GOT |Gz- * 0-86 ge 8-16 |GI- FZF-96 Ri 9¢-F09-081| FF-F69-9E1 |SV-LFL | GL-Eel | L8-PS1 |TIP-EFT | 88-LPL | LF-OGT |Tg- F 19-071 | 2 6PL |6G- F 2L-OFT |9T- F9G-6ET ¢ 9-F9G-e81| FG-FPP-EST |8-G8T | G8-Z8I | ST-S8I | 96-6LT | 96-6LT | 8G-081 |9E- F 90-68T | 8-98T |6E- F 8I-68T |0%- F9-L8T he eel 96-E81 43 ue "+ |ge. =F Ge-18T re L6-L8T |0G- FLL-G81 ori, S6ZI| (iL) GEZI| OPT] LLP | 0-181 ELFL| LL-S6FT| G-GOGTICL-6+ F6-9LF1) SLFL| 9-8 $-LISI|LL-bFEF-697T eer AG) : . "qostra : ; (Gor) "10949'T ‘URI[eI{SNYy| “UBIUBUUSBY, | “OULy ee “epenben pene i A -Avqiyv ne “MaNAA, met DOD *SOlI0G eouolefory “UBUIIO+) : “TR ce 44 > 178199009 480M! PUB JeZOBIVY) ; paige di A ley be han “payeys Jou Yay “F1qIo euO Wo yuEUT q(2¢- ¥ OL-18) BSL 08-ZL 6S-PL 00-92 e Ih F LGLL ti /*5) O01 Z-G8 CELL I1-S8 09-98 Lg: F 9F-Z8 ¥-68 ag- F eg-s8 | ° u'o/0 oor af . G-0G 1-2 €¢-0G 06-67 ee: F 6L:LF LFF Og F 119% |° HN/AN OOL i 1-19 9¢-0L IL-FL OV PL LG: F F6-LL a oe + GFLL |° @&9/H,D OOT 69: FF9-96 s 99-86 97-88 08-88 FG: F Ge-26 ‘i 6G: FEG-z6 | q/H OOL GG. 46-01 x 1-9) Bue LT-&L 06-EL SI. + €1-69 G-OL OZ: FIg-0L | ° 1/H OOL eP-F9G-eL | opeut apry 9-LL — 6L-FL 81-28 O1-€8 SI: F EL-FL TLL 6l- F COOL, | + an/qxO0T 06811 es * re: FEL-S1 a 0G: FolG6GT | > - = * Spe 60°63 As | mA FG: Fo1T-8Z 2 GI Feheoe “| "> 286-07 e ee Dige ecoe Tee | at SI- F6LZrF f° a> >0F-99 es #8 6% FOLI9 | 7° CG FG879 | * N> 069°C ze < 6G: F.06-EL os To te OrGlee v> oI8 69-8 o€6°F8 28:88 LG: F118 iz SI: F.1988 |: - d> 1-001 20-16. < bee OF: * ZF-06 2-16 LeeE LOG et eros te STEN) £ PLE 18-8 0G-L¢ OL-ZE ¥- F B2-9e x Tisceele OGuile eee + SS) 7, P-1E 68-€¢ 0-0 08-2F QZ: + EL-G¥ a 6leseosy s&h Dp =: 13-Z& a? Re El: F GL-E “4 Leelee aecer | WO . 6-8 9L-ZE 1G-ZE | gree ZI: + 6G-8E FEE El: F F6-ZE eo = 08: TP 2 * FL F G6-0P a C0" Pole i wo > 8-68 OL-LF G6LE ZE-8E SI- F LVIP PLE Ol) fe see foo = L-VG 1€-¥ GO-E% OL-€Z FL: = 61-83 0-2 Ii 9¢¢c |= aN a LP €L-9F PL-GP 0G-8F ZG: F 89-8F 6-6F 0G Iper ie 7 HN a 1-631 FO-LTL FP-PSL 0€-91 8G: F 12-081 9-1Z1 O22 Olelin ie 1 eset s 1-96 1-6 04-88 * 01-68 I? F 98-48 se 7 F60G8 |° * ao G-¢9 78-49 08-49 08-99 OF * 6-89 6-99 we FE8Go9 |° - HD ea T-LIE 97-96 G6-F1E OL-61E IL: ¥ 16-86% a 09°F 1Z660 es A 1-09 F9-E9E 9T-8G8 - OF-EGE $8: F 91-298 ess 99° Fares9e |- = we 1-10 PL-S6F 8G-00¢ O7-10¢ + | Ge F F8-E0G |} 421-908 Tle oe, T9608" 1 2 Pe ae €-001 98-F6 9-86 08-46 FC: F FE-G6 €-G6 0G. Zeee “T= qT 0-S1T IL-€11 89-ZI1 LL-FIL GZ: F 12-601 pe i a2 HO 18-FOG-PSL | 99-FL6-9Z1 | T-GET L¥-631 9-981 10-821 0g: F 9G-FZ1 a [ei TOSEE |. ot eL +26 5-88 06:86 0-96 GG: *F ZIL-€6 % qe. F99-7%6 |° * AS GG-FEL-BZL | FP-FEF-ZEL | 8-981 OG: 1ST 06-ZFL 86-EF1 L& ¥F 89-F81 SLE GeOLene ls) 4 8 ag 89:-F6G-GLT | OLFOO-GLT | S-LLT LP-LLI PL-ZLT GP-ELT Ge: F 98-081 | 6-8LT CA Sous 120 | i a TP-LLT = 7 9¢: F FL-O8T sk sf i (4) GLIT 80ST 6:L8Z1 €8-9E8T 0S-EeT IG-8F 6621 EZET O¢-9F GPIEI]*° ° 9) (uzepoy | : "10949'] ‘uBl[eIysny | ‘UBIUeUISeT, ‘ouly ‘epenbeny (caedimerat NN) oat ena Cause) : (oor) aouerejoy UBULIOS) -qostraAeqaty Jodeyqoogry AA YS19900Q SeLIeg YSI49099 pure qoqowrey) oO 2 o9 ‘ajnwea 7—"TTXX WAV, A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 371 These tables enable one to see at a glance the differences between the mean values of the principal dimensions of the skull as they are exhibited in the Scottish skull of the present series and in TuRNER’S Scottish series and modern advanced types, Bavarian, Wiirtemberg, and French skulls and primitive types, including the Negro, Australian, Tasmanian, and Aino, as well as the Naquada, the representative of an ancient civilisation. I include Australian and Tasmanian for comparative purposes at this stage, to show how they differ from the present Scottish series as shown by the application of the ordinary methods of measurement, because at a later stage I shall institute a com/parison between these types and the Scottish skull by other methods. Male Skulls.—Reviewing the above tables of measurements briefly and taking the male groups first, we find that, as regards capacity, the present Scottish series only differs in its mean value from the Whitechapel English series by an amount which may be wholly due to different observers utilising different methods of measurement. As was previously mentioned, this may account for a difference of 20 ¢.c., which is greater than the difference between the Scottish and English mean capacities. The mean cubic content value of TURNER’s Scottish series is practically identical with that of the English Whitechapel group. The mean capacity of the two German series of crania is distinctly greater, but they each consist of a group of about 100 skulls, and one would like to know if, in the case of the Altbiyerisch group, which was taken from a much larger group, the same influence has not been operative as has made itself apparent in my selection of 100, and with a like vitiation. In selecting male skulls, did they not select skulls with male characters distinctly marked? If so, these are, in general, skulls with a capacity above the mean value for the whole series, in fact, large skulls, and 1503°5 c.c. may not be the mean capacity if they take an average of all the male skulls. The mean value for 100 male skulls, chosen at random from the whole series, except that male characters were as a rule distinct, gives a mean capacity of 1511°3 ¢.c., which compares very favourably, as previously mentioned, with that given for the German Altbiyerisch group. The mean capacity of the French group is practically equal to that of the Whitechapel series, and differs ‘from the Scottish series by an amount which may be again due to the personal equation of the measurer. The mean capacity of the Scottish skull is distinctly greater than that of the Negro, Naquada, Tasmanian, and Australian, but practically equal to that given for the Aino—a very primitive type. The absolute lengths of the various types show distinct differences. There are different methods of taking this measurement : (a) Glabello-occipital maximum lengths. (6) Ophryo-occipital maximum lengths. (c) Horizontal length (7.e. the distance between vertical blocks in contact with the skull in front and behind when it is oriented in the Frankfurt horizontal plane). As has been pointed out by MacponneE.u and Fawoert, there is very little advantage to be gained by recording all these measurements. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 53 372 DR MATTHEW YOUNG. - Parsons (7) favours the ophryo-occipital. length as advocated by FLower, as he 4 says that by this means we avoid the projection due to the frontal sinus. He puts — the other on record, however, the comparison between the two giving one some — idea of the development of that sinus. Still it is remarkable how frequently the — frontal sinus extends upwards in a well-marked form beyond the point where the — extremity of the one arm of the calipers is usually placed in taking the ophryo- maximum length, as is evident from our series of sectionised skulls. He says that — on an average the ophryo-maximum length is about 2 mm. less than the glabello- maximum length in the male Hythe crania and equal in’ the female, and this is confirmed by the Whitechapel series and the two groups of the present Scottish series. I have not recorded the horizontal length. The glabello-occipital maximum mean length in the whole Scottish male series is somewhat below that of the White- — chapel series, but the subsidiary male group of 100 has a mean practically identical with the latter.. The German and French absolute mean lengths are practically — equal, but distinctly less than the two means above mentioned, while the Tasmanian, — Australian, and modern Negroes are all distinctly longer than the German, and also, but to a less degree, than the Naquada and Aino. : As regards greatest breadth, we find that the Scottish series and the Whitechapel — series show practically identical means, while the German and French are notably — higher. The mean value for the Aino is slightly greater; then, in a descending — series, follow the Tasmanian, which is slightly less in mean value, the Naquada and 4 modern Negroes about equal, and less still and lowest of all, the Australian, which has a mean value of about 10 mm. less than the Scottish skull in this group. The influence of the relative lengths of these measurements is apparent in the — cephalic index, #.e. 100B/L. The mean cephalic index in the Whitechapel series is practically identical with that of the West Scottish series of 400, and this relationship — is not appreciably altered if we take the mean of the “K” Scottish series of 100 — specimens. The cephalic index in the French series is definitely higher than the above, and lies at the upper limit of the mesocephalic class, while the two German — series show a mean value well above the lower limit of the brachycephalic class. The Naquada and modern Negroes give a similar mean value, while the Australian — gives the lowest value, dependent to some extent upon its well-marked glabello- occipital maximum length but to a larger extent on its extremely small maximum | breadth. MacponnELu (12), from a comparison of modern English skulls, obtained by reduction of the measurements from large numbers of living individuals, comes to the conclusion that there has been a decided change in the shape of the English — skull during the last two or three hundred years—a change from dolichocephaly as exhibited by the Whitechapel series towards a more brachycephalic type; the length has been decreasing and the breadth increasing. This change, he says, is more marked when we compare English with modern Germans. According to Fawcert — A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 373 (22), a similar change has been taking place in the development of Egyptian skulls. TURNER in his memoir says that in Scotland there is exhibited at the present time “a distinct brachycephalic tendency” ; while Pearson (8) states, with regard to the cephalic index, that “it appears to be a quantity closely associated with degrees of civilisation and capacity for racial survival in the struggle for existence. It is a measure of round-headedness, and, in a certain rough sort of way, round-headedness gives the maximum of skull capacity for the same amount of material.” He is of the opinion that the extra group struggle for existence has gone in favour of the brachycephalic races, and that in most continents we find the mainland occupied by the latter races, while the promontories, outlying borders, and islands are occupied by dolichocephalic races apparently driven out before victorious brachy- cephaly. His final conclusion, from a table comprising a long list of cephalic indices, is that “the dominating and most highly civilised peoples of the world, together with the races from which they have sprung, fall into the brachycephalic portion of the table.” He admits there are exceptions to this general statement, including the Anglo-Saxon and the Whitechapel English. The position of the Anglo-Saxon type he ascribes to the group being mixed with long-barrow British; but “the White- chapel English will still, however, remain a striking anomaly.” It is interesting to be able to range alongside the Whitechapel English another series of skulls, more extensive and derived from an entirely different source, but almost identical in their mean characters with the above series in support of the position of dolichocephaly. It is PEaRson’s contention that the Whitechapel skulls were those of the poorer classes, and that this accounts for the long-headedness, as he suggests that there is a class distinction to be noted in Huropean countries—the higher classes having a dis- tinctly greater tendency to brachycephaly than the lower classes. Such an explana- tion would not hold in the case of the present series, as there is no evidence that the skulls are derived from people in the lower walks of life. It has been asserted that town populations are more brachycephalic than country populations, but the evidence adduced by the present series does not support such a view, as the series—manifestly a town series—is, if anything, less brachycephalic as regards its skull form than the country around. Parsons (7) writes of his Hythe crania that “they entirely fail to substantiate the theory which Dr Macponnetu advances, that during the last two or three centuries a marked change has been going on in the shape of English skulls; that their length has been decreasing while their breadth has increased.” He reminds Dr MacponneEtt that the latter's Whitechapel series cannot be regarded as repre- sentative of the English skull as a whole, but only of the type of the seventeenth- century Londoner and nothing more, and says that “we must realise the fact that we are still in complete ignorance of the skull shapes in other parts of Kngland in the past.” 374 DR MATTHEW YOUNG. With regard to the alleged dominance of the brachycephalic over the dolicho- cephalic type, TURNER writes that “so far as quality of type is expressed by the amount of cranial capacity, skulls of Scotland do not sustain this claim, as those with dolichocephalic proportions were of greater mean capacity than the brachycephalic.” _ In 25 dolichocephalic crania the mean capacity was 1516 ¢.c.; in 13 brachycephalic the mean was 1469 c.c. E Comparing our series “K” of typical male skulls and dolichocephalic with the — Altbiyerisch German group, distinctly brachycephalic in type, we find that the # mean capacity of the dolichocephalic is greater than the mean capacity of the. brachycephalic type—a result which supports TuRNER’s statement. We are re- minded by the latter that “many other factors besides volume of cranial capacity — have to be taken into consideration in the estimation of the intellectual power — either a individuals or of a collection of individuals belonging to the same pea or race.’ The mean heights of the different types in the present West Scottish ‘series, TURNER'S series, the Whitechapel group, and the Tasmanian series are neatly identical—slightly greater than the Wiirtemberg series, slightly less than the — Altbiiyerisch. The French mean value is approximately equal to the Wiirtemberg \ value, with the Australian in close proximity to it. The Naquada and Negro are practically equal, and a little higher than the Altbayerisch, while the Aino shows — the highest value of all. It is interesting to find that the mean value for the Scottish subsidiary “K” group of 100 is identical with that of the a group of 100. The relative values of these mean absolute dimeneane of breadth and height ad well as length are shown in the height and breadth-height indices. As regards the first of these, the full Scottish male series, the subsidiary Scottish series and TURNER'S series are identical, and less than one point above the Whitechapel English ie The Altbiyerisch group is practically equal to the Negro in this respect, about four — points higher than the Scottish, as in it we find, associated with the occurrence of a nearly equal height, a smaller maximum length, while the Wiirtemberg and Naquada — are nearly equal and slightly less in value. The French value is just under one | point less than the last two, the Australian one point under the French, while im the Aino, owing principally to its relatively great height, is highest in value of ¢ i the groups. The height-breadth index (100 H/B) is hypsistenocephalic, 7.e. over 100 in three types, viz. Australian, Naquada, and Negro, and of these the Negro shows the highest value, while the other two are practically equal. This depends principally, in the Australian, on the extremely small maximum breadth, associated with a moderate height, while in the other two it seems dependent to a greater extent on the increase of height, which is associated with a breadth less diminished than in the Australian. The height-breadth indices in the Scottish series and in the Whitechapel series corre-_ A. CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 375 spond very closely, and are separated by a definite and distinct gap from the two German groups which, owing to their marked increase in maximum breadth, have distinctly lower indices. The auricular height in the subsidiary West Scottish series of 100 has a mean value slightly greater than in the English series, and intermediate in value to the two German series. The relationship of the mean values in the various groups can be readily seen and compared by referring to the table for male skulls described above. The basi-nasal length is very similar in the West Scottish and Altbiyerisch groups, slightly greater and practically equal in TurNer’s Scottish and the Whitechapel groups, while the relatively great length characteristic of primitive types is evident in the values for the Negro and Aino. While the upper face heights are approximately equal in the Scottish male group, the English group, and the Altbéyerisch group—a relationship which also obtains between the breadths in the two former groups,—the breadth in the Altbayerisch group is distinctly greater than the height, and so we find in this group of German skulls a smaller index and a relatively broader face than in the other two groups. The mean nasal heights are not markedly different in the four groups, Scottish, _ English, Altbayerisch, and Wiirtemberg, and the same may be said of the respective breadths. As regards the nasal index, the mean value in the Scottish series more nearly equals the Wiirtemberg value than it does the Whitechapel one. These are all leptorhine, while the Altbiiyerisch value is at the lower limit of, but within, the mesorhine group. With regard to the mean value of the orbital width, which for all practical purposes is the same on the two sides, it corresponds more closely with that found in TuRNER’s series and in both German series than with the English series, which is markedly greater, while the value for orbital height is very similar in the same groups. We see this relationship reflected in the orbital index, which is practically equal in the West Scottish male eroup and in the Wiirtemberg group and in the mesoseme class, slightly greater than that in the Altbiiyerisch series just within the mesoseme group, and distinctly less in the English Whitechapel group, which falls into the microsemice class. As regards palate index the West Scottish group is more nearly equal to the two German groups than to the English group. With regard to the profile angle in the various groups we observe that the angle in the West Scottish series, with the excep- tion of that in the Altbayerisch group, shows the highest value of all. It is on the average about 2° less than in the Altbiyerisch group, and about 1° higher in value than the English, Wiirtemberg, and French group angles, which are approximately equal. The angles of the triangle formed by the basi-nasal, nasi-alveolar, and basi-alveolar lines (fig. 1) are shown in the table for the ““K” group of 100 male Scottish skulls, and are not very different from those in the English and Naquada groups, which are the only two other groups where the values are given. The angle 9, is nearly identical in 376 _ DR MATTHEW YOUNG. the three groups above named, while it is slightly less in value in the Naquada than in the other two groups. 3 Female Skulls.—Fewer groups are available for the comparison of the values. of the various dimensions in the female skulls. An analysis of the groups arranged — in Table XXII, however, shows that many of the characters in the various series have a similar relationship to one another as they have im the corresponding male | groups. While the maximum breadth dimension is markedly greater in the German — groups than in the English and Scottish groups, we find that in the last group the proportional values of length and breadth have altered, and as a result the Scottish skull has moved into the mesaticephalic class, while the index for the English skull Nasion. f) ; ALVEOLAR Font ~al ann Basion, 4 ; Fie. 1. F.H. Frankfurt horizontal plane. P. Profile angle. A. Alveolar angle. 6,=(180° —- of GRO | sels) Beate I Wea ee ns Bor |} oieilte! Tasmanian . F ; Briss || ay 7/ 3°49 2°47 3°60 | 4:21 Australian . : ees OAs eo OM morOOmias000) 4:20) | 4586 Scottish (T) . 5 » | &Or } eo | abil jj weal yp eee) |] ais) 382 DR MATTHEW YOUNG. The Scottish male, as represented in the present series, is the least variable as — regards length, with the Naquada and Tasmanian almost equal to it; the English ; male is distinctly more variable, and the Australian shows the highest variation. In the case of the female, as regards length the Scottish group shows the smallest co- efficient of variation, the Australian is next lowest and almost equal to the Scottish, — while all the other types are distinctly more variable. In respect of breadth, the Scottish male is less variable than the English male, which is less than the German; the Tasmanian shows an equal variability with the Scottish skull. In the case of the female, the Scottish skull is less variable = than the English and German but distinctly more variable than the Tasmanian and Australian. As regards height, the English skull is again more variable than the West Scottish, which is practically equal in variability to the Naquada crania and distinctly less so than the Australian. Auricular height I have calculated only for 100 male skulls, — and the variability shown for this measurement is greater in the Scottish than in alles the other races contained in the table, except the Germans. For each of these dimensions, as before mentioned, the male is more variable than the female. j For circumferential measurements we can compile a table such as the following from the figures available :— TaBLE XXVIII. Horizontal Circum- Vertical Circum- ference. ference. Sagittal Are Race. ‘ aaa Glaieeee M F M. F M F Scottish . rr R : 2°69 2:11 3°52 3°02 3°40 2:79 English . : : , : 2°87 2°92 uO) anon 3°63 3°90 Bavarian : : F : 2°86 3°09 Naquada : : ; 2°54 2°27 3°32 2°72 3°19 3°51 Row Grave @annane : : 2°70 2°40 From an examination of the above table we find that, for the horizontal cireum- ference in the male, the Naquada crania show the least variability, the Scottish and Row Grave Germans practically equal variability and sensibly greater than that of the Naquada, while the English and Bavarian coefficients of variation are equal and i distinctly higher than that shown by the Scottish series. In the female, on the other — hand, the Scottish group shows the smallest coefficient of variation; then the Naquada, Row Grave Germans, English, and Bavarian in an ascending degree. With regard to the vertical circumference, we observe that the Scottish groups, both male and female, are more variable than the corresponding Naquada groups but less variable than the English Whitechapel series. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 383 For the variability of the sagittal are the same relationship obtains in the male sex as for the transverse arc, but in the female the sagittal arc is less variable than in the Naquada crania and the Whitechapel English. No figures are available for the sagittal arc and the transverse are in the Bavarian series. Fewer figures are available for the remaining absolute measurements in the various races, but we may compare the following :— TaBLE XXIX. Profile Angle. Standard Deviation. Race. M. F. Scottish . , : 5 ; 3°09 2°70 English P F : 2 ; 3°92 2°85 Bavarian . ; , ; 4 2:79 3°59 Naquada . 3 : i : 2°87 3°66 The English skull shows the highest variability in respect of this angle in the male sex, the Scottish is next in magnitude, while the coefficients of variation in the Bavarian and Naquada are distinctly less and about equal in value. In the female there is a rearrangement of the groups. The Scottish and English are approximately equal in variability, while the Bavarian and Naquada are sensibly more variable and are also about equally so. In the Scottish and English groups therefore the male is, as regards this dimension, more variable than the female, while the opposite relation- ship obtains in the Bavarian and Naquada. Comparing palate lengths and breadths with regard to variability, we have the following table :— TaBLeE XXX. Length of Palate. Breadth of Palate. Race. M H- M F Scottish . : : 6:47 6°29 6°84 577 English. : : 5 ; 5°65 6°53 775 7°68 Bavarian . : : ; : 6:42 6°85 Naquada . : ; : : 6°49 741 9°29 8°55 The above table shows the high degree of variability that holds in all the series named in respect of palate length and breadth. For palate length the Scottish 384 DR MATTHEW YOUNG. male is more variable than the English male, while the female is less variable than the English female. Exactly the opposite arrangement is shown in the values for the variability of palate breadth in these two series of crania. In both male and female the palate length is more variable than the palate breadth in the Scottish skull, while in the English series, in both male and female, palate breadth is more variable than— palate length. The Naquada crania show a very high degree of. variability for palate breadth in both sexes. The variability of palate length in the male Bavarian is practically equal to that of the Scottish male, while the variation in the female 4 greater than the male. The following table shows the variability of certain absolute measures in Scottish, English, and Naquada skulls :— TaBLe XXXI. Scottish. English. Naquada. Reference Letter. Character. M 15} M. F M F B Minimum frontal breadth | 4°63 4:05 4:29 4°55 5:29 4:47 LB Basi-nasal length 4:03 3°31 4-07 4:11 4°88 4°68 G’H Nasi-alveolar length 6:69 557 5-50 714 6:08 6°87 GB ' . | Intermalar breadth 5°42 4:25 5'58 5°40 | 518 4°77 4 ee : . | Interzygomatic breadth 4:22 B15) 4°28 4:13 4:16 4°77 NH Nasal height 6°78 6:19 5:08 5°55 6-13 6°81 NB 5, breadth 7:58 758 8°89 7-06 7:89 7:28 O,L Orbital breadth (left) 4°67 4:17 4:20 3°53 4:97 5°30 ie. : . (right) 4°47 3°97 4°69 4:00 5-02 5°38 O,L. : _ | Orbital height (left) 6°50 6:22 561 4°31 7:06 6:58 Osi ne : » (right) 6°38 6°28 6°65 4:47 C20 6°85 Gi. : | | Basi-alveolar length 519 5-21 4°68 4°95 4°84 5:09 The above table at first sight seems rather complex, but it shows a few notable features. First of all, in the Scottish series the variability for the male is greater than that for the female in all the characters except two, viz. nasal breadth and basi-alveolar length. In the former the variabilities shown by the two sexes are equal; in the latter the female is slightly the more variable. In the English series the female is more variable than the male in several of the characters, namely, minimum frontal breadth, basi-nasal length, nasi-alveolar length, nasal height, and basi-alveolar length. In the Naquada the female is more variable than the male in the following characters ; nasi-alveolar length, interzygomatic breadth, nasal height, orbital breadth (both orbits), and basi-alveolar length. Examining the male coefficients of variation we find that, for minimum frontal breadth, it is greatest in the Naquada and least in the English ; for basi-nasal length it is greatest in the Naquada and smallest in the Scottish; for nasi-alveolar length A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 385 the Scottish series shows the highest value, the English the least, and Naquada is intermediate; for intermalar breadth the Hnglish has the highest coefficient and Naquada the smallest; for interzygomatic breadth the same relationship holds; for nasal height the Scottish shows the highest value, the English the least ; for nasal breadth the English group has the highest value and the Scottish the smallest. For orbital breadth (left orbit) Naquada is most, and English least, variable ; for orbital breadth (right orbit) Naquada is most variable, but now Scottish is least variable ; for orbital height (left) the Naquada is most variable and the English group least; for the right orbital height Naquada is still most variable, but now Scottish is least variable; while for basi-alveolar length the Scottish group is most, the English group least, variable. For the female the relationship of the variabilities is equally changeable, and I think we can represent it best in the following way :— TaBLeE XXXII. : Male. Female. Character. Most Variable. Least Variable. Most Variable. Least Variable. a ; , ; Naquada English English Scottish Bee : , , iS Scottish Naquada . GL. : : : Scottish English English 5 GB. : English Naquada * cl ig Oa : : ; 53 9 Naquada 5 NH. : ; : Scottish English . English NB. j ; : English Scottish English. Scottish Orl . ; ‘ Naquada English Naquada - English O,R. . : 3 a Scottish r Scottish OF. : 5 ; is English 2 English OR. ; 3 : es Scottish 3 a GL Scottish English Scottish Ms The English group is most often least variable in the male skulls as regards those dimensions, and the Scottish is least variable of the three most frequently in the female sex. We have still to compare the variability of the various indices, and material for this purpose is more abundant. The variability is determined in the case of indices by the standard deviation. [ TABLE. 386 DR MATTHEW YOUNG. TaBLE XXXIII. Cephalic Index. Breadth and Length e x 100). Series. M. He Scottish . é 2 : ; 2:53 2°89 English . ; ; : : 3°26 2°98 Bavarian : : ; : 3°50 2°97 French . : : ; B 4°43 4:19 Naquada ; , : : 2°80 3°12 Row Grave Germans d . 2°28 2°35 Aino ; : é ; ; 2°41 2°54 Egyptian mummies . ; ; 3°35 3°36 Modern Egyptians . ; ‘ 5°42 5°10 Negroes . : 5 : : 277 3°52 , Punjabi low caste. ; ; 2-98 3°75 Australian. 5 . : 3°21 3:05 English (upper classes). . 3°03 Of the various series included in the table the Scottish male is less variable than all the others as regards the cephalic index with the exception of two, namely, Aino and Row Grave Germans. The standard deviation in the Scottish male is sponding sexes in the West Scottish series. The following table shows the data for the vertical or altitudinal index :— TaBLE XXXIV. H 100 x T Race, M. F Scottish : : : 2°92 3°00 English 5 ; : 3°22 2°83 French : ; ; oa - 3°67 Naquada : ; 2°73 2°96 Australian . : : 3°39 3°83 From the above we see that in the male sex, as regards the vertical index, the Scottish crania are less variable than the English, French, and Australian, but more variable than the Naquada crania; whereas in the female sex the English female is A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 387 less variable as regards this index, then comes the Naquada, then the Scottish, while the Australian is most variable. The Scottish male is less variable than the female as in the Australian, unlike the relationship which obtains in the English series. The following table represents the variability of the height-breadth index in the various types :— TABLE XXXV. H 100 x =. 00 3 Race. M. F. Scottish . ; : 427 4°49 English . : ; 4:58 3°84 French . ; ‘ 4°74 4°31 Naquada : 4:73 4°66 Australian 3 ] 4°59 4°81 In the male groups, as regards the height-breadth index, the Scottish series is the least variable ; then follow the English and Australian, sensibly different and showing equal variability, while the French and Naquada are still more variable and to an equal degree. In the female groups the least variable is the English series, then comes the French, then the Scottish, and last of all the Australian—most variable, with the Naquada between it and the Scottish. The Scottish male is less variable, as regards the above index, than the female— a relationship which also obtains in the Australian series, while the opposite relation- ship holds in the English, French, and Naquada series. The standard deviations of the remaining indices in the different series are as follows :— TaBLE XXXVI. Upper Face Index or! Masiliealeia Inder: Valbin cas Race. M. F, M. He Scottish . ; : 5:91 4°79 6°33 6°17 English . ‘ : 5:39 6°26 8°39 8-52 Bavarian . : : 3°26 3°33 Naquada . , 4°52 4:15 10°23 8°20 The Scottish male is more variable, as regards upper facial index and also palatal index, than the female; and, as so often occurred in the other characters examined, we find again the opposite relationship holding in the English series. TRANS. ROY, SOC. EDIN., VOL. LI, PART II (NO. 9). 55 388 DR MATTHEW YOUNG. The Scottish palatal index shows a remarkably small standard deviation compared ~ _ with that shown by the English and Naquada crania. | As regards the maxillo-facial index in the male, the Bavarian shows the least — variability, then the Naquada, while the English and Scottish are appreciably more variable. In the female the Bavarian is again the least variable; follow- — ing this comes the Naquada; then the Scottish, while the English female is the most variable. | We have still to examine the orbital and nasal indices, and the following table — contains all the figures available :— TABLE XXXVII. Orbital Index. , Nasal Index. Race. Scottish 5:25 (R.) 5:21 (R.) 4°62 4:49 English . 4°66 (R.) 4:33 (R.) 4°58 3°90 Bavarian . ; : 6°66 5-22, 4°43 4°61 Naquada . 5:00 4°78 4°18 4°86 In the Scottish series the orbital index was not calculated for both orbits, as — there seemed to be so little difference between the mean values of the respective — lengths and breadths. The orbital index in the above table is obtained from the — right orbit. ' a Examining the above table, we observe that the English crania are least variable = in both sexes for the orbital index» In the male the Bavarian is highest in variability, ’ the Scottish distinctly less variable, and the Naquada less than the Scottish. In the — female sex we find the Naquada next to the English in variability of orbital index, — while the Scottish and Bavarian are distinctly more variable than the latter and show — an equal degree of variability. The variability for orbital index is practically equal — for the male and the female in the Scottish series. 4 In respect of the nasal index, in the male sex the Naquada shows least, the Scottish most, variability, while the English series is just a slight degree less variable than the Scottish. In the female the English is the least variable, the Scottish comes next, and the Naquada is the most variable. The variability of the nasal — index is greater in the male than in the female in both Scottish and English series, — while the reverse is the case in the Bavarian and Naquada groups. In this Scottish series, as in TURNER'S, the English, and Naquada series, the nasal index, as measured — by the coetticient of variation, is much more variable than the orbital index, although, as MACDONNELL points out, TURNER says that the latter “shows a great range of variation in the same race, and that it possesses only a secondary value as a race i ia A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 389 character,’ while he says of the former “ that it constitutes one of the most important anthropological characters of the face. The results shown in the foregoing tables with reference to the variability of the different characters in different races are so diverse that it is difficult, in fact sometimes impossible, to find evidence therein for two important conclusions of PEARSON :— (1) That the male is not more variable than the female. (2) That the more highly civilised races are the more variable. - Before PEarson’s essay on “ Variation in Man and Woman” appeared, it seemed to be the general belief—a view supported by Darwin (33)—that the male was more 5) HH ae it || 20 HH iol Zz = ba Ea if a et Bie | | i Eft “lle eee ales fee eee eee eee ae @ : = tol _| aaa ae eee | Dit ies ae] mee | if | Hawa Pepe ee cce ae co ein) | BRR SAWEe mene | | ee (a a es (EN a mee — ae SCECC s : “ag Fe to JIS esta go 165 ho Ins 180 195° 190 igs 290 295° cae deel CEWBTA | TT TTI aa (dbeiael Fic. 2.—Frequency curve for glabello-occipital length in 100 male and 100 female skulls. variable than the female. This belief PEaRson ascribed to the use of non-scientific methods to measure the variability, and, in the course of his paper, furnished numerous instances where, in the most diverse characters, the female is as variable as, if not more variable than, the male; so that his conclusion is, with regard to the characters he has examined for both sexes, that there is “‘ no evidence of greater male variability but rather of a slightly greater female variability.” In the paper above mentioned the measure of significant variation is the coefficient of variation a when the sexual means differ considerably, and the standard deviation when the means are practically identical. As previously mentioned, in the different characters for which the variability has been estimated by these criteria in the present collection, with very few exceptions, the male skull shows greater variability than the female. Respecting the second conclusion, 7.e. that the more highly civilised races are the more variable, we find 390 DR MATTHEW YOUNG. that where measurements for certain characters of the Australian are available for comparison, the coefficients of variation are sensibly greater than in the present Scottish series of skulls in all cases, and than in MacponneELi’s Hnglish series in subseataestaveeti tafe: occ ne ale aT at A HH 3: CI a : iz = a = Lj ee ai ==) ia E c iB a LJ |_| a Pas a7) | A] CANOE Ee ae se SESS S4eR RE @ReS ae! SReeo CSSC VANE SESE DE oe os st |_| See SEWERS Seas & a [4 B ( t =~] [==] |_| B i al Ly |_| Race / SEER EEE Ee ie le Boe Cease RRP ERE Re BS USIARAMEERESSCCGZERUUGusizeeeeeeueeeeeete PEEEEEE EEC EEE EE EECCA atta ne Bee Sees ce eee sees PURARRSEAReeE SL. ROCCO SCE e See eee ees tt 4 ea 4 Case hie Ce Wee) ae bebe) ha iealia REDE SRR STAC Ae Os ie a a a eh, | LS | LTT TT folsdeuo[vsalea} | | | | | | fi cases eel Pees Ah VST Se is Ge Bee PER aE MPS SRNR eee sae aee PES DSRENe AlAs Sessa oe Same BARE SaD See Shirley TRS ZetHePAeee DE PHRASES ARPA Bea eres tI Be Bae RASA Bie Blea o4 Creeueere oLager 2 eon Ae See Ceneeee HN | AGERE "| SERA eeAL aaa ee eee Wo EN eee ee eRe ees | eet rs tol |) Page| || role [ee dei eer a | i] Fic, 4.—Frequency curve for basi-bregmatic height in 100 male and 100 female skulls. most cases. As we shall see later, the same relationship holds in the Tasmanian and Scottish in regard to certain dimensions in Kiaarscu’s scheme for the examination of skulls. The Tasmanian not infrequently shows greater variability than the Scottish — skull for the same character. Are we to regard these in the light of “exceptions — A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 391 that prove the rule,” or has there been any special selection of skulls made in our series, especially as regards the female group? Of such a contingency I am not aware, and the only conclusion I can come to is that there is room for further research as to the relative variability exhibited by the two sexes in different races. The variability of the foregoing dimensions and indices may be graphically repre- ‘sented by constructing frequency polygons; and, from the data given in the tables, it only requires time and patience to calculate the analytical constants of the curves which graphically represent the true nature of the frequency distribution. I have constructed frequency polygons for several of the dimensions and indices representing the variability of these in both sexes (figs. 2 to 6). One of the pro- i pest eteehs | Laas {20} _ | | | ii | iz Cer “lacie ESET AEH TEA cee ee aan amie L i Bee eee SEEAEEE ECE rE etedte pales : EEEEEEEE | | sCeueeeeah acany EEEEEEEE ea RC Vat a E EL cana wi ; Ry Je] A ECE eae 4 eb a | ay |? | linbek —4 | Fre. 5,—Frequency curve for cephalic index in 100 male and 100 female skulls. minent features of these frequency polygons is the irregularly peaked or multimodal appearance they present, which may appear to point to heterogeneity in the series of crania from which the measurements were taken. Both C. D. Fawcerr and Mac- _ DONNELL urge us strongly to be cautious against using this “mountain range” appear- ance of the curve as an argument in favour of heterogeneity, as the peaks may be due to various other causes, among which the chief are :-— (a) That we are dealing in most cases with very small random samples. (b) That some of the skulls are attributed to, and included in, the results of the wrong sex group. (c) That “foreign” skulls are included. (d) That the ages of the skulls in the collection are very varied. (e) That some of the skulls may have been altered by post-mortem changes. (f) That clustering may occur due to family groups included in the collection. 392 DR MATTHEW YOUNG. There are other causes that operate against uniformity in the curve of frequency ; _ and one that is specially noticeable in the curves given at the end of this paper for the indices is the small size of the unit compared with the number of skulls in the series. The indices have been calculated to °5 as well as to whole numbers, which gives “5 as a difference between successive groups. As may be easily seen from the |_| all allel | i il BRS eee PEE EEE PEER PERE eCBBE ANI © Sie Sane PEt 4 Ep Sao SSeS TTT ee cress ne 15 ALE (1 cues 7 see eee eeeeEesnosecloe Seattle Cae eae aie zi - aeana} rE ee - nua Nae ERBGeccc Ha ach ea Sere noe : . a / StiWinGeekeceseee |_| axl |_| ER tH Et a eal ONE LTT TAN IN LAR I tt pt eee ewe TT TNA WW A NNT VN & fi a pNP Ne at EN a pt | BRA Ne (a3) VA 1 BELO ctartirbtt henna 6le | ols: 7 yoy Ww lok | ye | ve | 76 | ¥6 | vy | va | via | elo ree iis aaa eect aeelaal mex tt EH & SEB Re: Fic. 6.—Frequency curve for height index in 100 male and 100 female skulls. curves, many of the indentations would not have been present had one point been the difference between one index group and the succeeding one. According to Fawcert, “it would be quite unjustifiable to argue from a series of peaks as to racial mixtures. Even if like peaks occur in the two sexes for several characters, the sources of hetero-— geneity given above are often ample explanation.” (C) The Correlation of the Characters of the Scottish Skull. Correlated variability is such a relation between the magnitudes of two or more — characters that any abmodality of the one is accompanied by a corresponding — abmodality of the other or others. The length of this series and the remarkable — uniformity of the type of skull represented presents an almost unique opportunity for the study of the correlations of cranial characters. ; As the possible combinations of the various measurements I have made of the different skulls, taking two at a time, would amount to a very large number, in fact some thousands, it is necessary to make a selection of the characters for the purposes of correlation. My choice has been regulated largely by those for which I can find comparative figures in other races, and so resembles closely, in so far as it goes, the list given by Fawcrrr and Macponne.t; which includes those characters of special A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 393 interest. Although the list at present is not so complete as theirs, lack of time and the labour and calculations involved in determining the coefficients of correlation of the pairs of characters selected by the method which gives the most reliable results, a.e. by estimating the sum of the products of the actual pairs of measurement, has made the list smaller than intended. The data for a longer list are provided in the Appendix. I give tables showing the pairs of characters correlated and their coefficients of correlation, with the probable errors of that criterion, and shall proceed to examine the various coefficients and compare them with those that have been calculated for other races. In the first place, we may construct a table (XXXVIII) for length, breadth, and height correlations such as the following :— Taste XXXVIII. Length and Height. Length and Breadth. Breadth and Height. Race. No. M. No. FE. No. M. No. F. No. M. No. F. Scottish . . 405) -356-+-029 |100}-089-+--066 | 405)-454-+-026/100) -187 + -064/405|-264-+-031 |100)-053 + -066 English (W.) . |120) -255-£-058 |117|-425-+--051 | 131]-240-£-055]130| -350-L-052]116)-233-+-059 |115]-340+-056 French fF . . 860} -294---022 |340}-132 +-036-| 860]-089 + -024/340) — -042 + -037/860|-224-+-023 |340|-229-+-035 German { . |100| — -097 + -067*| 99]-314-+--061*| 100)-286---062|} 99) -488---052}100]-072 -.-067*| 99]-276-+-063* English § (criminals) .].. ne ae 58 3000]-402-+--010} .. We | 87|:345---064*) 63)-178 +-082* English § (mid. classes) 0 a ar 1000)-345 +-019} .. fe 129}-273 +-055 |163]-119+-052 Aino||. . | 87) -501-+-054*) 63)-349-+-075*| 87]-432-+--059| 63) -376-+4-073 Naquada 4] . |1384) -489-+--044 |'163)-283-+-048 | 139]-344-+-050/183} -143-.-049 Australian tf .|..| °337-+-:060 | .. |-083-+-:143 | .. |-313-+-069)..| -O16-—--144) .. |-314-+-069 | .. |-311+-130 Tasmanian tf .|..| ‘577+-061 | .. |:178+:119 | .. |-452+-073) ..| -256+-115) .. |-268+--085 | .. |-151+-120 Sioux Indians §§}..| -36 +-08 |.. ot go 4h SECO In the above tables the correlations for length and height and breadth and height are marked with an asterisk in the Aino and German to indicate that they are not comparable with the other results, as the auricular and not the basi-bregmatic height has been made use of; whereas in the Naquada crania the height was measured from the basion to the vault vertically above, but perpendicular to, the German horizontal plane, instead of the distance between the basion and the bregma, which was introduced by Broca and adopted and advocated by TuRNER, FLOWER, and many others. From the median sagittal tracing of 100 male skulls I was able, having also traced the opening of the external acoustic meatus and the + Paris Catacomb Crania. + A. Luz, Phil. Trans., vol. excvi, A, p. 23. § Measurements on living head. || A. Lun, Phil. Trans., vol. cxevi, A, p. 238. gq C. D. Fawcrrr, Biometrika, i, p. 457. tr Roperison, Proc. Roy. Soc. Hdin., vol. xxxi, pt. i. tt Burry, Ropertson, and Cross, Proc. Roy. Soc. Hdin., vol. xxxi, pt. 1, No. 2. §§ Boas, Amer. Anthropol., vol. i, p. 453. 394 DR MATTHEW YOUNG. lower margin of the orbit, to delineate on the tracing not only the line of the — Frankfurt horizontal plane, but also to get the auricular height at right angles to that plane and the distance from the basion to the vault in a plane at right angles — to it. In the 100 male skulls I was able to compare the basi-bregmatic height (Broca’s) with this other height as used in Germany, and was able to convince myself that the two were practically identical in the group examined., In the great majority of the cases, in fact with the exception of a very few, the two measurements — were exactly equal; in some cases there was a difference of a 3 mm., and in only practically a negligible number of cases did the difference in the two heights equal — 1 mm. in amount. So far as the present series goes, therefore, I see no reason why we should not regard the correlations of length and height and breadth and height _ in the Naquada and Scottish crania as truly comparable, the difference produced in 3 the result due to the small difference in the heights being so small as to be of no ~ account. It is worthy of note that Huxuey (29) pointed out many years ago that the difference between the basi-bregmatic height and a more perpendicular height _ was not more than ;oth of an inch on the average. Having obtained the auricular — height by the method above noted, I was able to correlate this auricular height ; with the length, breadth, and basi-bregmatic height in 100 male skulls and thus get — Nee: LEE (13). The following table (XX XIX) shows the comparison :— TABLE XXXIX. Length and Auricular Height Auricular Height and Auricular Height. and Breadth. Basi-bregmatic Height. Race. Son M. M. M. Scottish (100) ch ae 45 +:053 39 +056 694-034 German (100) ; : — 096+ 067 ‘072+ 067 te Aino (87) : Sa -501+-054 345+-064 For the same 100 skulls the other correlation coetticients were Rede Length and Breadth and ; Basi-bregmatic Height. Basi-bregmatic Height. Scottish : : E 35+ 058 19+ "064 diminished in value. Taking the auricular height instead of the basi-bregmatic height to calculate the coetticient of correlation, we find that the auricular shows a distinctly higher correlation, with both the glabello-occipital length and the maximum breadth, — A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 395 than the basi-bregmatic height does, and the increase is more marked in the case of auricular height and breadth than in the other. Examining the above table of correlation coefficients, we find great diversity in their value and relationships. The first feature of the Scottish series that attracts our attention is the high correlation for these characters shown by the male com- pared with that by the female. The above fact is brought into greater prominence owing to the circumstance that, in the English Whitechapel series immediately underneath the Scottish series in the table (XXXVIII), the female is more highly correlated than the male for all these dimensions. In this relationship of the correlation of the sexes for the above measurements the Scottish skull resembles the Aino, the Naquada, the Tasmanian, and the Australian ; the French, in so far as length and breadth and length and height are concerned, also shows higher correlation coefticients in the male than in the female ; while, as regards breadth and height, the two correlation coefficients are practi- cally equal. In the German, on the other hand, the coefficients of correlation in the male for length, breadth, and height are very small compared with those in the female—in fact, in the case of the length and height in the male, the coefficient is negative. Of the three pairs of characters in the Scottish series we observe that length and breadth has the highest coefficient and breadth and height the least, and this holds for both male and female. In the English series the order of magnitude in the coefficients for both female and male is *LH, rLB, and rBH, the first being greatest, but, as before mentioned, the. coefficients for the female are in the case of all the pairs greater than for the male. So far as the results of the Scottish series go they would seem to support one of the conclusions of PEaRson and Les that “correlation is more nearly equal for the two sexes in uncivilised than in civilised races,’ when the Aino is regarded as the type of the uncivilised race; but, in the case of length and height in the Tasmanian, and length and breadth in the Australian, the difference between the male and female coefficients is greater than in the corresponding pairs in the Scottish series. With the second conclusion in PEarson and Lez’s (13) memoir, “ that woman tends with advance in civilisation to gain in correlation on man,” the results supplied by the Scottish skulls appear to be in direct contradiction. According to PEARSON and Len, their conclusions on this subject have been called in question by EH. T. Brewster (35), but they say that his results are not to be relied upon owing to his series being extremely small and his treatment of them not being quite satisfactory. They find confirmation of the suggestions of their paper “On the Relative Variation and Correlation in Civilised and Uncivilised Races” in a “ First Study of the Correla- tion of the Skull” (18). PrEarson and Lur’s chosen representative of the civilised race is the German, and their conclusions are largely based on a comparison of other races with that type. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 56 396 DR MATTHEW YOUNG. From an examination of the table of correlations of length and breadth, length — and height, and breadth and height, we observe that in all cases, with the exception — of length and height in the male German and length and breadth in the French ~ female, the coefficients of correlation are positive ; that in the case of each of these eroups of skulls of different types where one of these dimensions increases, there is a distinct tendency for increase in the other which is correlated with it. In the case of the above two exceptions, moreover, the value of the negative correlation is hardly — significant, considering the magnitude of the probable error; whereas in most of the © other cases the positive coefficients are scarcely, if at all, Semlees altered by the — deduction of the probable error. This series of Scottish skulls affords new proof that there is no scientific founda- tion for the theory usually defined as the “law of compensation.” This was first — enunciated by Vrrcwow (36), is firmly upheld by many anatomists, and, according to — Professor THomson, has an important bearing on the growth of the cranium. . THOMSON (37) writes: “As that distinguished anatomist (VrrcHow) has shown, among skulls belonging to the same type a breadth above the average is compensated for by a height and length below the average.” This, according to THomson, “is _ also associated with a difference in the activity of the growth along the line of the — cranial sutures as exemplified in the types of form described as scaphocephalic, . thyrsocephalic, and trigonocephalic, where each variety is associated with premature — synostosis of the sagittal, coronal, and mid-frontal sutures respectively. But this — principle of compensation seems to act independently of these factors, and appears — to be associated with the poise of the skull upon the summit of the vertebral — column.” The evidence supplied by 405 male skulls and 100 female skulls of the same type is confirmatory of the contention that we do not find between the lengths and — breadths of skulls in the same race a relation which is in accordance with the principle -- spoken of as that of * ‘compensational erowth.” If one increases the other also in- creases. It is otherwise if, according to TscHEPouRKOWSKY (27), we take dolicho- — races, “If we pass from one race to another we find that with a growth of length there 4 is a decrease of breadth, and vice versa. This arises from the fact that the most — brachycephalic races have the greatest breadths and smallest lengths.” Boas (26) seems to favour the intraracial “law of compensation” theory. He hand and height and length on the other.” He also states: “If one of the diameters — differs from the normal in being excessively large, the other will tend to be too — small.” This view is not supported by the correlation of these characters in series of skulls of the same type. —_ eyes A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 397 TaBLe XL. Correlations of Capacity, with Height, Breadth, and Length. Capacity and Height. | Capacity and Breadth. | Capacity and Length. Race. Scottish . - «| 564-024 *39+-05 ‘734-017 634-04 -69+-017 -55+-050 English , . | ‘5014-059 | -600+-049 | -631+-048 | -646+-044 | 5974-051 | -691+-040 | Naquada é - | 6424-043 | -519+-046 | -4344-058 | -532+-044 | -501+-054 | -599+-039 Aino. . . | 5444-054 | 5214-068 | -561+-053 | -502+-070 | -893+-016 | -663+-053 German . | 2434-064 | -451+4-054 | -672+:-037 | -706+-034 | -515+-050 | -687+-037 Sioux . : . | -4402--070 ie 67+-05 “a ‘54+:06 For the 100 male skulls previously mentioned I calculated the correlation _ between the capacity and both the basi-bregmatic and auricular heights with the following result :— Capacity and Capacity and Bac Height. Auricular Height. Scottish (100) ; ‘58 +°045 60+°043 From this table it is evident that the coefficient of correlation between capacity and basi-bregmatic height in the 100 male skulls is practically the same as the coefficient for the larger group of 405 male skulls. The coefficient of correlation between the auricular height and capacity is sensibly greater than between _basi- bregmatic height and capacity, but only amounts to ‘02, with a probable error (the square root of the sum of the squares of the probable errors of the coefficients them- selves) of 06, which is definitely greater than the difference, so that there is no real difference between the coefficients of correlation of capacity and the respective heights, basi-bregmatic and auricular, in this series of 100 skulls. Examining the above table, we find that in the Scottish series in both male and female the maximum breadth and glabello-occipital length have each a distinctly higher correlation with capacity than has basi-bregmatic height, which latter dimen- sion has, as stated above, a correlation with capacity practically equal to that of auricular height. Comparing the Scottish series with the English one, we observe that in the former the three correlation coefficients are higher in the male than the corresponding ones in the female, a condition that obtains in the primitive type, the Aino; whereas in the English series the opposite relationship holds, as it also does in the German. In fact, looking carefully at the table, there is seen to be a very close resemblance between the Scottish male coefticient and the corresponding German female coefficient : were we to change in each German pair the coefficients for the sexes, making the male female in all the three, we would be at once struck with the approach to equality exhibited by the corresponding coefficients. Again, there is a greater difference between the male and female coefficients in the Scottish series than 398 DR MATTHEW YOUNG. ¢ in the Aino series, which, unless my treatment of the series is “not entirely satis- factory,” goes to disprove Pearson and Ler’s second conclusion, namely, “that — woman tends with advance in civilisation to gain in correlation on man.” The comment of Dr Ler (Phil. Trans., vol. exevi, A, 1901) on Dr F. Boas’ (26) — statement to the effect that “the relation between capacity and head diameters is found to be of fundamental importance, and among these the relation of transversal diameter and capacity is more significant,” is that she thinks it is only true for fairly brachycephalic races. He is dealing with 87 Sioux Indian skulls with a cephalic index of 79. Here we have a series of 405 Scottish skulls, with a cephalic index of 74°4 —distinetly dolichocephalic,—and the correlation between capacity and breadth is definitely greater than that between capacity and length or height. It would be comparatively easy from the data supplied in the above tables to construct a formula from the three absolute dimensions of the skull which would give the capacity in the Scottish type. Let us now compare circumferences and capacity :— f TaBLE XLI. C and U. C and Q. C and 8. Race. Scottish F ‘ . | 784-013 -73+-032 ‘764-041 “724-032 ; English : ; . | -82+-026 | -848+-021 | -812+-027 | -744+-034 | -714+-039 | -811+-026 a Naquada . 4 . | 6814-040 | -723+-030 | -656+-042 | -603+-039 a Theban mummies . | 8138+-016 | -826+-033 | -788+-018 | -673+-038 : Examining the above table, we observe that the coefficients of correlation for horizontal circumference and capacity and transverse arc and capacity are in the Scottish series in each case of greater value in the male than in the female, but that the — two male coefficients are approximately equal, and that the two female coefficients are — even more nearly so. The coefficients are smaller in value than the corresponding — coefficients in the English series, except that the coefficient for transverse arc and — capacity is approximately equal in both. There is a close resemblance between the — correlation in the English series and that obtaining in the Theban mummies, while — both show distinctly greater values for correlation coefficients than the Naquada. . In the English series the female is more highly correlated than the male with regard to C and U and C and 8, whereas the opposite relationship is evident in regard — to C and Q. I have estimated the correlation coefficients for some absolute measurements for which I can obtain no comparative data. With regard to the correlations of minimum frontal breadth (B’), we have the following table :-— A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 399 TasBLeE XLII. B’ and L. B’ and B. B’ and H. B’ and C. Race. Scottish -374:056 | -24+.-064 | -39+-056 | -26+-061 | -28+-061 | -23+-064 | -42+-056 Here again we observe that the correlation between the characters in the male is higher than that between the corresponding characters in the female. It is evident that the correlation between minimum frontal breadth and maximum length 1s approximately equal to that between minimum frontal breadth and maximum breadth, while both are slightly greater than that between minimum frontal breadth and height and slightly less than that between minimum frontal breadth and capacity. In the female the three correlation coefficients are practically identical. I calculated these correlation coefficients in the male skulls in order to compare them with the corresponding coefficients obtaining in the metopic male skulls, in the hope that some light might be thrown on the condition of metopism, but I shall refer to this subject in a later paper. It was pointed out earlier in the present communication, when comparing the relative lengths of the three components, frontal, parietal, and occipital, of the sagittal arc, that these varied very much, and that, according to TuRNER and CLELAND, their relative lengths being so variable they could not be regarded with any degree of certainty as having either a racial or a sexual significance. - According to TURNER also, the variability was as great in the brachycephalic type as in the dolichocephalic type. The question as to whether there was a difference in their relationship to one another has engaged the attention of other anatomists, and the idea occurred to me that the correlation of these dimensions might throw some light on the subject. To that end I calculated the correlation coefficients between each segment and the total are in 405 male skulls and in 100 female skulls, and the coefficients of each of the seomental arcs with the greatest length, and the results are not without interest. They are included in the table below :— | TABLE XLITI. Frontal Are and Parietal Arc and Occipital Are and Total Arc. Total Arc. Total Are. Reet: a ees i * SA te M. F. M. F. Scottish (400 M., 100 F.) ‘734-017 | -574:048 | -534+-025 | -544+-050 | -77+-013 | -50+-050 400 DR MATTHEW YOUNG. We observe from the table that, in the case of the male, the coefficients for frontal segment and total arc and occipital segment and total are are greater than the corresponding coefficients in the female; while, in the case of the parietal arc, the © coeflicients for male and female are practically equal, taking into consideration the — greater probable error in the case of the latter. Of these coefficients the greatest of | all is that between occipital are and total arc, this being slightly greater than that for frontal are and total are, which implies a slightly greater tendency on the part of the occipital are to increase corresponding to an increase in the sagittal arc than is shown by the frontal arc. The coefficient of the parietal arc and total are is distinctly less than that of the other two segments with the total arc. I then correlated the frontal, parietal, and occipital ares with the maximum — glabello-occipital length in 100 male (“ K” group) and in 100 female skulls, and the results show a definite relationship between these measurements to obtain in the Scottish series under examination. The results are included in the following table :-— ~~ TaBLeE XLIV. Frontal Are and Parietal Arc and Occipital Are and R Maximum Length. Maximum Length. Maximum Length. | | ace. M F M. F M F Scottish (100 M., 100 F.) .| ‘47+-05 -27+:06 36+:-05 *38+-05 “614-04 -46+:05 | In the first place, the correlation coefficients are in all cases smaller than those — between the segments of the arc and the total arc, as was to be expected. In the case © of the frontal and occipital correlations with the maximum length the coefficients are % distinctly greater in the male than in the female, whereas in the case of parietal are — and maximum length the coefficient for the male is approximately equal to that in the female. As is seen from the table, the correlation coefficients, in the case of the occipital are for both sexes, are distinctly greater than the corresponding coefficients for the other arcs and greatest length. It would not have been so surprising had the 4 parietal arc shown the highest correlation with the greatest length, but it does seem somewhat anomalous that the occipital, whose whole length does not take part in the — formation of that portion of the sagittal are which forms the are of the glabello- occipital chord, should show the highest correlation with the aforesaid chord. Un- — fortunately, there are no comparative figures for other races, but so far as these figures _ go they seem to show that in the Scottish series, when the greatest length of the — skull is increased, there is a definitely greater tendency in the male for the occipital — to increase in proportion; and that the parietal shows, of the three segments, the — smallest tendency to increase under such circumstances. In the female, with increase of glabello-occipital length the occipital are again shows the greatest tendency to increase proportionately, while the frontal shows the least tendency to increase. As_ A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. A401 before mentioned, the tendency of the parietal to increase when the glabello-occipital length increases is approximately equal in the two sexes. Amongst other correlations I have calculated for the Scottish series those for cephalic and vertical indices and cephalic index and capacity, and they are embodied in the table below :— TaBLE XLV. Capacity and Cephalic Index. Cephalic Index and Vertical Index. Race. : No. M. No. F. No. M. No. F. West Scottish . . | 405 091-+-033 97 -154+:065 | 405 | -3014-03 | 100 | -316+-061 6 5 K .| 100 | —-013+-06 we - 100 | -211+-064 German . ; . | 100 -202+-0647 | 97 | —-0307+-0677 Aino : : . | 76 |—-3069+:0701 | 52 | —-2466+-0878 Thebans . : . | 187 | —-1480+-0482 | 84 -0080+-0736 Modern Egyptians .| 56 |—-1410+-0883 | 23 | —-4960+-1060 Male French . .| 56) -1487+-0883 Male Malays . .| 76} -0331+-0773 Etruscans ; .| 78} -2157+-0729 | 38 1443+-1071 Analysing the above table, there seems to be great diversity in the value of the correlation coefficient, so much so that Dr Lex (13), from whose valuable paper most of them have been derived, applies the term “erratic” to describe its unstable and variable character. It will be seen that I give in Table XLV two coefficients of correlation for the male skulls, one for a group of 100 and the other for one of 405. First ‘of all, I calculated the coefficient for the 100 whose mean cephalic index was 74°4, and found that it was negative in value, but that the number was so small compared with its probable error as to be practically negligible. I then calculated the coefficient for the whole 405 male skulls with the same cephalic index, and found that a positive coefficient was the result and that the coefficient was sensibly greater than its probable error. The latter coefficient should be more reliable than the former, and we may take it as definitely representing the correlation between the cephalic index - and capacity in the West Scottish male series rather than the former, which is derived from a smaller number of skulls. The conclusion that one may draw from this positive correlation, even though it is small, is that in this Scottish series a tendency to brachycephaly is associated with — a tendency to greater capacity. There would epee to be some doubt, therefore, in regard to Dr Lez’s conclusion on this subject, i.e. “On the whole, I think, we must conclude that, while there is only a small relationship between cephalic index and capacity, yet that in brachy- cephalic races greater roundness points to greater capacity and in dolichocephahe 402 DR MATTHEW YOUNG. races less roundness points to greater capacity. In either case the emphasis of the racial character denotes an increase of capacity.” Here in the Scottish series we have a dolichocephalic group consisting of at least four times the number of specimens of any series on which Dr Ler’s conclusions are based where, if the accuracy of the calculation of the coefficient is to be relied upon, we have a decrease in the emphasis of the racial character accompanying an increase in capacity. Dr LEE gives figures for three fairly brachycephalic races (3 male and 1 female) in which the correlation is positive, denoting increased capacity associated with increase in the magnitude of the index. Dr L&E admits, however, that in three of the cases the correlation is so small that it is hardly sensible considering the size of the probable errors. There is apparently need, therefore, for revision of the statement that “ there appears to be little doubt that in dolichocephalic races those who possess the race character most markedly have the greater skull capacity.” The mean cephalic index for the male Aino, from which Dr Lrr’s coefficients were calculated, was 76°5, and for the female 77°4. For the female Scottish group of 100 the mean cephalic index was 76'03, practically equal to that of the male Aino and more dolichocephalic than the female Aino; yet we find in the case of the Scottish female a positive correlation between capacity and cephalic index, and in both male ~ and female Aino distinctly negative coefficients. The female German skull, as brachycephalic as the male, shows a negative correlation coefiicient which, however, — is hardly sensible considering the magnitude of the probable error. The results certainly are very erratic, and at present, if the Scottish results are reliable, too unstable to warrant conclusions regarding the relation of brachycephaly and dolichocephaly to capacity. THomson (37) claims to have demonstrated in a most forcible way with his ‘elastic capsule arrangement” the gradual.transition from the dolichocephalic to the brachycephalic type which accompanies increased cranial expansion, and states that — “it seems absurd after this exhibition of the influence of internal pressure to discuss the independent origin of the dolichocephalic and brachycephalic types.” He states a further that “the modification in shape consequent on an increase of the bulk of the — encephalon will tend towards sphericity : in other words, the cranium will become rounder.” This is in direct. opposition to Pearson and LEr’s conclusion, which is, as stated above, that an emphasis of the dolichocephaly in a skull of that ee is associated with increased capacity. In our large male series, however, there is a positive correlation, though it 1s very small, between cephalic index and capacity, which shows that in that series there isa _ very slight tendency towards brachycephaly, associated with increase in capacity. Amongst the skulls that comprise our Scottish collection is one specimen of very — large size, well formed, and remarkable in appearance only in so far that the facial part seems smaller than it normally does owing to the brain-box being so large. It may have been the seat of a hydrocephalic condition at an earlier stage, but of that a . ES ’ San Lee es: ap ps A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 4038 there is no evidence now, and the walls are of average thickness. The norma verti- calis presents an elongated elliptical form and the cephalic index is 74°5. I had not enough shot to estimate the capacity in the usual way, as after the introduction of fully 1950 c.c. there was still left a considerable space to be filled, but I was able to calculate the capacity by using Lexr’s formula, and found it to be well over 2050 c.c. It thus seems possible for a skull from a dolichocephalic group to attain to the above capacity without in any way “tending towards sphericity.” If future stages of evolution therefore should be characterised by an increase in the cranial capacity, it would seem that this end is attainable in a dolichocephalic type of skull without a necessary change in the prevailing form of the latter. There would seem to be some factors exercising an influence over skull form when the capacity is increased which are not demonstrable by the aid of a “mechanical appliance.” The shape of this large skull is proof of such an influence, at least till the skull capacity reaches 2000 c.c. I have also calculated the correlation existing between several of the indices, namely, 100 P with 100 #; 100 # with 100 B; and 100 3 with 100 % As has been pointed out by Prarson (23), even though there is no correlation between the absolute measurements taken in pairs, in arranging them in the above manner and correlating two indices with the same denominator a form of correlation exists to which he applies the term “spurious correlation,” and suggests that this correlation value should be deducted from the whole value or gross correlation obtained by correlating the indices in the usual way to get the real measure of correlation between those indices or what is called “ organic correlation.” Having found it somewhat arduous to calculate the correlation between indices by the method, said to be the most reliable, of multiplying the deviation from the means in pairs and summing the products (which requires to be done over 400 times in each calculation in the male group), and then applying the requisite formula to . obtain the coefficient of correlation, I decided to apply the formula which Kari Pearson has provided for calculating the coefficient of correlation between indices with the same denominator in terms of the coefficients of correlation and coefficients of variation of the three absolute measurements. I had already calculated the coefficient between the cephalic and vertical indices by the previous method, and found, as stated in a former table, that the coefficient had a value in the male of ‘301-°030 and in the female of ‘316-++'061. By the formula above mentioned the values obtained were °302 and ‘317 respectively, quantities practically identical with those obtained by the other method. I applied the formula to find the correlation existing between the = and Z indices and the a and Z indices, and also calculated H H B the spurious correlation values. The results are arranged in the following table, with corresponding values from other races for comparison. TRANS, ROY. SOC. EDIN., VOL. LI, PART II (NO. 9), 57 404 DR MATTHEW YOUNG. Taste XLVI. i: and 3 ; Race. M. F, Na Gross Spurious No Gross Spurious ; Correlation. | Correlation. Correlation. | Correlation. Scottish series . | 405 302-030 428+. -028 100 *317+-061 320+ -061 English _,, || ila -439+-051 -419+-052 108 3914-055 -494+-049 French ; . | 860 -489+-018 -464-+-019 340 ‘576+-025 -ATT+-028 Naquada ,, ; 130 2844-054 -432+-048 166 -371+-046 -438-4-043 | Baw eer it and i Scottish series . .| 405 | -670-+-020 484-025 | 100 | -6564-043 | -6074-043 | English _,, ; || iia! -580+-042 +598+-041 108 -581+-043 559+-044 — French es . 860 -419+-020 -527+-017 340 -417+:030 -541+-026 ‘ Naquada ,, ale ele -595+-038 *603+-037 163 *527+:039 ‘5524-037 | 2 and 2. Scottish series | 405 494-025 484-025 | 100 | -5074-050 | -5144-050 | English _,, ; 114 -4784-049 -478+-049 108 522+:047 ‘4444-052 | French _,, .| 860 | -5864-016 | -5084-018 | 340 | -5034-027 | -482+4-028 | — Naquada ,, : ugh -601+-037 -459+-046 163 -594+:035 -508+:040 | that for gross correlation and sometimes equals, sometimes exceeds, the latter in value — in the Scottish as in the other series. The coefficient of correlation (gross) between B L L French, and Naquada groups, a tendency to brachycephaly is associated with a— tendency to hypsicephaly. It would appear that neither the opisthio-nasal length (CLELAND’s base line) nor the transverse base in a series of skulls is entirely independent of the magnitude of the capacity, as is generally believed. Doar ee being positive in the Scottish series shows that in this, as in the English, In the female series of 100 I calculated the coefticients of correlation between these two base lengths and the cubage and found — — i. < that in the former case the value of the coefficient was ‘38-056, and in the latter. > '26+'061, showing that there is a tendency, in the female series at least, on the part — In the male of both of these diameters to increase when the capacity increases. series of 100 skulls I also found that the coefficient of correlation between the basi- nasal length and the maximum length of the skull was +°60-+-'043, so that there is a distinct tendency shown for the former to increase when the latter increases. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 405 The feature that attracts our attention in a survey of the coefficients of correlation in the foregoing tables, apart from the diversity shown in the values in different races, which is very marked indeed, is the fact that the coefficients between the magnitudes of various skull characters, with a few exceptions, 7.e. principally when cubic capacity is one of the quantities involved, as has been pointed out by MacponneELt, are almost invariably of comparatively small value. When we recollect the close approximation to unity shown by the correlation coefficients of the bones of the extremities—for example: right and left femur, 96; femur and humerus, ‘84; femur and tibia, ‘81; humerus and ulna, ‘75—'86,—this fact is brought into greater prominence. The comparatively low value on the whole of the coefficients in the Scottish series points to the infinite variety of minor differences in, or refinements of, skull form that may exist in such a homogeneous dolichocephalic series as the present one, because it is possible apparently to have an alteration in a certain dimension with only a comparatively slight, though in most cases sensible, tendency to a corresponding increase in another dimension. The divergence shown in the coefficients of correlation between certain skull characters in different types or races shows to what a small extent it is legitimate, in the present state of our knowledge, to apply conclusions drawn from the evidence supplied by one series of skulls to the building up of suggestions or laying down of laws as to what may have been the original form of certain defective parts of pre- existent or fragmentary skulls, which at the present day forms the basis of most of the schemes for the reconstruction of skulls. 3. Kuaatscu’s “ CRANIOTRIGONOMETRICAL METHOD.” Having completed the study of the Scottish skull by the recently introduced biometric methods, in so far as I have been able to do so with the measurements and time available, I have applied the scheme for the examination of skulls devised by Kuaatscu to the “K” series in order to see if any fresh features would be brought to light by its use, and to try to form an estimate as to the value of the method. In this connection I have again found it convenient, in fact necessary, to utilise biometric constants in establishing a contrast between the Scottish skull and other types. The method introduced and made use of by KuaaTscu is not entirely new, as some of the chords and angles included in the scheme have been utilised by other in- vestigators, including CunnincHaM, ScHwaLse, and Turner, and Kriaarscy himself admits that it is merely a continuation or extension of the pioneer research of von LissavER on the “norma sagittalis”; but the complete scheme of examination is distinctly original, and since its introduction has been known as KLaaTscu’s “ cranio- trigonometrical method.” The following is a brief account of the procedure adopted in utilising this scheme :— 406 DR MATTHEW YOUNG. A median sagittal outline tracing of the skull, previously centred and oriented in — the glabella-lambda horizontal plane, is obtained, and on this outline are marked care- : fully the positions of the glabella, bregma, lambda, inion, basion, and prosthion. In this method KiaarscH adopts the glabella-lambda line as his base line, and explains wl at great length in his paper why he does so. This explanation I will discuss later. — Within the outline tracing a quadrilateral figure is devised (fig. 7) with its angles situated at the glabella, bregma, lambda, and basion respectively, and this polygon — is divided by the two diagonals being inscribed into four triangles almost invariably — right-angled at the centre of the figure according to Kuaarscy; as the latter dis- 3 Fic. 7.—Median sagittal tracing of skull No, 26, series K, showing the inserted quadrilateral figure on the sides of which triangles are erected. The values of the various dimensions in this skull are approximately equal to the mean values for the complete series, (For lettering see fig. 8.) covered in his experimental examination of skulls, and as he first explained at the Berlin Anatomical Congress, that the angle which the two lines form with one another _ On the four sides of the quadrilateral figure as bases are erected triangles which are — named the frontal, the parietal, the chief occipital, and the upper facial; the chief — occipital with its vertex at the inion, the upper facial with its vertex at the prosthion, — and the frontal and parietal with their vertices on the corresponding ares at the points on those most distant from the frontal and parietal chords respectively. On the — lambda-inion line as base another triangle is erected with its vertex at the point on — the lambda-inion are most distant from the lambda-inion chord. This triangle is named the adjacent occipital triangle. Perpendiculars are dropped from the vertices A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 407 of the frontal, parietal, and adjacent occipital triangles to the bases of these triangles, and another perpendicular is dropped from the prosthion on the lambda- glabella line, continued forwards, meeting the latter at the point Pr. The lengths of the various lines and chords composing the figure so devised and the magnitudes of the various angles are carefully estimated and noted, and an analysis of the whole system of lines and angles comprises what is known as Kuaatscu’s “ craniotrigonometrical method” for the study of the skull. In applying the method to the group of 100 male skulls I have made use of the dioptrograph to Fie. 8.—Outline of skull No. 13 of series K. This skull shows the highest value for the angle at the zentrum, 7.e. 100°. Note the relatively long glabella-bregma and relatively short glabella-basion chord. G. Glabella. I, II, III, 1V. Inner triangles. Br.Z.G. Zentral angle. Br. Bregma. Vi Frontal triangle. Br.G.Z. Upper glabella angle. Lambda. VI. Parietal triangle. Br.G.Ba. Cranial glabella angle. I. Inion. VII. Chief occipital triangle. G.Br.L. Bregma angle. Ba. Basion. VIII. Adjacent occipital triangle. G.Ba.L. Basion angle, P. Prosthion, IX. Upper facial triangle. Br. L, Ba. Lambda angle. Z. Zentrum. obtain the median sagittal tracings necessary, the skulls being previously carefully oriented and centred. The various points required having been marked on the tracings, the quadrilateral figure, the triangles and the various perpendiculars were drawn as above described. The lengths of the various chords and their segments were measured and the values of most of the angles noted. The angles of the polygon were named by Kuaarscu, each from the point where its apex lay, that no confusion could arise—as does happen in ScHwaLBE’s scheme,—and the position of these angles will be readily understood from fig. 8. From a superficial examination of the method, one would think that the measure- ments are so numerous and so varied that one should be able to make a minute and detailed examination of the difference between any two skulls or skulls typical of 408 DR MATTHEW YOUNG. different races. Several points of interest, as we shall see later, have been brought to — light during the examination of the results. The only large series of skulls to which the method has been accurately applied, and of which the measurements are available, is, so far as I am aware, that of 52 Tasmanian crania by BicHNER (11); of these, about 30 were those of males, and he has selected and recorded 27 different measurements in each skull. It is between the measurements of this group of Tasmanian skulls and those of the series examined by myself that I have instituted a comparison to see how the application of the method will bring out the characteristic features of each type, but I shall also refer to the measurements recorded by Kiaatscu of an Australian, a Huropean, and various other specimens as figured and contrasted by him, and certain measurements of the Tasmanian and Australian given by TURNER in his memoirs, as well as certain other measurements, which I have obtained from FaLKENBURGER (41), from BoxK (42), and from Berry and Roperrson’s atlases (40 and 40a), of dioptrographic tracings of series of Tasmanian and Australian crania. F It will be observed that the types compared—West Scottish, Tasmanian, and Australian—are all of the dolichocephalic type, and the special interest of the inquiry in this and the next section came to be the answer to the question, What are the morphological differences in the cranial vault of a dolichocephalic skull of high status and one of low status in civilisation ? It will be seen that the general shape of the quadrilateral figure inscribed within the more or less elliptical outline of the sagittal tracing of the skull will be influenced by the altitudinal index ; when the minor axis or height of the polygon is diminished in relation to the major axis or length, then the internal angles at its extremities are _ correspondingly increased, while those near the extremities of the major axis are diminished in magnitude. Examining first of all the angles of the figure devised within the outline tracing, and beginning with the zentrum angle, or the angle open upwards and forwards and with its apex at the point of intersection of the diagonals of the quadrilateral, we find that in Bicuner’s series of 25 male Tasmanian crania it varies from 85° — to 93°5°, with a mean value of 89°3°. In 9 of these skulls the angle is exactly 90°, 13 of the skulls show a zentrum angle under 90° in value, and only 3 an angle over 90°. The coefficient of variation of the angle is 1°92. Kzaatscu describes the angle as being almost invariably 90° in the Australian skulls examined by him by this method, and figures one with that value in his paper above mentioned, as — well as a EKuropean skull which shows an angle at the zentrum of the same value. — Werzet (44) did not find the rule invariable in the Australian skulls examined by A him. In 2 very typical Australian skulls examined by me the angle was exactly 90° in each case. I have been enabled recently to estimate the size of the zentrum angle in the larger series of 90 unsexed Australian crania figured in Berry and Ropertson’s atlas A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 409 of dioptrographic tracings, and find that the angle varies in value from 84° to 94°, with a mean value of 89°45°; 19 of the specimens have an angle exactly 90° in value, and 27 have an angle greater than 90°. In BiicHNER’s complete series of 40 Tasmanian skulls, including male and female, like the above Australian group, the zentrum angle ranges in value from 85° to 94°5°, with a mean value of 89°6°; 14 specimens have an angle exactly 90°, and 9 are over 90° in value. Bork (42), from an examination of 70 skulls (of which the race is not stated), has found that the angle which the basion-bregma line forms with the glabella-lambda line varies between 84° and 98°. The top of the variation curve lies at 90° (ve. in 17 cases out of 70). He further states, however, that “the angle formed by the basion-breema and the glabella-inion lines is almost as constantly found at 112° (14 cases out of 70).” In the present series of 100 the angle that occurred most frequently between the last two lines was 107°, and it was found in fifteen instances. The coefficient of variation of the angle was 2°95. FALKENBURGER (41), from an examination of a series of deformed Peruvian and Mexican skulls, was able to convince himself that this perpendicular relationship of the basion-bregma to the glabella-lambda line was prevalent also in such skulls; the variation of the angle occurring in these skulls was about equal to that found in series of normal skulls. In the series of 100 Scottish skulls examined [| find that the zentrum angle shows a range of variation from 87° to 100°, with a mean angle of 92°5°; the angle most frequently found is 91°, and it occurs 21 times. The coefficient of variation of the angle is 2°79. The angle is 90° exactly in 11 per cent. of the skulls, 10 per cent. only are under 90°, while the remainder, or 79 per cent., show an angle greater than 90°. In the Tasmanian and Australian series, therefore, the tendency for the angle to be less than 90° is marked, while in the Scottish series the tendency for the angle to be over 90° is still more evident. A comparison of the coefficients of variation shows that, as regards this angle, the Scottish skull is slightly more variable than the Tasmanian. From an analysis of the measurements in my series of 100 skulls I find that the magnitude of the angle at the zentrum may be considered to depend on several variable factors :— 1. The length of the glabella-bregma chord. 2. The length of the glabella-basion chord. 3. The size of the upper glabella angle. There is a very close relationship between the tendency towards increase in size of the upper glabella angle and the tendency towards decrease in the zentrum angle. The relative lengths of the glabella-bregma and glabella-basion chords have also a close relationship with the size of the zentrum angle, a long glabella-bregma chord associated with a short glabella-basion chord tending to increase the angle, and we 410 DR MATTHEW YOUNG. versa, and a reference to the two diagrams (figs. 8 and 9), showing the two extreme ~ values of the zentrum angle in this Scottish series, will make this evident. In — former is decidedly greater than that of the latter; whereas in the Tasmanian and Australian series of skulls the tendency is for the glabella-basion chord to be longer than the glabella-bregma chord, and the mean of the former is greater than that of the latter. Fie. 9.—Outline of skull No. 78 of series K. This skull shows the lowest value for the angle at the zentrum, i.e. 87°. Note the relatively short glabella-bregma and relatively long glabella-basion chord. (For lettering see fig. 8 The size of the upper glabella angle depends principally on the position of the — bregma, which again depends upon :— 1. The length of the glabella-bregma line (representing the length of the frontal bone), and 2. The slope or inclination of the frontal bone to the glabella-lambda limes factors :— 23 1. The degree of elevation or depression of the bone. 2. The degree of its curvature. 7 He believes that the bregma could be displaced ree so that the frontal — became more or less vertical. The extent of the displacement he measured by his — frontal and bregma angles; the more open the angles, the more was the bone ~ elevated. * It must be noted.that the argument here applies only to KLAatscu’s inscribed figure, the skulls compared being oriented on the glabella-lambda line as the horizontal. In the section dealing with ScHwa.Br’s data, it will be demonstrated that another factor of importance is involved. - A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. All ScHWALBE’s angle, according to Souuas (46), is determined by three factors :— 1. The choice of a base line. 2. The angular position of the bregma. 3. The altitudinal index of the skull. All three combine to give it a low value in the Australian skull and in the Neanderthal group. For Scuwa.pr’s angle or the bregma angle, TURNER gives for Scottish skulls, 6 in number, a variation of 54°-58°, with a mean of 56°; for Tasmanian, 7 in number, a variation of 54°-60°, with a mean value of 57°1°.. KuaatTscH gives for Tasmanians, 8 in number, a variation of 54°-59°, with a mean of 57°3°; Berry gives for unsexed Tasmanians, 45 in number, a variation of from 51°5° to 64°, with a mean of 56°. In this Scottish series of 100 there is a variation of from 50° to 63°, with a mean of 56°33° ; for the Australian KLAATscH gives a variation of from 51° to 62°, witha mean value of 57°5°. The Scottish skull thus shows a lower value than the Tasmanian, according to Turner's figures, whereas, according to the figures derived from the larger numbers, the Tasmanian and the Scottish show a practically identical mean angle, while the Australian shows the highest value. Soizas (46), however, gives for the Australian an angle of 48°3°, which is lower than the Scottish mean angle and is the position one would a priori expect the Australian to occupy, and this is con- firmed. by Brrry’s results (59) for a series of 100 Australian crania (unsexed). In this the bregma angle varies from 49° to 60°, with a mean value of 54°7°. TurNER (47), making use of the nasi-tentorial plane to avoid the variability in the glabella, found in his examination of the angle between the nasi-bregmatic and nasi- tentorial lines that the Australians showed the highest angle, the Tasmanians the smallest, and the Scottish an angle intermediate in value. Berry's dioptrographic tracings have permitted me to estimate the upper glabella angle in 90 Australian crania, and its mean value in this long series was practically equal to that shown in BicHNeER’s complete series of Tasmanian skulls, while the mean angle in the Scottish series was lower in value to the extent of 2°. With amore sloping frontal bone we would have expected, as Kuaatscu believed to be the case, that the angle above mentioned would be smaller in the Australian than in the European (as represented by the Scottish skull). SCHWALBE’S view regarding the elevation of the frontal bone, where its upper border is said to move upwards and forwards in modern, as compared with paleeolithic, man, the resulting displacement of the bregma and more vertical direction of the frontal bone, has been criticised by CunnrncHam (48), who raised several objections to the bregma angle :— 1. Because the glabella was so variable. 2. Because the upper end of the frontal was affected by changes in the other bones of the vault independent of those due to elevation or depression of the frontal bone. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 58 412 DR MATTHEW YOUNG. According to CunnincHAM (48), different degrees of extension of the parietal and — occipital regions, as well as differences in the growth of the frontal area, modify the position of the bregma. The difference in growth of the frontal region, as we shall see later, is brought out in the Scottish series examined by comparing the mean glabella-bregma chord found therein with those found in the Australian and Tasmanian types, and I believe that this difference in growth has an influence on the size of the zentrum angle, while, at the same time, as will be demonstrated later on, growth of the occipital region of the vault is an important factor,’ not only as affecting the zentrum angle, but also the relative positions of the bregma, and of the lambda, and the size of the upper glabella angle. The mean value of the upper glabella angle in the Tasmanian is greater than in the Scottish skull, and the glabella-bregma length is considerably shorter on the average in the former than in the latter. The result of these various factors is that the zentrum angle in the Scottish skull is on the average greater than in the Tasmanian skull. In the Australian skull, as calculated from the series of 90 dioptrographic tracings, the mean glabella-bregma line is relatively shorter than the glabella-basion line, the upper glabella angle is practically equal to that in the Tasmanian, and these factors apparently so balance one another than the mean zentrum angle is approximately equal in the two complete series, namely, 89°45° in the Australian and 89°6° in the Tasmanian. It is interesting to note in connection with this angle that in the skulls of the — anthropoid apes of different ages as figured in median outline by Kuaatscu, with the basion-bregma and glabella-lambda horizontal lines inserted, at an early age in the chimpanzee and orang-utan the zentrum angle is 89° and 90° respectively, thus approaching the angle found in the human subject; the younger the animal the closer is the approximation in the angle, while in the young gorilla it is 84°. With increasing age the basion-bregma line slopes more and the upper end moves forward relatively to the lower end (KuaatscH), so that the zentrum angle is considerably smaller in the adult than at an earlier stage. The explanation of the alteration in the size of the angle is, without doubt, that it is due largely to the displacement backwards of the foramen magnum and so of the basion, the mid-point of its — anterior margin, that takes place in all primates, including man, in their passage from the juvenile to the adult stage, and which is due to the increase in length of the basi-sphenoid and basi-occipital that takes place at that time (Bork). Another factor that undoubtedly influences the size of the angle at different ages — is the fact that, as pointed out by CLELAND (49), ‘“‘the base line continues to elongate after the arch has acquired its permanent characters,” which is fully confirmed by a comparison of the juvenile and adult skulls in the present collection. It is found that the chimpanzee shows the largest zentrum angle when it has reached the adult state and is considerably larger than that found in the gorilla and A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 413 orang. It is generally agreed that of the members of the anthropoid group the chimpanzee stands nearest to man, 7.e. is man’s nearest relative. It would seem therefore that from the relative size of the zentrum angle we may arrange the groups we have considered in an ascending series as follows :— Baile anthropoids { a. Gorilla and orang-utan. b. Chimpanzee. 2. Juvenile anthropoids. 3. Tasmanians and Australians. 4. Scottish. ‘ The tendency for the zentrum angle to be greater than 90° in the last series is apparently evidence of the higher morphological status of the Scottish skull than that of the Tasmanian and the Australian. Berry (50) was at first of the opinion that the Australian stood on the “ evolu- tionary plus” side of the Tasmanian, but has since come to the conclusion (59) that the reverse relationship holds. While there is not a great difference between the mean zentrum angles in the Tasmanian and Australian complete series, it is note- worthy that the angle in the latter is on an average less than in the former. Kuaatscu refers to the angles at the glabella as the most important angles of the quadrilateral figure, and refers to the different values of the two component parts of the complete or cranial-glabella angle, in the Australian and European specimens figured by him. In the Tasmanian series the cranial-glabella angle varies from 69° to 80°, with a@ mean value of 74°7°, S.D. 2°84, V. 3°80. The mean angle in the Australian series of 90 (unsexed) is 74°92°, the range is from 68° to 83°, and the standard deviation is 3°13. In the Scottish series the angle varies from 67° to 80°, with a mean of 74°4°, S.D. 2°72, V. 3°65. The mean values of the angles in the three types are thus approximately equal, while the angle in the Scottish series is less variable than in the Tasmanian, and considerably less variable than in the Australian. With regard to the components of the cranial-glabella angle, we find that in the Tasmanian series the upper glabella angle varies from 34° to 44°, with a mean of 37°9°, S.D. 2°26, V. 5°9; the lower component varies from 31° to 44°, with a mean value of 36°8°, S.D. 2°86, V. 7°77. In the Scottish series the upper component varies from 32° to 42°, with a mean of 35°9°, 8.D. 1°99, V. 5°54; the lower com- ponent varies from 31° to 44°, with a mean of 38°5°, S.D. 2°63, V. 6°84. In the Tasmanian series of skulls the upper glabella angle is thus, on an average, larger than the lower glabella angle, whereas in the Scottish series the reverse is the case ; the upper component of the cranial-glabella angle being on an average smaller than the lower, and so approximating to the condition found in the Australian figured by Kiaatscu, and also to that of an Australian measured by myself. It may be shown thus :— 414 DR MATTHEW YOUNG. le, 33. 2 An Australian skull (Knaatscn) | Bppet nels 4 Lower ,, 40. Narieleaeas Anoenee A neeatereemiiCce = > "N68 erdanites jose Australians (average of 2). oe Europeans (average of 2) ; ; nel oer The number of individuals in the above table is, of course, quite insufficient, but so far as the data go it would appear that, while the mean angle for the Scottish series is less than a degree below that of the Australian aboriginal, the range of variation of the angle is so great that it includes values higher than those given for the gorilla and chimpanzee, and even attains a value less than 2° below that given for the cercopithecus monkey, which is very expressive of the great degree of variability in the extent of flexion in the cranio-facial axis even in a homo- geneous series. | A reasonable explanation for the well-marked variation in the above angle is some- what difficult to find even with the evidence supplied by such a long series of skulls. At first sight it seemed probable that a long basi-nasal chord would be associated with a wide angle, and specimens are obtainable from the series, e.g. fig. 16, illustrat- ing this. On the other hand, fig. 17 shows that it is possible to have another specimen with a basi-nasal length little diminished from that shown in the former figure, but with the angle diminished to the extent of 34°. Figs. 18 and 19 show two skulls with approximately equal basi-nasal and glabello-occipital lengths, but with spheno- ethmoidal angles differing by more than 20°. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 441 If we regard the size of the angle as a quantity which either increases or decreases throughout the series, and correlate it with the basi-nasal length, we find that the coefficient of correlation is + 16+ 065, 7.e. it is positive and sensible but not of great magnitude, so that while there is a tendency for the spheno-ethmoidal angle to in- crease when the basi-nasal length increases, such a tendency is not at all marked. Correlating the greatest length of the skull with the spheno-ethmoidal angle gives a coefficient of correlation ‘053--'067 positive but practically negligible considering the size of the probable error, so that the size of the angle in the series seems to be independent of the length of the skull, although the latter, as has been previously stated, shows a high correlation with the basi-nasal length. Both the middle base Fic, 20.—Impression of median sagittal section of skull, N.B. series F ; cubic capacity 2060 c.c. Spheno-ethmoidal angle, 156°; spheno-maxillary angle, 88° ; foramino-basilar angle, 153°. and the anterior base, 7.e. the boundary lines of the angle above mentioned, vary con- siderably in their degree of inclination to the Frankfurt horizontal plane, and the angle which the anterior base forms with this plane is more variable than that formed by the middle base. Apparently there is some doubt as to the value to be attached to the above angle, as has been stated by DuckworrtH in regard to the foramino-basal angle, and that the former should be used like the latter in comparisons of the Hominide with the Simiidee and lower forms rather than for the comparative study of members of different races in the human family. B. Spheno-Maxillary Angle (a). This angle in the series of 98 varied from 67° to 101°, with a mean value of 84°29° ; the 8.D. was 5°43 and the V. was 6°44. In Turner’s group of 5 the mean value was 86'4°, slightly greater than the mean value shown by our series. DucKWoRTH gives 442 DR MATTHEW YOUNG. for aboriginals of Australia (average of 7) an angle of 92° 20’, and for Europeans (average of 2) an angle of 75° 30’; but it is obvious that the latter value cannot be accepted as an approximate mean value for EKuropean skulls. The spheno-maxillary angle, as defined by Hux.ey, and slightly different from the above angle, was utilised by him to classify crania into orthognathous if the angle was less than 95°, and prognathous if it was more than 95°. The size of the angle is, to a certain extent, an index to the degree of prognathism ; but, as TURNER points out, it is influenced, as is also the spheno-ethmoidal angle, by the variation in the slope of the basi-occipito-sphenoid axis. In the series of 98 sagittal tracings under examination I marked in the Frankfurt horizontal line, and, this being intersected by the basi-occipito-sphenoid axis, | Fic. 21.—Impression of median sagittal section of skull Fic, 22,—Impression of median sagittal section of skull No, 26, series F; cubic capacity, 1040 c.c. No. 21, series F ; cubic capacity, 1000 e.c. Spheno-ethmoidal angle, 167°. : Spheno-ethmoidal angle, 156°. Spheno-maxillary angle, 90°. Spheno-maxillary angle, 88°. Foramino-basilar angle, 143°. Foramino-basilar angle, 153°. measured the angle between the two, which gave me the change of slope of the above axis to the Frankfurt horizontal plane. This angle I have named the “y” angle in my tracings. It was found to vary in the series from 29° to 52°; the S.D. was 4°14 and the V. was 10°29. From the above range of variation, 7.e. 28° in the slope of the occipito-sphenoid axis in a uniform series, it is quite evident that this factor cannot be neglected in the general statement that the magnitude of the spheno-maxillary angle may be used unreservedly as a criterion for the degree of prognathism. From the supposed constancy of the somewhat similar angle between the basi- nasal line and the Frankfurt horizontal plane, THomson and M‘Iver (55), who found that in 38 skulls it varied from 22° to 34°, with a mean value of 27°, made use of it in a scheme for the measurement of prognathism. First (56), however, found that the angle possesses such a degree of variation that a constant size cannot be accepted for it. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 443 C. The Foramino-Basilar Angle (6). This angle may be taken to a certain extent as an index to the degree of slope in the plane of the foramen magnum, not forgetting, however, that the variation of the inclination of the basi-cranial axis will also exert an important influence in modifying the size of the angle as well as variation in the slope of the above plane. The angle in the series of 98 skulls was found to have a range from 135° to 168°, with a mean value of 147°68° ; the 8.D. was 6°36 and the V. was 4°30. The following values are given for the above angle in Duckwortu (54) :— Dae =: : ; ‘ i 5 : 108" Cercopithecus monkey . : 4 cre Simia . ; : : : ; : eal ae Gorilla . : ; ‘ 2 20% Chimpanzee . ' 2 nso" Aboriginals of ene aerate of De mG 30: Europeans (average of 2). ; 2 ; ee L495 30" According to Duckwortu there is some doubt as to the importance to be attached to the angle, and that it is to be used “in comparisons of the Hominidz and Simiide with other families and other orders rather than for the comparative study of various members within the limits of these families.” : While the mean value of the angle in the series of 98 skulls is just one degree greater than the mean angle of two Australian specimens, the latter are too few in number to warrant a conclusion as to its importance, and the same applies to the Kuropean value derived from two skulls. The angle in the Scottish series includes in its range the values given for the Cercopithecus monkey and for Simia. According to TuRNER (2) the angle presents considerable variations in different cerania. In a Fuegian he found it lowest in value, 7.e. 138°; it reached its maximum in an Australian, 157°; while in three Australians it ranged from 139° to 157°. These values are all included in the range of variation exhibited by the Scottish series. - If we consider the spheno-ethmoidal angle as described above as comparable with the orbito-basilar angle defined by CLELanp, then, the mean spheno-ethmoidal angle being greater than the mean foramino-basilar angle to the extent of about 5°, the mean cranial curve in the Scottish series would be less than 180° by that amount, and would approach the lowest value found for adult Kuropeans, according to this author. It would seem, however, that the angles above mentioned can hardly be regarded as strictly comparable. If we accept the value of the spheno-ethmoidal angle obtained by followian TuRNER’s directions as comparable with the above angle of CLELAND, then the two mean angles, foramino-basilar and spheno-ethmoidal, would be equal, the anterior base would be parallel to the posterior base, and the cranial curve would be 180°. It seems somewhat anomalous that, while the cranial flexure of a skull which has TRANS. ROY. SOC, EDIN., VOL. LI, PART II (NO. 9). 62 444 DR MATTHEW YOUNG. a capacity of over 2000 c.c. shown in median sagittal section (fig. 20) should fall short of a value of 180° by about 3°, another specimen (fig. 21) with a cubic capacity of about 1100 c.c. should show a cranial curve of only 156°, while a third specimen (fig. 22), very similar to the last, and of an equal capacity, should show a cranial curve of about 190°. One would have expected that the effect of the increased cranial expansion in the very large skull would have been to bring the anterior part of the cranio-facial axis more into the horizontal than in the case of the smaller skulls, but this is not the case, as is shown by the specimens. The influence of the slope of the plane of the foramen magnum on the parallelism of the anterior and posterior Fie, 23.—Impression of median sagittal section of skull No. 41, series K, illustrating BoLK’s method of defining the position and slope of the foramen magnum. Spheno-ethmoidal angle=151°. F. Fronton, BR= perpendicular dropped on FO from basion. Ben Fa = basal index (Boux). S, Opisthion, < RBS=angle by means of which slope of foramen magnum is measured. base and the resulting size of the cranial curve will be very apparent in the diagrams above mentioned. s Much research has been devoted to the question of the position and the inclina- tion of the plane of the foramen magnum in mammals. Amongst those who have contributed to the subject are Davpentron and Broca, and the angles they utilised are still associated with their names. A more recent investigator of the same problem is Bork (42), and I have adopted the methods advocated by him in investigating these points in the present series of skulls. Bortk does not favour the use of angles for defining position, but prefers the use of an index to indicate the location of the foramen magnum, and he obtains the index in the following way :—Having obtained a median sagittal section of the skull and an outline of this on paper, he defines a point on the inner surface of the anterior cranial wall where it becomes continuous with the floor in the mesial plane, which point he names the “ Fronton” (F in fig. 28). _ A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 445 From this he draws a line to the point on the inner surface of the occipital bone furthest removed from the Fronton, which he names the “ Occipiton.” If there is an are equidistant from the Fronton he makes the Occipiton the centre point of the arc. This Fronton-Occipiton line (FO) is his base line, and from the basion, the middle point of the anterior margin of the foramen magnum, he drops a perpendicular on the line FO, meeting it at R, then the length FR divided by the length FO and multiplied by 100 gives him an index which defines the position of the basion and thus of the foramen magnum. ‘To measure the degree of inclination of the plane of the foramen magnum he joins the point B to opisthion 8, and the size of the angle at B formed by the lines BR and BS is a measure of the slope of the plane of the foramen magnum. I have followed the above procedure in regard to the 98 skulls in the series, and find that the index described above or the “‘ basal index” (BoLK) ranges from 39°5 to 51°5, with a mean value of 45°61, which points to the fact that, as a rule, the anterior limit of the foramen magnum is in front of the middle of the Fronton-Occipiton line ; in fact, there are in the series only two skulls with an index above 50. Bo.x’s important conclusion from the investigation of a large number of specimens is “that in man, as the anthropoids, the foramen magnum is shifted occipitally during development.” He gives values for the index of position of the foramen magnum, or index basalis as he names it, and it is interesting to compare these values with those obtained by me in the Scottish series :— Tasie LIL. Race. Index. Type of Skull. Zeelandians . ; 45°7 Brachycephalic Javanese : : 47:9 Ps Papuans i : 44°6 Dolichocephalic Frisians : ; 44°7 Mesaticephalic Negroes , : 46°6 Dolichocephalic Scottish : : 45°61 * The Scottish mean index is thus seen to be about equal to that shown by the Zeelandian specimens, slightly greater than that shown by the Papuans and Frisians, and less than that shown by the Negro and Javanese skulls. In the case of the Negro group the range of the index is from 42 to 50; in the Scottish it is from 39°5 to 51°5; so that specimens may be found in the Scottish series with the foramen relatively further back than in the Negro, although on the average the opening lies relatively further forward in the former than in the latter. According to Boxk there is “no definite relation between the position of the foramen magnum and the shape of the skull,” 7.e. the cephalic and basal indices are independent of one another. 446 DR MATTHEW YOUNG. There is a distinct interval between the index shown by the human skull in its various types and that of the various anthropoids, which have always an index well above 50, usually at least 60, although in their juvenile forms they may reach as low a value as 52. With regard to the inclination of the plane of the foramen magnum, Boxk (42) 7 proves conclusively in his paper that a change in the slope takes place during de- — velopment both in the human subject and in the anthropoids, but less well marked in the former ; the young anthropoid in this, as in many other features, agrees more closely with the human condition than the adult one. Man occupies a distinct position amongst the primates in that the opening is turned towards the front and not towards the back. | In Bork’s method if the angle is equal to 90° the plane of the foramen magnum is parallel to the base line and is considered to be horizontal ; if greater than 90°, the foramen looks forwards to a degree which increases as the angle increases. BoLkK gives the size of the mean angle for a number of races, and we can compare the angle of the Scottish series with these. In Papuans the angle is equal to. a A019" > In Negroes e A ‘ ‘ - OOD In Javanese F # ; & 99S In Zeelandians _,, os » L023. In Frisians os fig Ba ; »° 96°2° In the Scottish series the mean angle for the 98 was 94°80°, but showed a range __ : in value from 84° to 112°, so that the mean inclination of the foramen magnum was not so much forward as in any of the races for which values are quoted by Botk, although in one case the basal angle was 112° and in a number of others it was over 100°. It is generally believed that in the European the plane of the foramen magnum is directed forwards to a greater degree than in the Negro, but so far as my figures are to be depended upon it would appear that in this Scottish series of skulls the inclination of the opening is directed forwards on the average to a less-marked degree than it is in such a primitive type as the Negro. There is a distinct gap’ between the values shown by the angle in the human skull and that of the anthro- poids, in which the angle varies from 70° to 75°, although in the infantile gorilla an angle of 95° has been found. The smallest angle in the Scottish series was 84°, while two showed an angle of 85°, one an angle of 86°, five were 87°, and four were 88°. There are certain lines measurable on the mesial sagittal section, and we can compare the mean values of those given by TurNneR for the corresponding measure- ments in his series in the form of a table (LIII). A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 447 TaBLeE LIII. TuRNER’s Series. Present Scottish Series (98). Character. Mean. | Mean. 8.D. V. Basi-occipital sphenoid axis, : 65 mm. (5) 66°21 3°41] 5°15 Spheno-maxillary line . : : : 74, (5) 75°39 4°34 5°75 Cribriform axis. ! : ; JO (0) 23°74 3°22 13-56 Basi-lambda line . 4 : ! : ‘ 1186 _,, 120°58 4°58 4:00 Perpendicular radius. : : j 3 138 ,, 135°85 5°28 3°88 Basi-bregmatic length . : : ISB IB Berl 5:47 4-09 Basi-glabellar length . 3 : : ESS, | 10757 4°183 3°88 Basi-nasal length . / ; 103: ,, (5) | 100-96 4°113 4-073 Basi-alveolar length : 95-4" ,, (©) 94°87 4°76 501 Anterior pole distance from perpendicular radius Gis, (®) 84°88 6:91 8-14 Posterior * a Rs Sim () 84:61 5°32 6°28 Sir Witt1am TuRNER, in his investigation into the characters exhibited on the median sagittal section of some specimens from his series of Scottish skulls, pointed out that in them the distance from the perpendicular radius to the anterior pole of the cranial cavity was greater than that from the corresponding radius to the posterior pole of the cavity in all the skulls examined. He points out further that the two distances were more nearly equal in the brachycephalic than in the dolichocephalic specimens, and that he found the posterior pole distance greater than the anterior pole distance in the dolichocephalic Fuegian and other dolicho- - cephalic skulls described in his ‘‘ Challenger” Reports, while in the brachycephalic and mesaticephalic specimens from the same groups the contrary was the case. It would seem, therefore, he says, “as if in the formation of some dolichocephalic skulls the growth takes place backwards behind the plane indicated by the perpendicular radius to a greater extent than in front of the plane.” ; In the series of 98 Scottish skulls | have examined which are dolichocephalic (C.L., 74°4) the mean value of the posterior distance was practically equal to that of the anterior distance, while the latter length was more variable than the former, as shown by the coefficient of variability. As regards the individual skulls, we find that the distance from the perpendicular radius to the posterior pole was greater than that to the anterior pole of the cranial cavity in 44 cases and the distances were exactly equal in 4 cases, which, taken together, is just less than half the number of the specimens examined. There are other features of the skulls which only become obvious after a sagittal section has been made, and these include the extent and great variability in the degree of development of the various sinuses, especially the sphenoidal and frontal. They are seen to vary very much in extent, in the case of the latter from almost or even complete absence to a condition where it extends upwards in a well-marked 448 DR MATTHEW YOUNG. form well beyond a third of the length of the frontal bone, and this not in the skull of an aged person. We also observe, in such a section, the great variability in the degree of distinct- _ ness which the impressions corresponding to the various convolutions exhibit, being sometimes quite evident, at other times practically invisible. Another feature brought to light and shown well in the sagittal tracings is the variation in the thickness of the skull. As is well known, the skull may become much thicker in old age, but the great variations shown in the series examined is remarkable, even excluding specimens that might have become secondarily thickened owing to advanced years. [ estimated the thickness with the calipers at six places, viz. :— ¢ 1. At the glabella. At the inion. . At the middle of the frontal (at the cut edge of the sagittal section). . At the middle of the parietal (at the cut edge of the sagittal section). . In the vicinity of the frontal eminence. . In the vicinity of the parietal eminence. The figures shown in the following table are interesting, as they show that the thickness at the glabella is the least variable, using the coefficient of variation as the criterion of variability, in spite of the variation in the degree of development of the frontal sinus in the series. TasLe LIV. Position of Measurement. Mean Thickness. S.D. V. At glabella : : : F : 15°88 mm. 2°43 15°30 Atinion . E : 14°25 ,, 3°36 23-58 Middle of frontal (mid line) : ; TOT 1:88 26°81 Middle of the parietal oe Tas : 24 5, 1°42 19°61 At frontal eminence . : 560" 4, 1°61 28'5 At parietal eminence . : ; 641. ,, 1°65 25°74 The greater variability of the thickness at the inion is to be attributed largely to the fact that the external and internal occipital protuberances were frequently at different levels, as has been pointed out by various observers, including Sotzas (46), ANDERSON (57), and Kiaatscu (43), so that the two eminences did not always come into the measurement of the thickness. ANDERSON (57) measured the thickness of a large number of skulls, making measurements in twelve different places in each. I have taken his mean measure- ments given to the places where I estimated the thickness, converted them from inches into millimetres, and the following table shows the mean thicknesses compared :— A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 449 TABLE LV. Measurement. ANDERSON’s Series. | Scottish Series. Thickness at inion . ; 9-10 mm. 14°25 mm. About middle of frontal (mid line) : 6°59 ,, COM 5 About middle of parietal (mid pe) : DED Ou; 2a At frontal eminence . ; : Bis) DiOOl es, At parietal eminence . : : : 5:66, 6°41 ,, There is considerable divergence in the mean thickness shown by the two series at the site of the external occipital protuberance or inion; but whether the other measurements may be taken as indicative of a greater mean thickness in the Scottish than in the Irish skull, as the figures given would lead one to believe, would require further investigation. Both Turner (47) and Sottas (46) refer to the relatively great thickness of the cranial wall in the Australian skull. 6. OrIGIN oF THE West ScorrisH SKULL TYPE. In the course of my investigations into the characters of this Scottish series of crania, | was impressed by the remarkably close resemblance they presented in their general proportions to those of illustrations I had seen depicting crania obtained from long barrows in various parts of the country, generally referred to as “ long-barrow crania,’ and to Professor Brycr’s skulls from the chambered cairns of Arran ; especially was this the case in the series of plates in Biometrika, vol. iv, depicting long-barrow crania from the collection of the University of Oxford, described by Scuuster. The present series of Scottish skulls also presented an exceedingly close resemblance in their general form to the series of crania described by MacDONNELL and known as the Whitechapel English crania, which the latter regards as the typical skull of the Londoner of 200 years ago. MacDONNELL was convinced that the skulls which resembled his series most closely were the long-barrow skulls. ScHusTErR (62) gives some measurements for the principal dimensions of the series of long-barrow crania in the museum of Oxford University, and I have made use of some of his measurements and placed the corre- sponding values in the Scottish series alongside for comparison. From the table it is quite evident that the present series of male Scottish skulls in their general pro- - portions are not far removed from the long-barrow crania, and are undoubtedly of the same type. Their difference from TuRNER’S collection of crania containing specimens from the whole of Scotland—although principally derived from the south,—especially in regard to mean maximum breadth, has been already described at sufficient length. 450 DR MATTHEW YOUNG. TaBLE LVI. Male. Female. Reference Letter. Scottish (100), ee a No. | Scottish (100). ene aie: No. L : : : 4 189°18 mm. 190°6 mm. | 16 177:97 mm. 182°6 mm. | 13 oa, Day A el Maree S71). 7 re 184, ne Be en eels oe Saar ~ 1494 , | 18 13516 ,, 1386 , | 19 in acs gee ees 97°48 ,. 989 ., | 16 92°66 ,. 941 ,, amt ci) ce rs leciect. me (37°85 12 125-01 ,, 1351 GE es iets 120-7 kee i (iss ey ere bee Pen etal an (nie Tae 1009-8 |e 93-82 96+8 — 100; Bl, = ; tie 74:4 74:9 oe: 76:03 76:3 123" 100 H/L : : 70°7 BP 70°31 74:0 8 100 NB/NH : : 45°58 49:0 ae 46°77 49°] | P ; ; 87° 83° pape Are the skulls in this series, then, to be regarded as those of descendants of the primitive people whose skulls have been found in the long barrows and chambered cairns, who have remained isolated, as it were, and practically uninfluenced by the influx of invaders of different races who have had such an influence in producing diversity of cranial form in the Eastern parts of Scotland and England ? : I was led to investigate the subject further, and to that end have consulted many contributions of various authorities. We find that, between forty and fifty years ago, THURNAM (63), investigated the characters of the crania of the long-vnambered tumuli, or long barrows, and found there a long-headed or dolichocephalic type of skull, whereas in the round tumuli and short cists of Scotland were the remains of another distinct stock, round-headed or brachycephalic ; and so constantly was this found, that his phrase “long barrow, long skull; round barrow, round skull” may almost be regarded as an unfailing law. He believed that the dolichocephalic type was the forerunner of the brachycephalic type which, having come from Belgium and France, had conquered the earlier type and had in part displaced, in part mixed with it, as appeared from occasional admixture of the two types of crania in round barrows. The later round-headed type, which, accortane to THURNAM, was the Celtic, finally became dominant. It is important to note that the primitive long-headed stock was, and is still, regarded as immigrating from the Iberian Peninsula, and was hence called the Iberian stock. Srei (58) has examined the skulls from these British long barrows and compared them with ancient and modern Mediterranean skulls, and believes that the primitive Mediterranean stock, having occupied the Iberian Peninsula, invaded France and then Britain, constructing tumuli for its dead wherever it took possession. Osteological remains of this Neolithic people are distributed all over Britain from end to end, but are most numerous in the south-west. As Professor Bryce has A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 451 demonstrated from the distribution of their sepulchral monuments, they seem to have entered Scotland from south and west. The round-headed or Alpine (RrpLEy) race entered Scotland from south. and east. The earliest comers buried their dead in short cists, and introduced the beaker type of urn. As has been demonstrated by Rep, Low, and Bryce, they were short and highly brachycephalic in skull form. The later invaders, according to WrieuT, founding on his observations on the skulls from the Yorkshire round barrows, were already mixed before their arrival with a long-headed stock. This round-headed influx is stated to have overrun England and every part of Scotland, but not Ireland. The question arises—Did it reach the west of Scotland, the part furthest removed from the landing-place? Archeological evidence clearly proves that it did, though probably it did not overrun the west to the same degree as the east of the country. There is another point to consider, however. We must not forget the Teutonic invaders from Denmark, Germany, and Scandinavia in the ninth century, who settled down amongst the natives, and descendants of whom we find all over Britain to-day. The head form of these invaders was dolichocephalic, and quite similar to the prevailing type in the parts of the British Isles, where the Mediterranean substratum had maintained its position previous to their arrival. Apparently the collection of crania forming the present series of Scottish skulls may either be representatives of the Mediterranean stock or, on the other hand, may represent those of a colony of Teutonic invaders; they are Mediterranean in their dolichocephaly but, unfortunately, just as much Teutonic in that respect. It must be apparent that it would be very difficult, in fact impossible, to decide such an important question from the evidence supplied by the skulls alone, but the evidence of place-names and physical traits strongly points to the conclusion that in the west the prevailing type owes its origin to the earlier invaders, not to the Teutonic settlers. We therefore conclude that the present series of skulls derived from the Glasgow district are to be regarded as those of descendants of the Mediterranean or Iberian stock, represented in a primitive form in the long-barrow crania, and not as descendants of the more recent Teutonic invaders. They seem to have established themselves in this region at a remote period, and have preserved their skull form practically unaltered. The present series of skulls, as the comparison with TURNER'S series demonstrates, represents only one type of skull prevailing in Scotland. The mean values are not national but local values; but none the less the type represented is probably the great basal national type which has remained rather purer in the west than in the east, where it has been affected to a larger extent by the early intrusive brachy- cephalic element. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 9). 63 2 DR MATTHEW YOUNG. REFERENCES. (1) Turner, Sir Wm., “ A Contribution to the Craniology of the People of Scotland,” Trans. Roy. Soc. Edin., vol. x1, part iii. (2) Turner, Sir W., “ Challénger” Reports : Zoology, ‘Human Crania,” part xxix, 1884. (3) Barrer, Paun, Uber Geschlechtsunterschiede am Schddel, Inaug. Diss., Berlin, 1897 (quoted by ROBERTSON). (4) Extis, Havetock, Man and Woman, Contemporary Science Series, (5) Kiaarscu, Herman, “ Memoir on the Skull of the Australian Aboriginal,” Reports from Path. Lab, of the Lunacy Dept., New South Wales, vol. i, part iii, 1908 (quoted by Roprrrson). (6) Warren, Ernest, “An Investigation on the Variability of the Human Skeleton, with especial reference to the Naquada Race,” Phil. Trans., B, vol. clxxxix. (7) Parsons, F. G., “ Report on the Hythe Crania,” Journ. Roy. Anthropol. Inst., vol. xxviii, 1908. (8) Pearson, eo The Chances of Death, vol. 1. (9) Ropertson, A. W., ‘‘Craniological Observations on the Lengths, Breadths, and Heights of One Hundred Australian Aboriginal Crania,” Proc. Roy. Soc. Edin., vol, xxxi, part 1. (10) Brycs, T. H., Quain’s Anatomy, vol. iv, part i, ‘‘Osteology and Arthrology,” 1914. (11) Bicuner, L. W. G., ‘‘An Investigation of Fifty-two Tasmanian Crania by Kuaatscu’s Craniotrigono- metrical Methods,” Proc. Roy. Soc. Vict., vol. xxv, N.S., part 1. (12) Macpownnett, W. R., ‘A Study of the Variation and Correlation of the Human Skull, with ge reference to sneieh Crania,” Biometrika, vol. ii, p. 191. (13) Pearson, K., and Les, Autcs, ‘‘ Data for the Problem of Evolution in Man,” ‘A First Study of the Correlation of the Human Skull,” Trans. Roy. Soc., A, vol. exevi, 1901. (14) Rieuey, W. Z., he Races of Europe, 1900. (15) Cretanp, Jon, (a) “An Inquiry into the Variations of the Human Skull, particularly in the Antero-Posterior Direction,” Phil. Trans. Roy. Soc. Lond., vol. clx, 1870; (0) “On the Form of the Human Skull at Different Ages and in Different Nationalities,” Memoirs and Memoranda in Anat., 1889. (16) Wetcker, H., Areh. f. Anthropol. (quoted by CLELAND). (17) Ecxrr, Arch. f. Anthropol. (quoted by CLELAND), (18) Scuwerz, Franz, ‘‘ Untersuchungen iiber das Verhiltnis von Frontal-parietal und Occipitalsehne zur Schidelbasislange,” Arch. f. Anthropol., N.F., 9. . (19) Pearson, Karu, ‘‘ Mathematical Contributions to the Theory of Evolution, Regression, Heredity, and Panmixia,” Phil. Trans., vol. clxxxvil, A. (20) Gatton, Francis, Biometrika, vol. i, also National Inheritance. (21) Quersier, Lettres sur la Théorie des Probabilités, 1846 (quoted by Ropgrrson). (22) Fawcrrt, C. D., and Len, Auicz, “ A Second Study of the Variation and Correlation of the Human Skull, with sehoiul reference to the Naquada Crania,” Biometrika, vol. i, p. 408. (23) Pearson, Karu, Proc. Roy. Soc., vol. 1x. (24) Sriepa, “Uber die Anwendung der Wahrscheinlichkeitsrechnung in der Anthropologischen,” Statist. Arch. f. Anthropol., Bd. xiv. (25) Kurrn, A., “A Description of a New Craniometer,” etc., Journ. of Anat. and Physiol., vol. xliv. (26) Boas, Franz, “ Heredity in Head Form,” Amer, Anthropol., vol. i. (27) TscuerourKowsky, E., ‘Contributions to the Study of Interracial Correlation,” Biometrika, vol. iv, p. 286. (28) Berry, R. J. A., and Ropertson, A. W., ‘A Biometrical Study of the Relative Degree of Purity of Race of the Tasmanian, Australian, and Papuan,” Proc. Roy. Soc. Hdin., vol. xxxi, part 1. (29) Huxury, T. H., ““On Two Widely-contrasted Forms of the Human Cranium,” Journ. Anat. and Physiol., vol. i. (32) Pearson, Karu, “ Homogeneity and Heterogeneity in Collections of Crania,” Craniological Notes, | Biometrika, vol. ii. (33) Darwin, Cuas., Animals and Plants under Domestieation, vol. ii. A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 453 (35) Brewsrer, E. T., Proc. Boston Soc. Nat. Hist., vol. xxix. (36) VircHow, thon den Cretinismus, etc.,” Gesumm. Abhandl. Frankfurt, 1856 (quoted from ToPrnarD by Professor THomson). (37) THomson, A., “‘A Consideration of some of the More Important Factors concerned in the Pro- duction of Man’s Cranial Form,” Journ. Anthropol. Inst. Lond., N.S., 6. (38) Cross, K. S., “On a Numerical Determination of the Relative Positions of certain Biological Types in the Evolutionary Scale, and of the Relative Values of various Cranial Measurements and Indices as Criteria,” Proc. Roy. Soc. Edin., vol. xxxi, part i, No. 4. (40) Berry, R. J. A., and Roserrson, A. W., “ Dioptrographic Tracings in Four Norme of Fifty-two Tasmanian Crania,” Trans. Roy. Soc. Victoria, vol. v, part i, 1909. (404) Berry, R. J. A., and Rospertson, A., W., ‘‘Dioptrographic Tracings in Three Normz of Ninety Australian Aboriginal Crania,” 7’rans. Roy. Soc. Victoria, vol. vi, 1914. (41) Fatkenstreer, F., “Zur Craniotrigonometrie,” Anthrop. Gesellsch., Nos. T-12, 1912. (42) Bok, L., (a) “On the Position and Displacement of the Foramen magnum in the Primates,” Proc. Kon. Akad. Amsterdam, No. 12; (b) “On the Slope of the Foramen magnum in the Primates,” as above. (43) Kuaarscu, H., ‘‘ Kraniomorphologie und Kraniotrigonometrie,” Arch. f. Anthropol., N.¥., 8. (44) Werzet, G., “Die obersten Halswirbel und die Beziehung ihrer Ebenen zu denen des Schiadels beim Menschen und den Anthropoiden,” Zettschr. f. Morph. u. Anthropol., Bd. xiii, Heft 1, 1910. (45) Scuwa.pg, G. (quoted by Turner in 47, 0), (46) Sotnas, W. J., “On the Cranial and Facial Characters of the Neanderthal Race,” Phil. Trans. Roy. Soc., B, vol. excix, 1907. (47) Turner, Sir Wo., (a) ““The Craniology, Racial Affinities, and Descent of the Aboriginals of Tasmania,” Trans. Roy. Soc. Edin., vol. xlvi, 1908 ; (b) “The Aborigines of Tasmania,” T’rans. Roy. Soc. Eidin., vol. xlvui, 1910. (48) Cunnincuam, D. J., “ Anthropological Essays presented to Professor TyLor” (quoted by TuRNER in 47, 6). (49) Kuaarscu, H., “Das Gesichtsskelet der Neanderthalrasse und der Australier,” Verhandl. anat. , Ges., Berlin, 1908. (50) Berry, R. J. A., and Ropertson, A. W., “The Place in Nature of the Tasmanian Aboriginal as deduced from a Study of his Calvarium,” Proc. Roy. Soc, Edin., vol. xxxi, part 1 (No. 3). (51) Huxury, T. H., Man’s Place in Nature. (52) ScHwa be, G., “Studien iiber Pithecanthropus erectus, Dubois,” Zeitschr. f. Morphol. u. Anthropol., BO, L899, (53) Scuwauss, G., ‘ Der Neanderthalschiadel, Bonner Jahrbucher, Heft 106, 1901. (54) Duckwortu, L. W. H., orphology and Anthropology, Cambridge, 1904. (55) THomson, A., and M‘Ivsr, R., The Ancient Races of the Thebaid, Oxford, 1905. (56) Furst, C. M., “ Kiniges iiber anthropologische Winkelmessungen und iiber ein Instrument fiir ~ Winkel- und Indexbestimmungen,” Zeit. f. Morph. u. Anthrop., Bd. ix, 1906. (57) AnpeRson, R. J., “Observation on the Thickness of the Human Skull,” Dublin Journ. Medical Science, vol. ea 1882. (58) Ssrei, G., The Mediterranean Race. (59) Berry, R. J. A., and Rogperrson, A. W., ‘The Place in Nature of the Tasmanian Aboriginal as deduced from a Study of his Calvaria,” Proc. Roy. Soc. Edin., vol. xxxiv, part 1, 1914. (60) Bicuner, L. W. G., “A Study of the Prognathism of the Tasmanian Aboriginal,” Proc. Roy. Soc. Vict., vol. xxv, N.S., part i. (61) Bicuyner, L. W. G., “A Study of the Curvatures of the Tasmanian Aboriginal Cranium,” Proc. Roy. Soc. Edin., vol. xxxiv, part i, 1914. (62) ScuustEr, E. H. J., ‘The Long-barrow and Round-barrow Skulls in the Collection of the Depart- ment of Comparative Anatomy, The Museum, Oxford,” Biometrika, vol. iv. (63) THurnam, J., “On the Two Principal Forms of Ancient British and Gaulish Skulls,” Mem. Anthropol. Soc. Lond., vol. i. 454 A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. — DESCRIPTION OF PLATES. | eh Plate I. A typical male skull from five different aspects: norma lateralis, norma facialis, ‘no: occipitalis, norma verticalis, norma basalis, : ‘Th Plate II, A typical female skull from five different aspects: norma lateralis, norma facialis, no occipitalis, norma verticalis, norma basalis, 7 ih: > aa Plate III. A, A typical male skull: norma lateralis and norma verticalis; B, a typical female skull: norma lateralis and norma vertiecalis. a rans. Roy. Soc. Edin. Wein, Jul. Dr MatrHew Youne: “A Contribution to the Study of the Scottish Skull.”—Puates I. " s | ~ he Vs “y . ~ a . r » Trans. Roy, Soc. Edin. Vou, LI. Dr Marruew Youne: “A Contribution to the Study of the Scottish Skull.”—Pratw II. ~ Trans. Roy. Soc. Edin. Vou. LI. Dr Marruew Youne: “A Contribution to the Study of the Scottish Skull.”—Purare III. — 4 APP TN Dix: MEASUREMENTS AND INDICES OF INDIVIDUAL SKULLS. TRANS. ROY. SOC. EDIN., VOL LI, PART II (NO. 9) fe Trans. Roy. Soc. Edin., Vol. LI.) SERIES “K.” Number. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 81 32 33 34 35 36 87 38 39 40 41 42, | Age | Sex 5 2 : : M. M. M. M. M. M. M. M. Cubic capacity 2 < 1930 | 1580 | 1450 1560 | 1880 } 1410 1520 | 1500 Glabello-occipital length 198 | 190} 195 188 | 181 189 200} 186 Ophryo-occipital 197 | 188} 193 187 | 181 187 198 | 185 Maximum breadth 5 152) 140) 135 WO} 141 145 136} 141 Minimum frontal breadth 101 98 96 102 93. 92 95 95. | Basi-brezmatic height 189 | 136) 124 M0) 133 135 136 | 138 Opisthio-nasal length 137} 144] 134 142] 184] 138 142} 131 : 102) 109 96 104 9B | 104 107 | 101 92) 100 91 101 92. 95. 97 91 73 12) 72 75 65 70. 70 73 55-7 | 56:6 | 58:5 56:8 | 524) 51-4 62:6 | 59:3, matic breadth 131} 127) 123 132] 124) 136 133 | 123 ‘y-facial index (b) 80:2 || 79:1 | 86-7 824 | 75:5 | 80-4 76-9 | 77-6 Intermalar breadth c 91 91 83 91 86 87 91 OF Tnter-stephanic breadth 135] 122))) 112) 9] 117} 115 17) 119) 4 ionic diameter 119} 110) 106 117} 109} 116 115} 111 117 122 108 115} 113} 123 123} 115 g 36 38 39 39 36 36 33. Nasal height - 55 53. 50 54 56 AT 5b | ., breadth 23 21 23 23 23 23 25 | Orbital height (R.) 36 33. 31 33 35 33 31 , (L) 35 bre 31 32 35 PJ Bl breadth (R.) 40 39) 40 40, 40 Al 37 A 3 _{005)) 43, 39 40 39 40, 41 37 Palatal length (F.) 50 51 51 50 5D 50 49, 49 index 70:0. 64-7 | 80-4 | 72-0 61:8 | 79:1 | 80-0 81-6} 71-4 breadth F 35 33 41 36 BL 38 40 40 85 length (TURNER) 53 55 53 54 56 49 50, 54 52 Palato-maxillary index - 109-4 90-9 | 122-6 | 103-7 100-0 106:0 116-6 | 109-6. | Palatal breadth 50. 65 56 63 57 | Horizontal circumference 563 | 5380) 537 543 | 525 E 563 | 528 | 526 531} 520 3:00 | 2:58 | 2:82 278)|) 2:93, 412} 372} 388 395 ) 385 Frontal sezment 145 | 130) 137 130 | 182 Parietal 133,| 127) 126 141 138, Occipital ,, 1st} 115) 125 124) 115 Transverse arc 344) 310 | 301 302 | 317 Cephalic index 76:8 | 73-7 | 69:2 68:0} 75:8 Height index : 70:2 | 71-6 | 63:6 68:0) | 74:2 ght-breadth index 91-4 | 97-1 | 91:8 100-0 | 97-8 Nasal index 41:8 | 39:6 | 46:0 48:9 | 45:5 Orbital 90:0 | 846 | 77-5 80:5 | 83:8 | Alveolar,. - A 9 3 906 90-2 || 91-7 | 948 90-7 | 90-1 Spheno-ethmoidal angle 2 - | 150° | 155° | 156° | 148° | 158° | 158° 16°} 142° 153° | 153° | 152° 151° | 149° Spheno-maxillary ,, 3 -| 79°} 82°) sf) Boe] 84°) 86° 83° | 85° 82°!| 87° | 82° 85° | 82° | Foramino-basilar ,, 2 143° | 140° | 160°} 153°) 145° | 145° 150° | 136° 153° | 139° | 156° 147° | 143° Dz M. Yousc—(i) ot 53 54 57 M. 1620 192 189 10 93 181 136 100 00 83 66:0, M. 1590 197 196 116 02 129 139 101 99 12 57:6 125 87:8 82 118 109 118 30) b2 23 38 33 86 56 49 67:3 33 53 1094 58 out 543 2:80 390. 138 129 123, 515, vel 65:5 88:3 4-2, 91-7 98:0 152° 85° Tot? L— M. 1550 187 187 call ca a 67 | 68 85 | 86 | 91 | 92 | 93 | 94 M. | M. | 1690 | 1650. 9 | 198 188 03 194 193} 188 Ii 6 OL 12 14 | 95) 95] 105 | 102 | | WO) M3 12 130 6 Vt 192 | 105 | 108 08 oT 100 o 7 68 ts OL8 | | 606 HOO | He7 | 127 156 120 N00 68:0 | 74:9 02 | bn 100, 2 119 120} 120 | 4 M1 |) 112 109 | 121) 110 86 | 38 3b 51 50 | 28 24 ot 58 b wh 53 9 | 39} 40 | 30 40 5 iM) be 175 | 75:0 38 30 m2 53 2 110:0 | 2, | 124-0 | 120-7 | 107-5 | 1211 55. O4 OF 63 517 oA 635 | 578) 600 516 533 | O74 | Sil 2-98 300 | 2-92 2-99 | 2-84 | 200 387 103 400 386) 395} 409 | 398 135 b 10 128} 10} 186) 142 140 WO) 135 133} 188} 154) 123 112 | 131 125 } 126) 117) 199) 133 306 312 | 324 | 922) 820) 335 | 3% 75:0 TOL 73:6) 736 | 76:6 | 735 | 76 717 be 722 | 7A-1 73:6 | 74-9) 67-2 | 121 95:6 | G14 | 92-2 102-9 | 100-7 100-0 | 96-5 | 91-3 47-9 | 46-2) B74 466 500 549 | 4B-0 | 180 82-1) 87-5 | 725) 89-7 | S46 87-2 | B25 | 85-0 | 948 98:0) 97-1 | 93-2 951 | 92:6 | 929 | 9641 144° | 159° | 157° | 140° | 137° 153° | 150° | 163° 85° | 90°] 95°) 76°) 79°] 83°) sf° 88° | 83° | 95° us 154° | 151° | 157° 17° | 142° | 145° | 155° 145° | 147° | 1687 | 151° | 1 Trans, Roy. Soc. Bdin., Vol, Uf.) SERIES “0.” Number. 16 Age Sex c Qubic capacity . Glabello-occipital length Ophryo-occipital ,, Maximum breadth Minimum frontal breadth Basi-bregmatio height Opisthio-nasal length Basi-nasal Basi-alveolar Nasi-alyeolar Maxillary-facial index (a) Interzygomatio breadth Maxillary-facial index (b) Tntermalar breadth Tater-stephanic breadth Asterionic diameter Transverse base 5 c Length of foramen magnum . Nasal height . breadth Orbital height (R.) 3 xi (Cd) ,, breadth (R.) » » (Li) Palatal length (I'.) index breadth length (TuRNER) PalJato-maxillary index . Palatal breadth Horizontal circumference Ophryo-occipital circumference Arch to base ratio Sagittal arc Frontal segment Parietal on Occipital _,, Cephalic index Height index Height-breadth index Nasal index Orbital ,, Transverse are Alveolar index Tt, 1260 176 175 137 O4 125 129 95 89 58 18:7 119 69:5 83 118 Ny 1330 17 7 138 87 126 128 0D 86 68 B81 50. 1080 OL 504 i. 1340 179 125:0 503 284 120 133 117 743 743 100-0 47:8 86:5 305 87-5 361 128 119 1l4 19:2 m9 90:7 42:9 1000 308 92:7 353 122 116 115 13-7 68-0 92-2 47-8 83-3 288 103-3 505 369 135 125 109 TA0 724 97 10-0 89-5 2 87-0 58] 59 619 | 505 2.99 | 2-91 386 | 367 M44} 125 130| 113 112} 129 75-4 | 73:3 73-2 | 68-3 97-1 | 93+ 42-6 | 46-0 B16 | 868 316 | 207 | 90-4 | 1043 R 1020 187 188 189 oT 124 131 9 05 66 600 116 7125 91 122 107 106 82 ol 20 wu Bot 40 40 60 76:0 68 02 100:0 21) 21 288 378 125 125 128 44, 663 BOD 510 860 312 960 D 1300 5:05 573 16 Da M. Youxo—{2), 59 60 61 62 63 64 65 66 67 68 69 | 70 1 72 D. KF. F. Ky F. F, 1p F. 250) 1300) .. 1380 | 1230} .. 1320 1300 181} 189) 178} 179) 176) 175) 179 179 180} 189) 178} 178) 176) 175) 179 179 133] 185] 135) 139) 185) 131} 136 132 94 93 92 93 92 97 93 95 131} 120) 120] 118} 120) 122) 1242 127 130} 185} 132) 120) 123} 133) 129 133 99 98 95 87 92 95%) 949 99 95 90 95, 80 86 912) 922 912 60 66 68 66 65 61 63 5 5 67 18:7 | Bil |) 601 | 60:0) 54:6 | 53:0) 53:8 |) 52-8 | 57-1 | 52-0 | 52-4) 57-7) 56:0) 55-8 128))) 122)} 113)|| 110)) 119)}) 115%) 17 || 125)) 126) 121 |) 124) 128) 116) 120 BY | 77:6 || 83:9} B14 | 77-3) 72:6) 74-1 | 76-7 | 83-7 | 76:8 | 72:2 | 81-6 | 72:2} 77-0 87 85 81 81 84 84 85 86 86 82 90% 87 90 87 U1 }) 111} Wey 124) 10) T14) 113} 119} 116) 111} 119} 116) 117) 115 106} 102} 108) 100} 103] 104) 110) 111} 107) 104} 114] 110} 106) 112 108} 113} 108 99) 108} 114} 111) 116} 116] 113} 117} 112] 107) 112 82 387 87 Bie 33 39 43%) 36 34 32 38 40 36 34 48 52 47 48 48 50 46 AT 51 45 52 50 45 46 22 24 21 23 21 20 22) 23 26 22%) 20 23 25 24 30 Bt 33 3S 34 33 33 32 35 35 34 36 30 3 31 34 33 35 Bt 34 32 32 35 35 35 35 31 34 38 39 38 36 38 38 38 40 38 38 40 38 39 37 38 39 38 36 37 37 38 40 39 38 40 38 40 37 41 AT 51 45 48 47 51 51 49 50 46% 47 47 43 2:7 | 70:2 | 647 | 73:3) 66:6 | 80:3 | 70:5 | 66:6] 73:4) 76:0 | 86:9 | 78:7 | 91-5 | 83-7 32 33 33 33 32 38 362] 34 36 382] 40 37 43 36 62 48 ot 46 49 49 52 53 50 52 48 50 50 45? 1-9 | 108:3 | 101-8 | 110-8 | 112-2 }116-3 | 100-0 | 101-8 |104-0 |1115 |116:6 |110:0 | 120 |126-6 53 62%) bb 51 55 57 62% p42) 52%) 58%) 6 55?) 60 57 B05 | 521) 504} 512} 499} 495] 508] 526] 523] 500} 514] 515] 518] 509 oe 274 | 2:71 | 3:10) 2:84) .. 2:80) .. 2:86 | 2:89 | 2:86 | 2:69} .. 2-72 370 | 370) 358] 372] 350] 344] 362] 373] 367] 362] 369] 359| 378| 362 127 | 182] 181) 129) 119) 123] 128] 121] 128} 130] 180| 135] 130] 118 116} 125) 116) 129) 116} 102} 119) 121} 130) 124] 190) 115) 133) 117 127} 1S) 11) 14) 115) 119) 115) 121) 114} 108} 119) 109) 115] 197 3D} Tt | 75:8) 77-7 | 76-7 | 74:9) 760 | 769) 77:8) 75:7 | 79-4) 77-3| 70:8 | 73-7 (2-4) 63:5 | 67-4 | 65.9) 68-2) 69-7} 69:3} 66-1) 65-4) 68-4 | 70:0] 64-1} 69-2] 70-9 185 | BRB} 88:8] 848) 88:8 | 93-1} 91-1] 96-0} 84-0] 90:3] 88-1] 82-1 | 97-7 | 96-2 158 | 46:2) 44-7 | 47-9 | 43:8 | 40-0 | 47-8 | 48:9] 51:0] 48.9] 385 | 46:0] 55-6 | 52-2 [8:9 | 87-2 | 86:8} 94-4 | 89-5 | 86:8] 86-8 | 80:0} 92-1] 92-1} 85-0) 94-7| 76-9 | 91-9 303} 294} 303] 316} 283) 280) 293} 311} 301} 293) 305] 306) 295| 295 16:0 | 91-8 }100:0} 91-9 | 93:5 | 95:8 | 97-9 101-0 | 90-4 | 98.9 | 94-6 |100-0 | 93-8 | 91-9 73 4 15 768 77 79 80 86 m4 65:6 88-0 40-0 921 285 967 Hh Th ath 1080 | 1490 186 186 185 | 187 138 | 142 92) 98 136 133 182) 128 038 a6 92| 91 62) 69 ob6 600 120 115 TT | 102 83) BT 121} 124 104) 107 107 | 108 36| 33 41i| 47 26 | 22 93) 31 32| 30 37| 36 38| 36 47 | 40 744 | 760 35) 35 49) 48 106-1 |126:0 621) 60 520} 620 201] .. 385 | 386 136) 190 143 | 183 107 | 123 TA2 | 76-3 73-1 | 715 9B | 95-6 634 | 408 89-2 | B61 323 | 324 93-9 | 94-8 dD. 1160 179 178 131 5 120 127 7 05 713 O34 116 Bo AT 13 98 104 a1 60 un bh) bh) 99 | 100 i D. M. M20 | 1270 | 1340 176 | WT) 186 176 | 176 | 186 187} 185 | 136 i) 88 4 134} 120) 126 120} 129) 131 oT 93 7 98 90 oa 65 Got) 68 O70 |) 679) 552 WW4} 119} 128 72:2) 84-1) 73:9 110 82 02 WW) 17) 112 101} 107} 109 107) 107 | 108 ob 48 34 AT 63 1 u 20 pr 50 33 86 31 83 36 oT 35 a7 a7 3b a7 NT AB AB ThA) 72:9 | 77-0 6b 3b 87 49 62 61 120-4 | 1088 | 113-7 bo O4t) 5B 605 | 600} 620 279) as oy 800) 362) 375 120} 124) 128 123) 124| 134 108 | 104} 113 178 | 76:3 | 72:6 761 | OTB) 67-2 97-8 | BBB | 92:5 Bll | 387-7 | AT Bll | 94:3 | 97-3 807 | 805 | 303 95:9 | 968 | 96:9 Trans. Roy. Soc. Edin., Vol. LI.} E | SERIES “D.” j | Number. 1 2 3 4 5 6 7 8 9 | 10 | 11 | 12 | 13 | 14 | 45 | 16 | 47 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 95 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 88 | 34 | 85 | 36 | 387 | 88 | 89 | 40 | 41 | 42 | 43 | 44 | 45 | 46 || 47 48 | 49 | 50 | 51 | 52 | 53 | 54 | gs | = Age - - : M. Sex eee ; M | M | Mw | Mw | M | Mm. | oo 5 MD appre) M. M. M, Cubic capacity . . 1460 | 1480 | 1570 | 1300 | 1260 | 1450 | 1540 1390 | 1430 1400 LO 1350 | Glabello-occipital length 184 191} 184) 183] 189} 195 185 | 184 185 180 185 Ophryo-occipital ,, 184 190 | 183] 182] 186} 195 183} 181 183 176 181 Maximum breadth 141 145 | 14} 133] 143\|| 138 141 | 138 132 140 135 | Minimum frontal breadth 98 95 96 91] 102 99 101 97 93 104 O14 | Basi-bregmatic height 134 126 | 126} 136] 124] 133 136 | 130 131 138 128 Opisthio-nasal length 129 131} 13] 133] 146} 14 139 | 134 133 139 131 Basi-nasal 98 96] 99] 98] 105] 107 101} 98 99 106 99 Basi-alyeolar ,, 93 89 974] 97 98 | 100 91 97 95 103 99 Nasi-alveolar 5 76 66) 68] 58] 73) 67 66) 70 66 12 62 Maxillary-facial index (a) 59:8 51-9 | 54-8 [> 50-4 | 57-0)| 51-5 48:8 | 56:0 51-5 52:5 HO Interzygomatic breadth 127 127 | 124] 115 || 128] 130 135 | 125 128 137 128 Maxillary-facial index (5) 87-3 | 80:9 | 73:3 | 72:3 | 66:6 | 78:5 | 76-1 Tt | 80-4 795 7210 | Intermalar breadth 87 89 90 4 87 93 88 89 87 83 100 118} 117] 19] 119] 105} 124) 193 120) 111 ume 120 111] 109} 110] 110] 109} 119} 109 117 | 103 109 110 Sa x 112} 114) 112} 113) 110) 121} 115 122)} 117 113 116 | Length of foramen magnum - 32)|/ 35 | “3! |) 35 “35 | “401! “36 40) 35 35 3d | Nasal height - Bar|) 55 |) 52) 53/46) be 58 51 52 | breadth 3) |e 2) | |G | 22 || 2 22 a Orbital height (R.) aby] Bi] o]) 82) Sl) Bt B 35 33 30 | (L) 3)|| 33)|| 31|) 31) 3u| 34) 32 35 34 pu breadth (R.) 40 40] 39] 40 5 13 37 AO, x > (1) 39 41 13 38 39 | Palatal length (F.) 51 46 54 50. 53 | index 70:6 804 80:0 | 75:0 | 72-7 | 71-7 68:5 68:0 Gee | breadth 2 36 37 36] 36] 40) 38 3T 34 34 length (TuRNER) 53 50 48] 51) 56) 55 BT 52 56 Palato-alyeolar index 109-4 120-0 108-3 109-0 | 108-0 100-0 | 100-0 109-6 1058 Palatal breadth : 58 60 60 | 542 bi | 51? 57 ue Horizontal circumference 501 538 518) 539 534. 517 pee Ophryo-occipital circumference 499 537 516 525 512 518 Sagittal are : 359 394 379 381 371 Bis Frontal segment 126 142 127 126 122 128 Parietal 116 140 130 134 128 137 Occipital, 117 112 121 121 108 Cephalic index 76-4 759 745 7 7 17-8 Height : 69-1 66-0 6 i 70:8 SU Height-breadth index 90-4 86:8 100-0 90:9 918 9 99-2 pee Nasal index 44-9 46-2 47-8 | 43:6 | 50:0 | 44-2 ‘ 416-2 4 43-1 the Orbital 82:5 80-0 86:8 | 87-8 | 86-5 | 86:8 83:8 100-0 9 89-2 75:0 Transyerse arc 294 | 312 318 301} 308} 306 | 304 996 | 318| 296] 307] 3 303 e238 at Alveolar index 97-9 | 100-0 100:0 | 92:7 94:8 | 38-0 | 99:0 |100-0 100:0 | 92-3 | 92-0 |103-1 | 91- 96-0 97-2 |100:0'), 96:8 | Du M. Youxo—(3) 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | G | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 7 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 8 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 | 97 | 98 | 99 | too mia | a | me | am | me | om | om MM | M | Mw | Mw | Mm | Mw 1400 | 1590 | 1510 | 1290 | 1580 | 1470 | 1480 | 1550 1340 | 1510 | 1500 | 1450 | 1540 | 1870 | 1870 187 | 190] 190] 184] 195} 192) 189) 191 18f | 192)) 187) 180)} 198)) 182)) 186 186 | 190} 189] 184) 192} 189] 187) 191 | 180) 188) 185 | 179) 191) 179} 183 141 | 139, 131} 14) 143) we) 143)) 144 || 136 95 91 102 ye 100 89 96. OL 135 133 | 139] 134] 136} 142) 129] 130 130 137 || 188)) 130) 126) 137) 128 || 132 98 99 86 92 65 69 53:7 5b:2 121 125 18:3 ne) 83 86 115 107 1 102 108 Hoye 3d Bie 39 50 53 53 20 24 20 30 35 a 31 35 a2 38 42 40 oF 49 39 46 52 bi 78:2 67:3 a) 36 35 ei 48 § 52 52 120-8 120:8 100-0 115:3 58 58 60. 525 525 512 521 518 501 379 374 ey) 131 126 127 132 116 112 731 fie 66:3 eee 90:0, 101-5 m 412-9 at 82-1 B2+5, 306 306 hh 101-0 | 96:9 | 940 97-1 | 98:0 | 98-1 1021 90:1 100-0 916 wey Trans. Roy. Soc. Edin., Vol. LI.) SERIES “ i.” ] \ 7 Number. 1 2 3 4 5 6 a 8 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 93 | 24 | 25 | 26 | 27 | 28 | 29 | 80 | 31 | 82 | 93 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 |) 48 | 49) |) 50) | 51 | 52) | 58 |) 54 | ‘55 | 56 | 57 IL | | —| — | | Sex 3 a Dd. M. M. M. M. M. L G - M. M. M. M. M. M. M, WG | NE 9) a, | Tw), | | Cubic capacity 1150 1400 | 1450 | 134 : 310 | 142 32 3: 3 350 | 1800 1190, 1700 1680 1330 | 1630 | 1390 | 1840 | 1610 1130. Glabello-occipital le 176 187 | 186 f 5 | 188 99 86 | 196 192 192 197 181} 186} 18f) 173 191 na) Ophryo-occipital 175 184 | 186 32 36 | 183 7 | 1 8 3 7 5 | 196 193 194 193 18f} 183) 179) 172) 190 176 Maximum breadth 134 138) 12) 13% 36) 139))| 13! 38 3 151 13, 147 143 139) 12) 138) 13) 1a 128 Minimum frontal breadth 95 99 | 100] 95 89 95 : 9: ‘ 93.) 105 102 92 96 89} 100} 96) 98 99 95 5 130] 130 137 137 130 15} WO) 139) 15!) Te 120 35 | 133] 126 133 110 138 130)} 142) 133) 125) 145 125 102 93 101 104 103 97 | 106) 101 92 106 HP] | 96 91 99 OL} 99} 102) 92) Jog 86 | Nasi-alveolar 66 713} cet 664] TL) G8) 652) Go 73 Maxillary-facial index (1) 51:9 58:2 56:0 50.0] 578 | 1 | 515 | 530 61:8 | Interzygomatic breadth 127 123 132 120] 128) 128} 126) 130 118 | Maxillary-facial index (/) 764 16-7 91:0 891 75:3) 81-1) 72:3) GO) 74-2 82:0 | Intermalar breadth 89 86 78 83 87) 88) -OL) 94) (98 89 | Inter-stephanic breadth 116 124 129 121 1b} 122) 115) 192) 129 113 Asterionic diameter 115 lit | 106 108 iit 103} 11G} 108} 107) 132 103 Transverse base - 116 119\| 119 11 117 107 | 120) 115) 118) 122 103 Lensth of foramen masnum Bt 83) 35) 37 35 82) 36) df) 33 | 30 Bi Nasal height - 49 47) 59 51 56 AZ) |) (525) 52h )) 1) 50 53 Treadth 25 2i|) 25) 22 23 BY) |) BP BA) SN) wp 28 Orbital height (R.) 32 30) | rab) 33 3D 30) |) 83))) 80) 183) 32 34 31 30] 36 33 85 qO|| BBN) By) BPI!) Bal 30 | 38 36 39 42 3) 401) 37) 38) 39 41 | 38 39 12 5 LO) |e dSs Ta eC) 40 | 50 A 50 49) 56) 51) 48) 55 48 | 72:0 86:3 78:0 734 | 60:7 | 70% | 791 | 72:7 58:3 : 36 38 ot) 36) 3£) 36) 38| 40] 28 length (Tu 187 5 5D 50 18 5A 50] 57) be) bl] 59 50 Palato-alveolar index 116-6 109-4 100-0 120-0 1208 109-2 120:0 | 1052 | 112-9 | 121-5 | 105-0 108:0 | Palatal breadth 58 55? 60 58 59 GO| GO} Gl) 162) 62 iit | Horizontal circumf 504 539 5fs bit} 532) bIl} 509) bu 501 Ophryo-cecipital circumference 539 511} 520) 503) 504) 535 | 496 | Sasittal arc 390 377 | 380) 373) 3b 350 | Frontal sezment 138 ¢ 130) 130') 127 5 | 125 | Parietal 132 140 180)} 1277) 185 125 | Occipital 120 117 111 100, | Cephalic index 706 12:3 72:3 Height = Tel Tb | te | Height-breadth index 90-6 93:2 90:9 1045 | | Nasallindex ; 1G | sla | 424 rest 51-9 | 3 | Orbits] .. - 5 | 84-6 90-0 90-0 83:3 | 87-5 5 83:3 83:8 | Transverse atc 308 : 306 303 311 | 306 315 310} 307} 319 285 Alveolar index § 87-8 | 91-3.) 95:8 | 96-1 100-0 100-0 99-0 95-0 | 96-2 96-1 | 90° 101-0 |100:0 | 96:2 | 101-0 | 93:5 | x | t | | De M. Yoruxsc—{4) 05 56 57 59 60 61 62 63 65 66 69 70 71 72 73 74 75 76 17 78 79 80 81 Y 82 83 84 85 86 87 88 89 90 ot 92 93 94 95 96 97 98 99 100 1 19 2 312 101-0 370 | 132 | 118 120 1 10:0 | 85:0 | 306 | 00-0, 383 128 138 Ltt 816 312 95:3 105-( 63. 529 119 80-0, 85 7 105 111 Gre 47 24 29) 29, Bye M. 1380 47-1 80-0 307 95-1 M. 1400 194 11 138 oo 131 10 130, 118 71 675 | O49 50-0 78:0 802, 101-9 Trans. Roy. Soc. Edin., Vol. L1.] SERIES “Ff.” Dr M. Youxo—(5) Number. 1 2) 3 9 10 11 12 13 14 15 16 17 18 19 20 24 22 23 37 38 89 40 41 42 47 48 Age > Sex 3 M. | M | M | M | M | M | M |] M | Mw | Mm | Mm | M M. | M. M. Cubic capacity 1680 | 1530 | 1630 | 1570 | 1560 | 1370 | 1550 | 1570 | 1350 | 1550 | 1320 | 1320 | 1000 | 1660 | 1240 1270 | {approx Glabello-occipital length 195 | 190) 192) 19£]) 194] 186} 192) 195 168 | 193] 185 182 Ophryo-occipital 194 | 187] 191] 193] 186] 180] 191] 188 166 | 192) 184 179 Maximum breadth 145 | 143') 149] 146) 139] 138) 142] 145 125 | 144) 135 130 Minimum frontal breadth 96] 105) 103) 100) 90) 90) 99) “93 88 | 101} 102 Ot Basi-bregmatic height 145 | 133) 128] 137} 138] 128) 131] 126 118} 138} 134 123 Opisthio-nasal length 142)| 13) 187) 135} 136) 132) 135)| 139 126 | 137} 140 134 Basi-nasal 108} 101} 100] 103} 101] 99} 98} Io1 93 | 101} 103 98 Basi-alveolar 107] 103} 99))/ lof] 94) 93] 991) 95 89| 97] 97 93 Nasi-alyeolar : |) Bl TS] i) TW) wo] 1% 62} 74] 76 69 Maxillary-facial index (a) 55-8 | 52-2 | 624 : : 5 | 52-6 | 52-9 538 | 57-3 | 60-8 515 Inte matic breadth 136 | 134 136? 115 | 129) 125 120 Maxillary-facial index (b) 82:6 | 744 818 73:8 | 77:9 | 83:5 72:1 | 82:3 | 70:0 83-1 Intermalar breadth 92] 94 882 BL) 95] 91 97] 96] 90 83 | Inter-stephanie breadth 121 || 123 122 107 | 127} 117 133] 119} 113 118 Asterionic diameter 110 | 110 116 104 | 109} 110 113) 112] 108 108 Transverse base : 121 || 122 117 101] 118) 110 125} 116) 113 110 Length of foramen magnum 37 | 132 33| 36] 39 34] 37 || 32 37 ¥ ight 56 | 47 46) 55] 52 49] 56] 48 53 F : 23) 24 22) 93) 25 24 || (23)| 22 20 Orbital height (R.) GB] 58 ail) Sei 29) 32] 3) 80 +. (¢bs) 33) 33 31 | 30] 34 39] 30) 31 31 breadth (R.) 40) 41 36] 40} 35 39 | 41) 38 89 : » -(L.) 40} 40 34} 39) 36 39} 41) 38 59 Palatal length (F.) 57 | 56 42) 5) ot 49) 56] 50 51 index 68-4 | 75:0 840 | 78-4 | 66:6 85:7 | 67-8 | 740 66:6 breadth 39 | 42 37 | 40) 3 42) 38) 37 Bf length (TuRNER) 58) 58 48) 52) 54 62] 58) 502 53 Palato-alyeolar index 105-1 | 110-3 116-6 | 125.0 | 105-5 123-0 | 110-3 | 102-0 107-5 Palatal breadth G1] 64 56 57 64) 64] 517 57 Horizontal circumference. 545 | 540 476, 518 641 | 555) 515 510, ecipital circumference B41 | 53d. 171 516 543] 543) 508 503 al are 391 | 389 334 362 386 | 384] 370 351 Frontal sezment 143 | 137 115 128 133 | 136} 128 121 Parietal 130} 131 108 124 128 122 124 Wess 121 118 | 121 11 110 125 120 106 halic index 73:3 TAA | 75-3 (een 73:0 76:3 Tt . : 69:0, 74-4 | 70:0 70-2 13°2, 67-6 ht-breadth index 94-1 100:0 | 93-0 O44 95:8 oye Nasal index 45:0 41-1 | 51-1 || 50:9 47-8 49-0 377, Orbital | 85.0) 82:5 | 80:5 | 90-0 86-1 TAA 76:9 ‘Tran are Sal | 323) 341 276 326 290 Alveolar index 99:0 99-1 |102-0 | 99-0 101-0 95-7 93-3 O44 75 76 171 78 79 43 | 49 | 50 | 51 | 52 | 53 | 54 | 55 || 56 | 57 | 58 | 59 | 60 | 61 | 62 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 oo, 80 | 81 | 82 | 83 | 84 86 | 87 | 88 | 89 | 90 | 91 | 92 | 98 | 94 | 95 | 96 | 97 | 98 | 99 | 100 | No. | Me) Me | ak fan ae) eM | || |) ow |) EATEN | eT op | eens | sop | MTom | econ | uecCena | MT |e M | Mw.) M | Mw | M | Mw 1270 | 1710 | 1360 | 1740 | 1670 | 1770 | 1460 | 1380 | 1350 | 1400 | 1410 1350 | 1180 | 1580 | 1350 | 1410 | 1550 | 1290 | 1450 | 1490 | 1090 | 1350 | 1470 | 1440 | 1330 | 1680 | 1220 181] 175] 192] 183] 190] 196] 178] 189) 186] 172] 184] 200] 181) 191) 190) 179 179 | 174] 186) 174] 184] 192] 172) 185] 187} 169] 184] 900) 181) 190] 188} 180 135 | 134] 141] 136) 141 | 142] 138] 132) 144 || 129') 133 143) 137] 135) 148]! 126 g2| ss] 96| 90] 94] 104) 989) 92) 97] 90} 90] 106] 105] 93) 96) 87 137 || 130] 141 | 132) 132] 195] 131] 1385] 129)| 120) 129') 135) 134] 126] 127 | 126 140 | 193] 136] 138) 140] 138 | 140] 135] 129) 127) 132) 142) 133 132] 133] 130 103 96 | 101 100} 107] 103) 104) 102 98 93, 99 | 104) 101 101 96 96 100 95 | 96] 102 86] 102} 97) 94) 101 73 7B\| || iP) ff) || io || 7A) Ol te 58-4 57-0 | 58-7 | 52-1 48:3 | 63:5 | 56-4 | 53:8 | 57-8 125 128 131 138 122) 118 140 130 121 811 81-1] 89:5 | 77-4 68-6 | 842] 85:8 | 71-4 | 76:0 90 90| 86] 93 86| 89] 92] 98) 92 106 119'|| 117) 107 108] 112} 122} 122] 118 104 11 | wt} 117 106 | 107) 118} 105} 103 115 iif) 114) 129 110} 105] 119} 115] 110 40 37 | 39] 36 Bl] By esi] BP] ai 53 54) 60) 55 44) 50) 59] 48] 4s 23 26) 25] 95 SY)|| PBS] Phil] pl)|| on 32 31| 36] 36 32) 35 31) 35 32 31] 36] 36 32) 35 32 | 34 38 40) 40] 39 37, | 39) 39 | 39 37 40] 39] 39 36) 39 38] 39 52 51} 56] 55 45 | 59 53 | 55 75:0 | 81-8 | 76-4) 69:6 | 70-9 75:5 | 52-5 66:0 | 67-2 39 | 36] 39] 391]| 39: 35 |r 18 |e) (eT T 54] 48] 53] 60) 55 47 || (Gl) 58 | be) 57 100-0 116-6 |116:6 |118-8 |103-3 | 116-3 112-7 | 86:8 |108-6 |112-9 |107-1 59 63| 56] 63] 62) 64 BB} || fi 63] 61] 61 537 503 | 493) 532] 516] 528 484 5p5 | 515 | 525 500 Oz 3 532 498} 490] 524) 500} 518 476 Bol} 514} 524 498 36 405 354] 358] 384] 360) 367 340 391 | 369 | 379 361 121) 138 129 | 128') 142'| 122) 124 113 138 | 129)|| 133 123 124 | 138 118| 126] 123] 190] 196 114 139 119 122 106) 129 | 107 | 104] 119] 118] 117 113 Wz 127 116 74-6 | 76:6 | 73-4) 74:5 | 74-2 75-0 715 10-7 0-4 75-7 | 74:3 | 73-4) 72:6 | 69-5 69:8 | 70-1 | 67-5 66-0 70-4 101-4 | 97-0 |100-0 | 97:9 | 93-6 93-0 | 96:9 | 94-4 93:3 100-0 43-4 | 50-0) 48-1 | 41-3 | 45-5 50-0 | 46-0] 42-4 50-0 48:9 8£2') 88:6 | 77-5') 90:0} 92:3 86:5 | 89-7 | 90:2 89-7 71:0 299 | 299'| 317 | 302] 301 279 | -302| 310) 315 || 302 291 97-1 | 90-6 | 94-1 | 96-0] 95-3 92:5 |103-0 | 93-3 | 93-1 | 100-0 106-2 Trams. Roy. Soc. Bdin., Vol. LI.) SERIES “G.” ' Number. 5 6 9 | 10 | 1 | 12 | 13 | 14 | 5 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 93 | 24 | 25 | 26 | 27 | 28 | 29 | a0 | 31 | a2 | 3 | aa | a5 | a6 | a7 | as | go | ao | a1 | a2 | be 47 48 | 49 | 50 | 61 | 52 54 | 55 | 56 ES “ Age : Sax. eee i), |) 1), |} 3) |) 10), D) |} Os |] Wh | Th | He | HM |} Me] a |), | De | me] a] oD, ||, |}, |), |) |) my, | om |] mW, || DD, |) m. | wD, || D | mw, |] m | wD D. D. | vp. | 9 Cabic capacity 1490 | 1180 1270 | 1160 | 1420 1300 | 1350 | 1200 | 1350 | 1280 | 1400 | 1410 | 1400 | 1430 | 1400 | 1210 | 1330 | 1200 | 1290 | 1200 | 1430 | 1400 | 1420 | 1570 | 1240 | 1380 | 1530 | 1300 | 1270 | 1870 | 1280 | 1350 1420 1270 | 1870 | 19280. | 1 Glabello-occipital length 184 | 172 178} 180) 178 183} 183} 174] 175) 181] 178} 189] 185} 183} 184] 181] 176] 178] 173) 179] 190) 185) 186] 189] 177) 189] 189) 178) 178) 176) 177] 179 187 176 | 189| 77) 1 Ophryo-oecipital 184} 17: 176) 178) 179 181} 183} 173) 174) 179] 180] 189] 183) 181] 182] 180] 176) 175] 173] 178) 189) 184) 184] 188] 176] 188] 189] 175] 176) 176) 178] 179 185 | 189| 177 Maximum breadth 146 134] 134) 142 143 | 135) 135} 140] 143) 145) 136) 136) 138] 142] 133] 133] 137] 128] 133] 140) 139) 136) 143] 130] 130] 141] 136] 180] 189) 132)) 134 139 136 | 192] 131 Minimum frontal breadth 95 93] 89 | 90 99] 97] 88] 94] 99] 92] 94] 95] 94] 96] 93] 95] 92] 91] 95] 94] 95] 93] 98] 93] 89] 95] 90] 94] 89) 90] 96 97 91] 97) 85 Basi-bresmatic height 133 124 || 122')) 128 116} 127} 128} 120] 127) 133) 137] 131/| 126))| 126 || 121} 128)) 121) 132] 126)) 120) 180) 192] 187] 120} 185 128) 126] 129) 123) 129) 132 130 12t] 1272) 198 Opisthio-nasal length 130 126] 128))| 123 125 || 132] 129} 129} 133') 124] 136] 134] 131) 125] 128] 125 126) 124] 134} 134] 182) 197] 180] 123] 140\| 121 || 198] 128) 124) 128) 125 134 124 | 137) 124 Basi-nasal 5 96 o£] 97 || 91 90| 95] 98] 94] 102) 95] 102} 99} 95] 93] 96] 94] 93] 92] 98) 98] 98] 96] 99] 92) 103] 88] 93] 98] 89) 98] 9% 101 93) 1098] 93 Basi-alveolar 92 2] 89} 92 80) 87] 91] 91] 95] 81) 97]| 921) 92 89 | 8427] 92] 90] 94] 95] 88] 85] 92] 86] 9£]) 88] 87] 90] 88]! 92) 88 90, 89 | 100%] 86 Nasi-alveolar : 63 59 | 60) 68 63} 70} G8} 63] 68) 65) 73] 75) 70) 68] 67] 65) 69] 65] G3) 75) 70) 72] 67| 67) 67) 66) G8} 7) 70) 68) TW 66 6L| 66] 73 Maxillary-facial index (a) 543 47-5 || 52:1) | 57-1 50-0 | 57-3 | 56:2) 52:5) 65:7 | 53:7 | 584 | 62-5'| 56:9 | 52-3] 58:2] 53-7] 58-9] 53-2] 51-2] 64-1) 57-3) 620) 55:3} 57-2] 56-7 | 56-4 | 56:2) 58-2) 57-3 | 562 | 57-7 51-9 5b7 | 53:0) Gat erzygzomatic breadth 116 124} 115] 119 126 | 122) 121) 120)) 722)|| 121 |) 125} 120))| 128)|| 130)) S|) Tan) Wy} 129))| 123))) 7 }) 122) 16) Te1)| 117) 1s |) 17) 121) 122))| 122) 121} 128 197 Wr} 123} 115 Maxillary-facial index (b) 75:9 70-2 | 73:1 | 76-4 75:9 | 84:3 ') 73:9) 72-4] 85-0] 79-2 | 82:9] 83:3] 80-4] 781] 843] 76-4] 81-1] 76-4) 68-4) 88-2) 85:3) 81-8} 705 | 77-9] 77-9 | 79:5 | 79:0 | 82:5 | 89-7 | 73-1 | 81-6 79: 83-1 | 70-0) 82:0 Intermalar breadth 83 Bi] 82] 89 83] 83 | 92] 87) 80] 82) 88] 90) 87] 87] 79) 85] 85] 85) 92] 85] 82) 88) 95] 86] 86] 83| 86) 86] 78] 93] 87 83 | 93)\) 189) ephanic breadth 118 108] 111] 118 118} 117] 110) 116) 125] 121} 119] 118} 118] 118} 114} 118} 116} 109} 112} 120] 117] 118] 126] 114) 110) 119] 114} 115] 115) 110) 119 127 118} 115 Asterionic diameter 106 107 | 105) 111 114] 106} 103} 116} 111) 110) 112} 104) 107) 108} 104} 110] 101} 98) 102] 113] 111) 103) 110] 98] 106} 114) 106) 105] 118] 105} 104 107 105 Transverse base Se 108 11} 109} 114 114} 109} 110} 111} 117] 111) 115) 113] 112) 119] 108) 117} 104) 110} 115] 111) 117] 106) 110) 107} 108} 115} 11) 110) 112} 108} 109 113 105 108!| 107 h of foramen magnum 4 yl] 6 35) || $86))) ¥32) 1)" ¥35)|) 931) §30)|/ ¥a5))) ¥37/)|/ 136)|) ¥32) 1) 932))) 33))) Fs) 33)| fse) 9 is5)|) FSb)|) 932) 52) 8s) 38)\Nts2)|nesbs| eco)! nd0)ls) SO) |e 3b 31 sail! 720 height 50) 52) 49) 47) 48) 52] 53] 56) 50) 51 51 48] 50] 51 49] 54) 45] 55] 50} 53|/ 50} 48) 48} 53) 52) 60) 49 49% AG 52 pres 26) 21 211) 238 fe) |e) | ee 22) | | 2303 | 22 | | 2c 7 || 3 |S) | 97) 22) 20) 23) 25 22 20 98 Orbital height (R-) 33)|/ 138)! 33) 781) 34)\) a4) 3s) 30) sl) 3) S|) saa) 3) 29) v3 33)|) 371 S3)|) S1)\) sb] S11) 180) |ir93))/S6)| S45) tSO) iss 34 31 30 - (L) 33] 37} 33] 31} 35) 31} 34! 36) 34] 34)|| 33) 37] 34) 30] 32!) 35) 36) 33] 381) 33] 32) 27) 33)|| 36) 34) 30) 32 34 30 31 breadth () 38} 39] 38] 37] 38] 37] 37] 40] 38] 38] 38 38) 40) 37] 37] 39) 38] 38] 36] 387] 387) 36) 36) 39) 36] 36) 37 42 38 38 eet f 5) 88] 39) 36) 37] 38) 37] 37) 38) 37) 37) 38] 37) 39) 37] 37] 38] 38] 37) 36] 37) 37] 36) 35) 39) 36) 36) 37 AD 38 458 iste enet Oe 35] 46') 47) 49)] 48) 42)) 53) 52) 50) 48) 48) 45) bal) 47] 50) 49) 48) 46) 41) 49) 48) 48) 47) 47) 48)) 47) 50 474 50 53 ae B44 | BLT | 76:5] 77:5 | 77-0'} 73:8 | 64-1 | 75:0] 74-0\| 79:1 | 70:8') 77-7 | 69:8 || 68-0 || 72-0}| 79:5 || 72:9) 869 | 97-5 | 77-5 | 79-1 | 87-5 | 765 | 80-8 | 79:1 | 89:3) 68:0 HA 60:0 17 | ete 382) 39) 36%) 38) 37] 31) 34] 39) 37] 38] 34) 35) 37] 32) 36 | 39% (352 40\| 40} 382) 38] 42) 36] 38) 38) 42:) 3f 3p 30 38 ee 472) 48 50), 49) 51) 442) 56) 54] 53] 48) 49) 47 55) 50] 53% 52] Bl} 48) 45] 51%) 50) 50] BO) Sl} BO) 49) 50 50 50 i Palatal breadth é 10:6 }125:0 | 102-0 | 124-4 )117-6 | 127-2 ) 107-1 | 111-1 | 10725 |116-6 |102:0 )108:5 }105-4 |116:0 |116-9 | 117-3 | 103-9 |127-0 |185-5 | 107-8 |110-0 }124-0 |116.0 | 123-5 114-0 | 1265 mE 112.0 102:0 109:0 Teer REE ie assis 524 60 51 61 60 56 60 60 574] 56 50 51% 58 58 62 617} 53%) GI 61 5b?) bb) G2 58 63 57 62 abt alts 51 60 Ophea nia ican 521 51d 497 | 503) 518) 512] 521) 518) 515] 519] 506) 507) BOL} 494] 505] 529] 520) 517) 533] 503] 515 | 529] 506] B00) 506) 503) 506 520 498 194 Wee . 518'| 512') 493 Ga 517 4) 518] 516] 513] 515] 502] 503] 499] 492] 501} 529] 517) 514) 532) 500) 512 || 529] 503] 496] 502] 501} 504 520 500 191 Arcl : 20 |] no |] BED || co S| eee! | etal Ny PTB TG 229.0 | aes a 3 a || oo ERI) cao |HiReo. || eat IlconifMeo. || con) 20 BS an - pene ee 873 | 850) 347 | 359 385 366 | 380] 360] 364) 358] 366] 352 378 | 381} 392] 353] 375] 393] 367] 360] 363) 365) 372 372 pity! 318, Pee 132} 124) 117 || 132) 134] 135 130] 182] 126) 129] 724) 126) 118) 120} 132] 392 )| 137] 123) 131] 181} 728] 127) 121) M7) 127 ey) 131 120 Ges 123} 115} 108} 120} 124) 124 110] 130) 119] 117) 122] 131) 117] 143] 116) 198] 135] 120] 124] 139] 136] 125) 124) 127) 125 wie 122 124 Ce 118] 111} 122} 107) 118) 196 126) 118) 115) 118) 112) 109) 117} 113} 130] 191] 120] 10} 190] 123] 103) 108) 118} Jal} 120 122 11 104 Heat = 78-1 | 73:8 | 77:6 | 80:0} 79-0} 81-5 | 72-0 75-4 | 77-2) 73:5 | 75:6 | 77-0 | 74-0 | 74:3 | 73:7 | 75-1 | 731 | 75-7 | 73-4] 68:8 | 746] 76-4 | 73:0) 79-0) 746) TLD 74-8 17-3 49 Heee aE 63-4 | 69-4 | 73:6 | 68:6 | 70:2 | 74-7 | 72-5 68:9'| 68-5 | 66-9 72:7 | 68:0 | 76:3 70-4] 632| 70-3) 710] 72:5 | 67-8] 71-4 | 67-7) 708] 72:5 | 69-9) 729) 7137 OG 705 7A Nacal index ; S11 | 94-0) 94-8 | 85-7 | 88:8 | 91-7 | 100-7 91-3 | 88-7 | 90-9) 96-2) 88:3 ]103-4 | 94-7 | 85:7 | 93.5 | 970] 95-8 | 92-3 |103-8 | 90:7 | 926 | 99:2) 88-4 | 97-7 | 9B fae ie Bape Orbital 52-0 | 40-4 | 42-9] 48:9 | 47-9) 38-5 | 41-5 46:0 | 45-1) 431 | 47-9) 44-0) 434 | 49:0 | 50:0) 51-1] 418] 50:0] 43-4} 46-0] 43:8) 56-3) 41-5 | 38:5 | 46.0) 51-0 aes 43-5 538 Teaver as 86:8 | 97-4 | 86:8) 83:8 | 89:5 | 91-9 | 91-9 895 | 89:5 | 89:5 |100:0| 85-0 | 784 | 86-5 | 846 | 97-4] 868] 86:1 | 94-6 | 83-8] 83:3 | 91-7 | 92-3) 944 | 83-3) 919 81-0 B16 78:9 Meese 293 | 307) 305 | 295] 315 | 320) 309 308 | 306] 300] 302] 305] 301] 295) 312] 307] 311| 335] 290) 905] 315| 295) 292) 290) 295) 06 303 303 398) 88:8 | 91:6 | 92-9) 96:8 | 93.1 | 85:2) 95-1 96:8 | 91-4 | 92-7 | 89:3) 98-9 | 97-8 | 95-9 | 96-9) 89-8) 885 | 92:9] 93:5 | 91-3 |100-0 | 93-5 | 91-8 | 989) 93.9 | 926 89-1 95-7 99:0 De ¥. Youso—{6) 100 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 68 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | of | 92 | 93 | 94 | 95 | 96 | O7 | 98 | 99 F. | D. |} 1 |} 1), |} ww) |] 1 |} my |} ah io), |} i), |) 1) | am | wD), 1) | | 1290 | 1540 1260 | 1190 | 1140} 1120 | 1260 | 1210 | 1400 1390 | 1460 | 1200 | 1190 | 1380 1300 | 1270 | 1270 | 1300 | 1320 182 | 183 179) 177) 166) 174] 175 || 175) 173 180 173) 180 182) 175) 172) 179) 175 182] 183} 178] 178 | 175} 163) 175] 183] 175), 176| 179] 166) 173) I74) 174} 172 178 172] 178 182) 175] 17} 179! 174 133] 138) 133] 141] 136)| 126) 135] 145) 135] 132] 130} 130) 131] 133] 135) 143 139 130 | 136 141 | 135) 134] 142) 149 96| 93] 87] 92] 88] Sf] 87] 97]| 95] 99) 87] (88) 95] 93] (89) (89 88 o1| 91 88] 93] 83] 95] 96 131 | 129'| 127] 127) 115] 121} 130)| 124) 123) 132) 121) 127) 132) 126] 124 127 131] 131 132) 129) 128) 128] 125 126] 134) 122] 127] 117) 127] 183] 126) 131] 124) 115) 129] 130) 123) 124 131 128] 127 121} 119} 123) 131) 119 91| 101) 91] 95) 86} 91] 99] 92) 99] 93) 86] 97) 96] 9£] 92 95 97 | 97 93| 89] 90) 97] 91 88] 95] 93] 95] 79) 92] 86% 91] 94] 89] 88] 95] 86] 83) 90 88 90} 91 87] 86] 87| 95) 81 68] 76) 62] 70) 60} 74] 65) 69) 69] 65| 67] 66) 64) 65] 69 62 72) 69 67) 65) 61) 66) 65 548 56-0 | 58:7] 55:0 | 524) 53:7 | 57-0 55:8 59:5 | 54-7 563) 566) 512) ol | 55:0 124 116 | 14) 120} 122) 121) 121 111 121) 126 119) 119) 119] 1223) 118 79:0 78-3 | 81-7] 76:8 | 77-1 | 764°] 80:2 384 7 | 79:4 80:0) 78-4 78:8 | 80:2) 622) 741 | 78:3 83| 82] 87) 83] 85] 86] 862 5 5 78 90 | 88 85| 81 /98Nn] 89) 83 106) 114) 115} 113) 108] 126) 118 5 m2 12) 111 113) 117} 112) 115) 120 108 103 106 102 105, 112, 104 5 h f 5) 0 108 105, 106 99 103, 97 110, 107 109| 109) 110) 111} 110} 113} 111 09 2 j 5 107 109 | 114 107 | 106} 106] 113] 141 B10) 929)/)933)|) 882)|) 730)|) 832) |) 136 36 32] 36 22) 137 34 | 32 Ru) Bell SBN BEEN Bh) 48] 45) 50) 43] 46] 50] 48) 49) 4 5 | 5 : 4 50} 49 50} 42) 45) 47) 50 22) |eo5) ees) ee ee 23 : 19 20 a1 | 20) 24) 24)! 90 312] 38) 33))/ 33)|| Sf) 35)| 342) 33 3 33) 32 34 31] 31 32 S31) |) 27\|) 33) = 837) 34)\) 35)|) 3d 332) 3! p & 31 3t By) 8B 32 38 34) 38] 40) 36) 38) 382 2s 382] 35 : 34 37 36| 37 37 38] 35) 38] 40% 36] 39] 38| 382] 37%) 3 i 5 3 37 35] 37 37 48} 47) 53] 47) 45] 48) 45) 46 |b 5 46] 41 18 {8| 47 43 75:0 | 80:8] 73-5 | 74-4) 82:2) 79:1 | 80:0 | 82:6 | 76-5 et | 72: 0: Hf | 80:5 79-1 72:9 | 76:5 83-7 36] 38] 39|) 35] 37] 38] 362 ) 363] 36 : 33 38 35] 36 36 50] 51 || 55] 49] 48) bli! 48) a 49 5 48 | 46 52 50 51} 51 154) 116-0 |119-6 | 110-9 | 112-2 | 118-7 |109-8 | 120-8 119-5 103-8 120.0 113+7 | 109-8 1266 58 | 61) 61] 955) 57] 56) 58%) Ar? 50? i 55 BL] 60 58] 56 57 495 | 478] 489, 496] 494] 510| “502) 46 9 | 49 5 94 | 505 510 | 510 508) 501 507 455 | 476] 486) 494} 193} 509 | 5u0 | 504 606 507 507 | 503) 492] 510} 508 SHON |] co |} oo |) BAD] oe ape ile ae a poy |e a .. || 3:10} 2:98 374 | 353] 349 | 363\| 353: 368 36 8 358 | 260 350 | 369 | 867 136 | 121) i214) 123} 130 131 G 32 | 125 | 130 125 | 138 T2£) 119} 124] 131] 110 125) 129 130} 112 114] 113) 111} 109] 118 115 14 73-4 | 78:3\| 75:3 | 76-0} 77-1 ee} 771 74-6 | 72:9| 73-0) 75-4 | 72:0 | 706 137 101-5 | 93:0) 96-9 | 99-2 | 93:3 | 9 91-3 | 86:7 95-5 45-8 | 51-1} 46:0} 51-2] 50:0 | 4 40-4] 385 57-1 81-6 | 82-4 | 86:8) 82-5 | 94-4 94-1 | 91-9 83:8 301 | 288] 295 301] 299 303 | 307 803 95:7 |103-5 | 97-9 | 89-6 | 88-3 92-6 | 98-1 96:6 | | | Trans. Roy. Soc. Edin, Vol. LE} SERIES “ Y.” —FACIAL PORTION DEFECTIVE. JUVENILE SERIES “J.” SPECIAL SERIES “X.” Number. | 1 2 3 4 5 6 il | se, years -| = ? M. M. F. M. F. M. M. (metop.) Cubic capacity 1430 | 1280 | 1380 | 1450 | 1350 | 1300 | 1530 Glsbello-occipital length 191 | 182] 182| 185] 180) 183] 189 Ophryo-occipital 189 | 182 182 182 180 | 182 191 ) Maximum breadth 135 137 134 148 138 138 | 144 Minimum frontal breadth 107 94 94 95 92 93 98 Basi-bregmatic height 138 | 129] 128} 127] 118] 123] 122 thio-nasal length 139 123 | 130] 138] 126 | (135) 129 Basi-nasal 109 93 98 | 100 92 99 94 | Bas-alveolar 102 86 90 96 90 93 94 Nasi-alveolar 80 73 65 69 66 75 67 Masillary-facial index (a) 63-4 | 61-8 | 56:0 | 51-8 | 55-4 | 57-6 | 57-2 Tnterzygomatic breadth 126 118 |} 116] 133) 119 130 117 Maxill: al index (6) 90:9 | 81-8 | 76-4 | 78:4 | 78:5 | 78-9 | 77-0 Entermalar breadth 88 86 85 88 84 95 87 Inter-stephanic breadth - 123} 119 | 120) 130} 119!) 113) 123 Asterionic diameter 107 | 100} 109 112 110 | 107 107 ‘Transverse base - : 119} 113] 109) 124] 110} 114 110 Lensth of foramen magnum - 33 33 33 38 34 35%) 372 Nasal height - 56 50 45 52 49 51 48 breadth - 23 21 22 227 21 28 26 (R.). 35 33 31 32 31 35 34 (L) 35 32 31 31 30 35 34 h (R.) 41 37 38 39 38 40 36 41 37 38 39 38 40 35 52 47 48 48 45 48 50 69-2 | 78-7 | 70:8 | 79-2 | 82:2 | 85-4 | 78-0 # 36 | 38 34 38 37 412 39 (Tvesee) 54 49 50 52 49 52 52 ry index - | 120-3 |112-2 |114-0 |117-3 | 118-3 |107-6 | 115-3 ~ 65 55 57 61 58 56 60 ircumference 532 | 510 | 508 | 534 | 507| 511] 538 I circumference 529 | 510 | 508 | 526 | 508) 508) 540 2 380 | 373) 367 | 363) 373 | 369) 390 Frontal segment 138 | 132) 127] 127| 1389] 133) 128 Parcial ,, 129} 123| 126| 120) 116] 120) 146 Occipital ,, 113 | 118] 114] 116] 118] 116] 116 Transverse arc 310 | 303 | 301} 314} 300] 300] 313 Cephalic index | 70-7 | 75:3 | 73:6 | 800 | 76:7 | 75-4 | 76-2 1 index | 72-3) 70-9 | 70-3 | 68:6 | 65:6 | 67-2 | 64-6 Height- breadth index = 101-4 | 941 | 95:5 | 85-8 | 85-5 | 89-1 | 84-7 Baca) index B | 41-1 | 42:0 | 48:9 | 42:3 | 42-9 | 54-9 | 54-2 | B54 | 89-2 | 81-6 | 82-1 | 81-6 | 87-5 | 94-4 } 93-6 | 92-5 | 91-8 | 96:0 | 97-8 | 93-9 |100-0 m5 a BBPEN oe 296] .. bn Number. 1 2 3 4 5 6 7 Age, years Sex é M. M. M. M. M, M. M. (metop.) Cubic capacity a 1600 | 1250 | 1570 | 1540 | 1400 | 1360 | 1330 Glabello-occipital length 193 | 187) 190} 194] 182} 179} 184 Ophryo-occipital ,, 190 | 186} 192 191 182 17 185 Maximum breadth 142 | 132) 143) 137] 140} 134] 131 Minimum frontal breadth 106 93 95 96 98 92 95 Basi-bregmatic height 129 | 131) 137) 180) 128} 126) 123 Opisthio-nasal length 136} 130 || 127 136 | 126 132 129 Basi-nasal length 102 95 94 99 95 93 95 Inter-stephanic diameter 132 | 114} 120) 118} 124) 115) 114 Asterionic diameter 11} 107) 107) 107} 107} 106 98 Transverse base : 112/} 108 | 109} 112} 112)| 113) 106 Length of foramen magnum . 35 BES 34 37 33 39 35 Horizontal circumference 540 | 515 | 530} 534) 518) 505] 511 Ophryo-occipital circumference 539) 512] 5381 | 526) 517) 502} 512 Sagittal are 383 | 383 | 407 | 386 | 375 |~361} 373 Frontal segment 136 | 132 148) 129 137 121 130 Parietal ,, 131 129) 130) 128) 127 122 130 Occipital ,, 116 | 122)) 129) 129)) 11} 118) 113 Transverse are 325 | 305 | 322} 310) 310} 301 | 295 Cephalic index 73:6 | 70:6 | 75:3) 70:6 | 76:9 | 74:9 | 71-2 Height x 66:3 | 70-1 | 72-1 | 67-0 | 70-3 | 70-4 | 66:8 Height: breadth index 90:8 | 99:2 | 958 | 94:8] 91-4 | 94-0 | 93:8 De M Youss.—i7) Number. 1 2 3 4 5 6 7 Age, years (5-6) | (4 or4-) | (4or4-)| (4or4-)| (4) |(4or4-)} (4) Sex : q a 2 2 2 2 2 Cubic capacity 3 00 96 ge fe 9a oe 20 Glabello-occipital length 165 164 155 162 176 161 Viz Ophryo-oceipital _,, 165 166 156 164 Viz 163 179 Maximum breadth . 126 130 120 130 137 123 131 Minimum frontal breadth 87 84 83 85 88 79 90 Basi-bregmatic height WME} |) = 90 36 ne ao Opisthio-nasal length 114 14 110 M7 122 108 122 Basi-nasal mH 81 00 ap 75 se oH Be Basi-alveolar yal a0 co ao oe Bn 66 Nasi-alyeolar i 56 51 49 48 56 48 58 Maxillary-facial index (@) 57-7 | 53:6 52:6 48:0 | 54-9) 53:9 | 55-2 Tnterzygomatic breadth 97 95 93 100 102 89 101 Maxillary-facial index (5) 83-5 72:8 71:0 69:5 | 86-1) 72-7 19-4 Intermalar breadth 3 67 70 69 69 65 66 73 Inter-stephanic breadth . 107 108 103 113 115 101 7 Asterionic diameter 102 99 be 93 99 102 98 101 Transverse base 2 92 91 92 96 98 83 95 Length of foramen magnum . 34 = ig 30 oc = oe Nasal height . 39 38 35 36 42 35 41 ,», breadth 5 20 19 19 17 19 16 21 Orbital height (R.) 31 28 32 31 30 27 32 e bse 31 29 31 31 30 27 31 », breadth (R.) 34 32 32 32 34 30 34 i (ita) 34 32. 31 32 35 30 33 Palatal length (F.) . 35 35 33 33 38 35 37 » breadth 28 30 26 28 26 27 30 » length (TURNER) 37 36 36 3d 40 36 38 » breadth 50 47 45 48 49] 44 55? Horizontal circumference 464 462 444 466 496 451 494 Ophryo-occipital circumference 466 467 446 468 499 455 28 Arch to base ratio 2:99 3:08 3:02 2-94 2-99 BOEE 3:07 Sagittal arc 341 352 333 345 365 339 375 Frontal segment 109 121 117 123 125 116 131 Parietal ,, 124 119 106 120 128 114 135 Occipital ,, 108 112 110 102 112 109 109 Transverse are 286 302 276 300 320 280, 312 Cephalic index 76-4 79:3 11-4 80-2 178 16-4 ves) Vertical ,, 71:5 oe aye oe a aa ae Height-breadth index 93-6 20 on 58 As ae ae Nasal index 51-3 50-0 54-3 46-4 45-2 45-7 51-2 Orbital ,, 91-2 87:5 100-0 96:8 88-2 90:0 94-1 Alveolar ,, 80 6 Se 30 OG ad 50 me ee 2O5 a : ah i we Trans. Roy. Soc. Edin., Vol. LI.) JUVENILE SERIES “kK.” Number. 1 2 3 4 5 6 7 8 9 10 41 12 13 14 15 16 17 18 19 20 : 5 = (3 - — 1-12} 11-12} 9-10 7) | 9-10} (8 7 8) 8 2 a 3 . " Ase, years 9 ae 9 a9 11 az 9 ag 9 10 1 a ; F ) ; o ) (8) 8) ( ‘ (7) Met (7) Met (7) Met (7) Mer: Cubic capacity 2 ° 66 65 oe oy os a6 5 a6 90 09 5p a5 oo oe 50 ab 00 50 ae Glabello-occipital length 173. | 178 | 173| 169] 178] 176] 176) 173] 170] 172) 172) 181} 163} 169) 164 170 | 171 178; 167 Ophryo-cccipital 174! 179] 173] 170] 178] 174] 178] 175] 173] 173} 173 || 182'| 164} I71} 165 172 173 | 179 || 169 Maximum breadth 134 | 134] 199] 197] 180] 198] 135] 184] 185] 127) 185] 138] 127) 181) 132 137 | 133 | 134 | 180 Minimum frontal breadth 90! 90) 92) 88| 93| 91) 91) 94] 98] 88] 86] 87) 88] 90] 8 98 94 95 90 Basi-bregmatic height 126 | 122 122 | 125 | 128) 125) 133) 116 126 TSH) 12/7 120 | 123 125 119 125 126 121 129 Opisthio-nasal length 96 | 124] 199] 191] 129] 119] 126] 124] 124] 120] 121) 118] 121] 117] 14 120 | 192 118 | 118 Basi-nasal * 91 86 93 86 93 | 87 94 92 87 90 85 85 90 85 80 84 89 84 86 Basi-alyeolar s3/ so] 92] 81| 90] 83) 8 | 86] 80] 83] 74] 792] 85] 79] 73 15 80 75 75 | Nasi-alveolar 4, 60| 582] 58] 53| 57] 62] 63] 61) 58| 61} BI] 50] 53) 658] 51 51 56 51 52 Maxillary-facial index (a) 560 | 55-2 | 50-4 | 51-4] 51-3] 548] 562 | 55-4 | 547) .. | 51-0] 485) 486 | 55:7 | 531 481 | 56:0 | 485 Re Tnteraygormatic breadth . 107 | 1052) 115 | 103) 111] 113| 112] 10%) 106%) .. | 100) 103] 109] 104) 96 106 100 105 ny Maxillary-facial index (5) 75-0 | 80:5 | 69:0 | 70:6 | 70:3 | 78-4 | 787 | 76-2 | 72:5 | 76:2] 70:8 | 69-4] 68:8 | 75:3 | 69:8 65:3 | 75:6 | 62:9 | 75:3 | Intermalar breadth 80 72 84 75 81 79 80 80 80 80 72 72 17 7 73 78 74 81 69 | Inter-stephanic breadth - 116 | 114 112} 109] 113) 118) 118) 112 112 110} 111 110} 210) 109) 111 120 115 119 116 | Asterionic diameter 104 | 100] 106| 101] 108] 103] 109] 101-| 102] 972] 104] 103] 100| 97) 99 98 | 101 97 98 | Transverse base ; 103 | 103) 109) 99] 106] 103} 106] 103 \100(4)) 106] 98] 98] 101| 97) 94 100 98 99 96 Length of foramen magnum .- 35 36 36 35 37 33 33 31 36 31 37 32 32 33 32 36 33 34 34 Nasal height . ; 45| 38) 43| 40] 42) 42) 47] 43] 43] 49] 991] 36] 40|| 42))| 38 40 44 39 38 >, breadth 20 18 23 19 20 19 21 21 21 21 20 19 22 20 19 22 92 20 19 Orbital height (B.) 32 31 30 28 32 29 35 31 32 32 29 30 29 31 30 33, 3l 31 31 a5 % (065) 32 31 31 28 31 29 35 31 32 31 29 31 29 30 30 33 31 31 30 5, breadth (R.) 33 33 34 32 37 33 35 37 34 34. 32 34 33 35 32 34 33 34 31 BS =~ (0) 33 33 34 32 37 33 35 37 34 34 32 3d 33 35 32 34 33 34 30 Palatal length (F.) 41 40 44 39 43 43 40 42 38 39 35 36 39 42 35 31 38 36 36 » index. 75:6 | 72:5 | 81-8 a6 TAA | 74-4 | 87-5 | 71-4 oD oo 5p 20 20 76-1 29 00 56 60 e > breadth 31 29 36 31 32 32 35 30 31 31 28 32 32 32 27 29 29 29 26 ,, length (TupRNER) 43, 41 46 42 45 46 43 447 39 41 38 384) 40 Ad 36 35 40 38 38 Palato-maxillary index . ae co | PAO |) oo -- | 115-2 o0 6 aS a0 a6 00 oo 7 lt obo ae fe ao 60 bo Palatal breadth 56 53, 58 54 58 53 57 52 56 50 51 53 55 54 49 52 53 53 47 Horizontal circumference 492 | 498 | 487] 471 496 | 497 | 497 | 491 | 489 | 483) 485 | 496] 469] 482) 470 489 488 499 479 Ophryo-occipital circumference 00 30 b0 471 | 496 | 495) .. a oo fs 30 80 36 oo 471 ae 20 ae ay Arch to base ratio . 2:84 | 2-96 | 2-69 | 2-95 | 2:83 | 3:06 | 2:89 | 2:82 | 288 | 2:91 | 2:05 | 3:16 | 2-81 | 3:03 | 3:06 3:03, 2-94 3-14 3:01 Sagittal are 358 | 368 | 347 | 358) 366) 365) 365 | 350] 358} 350) 358) 373] 341 3D5 | 349 364 359 371 356 Frontal segment 122 | 126) 113) 122 124 | 125) 135 125 123) 117) 117 131 111 124 123, 125 118 125 120 | Parietal ,, 125 | 127 122 | 123} 125) 184) 118} 120) 122') 126) 127) 127 119 120} 115 131 130 127 124 Occipital ,, 11) 115) 112] 118] 117) 106) 112} 105 /113(2)) 107) 14) 115) 11) 112) 1 108 111 119 112 Transverse arc 303 | 292 | 286 | 294 | 300 |) 309] 309] 297 |326(x) |290(})| 304) 298 | 287 |) 293 |) 298 312 302 304 306(4) Cephalic index 77-5 | 75:3 | 74:6 | 75:1 | 73:0 | 78:4 | 76-7 | 77-5 |79-4(4)|73-8(x)| 78:5 | 73-5 | 77-9 | 77-5 | 80:5 80-6 17:8 15:3 17:8 | Height sf 2 72°8 | 68:5 | 70-5 | 74:0 | 71-9 | 71-0 | 75:6 | 67-1 | 74:1 | 68:6 | 73:8 | 66-3 | 75-5 | 74:0] 72:6 73:5 73-7 68:0 1-2 Height-breadth index 94:0 | 91:0 | 94:5 | 98-4 | 98-4 | 90-5 | 98:5 | 86:5 | 93:3 | 92:9 | 94-0 | 90:2 | 96-8 | 95-4 | 90-1 9-2 94-7 90:3 99-2 Nasal index 44-4) 47-4 | 53-5 | 47-5 | 47-6 | 45-2 | 44-7 | 48:8 | 48:8 | 50-0) 51-3 | 52:8 | 53-7 | 47-6 | 50-0 55-0, 50:0 51:3 50:0 Orbital ., 97-0 | 93:9 | 88-2 | 87-5 | 86:5 | 87-9 |100-0 | 83:8 | 94-1 | 94-1 | 90:6 | 88:2 | 87-9 | 88:6 | 93-7 97-1 93:9 91-2) 100-0 Alveolar ,, 91-2 | 93:0 | 98:9 | 94-1 | 96:3 | 95-4 | 90-4 | 93:5 | 91-9) 92:2 | 87-0 | 92:9 | 94-4 | 92-9 | 91-2 89-2 89:8 89-2 87-2 De M. Youxc —{8) Trans. Roy. Soc. Edin., Vol. LI.] JUVENILE SERIES “1.” Number. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Age, years 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 11-12 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 13-20 | 18-20 | 11-12 Sexe ic, 2: 35 20 58 &6 a fe os o0 te a0 +» | (imetop,) | Gnetop.) Cubic capacity - 5 Sc De $5 30 a3 30 60 20 20 90 oe o6 06 ve on we me Glabello-occipital length 185 | 182] 181} 170) 183) 187] 173} 177} 184) 178) 172) 176] 178} 181 | 175) 175) 171) 176 Ophryo-occipital _,, 182} 183) 183 |) 171 184 | 188} 172) 178) 184) 173} 171 Wiz} 179) 181) 175} 175). 171 | 174 Maximum breadth 3 140) 138) 144) 137] 141 139 | 136) 130) 137) 185) 131 132 | 133) 1386) 1385] 127) 128] 135 Minimum frontal breadth 94 94 95 96 99 93 89 91 96 90 89 92 82 88 93 93 96 95 Basi-bresmatic height 138 | 1382] 125) 126) 185} 127} 128) 125) 183) 1382) 117} 125) 129) 129) 119 | 129'} 127] 134 Opisthio-nasal Jength 138 | 183) 1380) 124] 186) 128} 124) 128] 134] 129) 122) 117) 124] 131 | 122) 180) 125] 131 Basi-nasal 10] 99 94 89 | 100 92 92 92 99 97 90 88 91 96 91 96 95 96 Basi-alyeolar 89 94 89 82 97 92 91 83 83 862} 89 84 84 89 87 88 90 84 Nasi-alyeolar 74 65 67 57%] 63 62 61 65 63% 622) 59 56 66 63 65 57 65 74 Manxillary-facial index (a) 57-3 | 56-5 | 57-2 | 50-9 | 50:2} 54:8 | 52-5 | 63-5 | 55-7 | 54:8 | 52-6 | 50:0 | 60-5 | 55-7 | 58:5 | 50-9 | 58-0 | 63-8 Interzygomatic breadth 129) 115] 117) 112) 125} 113) 116) 107) 113} 113) 112) 112) 109} 113) 111 112} 112) 116 Mamillary-facial index (6) 78:7 | 80-2 | 77-0 | 65:5 | 69:2 | 72-9 71-7 | 86:6 | 72-4} 73:8 | 71:0 | 70:8 | 84:6 | 78-7 | 83:3 | 67-8 | 80:2 | 89-1 Intermalar breadth 94 81 87 87 91 85 85 75 87 84 83 79 78 80 78 842) 81 83 Inter-stephanic breadth 118 | 122) 125) 117) 122) 115) 113) 117) 117) 116) 113) 112) 112) 113) 117) 110) 113) 121 Asterionic diameter 110} 109] 106} 104} 111} 110] 107 99] 111) 105} 102) 103} 103) 110] 105) 101) 101} 104 Transverse base 5 117 | 105} 113) 112} 115) 108} 106 98} 107) 105} 108) 104} 100) 110] 102) 103) 102} 110 Length of foramen magnum 38 37 35 35 37 37 32 36 37 34 32 29 35 36 33 35 33 38 Nasal height - 53 48 50 45 47 43 45 48 49 46 49 41 50 45 46 42 48 55 , breadth 20 21 23 22, 23 22 23 20 19 19 24 23 21 23 20 21 25 24 Orbital height (R.) 34 31 33 32 30 32 31 3D 33 32 30 28 30 29 31 29 31 36 = (ty) 35 31 33 32 31 32 31 34 33 30 30 28 32 29 31 29 32 36 breadth (R.) 38 36 38 36 36 36 38 35 36 35 34 34 33 36 36 36 36 35 » +» {@t,)) 38 36 38 35 38 36 38 36 36 35 35 34 34 36 35 36 35 385 Palatal length (F.) 47 44 45 39 49 45 48 43 42 422) 44 43 44 45 45 42 48 48 index 80-8 | 77-2 | 82-2 | 87-1} 71-4] 84-4 | 72:9 | 72:0 | 83-3 | 85-7 | 84:0 | 72:0 | 79:5 | 80:0 | 66:6 | 76:1} 68-7 | 72:9 » breadth 38 34 37 34 35 38 35 31 35 36 37 | . 31 35 36 30 32 33 35 + length (TURNER) 50 46 48 422) 51 49 50 46 442) 442) 47 45 46 47 48 45 50 50 Palato-maxillary index 120-0 | 119-5 | 131-2 | 142-82) 117-6 |126:5 |118-0 [115-2 |138-6 |136:3 | 125-5 | 117-7 | 121-7 | 121-2 |108-3 | 120-0 |120-0 | 120-0 Palatal breadth 60 5D 63 60 60 62 59 53 61 60 59 53 56 57 52 b4 60 60 Horizontal circumference 518 | 512 | 520} 487) 516] 518 | 490] 497] 515 | 490] 482) 490} 494) 503] 494) 485 | 486} 497 | Ophryo-occipital circumference 515 | 510} 521 | 489] 516) 520} 491] 498) 515 | 490) 480) 492} 495 |) 504) 494) 484) 484) 495 Arc to base ratio 2:68 | 2-74 | 2-86 | 2-84 | 2-74 | 2-94 | 2:82 | 2:83 | 2-75 | 2-76 | 2:88 | 3:13 | 2:96 | 2-79 | 2:85 | 2-73 | 2-76 | 2-71 Sagittal are 370 | 365 | 372 | 353) 373 | 377 | 350] 363) 369) 357 | 352) 367 | 3868] 366) 348 | 356] 345 | 355 Frontal segment 120} 126) 128) 127) 183) 129] 120) 134] 131 | 123] 114) 128] 120) 125] 119) 124) 117) 126 Parietal 129 | 124) 132] 119} 127) 126] 120) 126} 126) 125] 126) 108} 183) 120) 121) 123) 111) 115 Occipital ,, 121} 115} 112) 107} 113) 122) 110] 103) 112) 109) 112) 132} 115) 121 108} 109, 117} i114 Transverse arc 310 | 312] 314} 300} 325) 310) 305] 315) 312] 307} 283) 293] 302) 302) 298] 291} 291 | 303 Cephalic index 75:7 | 75:8 | 79:6 | 80:6 | 77-0) 74:3 | 78:6 | 73-4) 74-5 | 78:0) 76:2 | 75:0) 74-7 | 75-1) 77-1 | 72:6) 74-9 | 76:7 Height % 746 | 72:5 | 69-1 | 74:1 | 73:8 | 67-9 | 74:0 | 70:6 | 72:3 | 76:3) 68:0 | 71:0 | 72:5 | 71-3 | 68:0 | 73-7 | 74:3 | 76:1 Height-breadth index 98:5 | 95:6 | 86:8 | 91-9 | 95:7 | 91-3 | 94-1 | 96-1 | 97-0 | 97-7 | 89-3 | 94-7 | 96-9 | 94-8 | 88:1 | 101-5 | 99-2 | 99:2 Nasal index 37-7 | 43:8 | 46:0 | 48-9 | 48:9 | 51-2} 51-1 | 41-7 | 38-8 | 41-3 | 49-0 | 56-1 | 42:0 | B1-1 | 43:5 | 50:0 | 52-1 | 43:6 Orbital ., 89:5 | 86:1 | 86:8 | 88-9) 83:3 | 88:9 | 81-6 |100-0 | 91-7 | 91-4 | 88:2 | 82-4 |) 90:9) 80:6 | 86:1 | 80-6 | 86-1 |102-9 Alveolar .. 88:1 | 94-9 | 94-7 | 92-1 | 97-8 |100:0 | 98:9 | 90:2 | 83:8 | 88:7 | 98:9 | 95:0'| 93:1 | 92-7 | 95:6 | 91-7 | 947 | 87-5 De M. Yousa—{9) Trans. Roy. Soc. Edin., Vol. LI.) METOPIC SKULLS.—SERIES “A.” Sie 10 | 14 | 12 | 18 | 14 | 45 | 46 | 47 | 18 | 19 | 20 | of | 92 | 28 25 ae MM | wm | we) Mw iw | wm | wm | we | mw | we | Me || we | om D. Ganges ae 1450 | 1640 | 1220 | 1350 | 1420] .. | 1400 | 1420 | 1360 | 1630 | 1430 | 1520 | 1350 | 1640 1210 Glabello-occipital length 186 | 185 | 177] 186 187] 184) 175 188 | 180] 191) 180] 196] 186) 196 172 Gintama 184 | 185] 176] 184] 185 182) 176 | 187] 178] 189] 180] 194) 195 |* 196 172 atasimuan breadth 138 | 188] 137 | 133} 135) 140] 142) 134" 187) 145 || 138 | 141 | 135 | 14d 136 | Minimum frontal breadth 95| 100] 92] 101] 95] 101} 103; 88] 98] 105] 104] 102) 98) i111 104 Basi-breematic height 130 | 138] 125 | 127] 132] 130] 126] 191] 130) 134] 133] 134) 124) 140 123 138 | 128] 127] 131] 138] 134] 128] 135 133] 132) 132] 136] 133] 146 124 = 103] 95}. 98] 97] 103) 102] 93) 99] 97) 99] 99) 104] 99) 105 93 Basi-alveolar ,, 99 | 91] 94] 912) 99] 92) 92) 98] 89) 86] 93] 100] 97] 95 87 Nigeioare 2 | 72) 75) 72) 72%) 67] 79] 66) 65) 66] 662] 7) 74| 69) 738 63 | Maxillary-facial index (a) | 56:6 | 59:0 | 60:5 | 55:3 | 540] 62:2 | 52:3] 546] 50-7 | 50:0 | 57-2) 57-8) 57-9 | 545 50:8 Interzygomatic breadth . 127] 127] 119] 130] 124] 197] 126} 119] 130] 132] 194] 1298] 19] 143 124 Maxillary-facial index (6) 75-7 | 78:9 | 80:0] 77-4 | 72-8 89-7 | 7-6] 81-2 | 725 | 81-4 | 79-7 | 81-3 | 8-1 | 75-7 73-2 inks Ee 95| 95] 90] 93] 92] 88] 85] 80! 91) sil] 89] 91] 85] 103 86 Inter-stephanic diameter | 112] 114} 105) 115] 115] 115] 119) 116) 118] 133] 123) 120] 121) 198 123 Asterionic diameter 103 | 111} 100} 110} 113} 116) 103 112) 110} 100] 108] 107] 109] 112 105 Transverse base s 109} 110} 107} 115] 112)-118) 112) 117) lie] 120] 114] 114] 112} 198 111 Length of foramen magnum . 352] 933) 31 936) “35 | 36 36 || 35 || 37 35) 34) 34)! ‘35 | “47 31 Nasal height - bb] 55) bl] 52) br} 59) 47) Bi) 40)) 53) 521) 55] Bl] GD 46 } ., breadth 92) 21] 93| 923] 93) 290] 92) 90] 21] 20) 99] 23] 91) 99 21 Orbital height (R.) 35 | 32 || 32)|| 35) 33] 35] 391] 34] 32] 34] 36) 34) 34) 37 32 = Gta). a] SB) SB] Se) si) aol] Si Bel Bi} eB) sal se) sel sy 32 breadth (R.) 40)| 391) 36] 37 38! 387 || 35 || 387 371] 38)| 361) 401) 40) 43 37 S (ata) | 39 || 38 | 36) 40) 39'| 37] 35) 37) 38!) 38\|| 36)| 40)|| 40) 44 37 Palatal length (F.) . 50) 52] 49) 4521 49) 53] 48) 53] 47] 48] 49] 56] 6502] 51 46 index . 76-0 | 61-5 | 71-4) 866 | 77-5 | 69:8 | 68:7 | 71-7 | 74-4) 93-7 | 71-4) 57-1) 78:0) 86:2 76:0 breadth 38| 32] 85] 391) 38) 37] 33| 38] 35] 36) 35) 32)| 39.) | 45 35 length (TURNER) 52| 55] 50) 472] 52) 53] 52) 53] 49] 47) 51] 59] 52%] 52! 472 A , 123-0 |103:6 |112:0 |129-7 |119-2 |107-5 }111-9 |109-4 |108-1 | 112-7 | 107-8 |103-4 | 105-7 | 128-8 114-8 64| 57] 56] 61] 62] 57] 53] 582] 5321 5321 55] Gl] 55| 67 5d Horizontal circumference 528 | 525 | 508| 516 | 523] 524) 513| 521 | 513 | 543 | 516] 549) 518] 557 505 Sagittal arc 367 | 400 | 352) 367] 370 | 362] 357] 373| 360] 390) 365] 388| 374] 393 359 Frontal segment 128 | 135 | 120] 130) 120) 130) 122] 125] 190} 135] 127] 180} 198 || 143 123 Parietal 112) 140] 125) 125] 125 115} 123 125] 120] 125] 193) 127] 129) 196 128 Occipital ,, 127 | 125] 107] 112] 125] 117) 112) 123) 120] 130] 115] 131) 117 | 124 108 Transverse arc 312 | 315 | 300) 293] 302] 300] 310] 303] 303] 320] 316] 305 | 302] 329 305 feat index 74:2 | 74-6 | 77-4) 71-5 | 72:2) 761 | 81-1 | 71:3] 761) 75-9) 76-7 | 71-9) 72:6 | 73:5 79-1 69-9 | 74:6 | 70:6 | 68:3 | 70:6 | 70-7 | 72:0 | 64-4 | 72:2 | 70:2 | 73-9 | 684 | 66-7 | 71-4 71-5 He Sit breadth index 94-2 1100-0 | 91-2 | 95:5 | 97-7 | 92:8 | 88-7 | 90:3 | 948 | 92-4 | 96:3 | 95:0] 91-8 | 97-2 90-4 Nasal index 40:0 | 38:2 | 45-1 | 40-7 | 45-1 | 33:9 | 46:8 | 39-2 | 42:9 | 37-7 | 42:3 | 41-8 | 41-2 | 49-1 45-7 Orbital ,, 87-5 | 82-1 | 88:9 | 94:6 | 86:8 | 94-6 | 91-4 | 91-9 | 86-5 | 89:5 |100:0 | 85:0 | 85-0 | 86.0 86-5 Alveolar ,, 96-1 | 95:8 | 95:9 | 93:8 | 96-1 | 90:2 | 98:9 | 99:0 | 91-8] 86:9 | 93:9 | 96:2 | 98:0 | 90-5 93:5 Dz M. Youse—{10) Age MUSE Uy f AY > f -~ fe. oe F am See - Trans. Roy. Soc. Edin., Vol. LI-] METOPIC SKULLS.—SHRIES “B.” Nanri fae 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23, 24. 25 | | Age | Sox ey oe 1 M | F. F. 1 D. Fr. F. D. F. F. R D. D. M. | M. R. Kr. F. Gabic capacity i Z 1530 | 1680 | 1520 | 1300 | 1300 1390 1280 | 1420 | 1470 | 1320 | 1270 | 1370 | 1550 | 1290 | 1520 | 1200 | 1270 | 1310 | 1310 Glabello-occipital length 1g] | 192] 175 | 175) 178] 180] 177 || 179) 185) 179) 175 || 185) 186) 174] 183] i77 || 172) 181 | 174 Ophryo-occipital 1s1| 191] 175] 176] 178| 180) 177} 180] 185| 179) 175) 185) 186} 174] is4) 178) 172 || 182) 175 Masimum breadth. 147 | 142) 144) 138] 138] 134] 130) 137] 133 | 181 | 135) 140) 139 || 139 | 145) 134) 138] 138] 134 SR 99| 101} 96) 101} 95] 96] 93) 100) 97] 97] 95) 104) 106) 99| 99] 100) 89) 103] 93 Bassbeamatiohaeht 134] 198) 138] 125] 129) 135 | 197 |.124| 130] 125] 126) 121 | 136] 181} 180) 125) 194] 195 || 127 Opisthiosnasal lence 127 | 143 | 127] 124] 199) 199) 193) 124] 130] 181 193) 199] 182] 129) 126) 191 | 1297] 185 || 198 asal 94 104 94 95 91 98 91 96 95 94 91 Chir Gs | Shi 93 90. 92) 100 94 lveolar 39| 972] 86| 88| 86] 90) 88] 89, 88] 862| 82%/ 88| 862] s8| 90] 85] 85) 99] 92 imal os : 66| 70!) 65] 69| 67] 69] 68| 6s| 72) 60% 65] 6r| 66) 67| 74) 70) 64) 68| 66 eee liniles (a) 550 | 560 | 55-5 | 55:2] 56-7 | 55-6 | 586 | 57-6 | 61-5 | 53-1 | 52:0] 53-6] 53-6 | 55-3) 59-2] 58:8 | 53-7 | 57-6 | 55.0 Tatees eoatie bread this 120 | 125| 17) 125) 118} 194) 116} 1182) 117} 113) 125) 125] 123] 191) 125] a9) 119] 18] 1920 Maxillary-facial index (B) 795 | 853 | 77-3 | 76-6 | 827 | 75:8 | 79:0 | 781) 81-8] 73-1 | 74-7 | 72:8 | 69-4 | 79:7 | S81 | 79-5) 75:3) 809 | 77:6 Titernalar Booed g3| 82] 84| 90| sl] 91] 86] 87| ss! 82! 87] 92) 95] 84) 84] 88] 85) 84) 85 Inter-stephanic depth 125 | 120) 120 |- 119} 123) 121) 115) 126] 195) 117; 119] 120] 199] 191) 126) 120] 12) 123) 116 Asterionic diameter 112] 114) 113) 105] 105] 105] 101 | 107| 102) 102] 107) 106] 110) 110] 104) 106] 106| 107) 110 Transverse base 3 114} 119} 112) 111} 10) 109) 103} 104) 109 10) 115) 110) 112) 112) 11) 107) Wi) 10 08 | Length of foramen magnum . 58) aol) Bel) al eal] a) sel) Bo] sai ar] sal at) eB a) sal sn] ee] gal ss Neealiedn: - 4g) 54] 48 | 48 || 491] 52)| 50) 46) 53] 47) 50|| 51] 50 50l| 53) 5O|| 47) 47) 47 breadth 22 22 24 24 22 24 22 23 23, 20 22, 22 25 19 20 21 24 20 26 Orbital height (R.) 63 || SBI S| ali) sai 6a) a2\| ga) a2) a) a a3) a2] sal ae] soi al) sal a = i) 34 32 31 30 33 31 34 34 34 32 35 37 35 33 32 3) 32 33 32 breadth (R.) 35 35 34 36 36 37 36 36 3D 36 39 40 37 | 36) - 35 35 35 36 36 = a (L.) 35 35 GEE 37 36 36 36 37 36 36 39 40 38 35 35 36 36 36 37 Palatal length (F.) - 46 482) 44 47 49 44 51 47 48 44% 472) 48 472) 50. 50 46 44 50 50 » index 5 82:6 | 77-0 | 77-2 | 80-8 | 65-3 | 86:3 | 70:5 | 74-4 | 66:6 | 77-2 | 80:8) 64:5 | 72:3 | 66:0) 70:0] 71-7 | 684 | 70:0 | 74-0 breadth 38 37 34 38 32 38 36 35 322] 342) 38%) 31 34) 33 35 33 30 35 37 ., length (TURNER) 47 492) ~ 46 51 51 49 52 50 51 45% 49%) 51 492) 51 52 49 45 52 51 Palato-alveolar index 127-6 |120-4 |119-5 | 121-5 |111-7 |124-4 | 113-4 | 112-0 |101-9 | 120-0 | 102-0 | 107-8 | 102-0 | 109-8 |109-6 | 118-3 | 117-7 |115-3 |113-7 Palatal breadth 60 59 55 62 57 61 59 56 52? 54? 50? 55%] 50 56 57 58 53, 60 58 Horizontal circumference 528 | 541 | 505} 505] 508] 508) 497] 509) 5l4 |] 505) 505) 529) 530] 503] 5bI8|) 505 | 490} 510) 503 Sagittal are 376 | 380) 371 357 | 366 | 372) 365] 371 378 | 362) 362] 372} 390) 357) 384] 368) 354] 364 | 362 Frontal segment 132/) 132 | 125) 125 133, 133 | 126 126 132 | 120) 180; 141 140 | 117 130] 128] 118} 115) 123 Parietal 120} 188) 120 115 117 120) 127 126 | 185} 110) 120) 117) 132) 128) 1380) 122) 125) 131 115 Occipital 124} 110] 126) 117} 116 119} 112) 119) 111 132% 112) 114) Ws} 112} 124) 118) 11t 118} 124 Transverse arc 320) 307} 328} 304) 296 | 304) 294] 307) 309) 290 ' 300} 300) 815 | 310] 322) 3807) 293 | 300] 298 Cephalic index 81-2 | 74:0 | 82:3 | 78:9 | 77-5 | 74-4 | 73-41 76:5 | 71-9 | 73:2) 77-1 | 75-7 | 74-7 | 79:9) 79:2 | 75-7 || 77-3 | 735 | 77-0 | Height Ss 2 74:0 | 66-7 | 78:9 | 71-4 | 71-9 | 75:0 | 71:8 | 69:3.) 70:3 | 69:81 72:0) 65:4 | 73-1) 75:3 | 71-0) 70:6 | 72-1 | 69-1 | 73:0 | Height-breadth index 91-1 | 90-1} 95:8) 90:5 | 88-4 |100-7 | 97-6 | 90-5) 97-7 | 95-4 | 93:3 | 86:4 | 97-8 | 94-2 | 89:6 | 93-2 | 93-2 | 93:9 | 94-7 Wasal index 5 45:8 | 40:7 | 50:0 | 50:0 | 44:9 | 46:2 | 44.0 | 50:0) 43-4 | 42:6 | 44-0 | 43-1 | 50-0) 38:0) 37-7 | 42:0 | 51-1 | 42:6 | 55:3 Orbital ,, 94-3 | 94:3 | 91-2) 86:1 | 88:9 | 86:5 | 94-4 | 94-4 | 97-1 | 88-9] 89-7 | 95:0 | 91-9 | 91-7 | 91-4 | 88:6 | 91-4 | 91-7 | 88-9 Alveolar ,, 94-7 | 93:3 | 91-5 | 92-6 | 94:5 | 91-8 | 96-7 | 92-7 | 92:6 | 91-5: 90-1 | 90-7 | 87-8 | 90-7 | 96-8 | 94-4 | 92-4 | 99:0) 97-9 De M. Youse—{11) Trans. Roy. Soc. Edin., Vol. LI.] METOPIC SKULLS.—SERIBS “H.” Number. i 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 } a ee M. M. M. | M. M. M. M. M. M. M. M. Dz. M. D. D. M. F. Fr D. D. D. F. F. F. F. Cubic capacity 3 1440 | 1320 | 1430 | 1500 | 1440 | 1410 | 1300'| 1280 | 1440 | 1250 | 1350 | 1380 | 1230 | 1270 | .. | 1280 | 1330 | 1180 | 1390 | 1180 | 1180 | 1270 | 1430 | 1470 | 1430 Glabello-occipital length 190 | 177 | 187] 181} 187 | 1824 174] 178] 191 | 179) 180.) 186} 175) 173] 175) 178) 174) 168) 181 | 168] 171 | 178 | “174 | Iso | “173 Ophryo-occipital 188 | 176} 187 180 186 181 173 176 | 193 178 | 180 184 176 171 178 178 174 | -167 182 | 167 172 180 | 176 181 174 Maximum breadth - - 141 | 132 145 | 139 140 140 131 132 | 142 135 140 131 131 135 139 137 139 | 135 134 126 137 130 135 136 142 Minimum frontal breadth } 101 97 100 | 102 103 | 100 93 93} 104 99 96 96 92 97 100 95 97 90 97 94 91 97 89 97 96 Basi-bregmatic height | 134 127, 129 | 132 129 127 129 123)) 129)) 124 126 123 123 124 131 119 128 | 128 133 120 122) 124 129 128 | 130 Opisthio-nasal length 135 | 139] 130] 182] 136 | 126) 128] 129] 129] 128) 131) 131 |} 128) 124) 126) 129 || 124) 192 || 181) 127) 126) 121 || 116) 198) 195 | Basi-nasal 101 | 100 98 96 | 103 90 94 93 97 97 94 99 92 90 93 94 94 92 97 93 93 89 87 95 93 Basi-alveolar 94| 94| 91] 86] 912] 88| 91] 91] 984%] 95] 90] 93) 97] 82] 90) 82] 988] 82] 92) 90] 92! 99) 77/| 89] 96 (Smear : 6 | 70) 64| 69) 6£| 70] 66] 70| 67| 68] 72) 72\| 65 | 70) 62) 67| 64| 68] 65 || 67 | 68| 65 | cal 64) 66 Maxillary-facial index (a) 52-8 | 58:3 | 51-6 | 54-7 | 48:8 | 54-2 | 55:0 | 57-8 | 54-4 | 55-7 | 55-3 | 61-5 | 52:0} 57-3 | 52:5 | 54:9 | 54-7 | 56:6 | 54-1 | 57-2) 55-2 | 55-0 | 54-7 | 55-1 | 55-9 Interzygomatic breadth - 123; 120 | 124 126 131 129 120 121 123 122 130 7 125 122 118 122 117} 120 120 117 123, 118} 115 116 118 Maxillary-facial index (b) | 74-7 | 81-4] 68:8] 92:0| 74-4] 72:9) 75:0) 83:3| 75:2] 74-7 | 82:7) 85-7 | 72:2] 7-7] 74-7 | 761 | 75:3) 80-0) 755 | 8-7 | 755 | 71-41 79-7 | 771 | sap Intermalar breadth . | 87 85 93. 75 862 96 88 84 89 91 87 84 90 90 83 88 85 85 862 82 90 91 79 83 80 Inter-stephanic breadth . } 117} Wy) 124) 127) 118) 122) 17) Wy) 128) 117) Wb) 119) 112) 11s) 127) 115} 118) 118 || 123) 116} 115) 117) 120))) 121) 126 | Asterionic diameter 115 110 | 106 106 | 110 110 106 113) 111 100 113 97 107 109 104 104 107 | 100 105 99 107 102 93 104 107 Transverse base : 12} 11) 115) 117) 7) 115) 10} 113} 113) 109} 117} 107) 114} 111) 110)) 112} 109) 109} 112) 103} 113:) 104) 100) 11} io Length of foramen magnum - 35 39 33 39 34 36 34 37 33 32 37 33 35 34 33 35 32 31 36, 36 32 32, 31 bd 34 | Nasal height - 47 50 49 51 53 49 48 50 49 47 54 53 47 53 46 49 51 48 49 49 49 48 49 49 49 | ,, breadth 23 22 26 18 24 21 24 20 23 24 23 22 24 22 23 24 22, 20 22 22 23 23 20 21 21 | Orbital height (R.) 32 37 32 36 362) 37 31 34 34 34 37 36 29 34 30 33 34%} 32 312) 35 31 28 30 32 34 = 5 _ (iy) 32 | 36 32 36 36 36 32 35 34 33 37 35 30 32 31 31 342/32 312) 35 30 28 28 32 33 breadth (R.) 39 39 40 41 4]? 40 36 37 41 35 40 36 37 37 37 38 37 38 38 37 35 34 35 36 35 3 (L.) 38 39 40 4] 41 40 36 37 4] 35 40 36 37 37 36 38 372) 38 38? 36 35 3B4 34 36 35 Palatal length (F.) . 49 47 46 48 497 47 50 46 47 52 47 48 53 47 45 48 50 44 46? 47 50 49 41 44 45 index 75:5 | 78:7 | 97-3 | 66:6) 73-4 | 72:3 | 66:0 | 76:0) 68:0 | 69-2 | 78:7 | 75:0 | 73:6 | 68:0 | 77-7 || 72:9 | 72:0°} 77-2 | 82:6 | 70:2] 66:0 | 75:5 | 80-5 | 79:5 | 77-7 breadth 37 37 45 32 36 34 33 35 32 36 37 36 39 32 35 35 36 34 38 33 33 37 33 35 35 length (TURNER) 51 50 482) 50 507 50 52 51 50% 54 52 51 55 47 48 51? 52? 462 48? 48 50 51 44 47 48 | Palato-mazillary index . 4 2-0 2 2 0 ¥, ‘ 4 * 6 : H . +4 |116-6 | 114-5 | 114-0 | 113-7 | 120-4 | 119-1 | 116-6 Palatal breadth 56 55 57 58 53 56 56 Horizontal circumference 5O7 | 482 | 494) 498] 496} 508} 504 Ophryo-occipital circumference 506 | 479} 498} 500} 499] 509} 505 Sagittal arc 365 | 330] 348 | 372) 375 | 372} 361 Frontal segment 126) 123) 118} 132} 131} 129) 133 Parietal, 126 | 107} 115) 128%) 119} 133) 117 Occipital ,, 113 | 100} 115) 112} 125) 110) 111 Transverse arc 303 | 283} 294) 298 | 308] 299} 320 Cephalic index 74:0 | 75:0 | 80:1 | 73:0 | 77-6 | 75:6 | 82-1 Height a z 73:5 | 71-4 | 71:3 | 69-7 | 741) 71-1) 75-1 Height-breadth inde 99:2 | 95-2 | 89:0) 95:3 | 95-5 | 94-1 | 91-5 Nasal index 44-9 | 44-9 | 46:9 | 47-9 | 40:8 | 42:9 | 42-9 Orbital ., 81-6 | 94:6 | 88:6 | 824) 85-7 | 88:9 | 97-1 Alveolar ,, 91:8 | 96:8 | 98:9 }100:0 | 88:5 | 93-7 | 92:5 Dz M. Yousa—{12) ( 455 ) X.—Skiagraphic Researches in Teratology. By Harry Rainy, M.D., F.R.C.P.E., and J. W. Ballantyne, M.D., F.R.C.P.E. (With Fifteen Plates.) (Read February 15,1915. MS. received November 5, 1915. Issued separately December 31, 1915.) The study of human teratology has always been rendered difficult by the relative scantiness of available material; consequently it is of great importance that such cases as can be obtained should be fully utilised and recorded. It has seemed to us that in this respect the employment of X-rays is likely to prove of service, for the abnormalities of the skeletal structures can then be readily and fully studied without interfering with the integrity of the specimen or its utility in the investigation of other elements. We therefore desire to place on record the skiagraphic findings of several series of developmental anomalies in the human subject, in order that they may be available for future workers. Some of the specimens have been already described from the anatomical stand- point, without X-ray photographs, and references are given in such cases ; others are presented in this communication for the first time, and we desire to express our indebtedness to those who have placed the material in our hands, as well as to the Trustees of the Moray Fund, who have aided us in meeting the expense of producing the skiagrams. We also wish to acknowledge our obligations to Dr Berry Harr for many valuable suggestions. One of us* has elsewhere pointed out that there are three main subdivisions of antenatal pathology, corresponding to the three main subdivisions of antenatal life. Of these subdivisions of the subject, the first is foetal pathology, characterised in ereat measure by the same diseases as those of the ehild and adult. The second is embryonic pathology, or, as it is more commonly called, teratology, which deals with the monstrosities of the embryo, for at this stage morbid agencies produce not diseases but malformations and monstrosities. These are naturally carried forward when the embryo becomes a foetus, and so the condition, though a product of early intrauterine pathology, is continued into the later stages of life. The third, and in some respects the most interesting, part of antenatal pathology is concerned with the action of morbid causes on the organism in the early germinal period. It is at this period that many of the teratological formations that are usually referred to em- bryonic pathology most probably arise, and if we tentatively accept the suggestion that “unit characters” are fixed by the fate of their “determinants” in the germinal stage of development (and especially at the moment when the polar bodies are expelled), we may obtain a scheme of classification of certain well-defined monstrosities * J.W. B., Glasgow Med. Journ., xlix, 241, 1898. TRANS. ROY, SOC. EDIN., VOL. LI, PART II (NO. 10). 64 456 DR HARRY RAINY AND DR J. W. BALLANTYNE. which, whether it is eventually found to be true or not, will at least help us to arrive at a clearer conception of their production. The examples whose skiagrams form the subject of this communication may be arranged, in accordance with the suggested classification, into: (A) cases where a complete system of “determinants” is absent; of this group achondroplasia may | serve as a type: (B) cases where parts of one or more systems of “ determinants ” are absent; this is exemplified in anencephaly and aprosopia: (C) cases where the “ determinants” though apparently present may, in consequence of some inherent defect, lead to anomalous developments; some instances of iniencephaly may be placed in this group: (D) cases where there is an excess of “ determinants,” as in polydactyly and dichirus. ; These various defects in “‘ determinants” are revealed in the soma of the cases that form the subject of this paper, but doubtless the defect is transmitted, at least in some cases, in the germ plasm also, and so may become hereditary, if the somatic deformity is not so serious as to preclude the fcetus from attaining to sexual maturity. There are many forms of monster which do not find a place in the above scheme, for, even where the character and grouping of the “determinants” are normal, the moulding of the embryo is interfered with by pathological occurrences during intia- uterine existence, and there is but little doubt that the amnion (whose formation in the mammalian embryo is very different from that described in the chick) is fre- quently responsible for grave disturbances of normal development. Group A. (Characterised by total absence of a system of determinants.) This type is represented by the following cases of achondroplasia. The first case is that of a female infant born in the practice of Dr Mowar of Edinburgh in 1903. The external appearance is shown in fig. 1, Plate I, and a frozen section of the body is reproduced in colour in Plate I, fig. a. The X-ray photograph (fig. 2, Plate Il) shows the characteristic defect in the formation of those bones of the basis cranii which are early prefigured in cartilage, causing marked indentation at the root of the nose. It also demonstrates the extreme shortness of the long bones, and the broadening of the hands. The second case is that of a full-time male foetus which also presents a very typical example of the condition. Fig. 3, Plate III, shows the general appearance at birth ; fig. 4 shows the broad, short forearm and hand; whilst fig. 5 is a stereoscopic photograph of the head, which*enables one to distinguish with great clearness the details of the defective development in the basis cranii. Fig. 6, Plate IV, represents the subsequent development of the wrist and hand bones in achondroplasia, as exhibited in an adolescent patient. Fig. 7 shows the final result in an adult case of the same abnormality. SKIAGRAPHIC RESEARCHES IN TERATOLOGY. A457 Further illustrations of this condition are figured in The Treasury of Human Inheritance, parts vii and viii, section xv, A (Hugenics Laboratory Memoirs, Dulau & Co., London). Group B. (Characterised by the absence of part of one or more systems of determinants.) Fig. 8, Plate V, represents a sagittal section of an anencephalic foetus, which occurred in the practice of Dr FERGuson of Anstruther in 1899. It is a female, as is so often the case ; and the spinal canal is closed save in the cervical region, constitut- ing the derencephalic variety of anencephalus. The X-ray photograph (fig. 9) shows a certain measure of that excess of bone in the upper part of the spine which has been recorded in similar cases. It will be noted, in this and the following photographs of anencephalic terata, that there is usually, but not always, an abnormal curvature of the cervical vertebral column just below its articulation with the skull. Fig. 10, Plate V, shows the external appearance, and fig. b, Plate I, the sagittal section, and fig. 11, Plate VI, the skiagram of an anencephalic feetus, which was sent for examination to Dr BatuanryNE by Dr Rorte of Cardenden in 1898. The foetus, a male, exhibits the characteristic features of anencephaly, along with marked lordosis of the cervical vertebre. There is. practically no chin-groove in front, and the spinal canal is closed below the region of the neck. The lower jaw and clavicle are shown to be well formed; the ossification of the spine is unusually good for anencephaly, except in the cervical region; the sternum is unossified, and therefore is practically invisible in the skiagram, thus contrasting markedly with the clavicle. Fig. 12, Plate VI, is the skiagram of another specimen of anencephalus. The foetus occurred in the practice of Dr James Smrtu of Edinburgh in 1892, and is described and figured in the Transactions of the Edinburgh Obstetrical Society (vol. xviii, p. 256, 1893). The X-ray photograph shows the presence of the lower jaw, temporal bones, basis cranii, and ribs. The sternum is mainly cartilaginous, but small ossific centres can be seen. As in all anencephalics, the cranial vault bones are wanting. Fig. 13, Plate VI, is taken from another anencephalic foetus, which shows an open condition of nearly the whole spinal canal. The curvature of the spine is cyphotic in the cervical and lumbar, and lordotic in the dorsal, regions. An ex- aggeration of this curvature would mark the beginning of retroflexion of the foetus. _ The more extreme degrees of this curvature are shown in the specimens of iniencephaly. The next case we have figured is one of a very interesting monstrosity, in which anencephaly, lumbar spina bifida, a caudal appendage, and deformities of the limbs are combined in the same foetus. The infant was born in the practice of Dr C. A. Burcnart, in January 1894. The naked-eye appearances are shown in figs. 14 and 15, Plate VII. The most noteworthy features shown in the radio- 458 DR HARRY RAINY AND DR J. W. BALLANTYNE. graph (fig. 16) are the acute curvature of the lower lumbar region associated with the spina bifida, the well-developed state of the lower and upper jaws and of the basis cranil, and the deformed condition of the feet and knees (genu recurvatum). The latter abnormalities are often considered to be of amniotic origin. The condition of genu recurvatum may occur by itself, and is shown in fig. 17, Plate VII, where it was present in an otherwise healthy infant. Fig. 18, Plate VIII, shows a fcetus from the practice of Dr ArtHur Witson, Edinburgh, in 1893, with anencephalus and exencephalus. The details are given in the Transactions of the Edinburgh Obstetrical Society, vol. xix, p. 49, 1894. The X-ray photograph gives additional information regarding the basis cranii, where the pituitary fossa is clearly outlined and the basi-occiput is well-developed. Fig. 19, Plate VIII, is from an anencephalic foetus, full details of which were — published in the Jowrnal of Obstetrics and Gynecology of the British Empire for 1902 (vol. ii, p. 521). There was practically no doubt that the foetus was post- mature, for the labour, which was due on April 25th or 26th, did not occur till June 13th, and the anatomy of the monster showed advanced development of all its parts except its head (anencephalic). The circumference of the shoulders was 42 cm.; the weight was nine pounds, which was high when it is borne in mind that there were no cranial vault bones; the length was 55 em. The X-ray features bear out the diagnosis of post-maturity. Fig. 20, Plate IX, represents a foetus aprosopus (faceless foetus), which was born to a woman, forty-three years of age, as her first child. She afterwards had a normal infant. The mother is described as healthy and active, but very neurotic. The pregnancy, which terminated at about the eighth month, was accompanied by some hydramnios. The external appearances are shown in fig. 67 (p. 434) of the Manual of Antenatal Pathology, vol. u.* The malformation affects the head and the left hand. The former is reduced to little more than a rounded knob, and shows no external trace of nose, eyes, or mouth; there are two large external ears situated laterally. The left hand exhibited marked irregularity of the digits, the rmg finger being shorter than the little finger. The X-ray appearances are as follows :—there is an irregular mass of bone at the apex of the vertebral column, in which the lower jaw can be easily recognised and is seen to articulate at the level of the ears. Fig. 21, Plate IX, is a second case of foetus aprosopus, which was sent to Dr BaLLaNnryneE in the end of 1905 by Dr Macponatp. Several of the vertebral bodies in the lower dorsal region show malformation (half vertebree, etc:); the right clavicle is irregularly bent in its central portion; and the ossific nuclei in the lower ends of the femora are absent. The cranial bones form an irregular mass, in which it is impossible to separate the constituent elements. * Manual of Antenatal Pathology and Hygiene, by J. W. Bauuantyne, M.D. Edinburgh: Williain Green & Sons, 1902-1905. SKIAGRAPHIC RESEARCHES IN TERATOLOGY. 459 Fig. 22, Plate X, shows the skeletal structure in a so-called siren fcetus or sympodia. It will be noted that the lower part of the vertebral column is im- perfectly developed, that the pelvis is exceedingly rudimentary, being represented by one small iliac bone, and that only one lower extremity is present. This contains two bones, the upper of which represents the femur, and the lower the tibia. It will further be noted that the radius is symmetrically absent in both arms. This last abnormality is by no means an uncommon one, and is shown also in fig. 23 and in fig. 24, Plate X, the latter being taken from a child aged eleven years. As the pelvis and lower extremity in this case represent only one limb, the specimen may be classed as an example of the variety known as monopodia. Group C. (Characterised by the occurrence of determinants with inherent defects.) Fig. 25, Plate XI, is taken from a foetus iniencephalus which was given to Dr BALLANTYNE for examination by Professor Sir ALEXANDER Simpson, to whom it was sent with no clinical details. It is described in the Manual of Antenatal Pathology (vol. ii, pp. 275, 276), where also its external appearances (fig. 44) and a sectional view of the left slab (fig. 45) are figured. This monstrosity is so extreme that the human form is lost, the head and trunk together making a more or less rounded mass from which the limbs project. Through the extreme backward bending of the head upon the trunk, the nape of the neck loses its external position. The three factors in the production of iniencephaly are imperfect development of the occiput in the neighbourhood of the foramen magnum, spina bifida of considerable extent, and retroflexion of the trunk. The X-ray photograph supplements the frozen sectional view in respect to the state of the lower jaw, the ribs, and the pelvis. The jaw is fairly well formed, the ribs are much crowded together, the ossification of the lower half of the spine is very irregular, and the pelvis is rudimentary. Fig. 26, Plate XI, represents an iniencephalic foetus which occurred in the practice of Dr Murray Carrns of Liverpool in 1900. The labour was prolonged by reason of the deformity. In this case the X-ray appearances which chiefly call for remark are the acute flexure which is present at the upper end of the cervical spine, and the sharp bend in the lumbar region. Fig. 27, Plate XII, is taken from a foetus with exomphalos and a congenital cutaneous band attaching the right wrist to the margin of the exomphalic aperture. It was born in the practice of Dr Mactacan of Sleaford in 1901, and it is figured and described in the second volume of the Manual of Antenatal Pathology (fig. 34, p. 180). The X-ray photograph shows the right arm in its attached position. Opposite the point where the exomphalos occurs there is an acute flexure of the TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 10). 65 460 SKIAGRAPHIC RESEARCHES IN TERATOLOGY. spinal column. There is an extraordinary degree of distortion of the ribs, which is probably secondary to the malposition of the viscera. The flexure of the spinal column in these ,cases should be compared with the sigma curve of the column of primitive segments in the dorsal region of the early embryo (Manual of Antenatal Pathology, vol. ii, figs. 10, 11, and 12). Fig. 28, Plate XII, is taken from a specimen which has been described in the second volume of the Manual of Antenatal Pathology (fig. 75, pp. 518, 519). It represents anencephalus and exomphalos, with an extraordinary degree of curvature of the. spinal column. There is spina bifida affecting almost the whole vertebral canal. The monstrosity was born in the practice of the late Dr C. E. UNpERHILL of Edinburgh, and the details are given in the Transactions of the Edinburgh Obstetrical Society (vol. xvii, p. 278, 1892). The X-ray photograph shows that the lumbar spine is fairly straight, whilst there is marked kyphosis of the dorsal, and lordosis of the cervical, regions. : Fig. 29, Plate XIII, is somewhat similar to the preceding, but the spinal curvature is still more marked and more extraordinary. The fcetus was born in the practice of Dr BatiantyneE of Dalkeith, and details are given in the Transactions of the Edin- burgh Obstetrical Society (vol. xvii, p. 241, 1892). The X-ray photograph shows — very clearly the extreme degree of spinal curvature. Fig. 30, Plate XIII, is somewhat similar as regards the curvature of the spine, but there is no exomphalos. ‘The specimen came from Dr Durr of Helensburgh, and - details are given in the Transactions of the Edinburgh Obstetrical Society (vol. xx, p. 18, 1895). The X-ray photograph shows the extraordinary dorsi-flexion of the head and the abruptness of the S-shaped curve of the spine. Group D. (Characterised by the presence of an excess of determinants.) In fig. 31, Plate XIV, a good example is presented of the arrangement of the bones of the hand in a case of dichirus, and in fig. 32 in the same Plate are shown the doubled thumbs from a case of symmetrical polydactyly. Fig. 33, Plate XV, is a radiograph of thoracopagous united twins that occurred in the practice of Dr Roperr Stewart of Edinburgh. The external appearances are given in the Manual of Antenatal Pathology (vol. ii, p. 657). They had to be separated from each other in order to make their birth possible. The most marked feature is the extreme distortion of the spinal column in each. ‘This case is of con- siderable intrinsic interest, and perhaps it may be explained by assuming that the determinants were doubled and segregated at an early age. Hach ordinary deter- minant is thus represented by two separate determinants. If the process is complete for all the determinants, one gets “identical twins”; if it is partial, the result is the formation of complete or partial double monsters. | Trans. Roy. Soc. Edin. Vor. LI. Drs Harry Rainy and J. W. Batnantyne: “‘Skiagraphic Researches in Teratology.” —Puate I. Fic. 0. Fic. a. ~ -\\ a “ Trans. Roy. Soc. Hidin. Vou. LI. Drs Harry Ratny and J. W. Baurantryne: “ Skiagraphic Researches in Teratology.’—Puare I. | Ce ire, 2 4 Trans. Roy. Soc. Edin. Drs Harry Rainy and J. W. BALLANTYNE: Vou. LI. ““Skiagraphie Researches in Teratology.”—Purave III. Fic. 4. Trans. Roy. Soc. Edin. Von. LI. Drs Harry Ratny and J. W. Batnantyne: “ Skiagraphic Researches in Teratology.” —Puave IV. Fie. 7. . & Xe Ue Fic . a q aie , - ’ & ‘ gt | ‘ ; —~ ' f ie ‘ | 7 ae AQ 5 i é ‘J c 4 ‘ 3 “ Weak —_ oF ’ Trans. Roy. Soc, Edin. Drs Harry Rainy and J. W. BaLLaANnrTyNE Vou. LI. : “Skiagraphie Researches in Teratology.”—P uate VY. 10, Fic. ih Vie. Fie. 8. ‘ s \ ' } 7 =i, - me 7 r Trans. Roy. Soc. Edin. Vou, LI. y Fic. 138. Fie. 12, ie, 1k, Trans. Roy. Soc. Edin. Von, LI. Drs Harry Rainy and J. W. Batiantyne: “Skiagraphic Researches in Teratology.”—Piave VII. Fic. 16, Vie, 15, Fic. 14. +a Trans. Roy. Soc. Hdin. Vou. LI. Drs Harry Rarny and J. W. Bauianryne: “ Skiagraphic Researches in Teratology.”—Puare VIII. ieee li. Fie. 18. Trans. Roy. Soc. Edin. Vou. Drs Harry Rawyy and J. W. Bannanryne: ‘“Skiagraphic Researches in Teratology.”—Prate IX. Fie. 21. S wr Fie. LI. Trans, Roy. Soc. Edin. Vou. LI. Drs Harry Rarny and J. W. Batuantyne: “Skiagraphic Researches in Teratology.”—Puare X. 0 = nN nN é g = i= Fic. 22, Trans. Roy. Soc. Edin. Drs Harry Ratny and J. W. BALLANTYNE : 6 Wroje, Iba. Skiagraphic Researches in Teratology.’—Puare XI. S na S d. < <5) 19 a ie} ee x. Trans. Roy. Soc. Edin. Vou. LI. Drs Harry Rarny and J. W. Bantantyye: “Skiagraphic Researches in Teratology.” —PLatE XII. Fic. 28. 27 Trans. Roy. Soc. Edin. Vor. LI. Drs Harry Rainy and J. W. Batuantyne: “Skiagraphic Researches in Teratology,”—Puate XIII. Fic. 30. Fie. 29. Trans. Roy. Soc. Edin. Vou, LI. Drs Harry Rary and J. W. Batnanryne: ‘“ Skiagraphic Researches in Teratology.”—Puate XIV. Fie. 31. Fic. 32. Trans. Roy. Soc, Edin. Vou. LI. Drs Harry Ratny and J. W. Bauuanryye: “ Skiagraphic Researches in Teratology.” —PuaTE XV. Fic. 33. (461) XI.—On a Small Collection of Terrestrial Isopoda from Spain, with Descriptions of Four New Species. By Walter E. Collinge, M.Sc., F.L.S., etc., Research ‘Fellow of the University of St Andrews. Communicated by Professor M‘Intosx. (With Two Plates.) (MS. received August 2, 1915. Read December 6, 1915. Issued separately December 28, 1915.) I am indebted to the kindness of Dr Lronarp Doncaster, F.R.S., for the opportunity to examine the present collection of Terrestrial [sopoda from the Cambridge University Museum of Zoology. The Terrestrial Isopoda of Spain have received considerable attention in the past at the hands of L. Kocu,* Buppe-Lunp,t O. pE Buen,{ and Do.trus,§ and present many features of great interest, one of the most striking of which is perhaps the large size of the various species, particularly in the genera Porcellio, Latreille, and Armadillidiwm, Brandt, and to these | am now able to add the genus Cubaris, Brandt. As Doutrus (op. cit.) has already remarked, the fauna of the Pyrenees, so far as the Isopoda are concerned, is characterised by a group of species quite distinct from the Mediterranean coast fauna, and one which well deserves further investigation. The present collection contains seven species, of which four are new, two of them coming from the Pyrenees. The list is as follows :— 1. Porcellio batesont, n. sp. 8. Spain. 2. , explanatus,n. sp. lia Massane, Pyrenees. DB: a rathker, Brandt. H. Pyrenees. 4 5 ms sp. La Massane, Pyrenees. ; 5 Sp: Madeira. 6. Armadillidium mitidulus, n. sp. Madeira. 7. Cubaris invenustus, n. sp. La Massane, Pyrenees, and S. Spain. 1. Porcellio batesoni, n. sp. (PI. I, figs. 1-6.) Body large, oblong-oval, covered with irregular coarse tubercles, dorsal face convex. Cephalon (fig. 1) with large lateral lobes, terminally truncate, median lobe small and slightly indented in the median line, epistoma convex. Eyes moderate in size, situated dorso-laterally. Antennee (fig. 2) covered with minute setz, joints * Die Thieren Andalusiens, 1856, pp. 418-423. + Crust. Isop. Terr., 1885, pp. 1-319. { Ann. de la Soc. Esp. Hist. Nat., 1887. § Ibid., 1892, t. xxi, pp. 161-190, 13 text-figs. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 11). 66 462 MR WALTER E, COLLINGE ON 1-3 short, 4th with groove on the outer side, the 5th being the longest; flagellum two-jointed, with terminal style. Oral appendages typical. The segments of the mesosome are convex dorsally and roughly tuberculate, with moderately developed pleural plates, the first of which partly surrounds the cephalon; posterior angles | acutely produced backwards. Thoracic appendages long, otherwise typical. Uropoda (figs. 8-5) extending considerably beyond the telson, basal plate grooved on the outer side ; exopodite long, lanceolate, and slightly curved upwards, setaceous ; endopodite short, slightly longer than the basal plate and telson, terminating in four stout spines and a long style, setaceous. Telson (fig. 3) with broad anterior portion, terminating posteriorly in a long, blunt point, slightly grooved dorsally. Length 17 mm. i Colour (in alcohol) blackish with a tinge of olive-green. Habitat.—S. Spain. Easter 1894 (W. Batzson). Type.—In the Cambridge University Museum of Zoology. This interesting species finds its nearest ally in P. nicklesi, Dollfus, from which, however, it is separated by the form of the lateral lobes of the cephalon, the antenne, the uropoda, and the telson. In the form of the cephalon it bears, at first sight, a superficial resemblance to P. magnificus, Dollfus, but is very distinct from that species when examined in detail. I have much pleasure in associating with this fine species the name of Professor W. Bareson, F.R.S., to whom the University is indebted for the specimens. 2. Porcellio explanatus, n. sp. (Pl. IL, figs. 7-11.) Body large, broadly oval, sparsely tuberculate, dorsal face somewhat flattened. Cephalon (fig. 7) with well-developed lateral lobes, deflected downwards; median lobe a narrow ridge between the two lateral lobes; epistoma convex. LHyes moderate in size, situated dorso-laterally. Antennule short and stout. Antennee (fig. 8) covered with minute sete, joints 1-3 short, 3rd joint with tooth-like process on the outer side, 4th joint with groove on the outer side, the 5th being the longest ; flagellum two-jointed, with terminal style. The segments of the mesosome are somewhat flattened, sparsely tuberculate, with moderately developed pleural plates, the first of which partly surrounds the cephalon, the posterior angles slightly produced backwards. Uropoda (figs. 9-11) extending beyond the telson, setaceous, basal plate almost cubical and deeply grooved on the outer side; exopodite broad, lanceolate, biconvex, with slightly raised median ridge; endopodite short, slightly longer than the basal plate and telson, and produced into a ridge on the ventral side, terminally there is a short style. Telson (fig. 9) anteriorly more triangular than in P. batesons, terminating posteriorly in a long acute point, with slight dorsal eroove. Length 20 x 10 mm. A SMALL COLLECTION OF TERRESTRIAL ISOPODA FROM SPAIN. 463 Colour (in alcohol) : body blackish grey, head pale green with three dark, irregular, trianeular patches. The middle portions of both mesosomatic and metasomatic seoments are occasionally marked with light-green patches. Habitat.—La Massane, E. Pyrenees. June 22nd, 1891 (D. SHarp). Type.—In the Cambridge University Museum of Zoology. In certain features this species is allied to P. expansus, Dollfus, but differs from that species in the smaller size, the shape of the cephalon and its lateral and median lobes, the antennze, and the uropoda. 3. Porcellio rathkei, Brandt. Habitat.—k. Pyrenees. June 1891 (D.S.). Two examples. 4. Porcellio sp. This is probably a new species; unfortunately, there is only a single imperfect specimen. Habitat.—La Massane, E. Pyrenees. June 22nd, 1891 (D.S.). 5. Porcello sp. Habitat.—Madeira, 1892 (J. W. CLarx). Two imperfect specimens. 6. Armadillidium mtidulus, n. sp. (Pl. I, figs. 12-15; Pl. II, figs. 16-19.) Body oblong, strongly convex. Cephalon (figs. 12 and 13) large, broadly quad- rangular, marginate, lateral lobes rounded, deeply concave below, epistoma vertical, with triangular shield from the frontal margin. [yes distinct, dorso-lateral. Anten- nulee (fi. 14) small, three-jointed, terminal joint with pointed end and seven strong lateral spines. Antenne (fig. 15) short, joints 2-4 flattened on their outer sides, 5th joint slightly grooved ; flagellum two-jointed. First maxille (fig. 16): the outer lobe terminates in four stout curved spines and five more slender ones, with numer- ous sete distally on the outer side; inner lobe terminally rounded, thin, with two setose spines. Second maxillz terminate in an inner dense tuft of setze and a blade- like outer lobe. The segments of the mesosome are strongly convex and shiny, pleural plates 1-5 distinct, the 1st only overlaps the segment behind, posterior margin not incised, those of 6 and 7 lamellar-like and truncate terminally. Outer palp of the maxillipedes terminates in spinous process with a multispinous termination. Metasome comparatively small. Uropoda (figs. 18 and 19) short, not extending beyond the telson, basal plate stout, somewhat triangular, with broadly expanded anterior surface ; exopodite spatulate, endopodite rudder-shaped, flattened ventrally and slightly raised on the dorsal surface. Telson (fig. 17) rather long, triangular, terminating in a truncate base. Length 16°5 mm. 464 MR WALTER E. COLLINGE ON Colour (in alcohol) dark shiny olive-green, with the posterior margins of all the seoments lighter. Habitat.—Madeira, 1892 (J. W. Ciarxk). Type.—In the Cambridge University Museum of Zoology. Some dozen or more species of this genus have been described from Spain, but I have been unable to identify any of them with the present species. In the elongated telson and the form of the uropoda, cephalon, antennule, and antenne it differs from any of them. Its nearest ally is probably A. vulgare (Latreille) ; it is, however, more elongated than that species, and differs from it in the characters above mentioned, as also in the structure of the mouth parts. 7. Cubaris invenustus, n. sp. (Pl. IL, figs. 20-28.) Body oblong-oval, strongly convex, smooth. Cephalon (figs. 20, 21) flattened, marginate anteriorly, lateral lobes small, median lobe absent; epistome almost vertical, with shehtly raised triangular shield. Hyes situated dorso-laterally. Anten- nul (fig. 22) short and stout, three-jointed, with number of stout sete on the lateral portion of the 3rd joint, terminal portion rounded and with short, blunt spine. Antenne (fig. 23) short, sparsely covered with fine sete, 2nd to 5th jomts grooved on the inner side ; flagellum two-jointed, the distal joint being a little over twice as. long as the proximal one. Segments of the mesosome with the pleural plates excavate anteriorly, not produced backwardly, 6th and 7th segments terminally truncate, lateral portions of Ist and 2nd segments notched and grooved on their lower inner margins for reception of succeeding segments (fig. 24). Maxillipedes (fig. 25), the outer palp terminates in a multispinous process on the outer side, with along spine below it and two smaller ones within; the inner palp has two blunt, tooth-like spines and a much smaller one on the internal border. Thoracic appendages typical. Uropoda (figs. 26 and 27) not extending beyond the telson, basal plate thick and robust, with raised dorsal face, posterior margin almost straight ; exopodite very small, situated on the lower inner border of the basal plate, endopodite large, setaceous. Telson (fig. 28), posterior margin straight, and as broad as length of seoment, expanded anteriorly with triangular convexity, and slight concavity in the median line. Length 16°5 mm. Colour (in alcohol) hght green with dark brown transverse stripe on the posterior of each mesosomatic segment. . Habitat.—La Massane, E. Pyrenees, June 20th, 1891 (D. SHarp); also S. Spain, May 3rd, 1894 (W. Bateson). Type.—In the Cambridge University Museum of Zoology. — 5 Fig. Fig. ig. 20. eal, . 22. Bzo: . 24, mm Oo DD aS oO KF Oooo sr pt at ig. 12. nls. a 14, Beto; ihe. alle 5, £8, ile. ig. 25. . 26. ee . 28. A SMALL COLLECTION OF TERRESTRIAL ISOPODA FROM SPAIN. EXPLANATION OF PLATES. PuateE I, Porcellio batesoni, n. sp. Dorsal view of the cephalon and Ist mesosomatic segment. . Left antenna. Last metasomatic segment, telson, and uropoda. . Right uropod seen from above. . Portion of same seen from below. . Terminal portion of endopodite of right uropod, showing terminal style and large sete. Porcellio explanatus, n. sp. . Dorsal view of the cephalon and Ist mesosomatic segment. . Left antenna. . Last metasomatic segment, telson, and uropoda. . Right uropod seen from above. - Portion of same seen from below. Armadilldium nitidulus, n. sp. Dorsal view of the cephalon and 1st mesosomatic segment. Anterior view of the cephalon. Right antennule. Left antenna. Puate II, Terminal portions of the 1st maxilla, inner and outer lobes. Last metasomatic segment, telson, and uropoda. Right uropod seen from above. The same seen from below. Cubaris invenustus, n. sp. Dorsal view of the cephalon and 1st mesosomatic segment. Anterior view of the cepbalon. Left antennule. Left antenna. under side. Terminal portion of the maxillipede. Right uropod seen from above. The same seen from below. Last metasomatic segment, telson, and uropoda. TRANS ROY. SOC. EDIN., VOL. LI, PART II (NO 11). 67 465 Lateral portion of the lst and 2nd mesosomatic segments, showing notches and grooves on the VoL. Be oc. EDIN. W. E. CoLLinceE: NEw TERRESTRIAL ISOPODA FROM SPAIN.—PL. I. A RITCHIE & SON, EDINE Soc. EDIN. W. E. COLLINGE: NEw TERRESTRIAL ISOPODA FROM SPAIN.—PL. II: Vor A RITCHIE & SOW, EDIW * Ek Gu4e7 —) XIJ.—The Lateral Sense Organs of Elasmobranchs: The Ampullary Canals of the Genus Raia. By Augusta Lamont, B.Sc., Baxter Scholar in Natural Science, University of Edinburgh. Communicated by Professor J. C. Ewart. (With Hight Plates and Nine Text-figures.) (MS. received July 5, 1915. Read July 5, 1915. Issued separately January 15, 1916.) CONTENTS. PAGE PAGE 1. Introductory. ; : ; . 467 | 4. Distinguishing Features of the System in the 2. Historical . : 3 ; 2 : ; . 468 Several Species . ; ; . . 480 3. General Characters and Distribution of the 5, Innervation : ; : : : ; . 484 Ampullary Canals. 0 : : . 473 | 6, Comparison with other Members of the Batoidei 489 7. Bibliography . : : ; : : . 490 1. INTRODUCTORY. For two centuries and a half the ampullary canal system of Elasmobranch fishes has occupied the attention of a series of investigators, who have treated the subject from a variety of standpoints, and have amassed a considerable literature represent- ing, in spite of repetition and some conflicting statements, a gradual advance to a fuller knowledge regarding the occurrence and the structure of the organs in question. | Harly researches suffered from the want of recognition of any distinction between the sensory canal system and the ampullary canal system, both being included under the term “mucous canals,” a name given on account of the secretory function which was ascribed to them. Later, electricity-producing and sensory functions were in turn attributed to them, and the disputed point of their physiology became one of the chief points of interest among the investigators of the first half of the nineteenth century. From the middle of the nineteenth century onwards, the histology and innervation of the ampulle and their canals attracted attention, and as a result these gradually came to be recognised as organs of sense. In spite, however, of continuous work and varied treatment, no one among this series of investigators has undertaken to give a complete description of the ampullary canals in any one genus, observations having been made in turn on Torpedo, some species of Rava, Scyllium, Acanthias, Pristiwrus, and others. It has therefore been the purpose of the present investigation to attempt what has hitherto been omitted, and to describe as fully as possible one genus—the genus Raia,—comparing with one another as many species of the genus as have been available. The present paper gives an account of the general anatomy of the ‘ampullary canal system, based on work which has been carried out in the Zoology Department of the University of Edinburgh. It is hoped that later an account of the histology of the ampulle may be contributed. TRANS. ROY. SOC. EDIN, VOL. LI, PART II (NO. 12). 68 468 AUGUSTA LAMONT ON Acknowledgments are due for assistance granted from the Earl of Moray’s Endowment for the Promotion of Original Research, University of Edinburgh. 2. HISTORICAL. ) The first observer to record the occurrence of the so-called ‘“‘ mucous canals” was Nixs Stenson, who in 1664 and 1669 makes mention of having seen their openings in Raia, Canis carcharias, and Galeus levis; but it was the Italian, LoRENzINI, who first gave a definite description of the “ampulle,’—as he named them—and their canals, and who assigned to these structures, in his work on Torpedo in 1678, a secretory function. The term “Lorenzinian ampulle,” therefore, first made use of by Bott nearly two hundred years later, with justice associates these organs with their discoverer. No further work on the subject appeared until Monro, after more than a century, published in 1785 his Structure and Physiology of Fishes. In it he described and figured for the skate a pair of groups of radiating ducts containing a viscid mucus and possessing a rich supply of nerves, and these structures he classed as glandular organs. The next investigator was Grorrroy Sr Hiuarre (1801), whose work led him to the conclusion that the gland-like masses and radiating canals of Rava rubus were the homologue of the electric organs of the Torpedo. The work of Jaconson (1813) presents a considerable advance on that of his pre- decessors. He was the first to record the occurrence of five pairs of “central organs” (groups of ampulle), to mention the presence of similar organs throughout the sharks and.rays, and, repudiating former theories, to class them as organs of touch adapted for the perception of undulatory motion. He compared these organs to the vibrisse of birds and mammals, and stated in support of this comparison that both series of structures are supplied by the same branch of the fifth nerve. G. R. Treviranus, who followed (1819), agreed with Jacopson in classing the ampullary structures as sense organs, but, demurring at attributing to them a tactile function, hinted at the possibility of a sense other than the five possessed by man. His observations were made on Squalus acanthias, Rava batis, and R. rubus, and he was the first to observe the partitioned structure of the ampulle. Another supporter of Jacogpson’s theory was found in BuatNvILLE (1822), who, in his work on Comparative Anatomy, treats of the “tubular organs” (ampullary — canals) of Selachii generally under the heading of “active organs of touch.” He — clearly distinguished between these “tubular organs” and the “lacunar system” (lateral canal system), but appears to have recognised only three pairs of ampullary capsules, and fell into the curious mistake of supposing that these capsules—apart — from the canals—were the possible homologue of the electric organ of Torpedo. Shortly after BuarnvitLE, Ropert Knox (1825) wrote on the theory of a sixth — THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 469 sense in sharks, tope-fishes, and rays, but he appears to have been rather a theorist than an original worker. Without assigning any reason, he contradicts TREVIRANUS’S statement regarding the division of the ampulle into compartments, and bases state- ments of his own on microscop.cal observations undertaken for him by another in default of original investigations. In spite of being aware of former confusion between the two canal systems, he appears himself to have fallen into the error of not clearly distinguishing between them. He considers the function of the “ tubular organs” (ampullary canals) to be intermediate between hearing and touch, and suggests that the shark’s facility for tracking prey is accounted for by the possession of this additional sense. Dette Cutase (1839), in his work on Torpedo, shortly describes the ampullary organs, introducing the subject in connection with a description of the electric organ and pointing out similarities between the two systems. He mentions haying seen the same system with its rich nerve-supply in Zygena, the Hammer- head shark. The investigations of Joun Davy, on the electric properties of Torpedo, published in 1839, were undertaken in fulfilment of a last request of his brother Sir Humphry. 4 His account includes a description of the “ mucous system,” the mucus of which the author considers to be a good conductor of electricity, while at the same time he holds that its secretion is stimulated by the electric function. No mention is made of previous workers other than LorEnzinr and Monro, and no distinction is drawn between the sensory and ampullary canals, both of which are figured in his Plate X as arising from the group of ampulle shown. Savi, who also worked on the electric organ of Torpedo, similarly gives an account of the so-called “mucous organs.” Writing in 1841, he inclines to repudiate the view of the secretory function ; but in 1843 he alludes to the organs as “ mucous follicles,’ and describes their contained humour as continually flowing out on the skin, although without committing himself to any theory regarding the purpose of the secretion. Maver, writing in the same year (1843), is said to have revived the theory that the ampulle of rays are the homologue of the electric organ of Torpedo, while Migscuer, early in 1844, writing on “the organs discovered by Professor Mayer in non-electric Rays,” again reverts to the view that they are mucus-secreting structures. He shortly describes for Rava clavata four pairs of these glands, evidently reckoning the ophthalmic and inner buccal groups of ampulle together as one, since the other three are unmistakably recognisable from the description given. The researches of Rosin (1846) had for their object the investigation of the electric organ of rays (Raia); and the ampullary capsules and tubes are chiefly considered in order to repudiate theories suggesting their homology with the electric organ. Rosin limits the number of pairs of capsules in Rava to four, and claims to have been the first to describe the fourth capsule (mandibular), maintaining that 470 AUGUSTA LAMONT ON Jacopson described only three while mentioning five, and thus ignoring the work of MiEscHEr. Rerzius (1847), who investigated “the supposed electric organs” in Raia batis and Squalus acanthias, also described four pairs, and concluded from their rich nerve-supply that Jacopson’s theory as to their sensory function was correct. StTannivus, in his memoir on the Peripheral Nervous System of Fishes (1849), describes for Plagiostomi the innervation of the ampullary system, which he alludes to throughout as a “‘ mucus-secreting apparatus.” He gives an account of the innerva- tion of the ampulle by the ophthalmic and buccal branches of the trigeminus, and by the hyomandibular and mandibular branches of the facialis. He describes a single nerve-thread, composed of eight or ten fibres, as entering each ampulla, and admits ) his failure to have discovered anything definite regarding the nerve-endings. Spinaw, Raia batis, and R. clavata are mentioned as subjects of the investigation. The work of H. Miiipr (1851) was in many respects an advance on that of his immediate predecessors. He appears to have written largely with the object of refuting the suggestion of the existence of a relation between the follicular apparatus of Torpedo and electricity production, by means of a comparison with the mucous canals of other Plagiostomi. He clearly distinguished between the different systems of canals, and recognised that the ampulle and their tubes were confined to car- tilaginous fishes, inclining to a belief in the sensory function of the system. He described in more detail than previous workers the distribution of the ampullary canals, especially in the genus Raia, of which he investigated three species. He points out the pigmentation and wide distribution of the canals of R. vomer* com- pared with the more restricted distribution and hyaline appearance of those of R. undulata and R. asterias.t - -Leypic, who did much to advance knowledge of the ampullary canal system, contributed between the years 1852 and 1868 several publications dealing with the subject. He investigated, in considerable detail, the structure of the ampulle in various Elasmobranchs, contributing much that was new regarding their histology, innervation, and vascular supply. He maintained, throughout, the sensory function of these organs, and concluded that they are organs of a sixth sense adapted for life in water ; suggesting, however, that secretion might take place in association with a sensory function allied to that of touch. In this connection he raised the question as to whether similarly among the sweat-glands of mammals there may not be some which have a more direct relation to the nervous system than to secretion. Contemporaneous with the work of Lrypie was that of LamBaLLE, Eckuarp, and M*DonneELL, the two first of whom both published the results of their investiga- tions in 1858. JOBERT DE LAMBALLE appears to have been ignorant of any of the work of his predecessors on the subject of the ampullary canal system, with the exception of an * Rava vomer, Mill. =R. oxyrhynchus, Linn. + Raia asterias, Mill.=R. maculata, Mont. THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. AT] account given by Cuvier and VALENCIENNES in their Natural History of Fishes. Dealing with the “ Electric Organs of Klectric Fishes,” he describes a single pair of ampullary capsules in the skate, and, reverting to the standpoint of Sr Hrarrg, concludes that these represent an elementary electric apparatus comparable to that of Torpedo. Ecxuarp, dealing with Torpedo alone, went into the microscopic structure of the ampulle in considerable detail, and concluded that the six spheres contained in each ampulla are peripheral ganglia. While considering it as proved that these organs are not secretory, he was non-committal as to what type of sensory function they might fulfil. M‘DonnELL (1861) rejected St Hitarre’s theory, and supported Leypie’s view of the tactile function of the muciferous tube system of sharks and rays. Bott (1868), in his work on Lorenzinr’s ampulle in Torpedo and an unnamed species of shark, devoted much attention to their histology, and confirmed Lrypic’s view as to their partitioned structure. The researches of Toparo (1870) are referred to by both MERKEL and Frirscu, who followed him, but his “Contribuzione” have unfortunately not been available for reference. Batrour (1878), who was the first to deal with the subject from the embryological point of view, pronounced in his work on The Development of Elasmobranch Fishes that the ampullary canals are products of the mucous layer of the epiblast. The work of Sappry (1879) contains more detail as to the distribution of the eanals than that of any of his predecessors. He gives detailed accounts of the grouping of them in a skate (/a raze) of which he does not give the specific name, but the description as far as it goes is applicable to several of the genus, although it is incomplete in giving an account of only three pairs of capsules. He repudiated LeEypie’s view as to the sensory function of these organs, and maintained the glandular structure of both the ampulle and their ducts, the function of which, he considered, was to pour mucus on the surface of the skin, although he admitted the use of the mucus to be still in doubt. : MerRKEL’s researches (1880) dealt with the nerve-endings and sensory cells of the system in Elasmobranchs generally, of which he investigated a number of different species. After a detailed study of the histology of the ampulle, he arrived at the conclusion, confirming that of Jacosson and his followers, that the sensory function of these organs is established beyond all question, and that the probability of this function being of a tactile nature has much to support it. Ten years later we find again in the work of Fritscu (1890) a reversion to the theory of secretion. In the section of his memoir on Electric: Fishes dealing with Torpedoes he expressed the opinion that the ampullary canal system is a part of the lateral canal system, from which it has become separated in structure and in function, and that thus, while the lateral canal system is sensory in function, the ampullary 472 AUGUSTA LAMONT ON system has become predominantly secretory. He, however, considers it possible that the ampulle may at the same time be organs of sense, although he holds that the variability of their structure in the different types of Selachii is to be taken as evidence against this view, seeing that sensory epithelia are of noticeably similar structure even in classes of animals whose relationships are only remote. As a result of the work of Ewart (1889-1891) on the cranial nerves and lateral sense-organs of Elasmobranchs, the innervation of both the sensory canals and the ampulle by nerves of the facial complex, and not by branches of the trigeminus, became established. His figures of the sensory canals of Lemargus and Raia bates, and their supplying nerves, also show the relative position and the associated innerva- tion of the ampulla, and his naming the groups of ampulle after their supplying nerve-branches instituted a system of nomenclature by means of which the ampullary groups can be homologised in different types of Selachians. — Fucus (1895) was the first since Davy in 1839 to institute experiments with the object of elucidating the vexed problem of the physiology of the ampulle. His experiments, which were carried out on live specimens of Torpedo and Rava, took the form of severing the supplying nerve-branch on one or both sides, applying pressure, heat, and acids, and stimulating the organs after they had been laid bare. The results were negative in all cases except that of pressure applied to the nerve supplying Savr’s yesicles, and Fucus therefore concluded that as a result of his experiments the ampullary system must be denied the sensory function. Later, the experiments were judged by Branpgs to have proved nothing at all. In 1895 CoLLINGE, in a paper on the sensory and ampullary canals of Chimera, maintained that the innervation of the ampulle is from the trigeminus only, and was severely criticised by CoLE for perpetuating this old error. In the following year CoLE published the results of his own work on the cranial nerves of Chimera, in which the innervation of the ampulle from branches of the facialis is clearly described. During the following ten or twelve years several works appeared dealing with the histology and development of the ampulle. PEaxsopy in 1897 published a well- illustrated abstract of results obtained from work on Galews canis, giving the most complete account that had till then appeared of the histology of the ampulle, and describing more fully and clearly than had hitherto been done the manner of the nerve-endings. BRANDES, in 1898, also published a preliminary note on his work on the ampulle of Elasmobranchs generally, confirming all MeRKEL’s histological results, including the presence of a ciliated epithelium (denied by Puazopy), but interpreting the ‘pear-shaped ” cells as being glandular and the “ cover” cells as sensory. FoRSSELL, in a publication which appeared in 1898, gives a very complete account of the anatomy and histology of the ampulle of Acanthias vulgaris. His conclusion is that a sensory function is to be attributed to the epithelium of the ampulle, but THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 473 that the gelatinous contents are produced by the activity of the lining cells of the canals, and have the function of protecting the sensory epithelia, and of affording a means of transmitting stimuli. Minckert (1901) and Coeer (1902) both worked at the development of the ampulle—the former in Spinax niger, the latter in Torpedo, Raia, Mustelus, and Pristiurus. Coeer points out that, since the lateral sensory apparatus appears while the ectoderm still consists of a single layer, BaLFour’s statement as to the mucous layer only taking part in its formation is incorrect. ALLIS’s memoir (1902) on the Sensory Canals, Eye-Muscles, and Cranial Nerves of Mustelus levis was the result of work undertaken primarily in order to investigate the innervation of the ampulle and the lateral canal, and during the course of it the author became confirmed in his previous impression that the ampulle of Selachians are homologous with the terminal buds of Ganords and Teleosts and the nerve-sacs of Acipenser rather than with the pit-organs of those fishes. This view is criticised by Jounston (1902) as being ‘“‘ wholly unsound and likely to lead to further difficulties.” The work of Mrs Hawkes (1906) on The Cranial and Spinal Nerves of Chlamydoselachus anguineus touches only indirectly on the ampulle, the chief points of interest in connection with which are their diffuse arrangement and the absence of a mandibular group. Further researches on the development of the Lorenzinian ampulle were carried out by BRoHMER (1908) in Spinaw embryos. He claims that his work confirms and supplements that of Minckerr, but he makes no mention of the more recent and very complete work of Coccr, with whose results as regards the stages in develop- ment observed, and as to the non-simultaneous appearance of the ampulle, BRoHMER’s was nevertheless in agreement. The very confusing nomenclature adopted by BrouMeER for the ampullary groups might have been avoided had he observed Cocers remark that “the most rational way to classify the ampulle is according to their innervation, especially since this accords with the facts of development.” 3. GENERAL CHARACTERS AND DISTRIBUTION OF THE AMPULLARY CANALS. The species of Raia investigated in the course of these researches were seven in number, viz. R. batis, R. fullonica, R. radiata, R. circularis, R. clavata, R. micro- cellata, and R. maculata, the nomenclature adopted being that used by Day, in his Fishes of Great Britain and Ireland. Of these the first four were obtained in Edin- burgh from the Firth of Forth, while the specimens of the last three species were obtained in Brixham fish-market, on the south coast of England. The results of the observations made have been summarised so that an account of the ampullary system could be given applicable to the genus as a whole; sub- sequently each species is reviewed in turn and the distinguishing features pointed 474 AUGUSTA LAMONT ON out, so that, with the aid of the drawings of both dorsal and ventral dissections, a comparison of the several species can be made (Plates I to VIII). Thus, throughout the genus, this system of organs consists of five paired groups” TEx?-FIG, 1.—A group of ampulle from the hyoid capsules, R. radiata. x5. A, Ampulla. C. Canal. N. Nerve. TExt-FIG, 2,—Ampulle of R. radiata. x21. N. Nerve. A. Ampulla. C. Canal. of ampulle, situated in the cephalic region, and innervated by branches of the seventh cranial nerves. In many cases each group of ampulle is enclosed in a fibrous covering, the so-called “capsule.” The ampullee themselves are irregularly sacculated vesicles, — each receiving proximally a nerve-twig, and passing distally into a thin-walled hyaline tube, or canal (text-figs. | and 2), which contains a colourless and trans-— THE LATERAL SENSE ORGANS OF ELASMOBRANCHS, 475 parent gelatinous substance, and opens on the external surface. The groups of ampulle, and the proximal portions of the canals, are more or less deeply embedded in connective tissue, and become visible only on first removing this. Distally, how- ever, the canals lie to a large extent immediately below the skin, on removing which they are revealed spreading out in all directions on both dorsal and ventral surfaces. Where they pass over the muscles of the pectoral fin their course presents a series of smal] undulations caused by the ridges and furrows of the muscle-plates. Some of the canals are short, and lead in a more or less vertical direction to the external surface ; others are long and horizontally spreading, in many cases diverging widely as they follow their course from the ampulle to the exterior. In all cases the canals terminate at the outer surface in apertures which appear as little round dots in the skin, irregularly scattered or arranged in definite curved rows. The jelly contained in the tubes does not appear to be naturally extruded through these openings, though it can by pressure be forced out. ReErztus’s supposition (1847) that the canal-openings are contractile, and capable of being opened and closed at the will of the animal, does not appear to have been either confirmed or refuted by any more recent physiologist ; but the frequently expanded appearance of the canals towards the openings, followed by a contraction at the opening itself, suggests at least a passing of the contained jelly towards the exterior, and a checking of further movement as the opening is reached. The groups of ampullee—or “central organs” as they have frequently been termed, since they are the centres from which these canal systems arise—are constant in position, and the five pairs in which they are arranged have been named by Ewart after the branches of the facial nerve by which respectively they are supplied. Three groups lie at no great distance from one another, embedded in the gelatinous tissue of the snout. These are in some species enclosed in capsules, but not in all. 1. The superficial ophthalmic is the most anterior group, and lies close to the rostrum about mid-way between the eye and the tip of the snout (text-fig. 6). 2. The inner buccal is situated posterior to, and to the outer side of, the preceding group, and lies between it and the nasal capsule. Cases occur where this and the ophthalmic lie in such close contact that, where there is no enclosing capsule, it is only by means of the innervation that they can be distinguished from one another. Text-fig. 3 represents a dissection of these groups in Rava clavata where this was the ease. The same figure also shows how the inner buccal ampulle may be distributed in two sub-groups. In Rava batis, however, a single capsule encloses all the ampulle innervated by the inner buccal nerve (text-fig. 7). 3. The outer buccal group lies immediately exterior to the nasal capsule—between it and the propterygium, and just in front of the antorbital cartilage (text-fig. 6). 4. The largest and most conspicuous group is the hyoid, situated at the angle between the branchial region and the inner border of the propterygium. This group is in all cases enclosed in a capsular covering (text-figs. 6 to 9). 5. The fifth or mandibular group is found on the lower jaw, slightly posterior to TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 12). 69 476 AUGUSTA LAMONT ON the corner of the mouth and nearer to the median line (text-fig. 7). This is the smallest of the five groups, and it is invariably enclosed in a capsule. It alone can be conveniently dissected out entire and completely separated from the surrounding tissues, when it appears as a neat oval body about 3 or 4 mm. in length (text-fig. 4). Text-fig. 5 represents an abnormal mandibular capsule, containing only two ampulle, obtained in a specimen of Rava circularis. ! i avin é Text-Fic. 3.—Ventral dissection of ophthalmic and inner buccal ampulle, A. clavata. x. 8.0. Superficial ophthalmic ampulle. I.B. Inner buccal ampulle. N.A. Nasal aperture. M. Mouth. / / / 1 Of the five groups of ampulle, three—the hyoid, the ophthalmic, and the outer buccal—have canals opening on both dorsal and ventral surfaces; while two—the inner buccal and the mandibular—have canals opening on the ventral surface only. The canals in all cases are distributed in definite groups, constant enough in relative position, course, and manner of divergence from the ampullz to be recognised in each species and to be known by some distinguishing name. The number of canals in a group is not constant for a species, nor indeed are the numbers necessarily equal in the corresponding groups on the right and left sides of the same individual. Nevertheless, there is an average number, characteristic of each group of canals in a species, from which there are no extreme variations, and which justifies statements regarding the numerical superiority of certain groups relative to others, and as to THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 477 the larger or smaller numbers of canals generally, characteristic of species relative to one another. (1) Superficial Ophthalmic Canals. The dorsal canals of the ophthalmic ampullee are distributed in two groups: (1) an anterior, and (2) a posterior group, both of which lie close to and parallel with the rostrum (Plate I). The canals of the anterior group are directed forwards towards N.S. Trxt-Fic, 4.—Mandibular capsule, &. radiata, x14. A, Ampulla. C.W. Capsular wall. C, Canal. N. Nerve. N.S. Nerve to sensory canal. TEXT-FIG, 5.—Abnormal mandibular capsule, R. cireularis, x14. A. Ampulla. C.W. Capsular wall. C. Canal. N. Nerve. the tip of the snout; those of the posterior group are directed backwards to a point in front of and slightly to the inner side of the anterior corner of the eye. These canals were overlooked by Sappry, who described a species of “la raie” to which he gives no specific name. He expressly states that canals arising from his “cephalic glands” (ampullee of the cephalic region) open on the ventral surface only, and in his figure of a dissection of the dorsal surface canals are represented arising from the hyoid ampulle only. Similarly, GARMAN, in his paper on the lateral canal system, omits them from a figure of the ampullary canals of Rava levis which he gives for purposes of comparison with the sensory canals. 478 AUGUSTA LAMONT ON The canals opening on the ventral surface are directed towards the tip of the snout, and are arranged in a single group, sometimes with a few detached scattered canals appearing scarcely to form a part of the group nor yet a group by themselves (Plate II). (2) Inner Buccal Canals. These canals open on the ventral surface only, and are distributed in two groups, one of which is directed backwards on the naso-buccal flap, while the other takes an out- ward course in front of the nasal aperture, round which it then curves in a backward direction (Plate IV, fig. 6). (3) Outer Buccal Canals. The canals from the outer buccal ampulle opening on the dorsal surface are — comprised in a single small group situated a short distance anterior and exterior to the front angle of the eye. They are few in number—in some cases not more than - seven or eight—and are directed from within outwards, so that they are readily overlooked, and even when searched for are only distinguished with some difficulty. They come to the surface immediately anterior to the mandibular muscle, and just on the outer side of the infra-orbital (sensory) canal, from the branches of which they can be distinguished by their soft consistency and hyaline appearance, which contrast with the tougher texture and opaque white appearance of the sensory canals, — characters which also combine to give the respective openings of the two systems a different appearance. These canals appear to have been overlooked by previous investigators ; like the dorsal canals of the ophthalmic ampulle, they go unrecorded by Sappey, who gave more attention than others to the distribution of the ampul- lary canals, and they are not represented by Garman in his figure of Raia levis mentioned above (Plate [). Three groups of canals open on the ventral surface :—(1) An outer group has typically—but subject to modification in some species—an arrangement in which — the canals diverge forwards, outwards, and backwards in radiating fashion, over- lapping the most anterior canals of the hyoid ampulle. (2) An anterior group is composed of canals directed forwards. (3) A posterior group lies external to the naso-buccal groove, and its canals are directed backwards (Plate II). Collectively the canals of the ophthalmic, inner buccal, and outer buccal ampullee opening on the ventral surface occupy the region between the tip of the snout, the mouth, and the propterygium. The openings appear to be irregularly scattered, and even after removal of the skin it is not always easy to say from which group of ampulle the canals arise, and they must be traced back and the ampulle exposed before this can be decided. As the ampulle are deeply situated, and as the canals are short, the course of these is directed largely from within outwards rather than along the surface, and they appear as little tufts radiating from within towards the exterior. Hxcept in cases where pigment is present in the canal-walls, they are so THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 479 like the gelatinous tissue of the snout in transparency and want of colour that they are not readily seen, and require more careful dissection and examination than do the conspicuous canals of the hyoid ampulle. (4) Hyoid Canals. The canals of the hyoid ampulle opening on the dorsal surface are arranged in four groups which, adopting the nomenclature of Sappry, may be distinguished as anterior, outer, inner, and posterior (Plate I). (1) In the anterior group the canals lie parallel to one another, and are directed forwards along the groove between the anterior inner border of the pectoral fin and the cephalic region. They open close to one another. where this groove approaches the outer border of the disc. (2) The canals of the outer group diverge outwards and fanwise over the anterior and middle portion of the pectoral fin, opening at short and approximately equal distances from its outer border. The most posterior canals of the group are from three to four times as long as those whose course is directed immediately outwards to the nearest portion of the edge of the fin. (3) The inner group comprises a compact fascicle of parallel or slightly divergent canals, directed inwards and backwards along the anterior border of the branchial region, and ending in openings which form a semi- circular curve near the median line between the two gill regions. The canals of this group are of approximately equal length. (4) The posterior group contains typically fewer canals than any of the preceding groups. They are directed backwards in a sinuous course along the junction line of the pectoral fin and the trunk. In some species these canals present a succession of different lengths, the openings occurring at intervals along the course of the group; in other species they are all equally long. The longest of them reach to near the posterior border of the fin, and these either equal or exceed in length the longest of the canals of the outer group. Usually the canals of this group are of larger calibre than others, and increase markedly in width towards their openings ; for instance, in a specimen of Raia clavata measuring about 21 inches across, the width of these canals was 1°5 mm. near the ampulle, and 3 mm. near the openings. The hyoid canals opening on the ventral surface, although showing a certain correspondence in arrangement, are somewhat less definitely grouped than those of the dorsal side (Plate II). They are best described as falling into two groups: an inner and an outer. (1) The inner group corresponds to the inner group on the dorsal surface, and extends inwards and backwards along the upper and inner borders of the gill region ; but instead of lying close together and having approximately equal lengths, the canals in this case diverge to some extent, and, taking a more curving course, end at varying distances between the ampulle and the extremity of the group. (2) The outer group of canals (including Sappry’s outer and posterior groups) may be taken as corresponding to the anterior, outer, and posterior groups of the dorsal surface, and like these collectively they spread out fanwise, extending forwards, out- 480 AUGUSTA LAMONT ON wards, and backwards, and covering the region between the propterygium, the outer anterior border of the pectoral fin, and the side of the abdominal cavity. The canals are, however, not subdivided into separate groups, but form a continuous series, diverging over the region described, and ending at varying distances from the ampulle in such a way that the area they cover forms a definite and characteristic shape which varies slightly according to species. (5) Mandibular Canals. The canals of the mandibular ampulle are uniformly arranged in one group opening on the ventral surface, and they are directed inwards towards the median line, just posterior and parallel to the lower jaw. Not only is the capsule, which in all cases encloses these ampulle, smaller than the others, but the number of its canals is less, and these and their ampulle are smaller in size than those of the other ampullary groups (Plate III, fig. 4). 4, DISTINGUISHING FEATURES OF THE SYSTEM IN THE SEVERAL SPECIES. Of the various British species of Raia collectively the most noticeable point is that they at once fall into two groups :—(1) Those with many, widely distributed, and pigmented canals. (2) Those with fewer canals of more restricted distribution, and without pigment. The first group contains only three species—Raia batis, R. macrorhynchus, and R. oxyrhynchus. Of these Rava batis only was included among the species examined during the present investigation, but it is obvious from Mi.uer’s description of R. oxyrhynchus (which he refers to under the name of FR. vomer) that it belongs to this group. Covucn’s reference to the pigmented tubes appearing as dusky lines and spots on the ventral surface of R. macrorhynchus is also conclusive with regard to — that species. “A The second group contains the remaining seven species, viz. Raia alba, R. fullonica, R. radiata, R. circularis, R. clavata, R. microcellata, and R. maculata, of which only the. first was missing from among the subjects of this investigation. Among these there is a remarkable similarity in the general distribution of the canals, the most distinguishing feature between the species occurring in the number, length, and distribution of the posterior dorsal group of canals of the hyoid ampullee. Following out GarMAN’s idea that the canal systems may be used as a basis for classification, one is at once led to the conclusion that Day’s division of the venus into “long-snouted” and * entails the classing of Raia alba and R. fullonica along with the types having pigmented canals. Coucn considered the grey colour on the ventral surface (due to short-snouted” species is artificial, since that grouping the presence of pigmented canals) of systematic importance, and went so far as to — THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 481 advocate the separation of the pigmented species into a distinct genus under the name of “ Batis.” 7 This grouping, arrived at by Covucn from external characters, and confirmed by dissection of the ampullary canal system, would, as he himself poimts out, be in accordance with the distinction made by fishermen, who, in Devonshire at all events, eall Raia batis a ‘skate,’ while they class R. fullonica, R. clavata, R. maculata, R. microcellata, and R. circularis all together as “ rays.” (1) Pigmented Types. Raia batis, Linn., or the common skate, taken as the only available representative of the types with pigmented canals, is easily distinguished from any species of the non-pigmented group. Contrasted with these it is characterised by the presence of a dark pigment in the canal walls at and towards their openings. The pigment causes the openings of the canals to appear on the external surface as little black dots, and also makes the course of the canal visible as a grey streak for a short distance through the skin. The large number and wide distribution of the canals are somewhat less pro- nounced on the dorsal surface (Plate I). The canals of the posterior hyoid group usually number ten or eleven, and are of varying lengths. It is convenient to take special notice of this group for purposes of comparison, its isolated and superficial position, as well as the small number, greater length, and larger calibre of its component canals, rendering accurate counts and correct delineation easier than in other cases. Another characteristic feature on the dorsal surface occurs in the distribution of the canals of the ophthalmic group of ampulle; these, as already described, lie parallel to the rostrum, one group running forwards to the tip of the snout, and another backwards to the nasal capsule; but in this case only canals of varying lengths terminate between the two extremities, so that their openings are roughly grouped in linear series along the edge of the rostrum. On the ventral surface (Plate II) the canals are so numerous and so widely dispersed as to defy enumeration, covering almost the whole of the cephalic, thoracic, and abdominal regions, as well as the whole surface of the pectoral fin, with the exception of a zone about 1°5 inch wide (in specimen 17 to 18 inches across) along the posterior edge. The canals of all the three anterior groups of ampulle are widely divergent, as are also those of the inner hyoid group, while the mandibular canals, although directed inwards, are not, as in other cases, closely parallel, but diverge somewhat and terminate at varying distances posterior to the lower jaw, forming an elongate fringe-like arrangement extending from the corner of the mouth inwards. It is, however, in regard to the outer group of the ventral hyoid canals that the most distinctive grouping is observable. The canals here are clustered in little sub- groups of about six or seven, those forming a group being of different lengths, but 482 AUGUSTA LAMONT ON lying in contact with one another as far as their course persists, except towards the extremity of each, where it diverges from the rest. The canals of a sub-group are thus separated by spaces from the next sub-group, but not from one another. This arrangement, and the fact of the exceptionally fine calibre of the canals, as well as their large numbers, suggests the possibility that each such sub-group may in reality be the equivalent of the larger and separately placed canal in the corresponding group of non-pigmented species. ; (2) Non-pigmented Types. The arrangement of the canals in the non-pigmented species is similar to that of the pigmented types, but their distribution is more restricted and their numbers are considerably less. Owing to the absence of pigment in the canals of these species, and also on account of the colours of the skin of the dorsal surface, the dorsal canals are practically indistinguishable before dissection, and their openings are detected only with difficulty. On the ventral surface, however, which in all these species is white, their course can often be traced through the skin, and the openings stand out distinctly as round pores dotted over the surface and predominating in the head region. ; On the dorsal surface the arrangement of the canals is more characterised by restriction in numbers than in distribution. On the ventral surface the canals are both fewer and of less extent, spreading only over the head region and over the basal and anterior portions of the pectoral fin (Plate III, figs. 3 and 4). Particular instances of a difference in the manner in which the canals are disposed may be observed in the case of those arising from the ophthalmic and the mandibular ampulle. The dorsal ophthalmic canals, instead of terminating as described for Raa bats at varying distances along the side of the rostrum, are all of approxi- mately equal length; an anterior fascicle directed forwards has its openings grouped close to the tip of the snout, while the canals of the posterior fascicle, directed back- wards, all terminate close together, anterior to the front angle of the eye. A similar difference is observable in the case of the mandibular canals which, being also more equal in length and running closely parallel towards the median line, avoid the fringe-like arrangement described for 2. batis. The general arrangement of the canals is better understood by a glance at the accompanying figures than by means of any description, but it must be borne in mind that the drawings do not claim to be exact representations as regards numbers of canals where these are very numerous, deeply embedded, or overlapping one another. The general distribution, and an approximation to the numbers visible is all that has been aimed at, for instance, in the case of the ventral surface of the snout region, where the clear colourless appearance both of the canals and of the gelatinous tissue in which they are embedded renders the former very inconspicuous. In the case of the dorsal posterior hyoid canals, on the other hand, the drawings represent the actual THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 483 numbers present in the specimens, since the size, position, and arrangement of these canals renders exact delineation possible. , With regard to the individual species of this group it is difficult to single out any definite features by means of which they may be distinguished from one another ; but attention may be drawn to one or two points which appear to constitute the most noticeable differences. These occur in the number, length, and divergence of the dorsal posterior group of hyoid canals, and in the length of the most posterior canal or two of the dorsal outer hyoid group. Differences as regards these points will therefore be noted as each species is passed in review. Another feature of the species of this group is the absence of a capsular wall surrounding the ampulle of the ophthalmic and inner and outer buccal groups (text-figs. 8 and 9). H. Miuuer remarked in this connection: “The central organs (.e. groups of ampulle) are frequently, but not always, surrounded by a strong fibrous capsule, which is pierced alike by the tubes and nerves.” In specimens dissected to make observations on this point, the ampullz were found to be merely embedded in the gelatinous tissue, but a more fibrous layer of connective tissue immediately underlay the skin, covering all the gelatinous area of the snout, and this had to be cut through before the ampulla were exposed. This fibrous layer did not, however, surround the groups of ampulle. R. fullonica, Linn., the “Sandback Ray” of the Brixham trawlers. In the single specimen examined the canals of the posterior group of the hyoid ampulle. on the dorsal surface were five in number. They were of varying lengths, and the longest diverged from one another towards their extremities (Plate III, fig. 3). The dorsal outer canals of the hyoid ampullz were considerably more numerous than in other “rays,” numbering about fifty-five, as against from thirty to forty, but there was nothing exceptional in the length of the last two canals of the group. On the ventral surface (Plate III, fig. 4) the outer group of canals from the outer buccal ampulle was less extended outwards over the pectoral fin than in other species, most of the canals being directed backwards along the pro- pterygium and lying close together, for the first part of their course at least in a compact fascicle. R. radiata, Donovan.—Specimens of this species (Plate IV, figs. 5 and 6) were examined in which the canals of the same posterior dorsal group numbered respec- tively four and five, and in another specimen there were six on the left side and five on the right. The specimen figured shows these canals of approximately equal length, all except two diverging from one another towards their openings. Rf. circularis, Couch, known in Edinburgh as the “ Medallion Skate,” and in Devonshire as the “Cuckoo Ray,” has four or five canals in the dorsal posterior hyoid group, of which one or two are shorter than the rest, while the longer ones are divergent (Plate V, fig. 7). In the specimen figured it will be noticed that the most posterior canal of the dorsal outer hyoid group is slightly shorter than those next to TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 12). 70 484 AUGUSTA LAMONT ON it, and that there are only five canals in the dorsal inner hyoid group—a decidedly lesser number than in other species. 7 On the ventral surface (Plate V, fig. 8) the outer group of hyoid canals is shorter and covers a less extended surface of the fin than it does in other members of the genus. Capsular walls were observed to be absent from the ophthalmic and inner and_ outer buccal ampullee. R. clavata, Linn., or the “Thornback Ray,” has the three or four canals in the dorsal posterior hyoid group all of equal length and lying close together for the whole of their course. The last two canals of the dorsal outer hyoid group terminate at a greater distance from the border of the fin than do the canals next to them (Plate VI, figs. 9 and 10). Specimens of this species also, dissected for the three cephalic groups of ampulle, showed an absence of enclosing fibrous tissue. R. macrocellata, Montagu.—The only specimen examined had in the group selected for comparison three close and parallel canals of nearly equal length (Plate VII, figs. 11 and 12). The last canal of the dorsal outer hyoid group ended at the angle where the other canals cease to cross the muscle ridges and commence to follow the groove between them. ! R. maculata, Montagu, has five canals in the dorsal posterior hyoid group, lying close together for the whole of their course, and all of equal length (Plate VIII, figs. 13 and 14). The features above enumerated have been selected for purposes of comparison where distinguishing features are neither obvious nor easy to describe. It. is not, however, suggested that these characters are necessarily constant, but rather that, owing to the variability which the canals exhibit, especially as regards numbers, the development of any given species might be expected to be in the direction indicated, with fluctuations about an average type. 5. INNERVATION. As stated above, all the five pairs of groups of ampulle are innervated by branches of the facial nerve, and have been named after the branches by which they are severally supplied. Hach of these supplying branches innervates, in addition to the ampullz, sense organs of. the sensory canals, as described and figured by Ewart in his paper on the Sensory Canals of the Common Skate, in which the relationship of the two systems as regards their innervation is made clear. For facts relating to — the nerve-supply of the sensory canals ‘reference has therefore been made to the above paper. Since a dissection of the dorsal surface best exposes the nerve-supply of the — superficial ophthalmic, outer buccal, and hyoid ampulle, while the inner buccal and THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 485 TEXT-F1G. 6.—Dorsal dissection of nerve-supply of ampulle, R. batis, natural size. S.0. Superficial ophthalmic ampulle. O.B. Outer buceal ampulle. H. Hyoid ampulle. S. Spiracle, N.C. Nasal capsule. A.C. Auditory capsule. ¢.b. Inner buccal nerve. 486 AUGUSTA LAMONT ON mandibular nerves can only be traced to their respective groups of ampulle from the ventral side, the accompanying drawings (text-figs. 6 to 9) aim at representing the TExt-FIc. 7,—Ventral dissection of nerve-supply of ampulle, R. batis, natural size. I.B. Inner buccal ampulle, H. Hyoid ampulle, M. Mandibular ampulle. N.A. Nasal aperture. L.J. Lower jaw. o.b. Outer buccal nerve. two aspects of the nerve-supply as actually seen on dissection, and not at a theoretical scheme of the whole as it would appear at one glance. . In most cases the supplying nerve divides on approaching the ampullary group, ae THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 487 and on entering the capsule, where one is present, each division again divides to send a small nerve-twig to each ampulla. (1) Superficial Ophthalmic. The main portion of this nerve proceeds to the ophthalmic group of ampulle (text-figs. 6 and 8), giving off slender branches to the supra-orbital (sensory) canal ear | pA as aw | { ae BY Hu ae aN an WW, = \ {|| \ @ 2 oe 1\' Nel aa oF \ on M2 4 -\ B a et a ; aw 7 ) SS ‘ / | | Text-Fi@, 8.—Dorsal dissection of nerve-supply of ampulle, R. cirewlaris, natural size. G. Gills. Other lettering as in text-fig. 6. on its way, and continuing as a slender branch beyond the ampulle to the tip of the snout for the further supply of the sensory canal. (2) Inner Buccal. The inner buccal branch of the buccal nerve, after its division from the outer buccal, passes under the nasal capsule and sends two branches to the inner buccal 488 AUGUSTA LAMONT ON ampulle (text-figs. 7 and 9). On its way it sends off a series of slender branches to the ventral portion of the infra-orbital canal, and after supplying the ampulle, proceeds to the tip of the snout for the further supply of the same sensory canal. Where the inner buccal ampulle are enclosed by a capsule, as shown in the accompanying TEXT-FIG. 9.—Ventral dissection of nerve-supply of ampulle, R. circularis, natural size. G. Gills. Other lettering as in text-fig. 7. sketch of a dissection of Raia batis (text-fig. 7), both the branches mentioned pass through it before breaking up for the supply of the individual ampulle. In~ Rk. clavata the condition represented in text-fig. 3 was observed. The two branches —much longer than those of &. batis—supplied each a separate little cluster of ampulle, The ophthalmic ampulle lay between, though a little anterior to these two clusters, and in close contact with them, and, since there were no enclosing capsules, THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 489 the two groups appeared as one, and the importance of the innervation in dis- tinguishing between the groups of ampulle was demonstrated in a practical way. (3) Outer Buccal. The outer buccal branch of the buccal nerve, after its division from the inner buccal, proceeds direct to the outer buccal ampulla, passing through the antorbital cartilage on its way (text-figs. 6 and 8). After the major portion of the nerve has broken up for the supply of the ampullz, the remaining slender portion continues forwards for the supply of part of the infra-orbital (sensory) canal. (4) Hyoid. The large hyomandibular nerve proceeds direct, behind the spiracle, to the hyoid group of ampulle, where it breaks up into numerous branches, some of which go to supply the ampulle, and others to supply both the dorsal and ventral loops of the hyomandibular (sensory) canal (text-figs. 6 to 9). (5) Mandibular Nerve. The mandibular branch of the hyomandibular nerve divides off from the main stem before this splits up for the supply of its sense organs, and then, curving round to the lower jaw, sends a short branch to the mandibular capsule, and a longer one to the mandibular (sensory) canal (text-figs. 7 and 9). 6. COMPARISON WITH OTHER MEMBERS OF THE BATOIDEI. In closing this account it may be well briefly to compare the structure and arrangement of the ampullary system in Rava with that of the most closely related forms which have been described. These are Torpedo and Trygon—the only other genera of the Batoidei of which accounts have been given. For details relating to the former, reference has been made to Lrypie’s Rochen und Hae, Fritscu’s Torpedineen, and Cocears Sviluppo delle Ampolle di Lorenzin. The ampullary system of Trygon has been referred to nowhere except in Leypic’s Rochen und Hae, where very brief allusions are introduced among references to many other Selachians. According to Leypic, Torpedo has the simplest ampulle, and he describes them as simple round vesicles without outward sacculations, although internally four- chambered by means of partitions. He describes the only two pairs of ampullary groups which he recognised as being enclosed in capsules. According to CoGer’s more recent account of Torpedo ocellata, there are three pairs of groups of ampullz, comprising 168 ampulle in all, and supplied respectively by the superficial ophthalmic, the buccal, and the external mandibular branches of the facialis. Of these groups the mandibular contains by far the largest number of 490 AUGUSTA LAMONT ON ampulle, the superficial ophthalmic contains considerably fewer, while only four to six ampullee are innervated by a single small branch of the buccal nerve. In Rava the total number may be taken to be in excess of this—in some species very greatly so. According to figures and descriptions of the ampullary canals in Torpedo, the canals appear to be less widely distributed over the surface than is the case even in those rays which have the shortest and fewest canals. The ampulle of Trygon pastinaca are described by Leypic as representing the next step in development above those found in Torpedo. Each ampulla has four big sacculations which bulge out especially in a downward direction, as shown in Plate II, fig. 4 of Rochen und Hate, and which are not indented like those of Rava. The ampulle are smaller than those of Raia clavata or Leviraia oxyrhynchus,* and are peculiar in having a yellowish colour. There are five pairs of ampullary capsules. The ampullary canal system of Raia thus appears in some respects to be more highly developed than that of either Torpedo or Trygon. The number of its ampulle and ampullary groups is in excess of those possessed by Torpedo, where the hyoid and one of the buccal groups are absent ; and the individual ampullz are more complicated in structure than those of either Torpedo or Trygon. On the other hand, the much greater development of the mandibular group in Torpedo— both relative to the same group in Rava, and to the other groups in Torpedo itself— invites remark, reversing as it does the condition found in Rava, where that group is always the least both in size and in number of the contained ampulla. , BIBLIOGRAPHY. 1664. Stenson, De Musculis et Glandulis Observationum specimen, etc., Amst., 1664. 1669. Stenson, Hlementorum Myologiz specimen, etc., Amst., 1669. 1678. Lorenzint, Osservaziont intorno alle Torpedini, Firenze, 1678; London, 1705, Angl. _1785. Monro, ALEXANDER, The Structure and Physiology of Fishes, 1785.: 1801. Sr. Hinarre, Guorrroy, “Mémoire sur l’anatomie comparée des organes électriques de la Rate torpille, du Gymnote engourdissant, et du Silure trembleur,” Ann. Mus. Nat. d’Hist. Nat., 1802, i, p. 392. 1813, Jacopson, “ Extrait d’un mémoire sur un organe particulier des sens dans les Rates et les Squales,” Nouv. Bull. Sci. Société Philom. Paris, 1813, vi, p. 332. 1820. Treviranus, G. R., “Uber die Nerven des Fiinften Paars als Sinnesnerven,” Vermischte Schriften anatomischen und physiologischen Inhalts, Bremen, 1820, iii, pp. 135-146. 1822. Buainviniz, Vraité des animaux: de leur organisation, ou principes d’unatomie compareée, Paris, 1822, i. 1825. Knox, Ropert, ‘On the Theory of the Existence of a Sixth Sense in Fishes, supposed to reside in certain peculiar Tubular Organs found immediately under the integuments of the Head in Sharks and Rays,” Edinburgh (Brewster's) Jour. Sct., 1825, ii, pp. 12-16. 1828. Cuvier et Vatenciennes, Histoire naturelle des poissons, Paris, 1828, i, p. 521. 1836. Deus Curasn, Instituzione di anatomia comparata, 1836. * Leviraia oxyrhynchus, Bonap. = Raia oxyrhynchus, Linn. 1870. 1878. 1878. 1885. 1879. 1880, 1880. 1888. THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 49] . Dee Cutage, “ Anatomiche disamine sulle Yorpedini,” Atti R. Instituto d’ Incorragiamento alle Scienze Nat. Napoli, 1840, vi. (Published separately in 1839.) . Davy, Jonny, “An Account of some Experiments and Observations on the Torpedo,” Researches Physiological and Anutomical, London, 1839, i, pp. 1-93, pls. x and xi. . Savi, Atti della terza Riunione degli Scienziati Italiani, Firenze, 1841, pp. 334-335. . Savi, Zraité des phénomenes electro-physiologiques des animaux, par C. Matteuci, swivi d’“ Etudes anatomiques sur le systéme nerveux et sur V’organe électrique de la Torpille,” par Paul Savi, Paris, 1844, chaps. vi and ix, pp. 329-331, pl. iii. . Mayr, Spicilegium Ubservationum Anatomicarum de Organo electrico in Raiis anelectricis, 1843. . Miuscuer, Bericht ciber die Verhandl. der Naturforschender Gesellschaft in Basel, vi, pp. 107-108, 3lst January 1844. . Romy, ‘‘ Recherches sur un appareil qui se trouve sur les poissons du genre des Raies (Raia, Cuv.), et qui présente les caractéres anatomiques des organes électriques,” Ann. des Sct. Nat., 3m série, Zoologie, 1847, vii, pp. 193-204. e . Rerzius, “ Uber die vermeintlichen elektrischen Organe bei den nicht elektrischen Rochen,” Froriep’s otizen aus dem Gebiete der Natur- und Heilkunde, 1848, No. 3, vol. v, pp. 54-55. . Stannius, Das peripherische Nervensystem der Fische anatomisch und physiologisch untersucht, Rostock, 1849, Dritter Abschnitt, pp. 20-74. . Miuusr, H., “‘ Die Nervése Follikelapparat der Zitterrochen, und die sogenannten Schleimkanale - der Knorpelfische,” Verhandl. der Physikalisch-medicinischen Gesellschaft in Wiirzburg, ii, 1852; No. 9, 1851, pp. 134-144; No. 10, 1851, pp. 145-149. 52. Eoxer, ‘‘ Jahresbericht iiber die Fortschritte der vergleichenden Anatomie der Wirbelthiere in den Jahren 1845, 1846, und 1847,” Miiller’s Archiv fiir Anat., Physiol. und wissenschaftliche Medicin, 1852, p. 27. 2. Leyvie, Bettrdge zur mikroskopischen Anatomie und Entwickelungsgeschichte der Rochen und Hate, - 1852, pp. 36-50. . Leypie, Lehrbuch der Histologie des Menschen und der Thiere, 1857, pp. 196-210. . Leypie, ‘‘ Uber Organe eines sechsten Sinnes,” Verhandl. der Kaiser. Leopold.-Carol. deutschen Akad. der Naturforscher (Nova Acta), 1868. . De Lampauue, Josert, Des appareils électriques des poissons électriques, Paris, 1858. . Ecxuarp, ‘‘Uber die Endigungsweise der Nerven in den Schleimkaniilen des Zitterrochen,” Beitrage zur Anat. und Physiol., Giessen, 1858, i, Fiinfte Abhandlung, pp. 85-94. . M‘Donvatt, “On an Organ in the Skate which appears to be the Homologue of the Electric Organ of the Torpedo,” Nat. Hist. Review, 1861. . Coucn, British Fishes, 1862, 1. . DumeériL, Histoire naturelle des poissons, Paris, 1865, tome premiere : Elasmobranches. . Bout, “‘ Die Lorenzinischen Ampullen der Selachier,” Schultze’s Archi fiir mikro. Anat., 1868. . Bott, “ Ein historischer Beitrag zur Kenntniss von 7'orpedo,” 1873; Reichert’s und Du Bois- Reymond’s Archiv fiir Anat. und, Physiol. (Miiller’s Archiv), Leipzig, 1874. (On Lorenzini’s Osservaziont.) Tovaro, Contribuzione alla anatomia e alla fisiologia dei tubi di senso dec Plagiostomi, Messina, 1870. Soucer, B., Uber die Seitenorgane der Fische,” Verhandl. der Kaiser. Leopold.-Curol. deutschen Akad. der Naturforscher (Nova Acta), 1878, Heft 14, pp. 74-80; Nachtrag, ebd. pp. 159-160. Batrour, 7he Development of Elasmobranch Fishes, 1878, pp. 141-147. Batrour, A T’reatise on Comparative Embryology, 1885, ii, p. 540. Sappry, Etudes sur Vappareil mucipare et sur le systome lymphatique des poissons, Paris, 1880. Market, Uber die Endigqungen der Sensiblen Nerven in der Haut der Wirbelthiere, Rostock, 1880, pp. 39-48. Day, The Fishes of Great Britain and Ireland, London and Edinburgh, 1880-1884, ii. Garman, “On the Lateral Canal System of the Selachii and Holocephala,” Bull, Mus. Comp. Zool. Harvard College, Cambridge, Mass., U.S.A., 1888-1889, xvii, pp. 57-119. TRANS. ROY. SOC. EDIN., VOL. LI, PART II (NO. 12). 71 1902. 1902. 1906. 1908. AUGUSTA LAMONT ON . Frrrsen, “Uber Bau und Bedeutung der Kanalsysteme unter er Haut der Selachier,” Sitzwngs- berichte der Kinig. Preuss. Akad. der Wissenschaften zu Berlin, Berlin, 1888; Jahrgang 1888, Erster Halbband. . Fritscn, Die elektrischen Fische. Zweite Abtheilung. die Torpedineen, Leipzig, 1890, pp. 78-92. . Ewart, “On the Cranial Nerves of Elasmobranch Fishes: Preliminary Communication,” Proce. Roy. Soc. Lond., xlv, November 1888—April 1889, pp. 524-537. . Ewart, “ The Lateral Sense Organs of Elasmobranch Fishes: i, The Sensory Canals of Lemargus,” Trans. Roy. Soc. Edin., vol. xxxvii, part i, 1892, pp. 59-85, pls. i and 11. . Ewart and Mrrcuett, “The Lateral Sense Organs of Elasmobranchs: ii, The Sensory Canals of the Common Skate (Raia batis),” Trans. Roy. Soc. Edin., 1892, vol. xxxvii, part i, pp. 87-105, pl. iii. . Fucus, §., “Uber die Function der unter der Haut liegenden Canalsysteme bei den Selachiern,” Pfliiger’s Archiv fiir die ges. Physiol. des Menschen und der Thiere, Band lix, Bonn, 1895, pp. 454-478, pl. vi. . CoutingE, W. E., “On the Sensory and Ampullary Canals of Chimera,” Proc. Zool. Soc. Lond., 1895, pp. 878-890, pls. li-liii, . Cor, F. J., “On the Sensory and Ampullary Canals of Chimera,” Anat. Anzeiger, 1896, Band xii, pp. 172-182. (Criticism of above.) . Coxz, F. J., “On the Cranial Nerves of Chimera monstrosa (Linn.), with a Discussion of the Lateral Line System, and of the Morphology of the Chorda tympani,” 7’rans. Roy. Soc, Edin., 1896, vol. xxxviii, part 11. . Pgasopy, J. E., “The Ampulle of Lorenzini of the Selachii,” Zoological Bulletin, Boston, U.S.A., 1898, vol. i, pp. 163-178, with 9 text-figures. . Branves, G., “Die Lorenzinischen Ampullen,” Verhandl. der deutschen Zool. Cresells. (Achte Jahresversammlung), 1898: Leipzig, 1898, pp. 179-182. . ForssELu, G., “‘ Beitrage zur Kenntniss der Anatomie der Lorenzinischen Ampullen bei Acanthias vulgaris,” Zeitschrift fiir wissenschaftlicher Zoologie, 1899, Band lxv, pp. 725-744, pl. xxxiv. - Mincxert, W., “ Zur Topographie und Entwickelungsgeschichte der Lorenzinischen Ampullen,” Anat. Anz., Jena, 1901, Band xix, pp. 497-527, 11 figures. . Auuis, E. P., Jr., “ The Lateral Sensory Canals, the Eye-Muscles, and the Peripheral Distribution of certain of the Cranial Nerves of Mustelus levis,” Quart. Journ. of Micro. Science, London, 1902, vol. xlv, new series, pp. 87-236, pls. x and xi. Jounston, J. B., ““The Homology of the Selachian Ampulle: A Note on Allis’s recent paper on Mustelus levis, Anat. Anz., 1902, Band xxi, pp. 308-313. (Criticism of above.) Coeat, A., “ Nuove ricerche sullo sviluppo delle ampolle di Lorenzini,” Atti R. Accad. Lincei, anno 299, 1902. Serie quinta, Rendiconti, Classe di Scienze fisiche, matematiche, natural, Roma, 1902, vol. xi: I semestre, Nota I, pp. 289-297 ; Nota II, pp. 338-340. Hawkes, Mrs O. A. Mzrrirt, ‘The Cranial and Spinal Nerves of Chlamydoselachus anguineus (Gar.),” Proc. Zool. Soc. Lond., 1906, ii, pp. 959-991, pls. Ixviii and lxix, and 2 text-figures. Broumer, “Die Sinneskanale und die Lorenzinischen Ampullen bei Spinax-Embryonen,” Anat. Anz., Jena, 1908, xxxii, pp. 25-40, 8 text-figures. EXPLANATION OF PLATES, Plates I to VIII, figs. 1, 3, 5, 7, 9,11, and 13, show the ampullary canals of the several species as seen on removing the skin from the dorsal surface. The hyoid and superficial ophthalmic ampulle only are shown, the outer buccal ampullz becoming exposed only when a deeper dissection is made. Plates II to VIII, figs. 2, 4, 6, 8, 10, 12, and 14, show the ampullary canals of the several species as seen on removing the skin from the ventral surface. In most cases the hyoid and mandibular ampullary THE LATERAL SENSE ORGANS OF ELASMOBRANCHS. 493 capsules are shown; the superficial ophthalmic and inner and outer buccal ampulla are not exposed by a superficial dissection. COMMON REFERENCE LETTERS FOR PLATES. S.O. Superficial ophthalmic ampullary canals. O.B, Outer buccal canals. I.B. Inner buccal canals, H. Hyoid ampulle. A. Anterior canals of hyoid ampulle. O, Outer canals of hyoid ampulle. P. Posterior canals of hyoid ampulle. I. Inner canals of hyoid ampullee. M. Mandibular canals. 8.C. Sensory canal. K. Eye. S. Spiracle. G. Gills. Mth. Mouth. M.R. Muscle ridges. PRESENTED 4 9JUN.191Z, 1 . v “ i ‘ w ' . j al v Vou. LI. Trans. Roy. Soc. Edin. Avausta Lamont: “The Lateral Sense Organs of Elasmobranchs.”—Ptave I, ———————— a be ee ee ee a 5 CS ) o Oo & a mH =] n = ss] a = io) xs) s = ~~ = > 8 iS S n = 3s i=] ss ‘ / Fic. 1.—Ampullary of ( Trans. Roy. Soc. Edin. : Vou. LI. Aveusta Lamont: ‘The Lateral Sense Organs of EHlasmobranchs.”—Pr ate II. SS \N PS j SQ y \ Se NY \ NN AN \\ Fic. 2.,—Ampullary Canals of Raia batis, ventral surface, 2 n.s. Trans. Roy. Soc. Edin. Vou. LI. Aveusta Lamont: ‘The Lateral Sense Organs of Elasmobranchs.”—Puare III. Trans. Roy. Soc. Edin. Vou. LI. Avausta Lamont: “The Lateral Sense Organs of Elasmobranchs.”—Pratr IV, Fi| 5.—Ampullary Canals of Raia radiata, dorsal surface, n.s. Fic. 6.—Ampullary Canals of Raia radiata, ventral surface, # n.s. Lei! Vou. LI. Trans. Roy. Soc. Edin. Prats VY, Aveusta Lamont: “The Lateral Sense Organs of Elasmobranchs.” Trans. Roy. Soc. Edin. Vou. LI. Aveusta Lamont: “The Lateral Sense Organs of Elasmobranchs.”—Puate VI. Trans. Roy. Soc. Edin. Vou. LI. Aveusta Lamont: “The Lateral Sense Organs of Elasmobranchs.”—Puare VII. Trans. Roy. Soc, Edin. Vou. LI. Aveusta Lamont: “The Lateral Sense Organs of Elasmobranchs.”—Puate VIII. \ BENG J ‘e _———~ a The Transactions of the Royat Society or Epinsurex will in future be Sold at 4h following reduced Prices :— =. j Price Price Price Price Price Price | Vol. to the to Vol. to the _ to Vol. to the to Public, Fellows. Public. | Fellows. Public. Fello VS, oh Cet Aeneas me. C4 Shsa, Ko Seen se £ os. dia VI. O1l 6/0 9 6] XXXITT. Pt 1/1 1 OO 16 Ole, XLII. 2 2 0 VII. 018 0/015 O 3 Biante ecu) oleedar 058 XLIII. 2.2 0) ie VIII. OL O10 14,0 yp eet: 73 Onl 200 ane e Gn Fic XLIV-Pt-1|\1 18. 63a IX. 1 0 0/017 O|] XXXIV. 7A a OM Me ak 0 |RP » « Pt.2 |i ‘1. 200s Ke OFS" FOHOMUG = Olli, XR Pt | Qe Oy eaten 0 XLV. Pt. 1) 1: 9 Sones XI. 014 6/012 0O a ety?) eael SO) leone » Pt 2)) 1 7 SO XII. Orla “6: /0y 12 0 a Pi3/2) VaeOlol eal = 0 5 Pt. 3) 1 13 one MILE 018 0/0 15 0 % Rt Als 1 OLOn16 10 ae Pt. 4/0 4 6/0 XIV. LD: Ol Le OW XRT BS al OOP iG ane XLVI. Ptul}1 1 TO XV. El eae ale GEO 9 Pte 1G: lel vib ., Pt. 2)1° 5b Sve Mo eb a0 18! 10/0) TAO 5 Pro OOO Ge Oy i Pt. 3) 1 “i 3iee XXII. Pt. 2/0 10 010 7 6) XXXVIIPt.1)1 14 6)|/1 5 6) _XEVIL Pt 1/0 19° 59)0a im Te ay GER jee 0) pale TES 20) 3 Pt. 2) 1-1 O00 16 0 =P Pt. 2/1 ° 3 OO XXVIII. Pt. 11/0 16 0/0 120 53 Pt3)0 16° 0/0712: 20% » . Pt. 3) 1) Oto L HS Pt.2/'0 6 OO .4.°6 3 Piss ver ol Omens . Pt. 4)1) 7 Vai < Pt.4)1 0 010 16 OVXXXVITT.Pt.1)/2 0 0/1 10 O} XLVILTL Pt. 1\i, 269) ROXOVLee ty Lsleeo) One) sllexO Pores ng aye COICO) SUEY (0) 3 Pt. 2 | 1 OSes s Pie preOs il ae Ag Pinot LO. Ol 2320 a Pt. 3.1 1 sOieeae : Pt.3/0 18 01/013 6 FP Bt e0) ve 26) 10} cone 5 Pt. 4/0 16 8/0 1: ORES Pid Oe 6 “0: XOXER XS SE! | SL OMeOll eos ao XLEX. Pt. 1.10 7eaG Ome “f Pt. 2/0 16 0/0 12 O x Pt.2/0 19 0/0 14 6 5 Pt. 2)1 12) Ghee ROX Pet 20) ae Gi 0 os ledta oper oo) ON Beatle ~ @) » Pt-31 LOS Onst 55 O22 OG O02 12550 =A Pt. 4/0 9 O10 7 O “5 Pt.4/1 1. 90 5 Pt. 3/0 5 0/0 4 O XL. Pt.1/1 5 0/019 0O L. Pt. 1-11 5 9a ss BeeAO a7 Ss63/10: mbes 99. E827 OG) ai ono 5, Pt. 2|1 7 Oe XOK Ks AWA Oo! none) » Et. 31 26" 7030-19 - 6 » Pt. 3) 1 eee i XXXII. Pt.1)/1 0 0/0 16 0O » Pt4/1 0 0/0 16 O » Pt 41 2) 00m oa Pt. 2/0 18 0/013 6 XL Pe Tide OOS: 9 LI..Pt. 1)1 5-0) a IPpsrd Nol O! 20) Slee pe ed oAlii bes) (yl ibe XO) » Pt. 2/1 13 0am a Bid: (Og ba Oh OnedaeO sh) Pt.20)|:2) Om Oi plant) 986 » bt.-31 17° 160i a, ( * Vol. XXXV, and those which follow, may be had in Numbers, each Number containing a complete Paper. January 1917.—Volumes or parts of volumes not mentioned in the above list are not for the present to the public. Fellows or others who may specially desire to obtain them must apply direct to the Soe: the Society reprints from time to time parts of its publications which have become scarce, the absolute correch of this list cannot be guaranteed beyond this date. * . 4 “J ‘< The papers published in this Part of the TRansactiIons may be had on application to the Publis ers at the following prices :— Price to Public. Price to Fellows. Price to Public. Price’ 0 Fe No. VIL, f : 7s. 5s. 3d. | No. X., : ; 4s, 6d. 3s. 64 No. VIIL., a : 1s. 10d. 1s. 5d. No. XI, Y, : Is. = No. IX., . 14s, 6d. 11s. No. XIL., ; : 4s, 3d, 3s, 3 TRANSACTIONS _OF THE ROYAL SOCIETY OF KBDINBURGH. VOLUME LI, PART III.—SESSIONS 1915-16-17. CONTENTS. XIII. Contributions to the Geology of Benyuella. By Professor J. W. Speen F.RB,S., DSc: (With ‘Two Plates), : (Issued August 14, 1916. ee XIV. A Contribution to the Petrography of Benguella, based on a Rock Collection made by Professor J. W. Gregory. By G. W. Tyrruit, A.R.C.Sc., F.G.S., Lecturer in Geology, Glasgow University. Communicated by Professor J. W. ’Grucory, F.R.S. (With One mm Plate), ’ . a ) (Issued August L, 1916, ) On some Cretaceous Brachiopoda and Mollusca from Angola, Portuguese West Africa. By \ KR. Buttey Newron, F.G.S., Geological Department, British Museum. Communicated . L by Professor J, W. Grucory, F.R.S. (With Two Plates), ; (Issued August 14, 1916.) if Nos on an Aly a8 Limestone from Angola. By Mrs M. F. Romanus, late Harkness Scholar, a . Newnham College, Cambridge. Communicated by Professor J. W. Grugory, F.R.S. sees ila One Plate), iheli«: are (Issued April 25, 1916.) e Cr etaceous Eehinoidea from the Neighbourhood of Lobito as “as Professor J. W. (Issued August 21, 1916. eke butions towards a Knowledy ye of the Anatomy of the Lower Dicotyledons. 1. The Anatomy y of the Stem of the Papaveracex. By R. J. Harvey-Gisson, D.L., M.A., Pro- sor of Botany, University of Liverpool ; and Minniz Brapuey, M. Se., Hartley Research, ar, University of Liverpool. (With Three Plates), . . * (Issued September 16, 1916.) oeleidus teretipes: A new Oxfordian Plesiosaur in the Hunterian Museum, Glasgow Tniversity. By Wittiam R. Smenum, M.A., B.Sc. Communicated by Professor J. W. Grecory, D.Se., F.R.S. Edin. and Lond. (With One Plate), , : t (Issued August 29, 1916.) ‘The Anatomy and Ajjinity of Platyzoma microphyllum, R. Br. By Joun M‘Luan THompson, ea A., B. Se., Chief Assistant in Botany and Robert Donaldson Research Scholar, Glasgow University. Communicated by Professor Bowrr, F.R.S. (With Four Plates), ‘ ‘ (Issued November 6, 1916.) XI. On Leaf-Ar chitecture as illuminated by a Study of Pteridophyta. By F. O. Bowsr, D.Sc., | et: ‘F. R.S., Regius Professor of Botany in the University of Glasgow. (With One Plate), . 1k : (Issued November 8, 1916.) XX. Contributions to our Knowledge of British Palxozoic Plants. Part I. Fossil Plants from the ca Scottish Coal Measures. By Dr R. Kipsron, F.R.S. (With Three Plates), : ‘ iby (Issued November 29, 1916.) XNIIL A Revision of the British Idoteide, a Lamily of Marine Isopoda. By Wanrer EK. CoLnince, D.Sc., F.L.S., ete., Research Fellow of the University of St Andrews. Communicated by Professor M‘InTosH. (With Eleven Plates), F } ‘ . (Issued January 17, 1917. ‘e Y XXIV. On Old Red Sandstone Plants showing Structure, from the Rhynie Cherl Bed, Aber ae af Part I. Rhynia Gicynne-Vaughani, Kidston and Lany., By R. Kipsrox, LL.D., F.R and W..H. Lane, D.Sce., F.R.S., Barker Professor of Exypicgemic pet in the Univer of Manchester. (With Ten Plates), : (Issued February 27, ‘1917. ) XXV. The Prothallus of Tmesipteris Lannensis, By Professor A. AnsrrurHER Lawson, D.Sc., F.LS. (With Three Plates), ‘ ‘ . (Issued April 21, 1917.) EDINBURGH: PUBLISHED BY ROBERT GRANT & SON, 107 PRINCES STREET, MDCCCCXVILI. Price Thirty-five Shillings. PAGE 495 589 609 631 657 709 AND WILLIAMS & NORGATE, 14 HENRIETTA STREET, COVENT GARDEN, LONDON. = =a” en (7495) XII].—Contributions to the Geology of Benguella. By Professor J. W. Gregory, F.R.S., D.Sc. (With Two Plates.) (MS. received June 2, 1915. Read June 28,1915. Issued separately August 14, 1916.) CONTENTS. PAGE PAGE I. Introduction and Previous Literature. . 495 G. The Bulu-Vulu to Huambo . F . 518 11. Geological Observations— (1) The Bulu-Vulu.. : . 518 A. The Coast Zone and the Cretaceous . 498 (2) Cambenge and the Umbal : . 518 . (1) The Catumbella Section é . 499 (3) The Maral to Huambo . ; . 519 (2) South-east of Lobito . ; . 501 | III. The Correlation of the Benguella Series— (3) From Lobito to Hanha . 5 502 A. The Mesozoic Rocks : 521 B. Benguella to the Upper Catumbella River 504 B. The Older eee Rocks on the C. Oendolongo Mountains . ; é . 507 Plateau ; : . 522 D. From Lepito Huambo . : é . 509 C. The Eozoic Gneicces : 528 E. Huambo to Ochilesa : : : . 511 D. The Volcanic Rocks and Taiamese Rooks F. Ochilesa to Saccanjimba . ; c . 515 of Undetermined Age . ; 2 . 528 IV. The Tectonic ere : . z : . 529 V. References. 5 : : é . 532 I. InrRopuctTion AND Previous LITERATURE.* Benguella is a province of Angola, the chief Portuguese colony in western Africa. From the chief town, Benguella (lat. 12°35° S.) and from Catumbella (lat. 12°25° S.) old-established trade routes go inland on to the Benguella plateau, and across it along the watershed between the Congo and the Zambesi to Katanga and the Upper Congo. The geographical pioneers in this part of Africa showed that the land rises in steps from the coast to a high plateau, which consists of a foundation of ancient erystalline and sedimentary rocks, and is flanked by younger marine beds along the coast. _ Angola is divided into three provinces, of which the northern, Loanda, and the southern, Mossamedes, are better known geologically than the middle province, Benguella. The geology of Loanda is the most varied, and the first important con- tribution to the interpretation of its structure was by Livinestong, who crossed it during his trans-African march from the Upper Zambesi to the coast at Loanda. Of Benguella he saw only the far interior, and he reached the western edge of the African plateau in Loanda. WetwritscH (1889), the well-known British Museum botanical collector, made important geological observations in the same region and * The bearings given are from true north. The variation has been taken as 18° W.; at Benguella it is 18° 10’ W.; at Huambo Fort 17° 53’; it decreases to the north-east. The spelling of place-names varies considerably, and K has been often used instead of C. I have followed in most cases the spellings adopted on the Portuguese map of Angola (2nd edit., 1910) and by the Benguella Rail- way Co. References are given by the author’s name and date to the list at the end. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 18). 72 496 PROFESSOR J. W. GREGORY. in Mossamedes, but his valuable notes and sections were not published until 1889. Meanwhile, the existence of gneiss, schist, gypsum beds, limestones, volcanic rocks, and copper ores on the Benguella coast-land was announced by Monterro (1875, vol. 1, pp. 145, 177, 189, 192, 197, 283; -vol. a1, pp. 655-70; 91, Lo2,.17 7, 196) Oot ae etc.), a mining engineer who was for many years engaged at the copper mines; his work also dealt mainly with Loanda. Loverr Cameron, who crossed the whole width of the Benguella plateau, showed (1877, vol. ii, pp. 231, 236, 254, etc.) the widespread distribution of granite in the interior, the presence of red sandstones on the plateau at Bihé (cbid., p. 317), and of limestones containing ammonites on the coast near Catumbella (cbid., p. 261). Major Serra Pinto (1881, vol. i, pp. 218-219) referred to the existence of lime- stone at Dombe Grande, 30 miles south-east of the city of Benguella, and of granite at nine scattered localities on the plateau to the east (ibid., pp. 54, 55, 72, 78, 85, 87, 121, 219, 221).. CaPELLo and Ivens (1882, vol. i, pp. 82, 48, 81, 217, ete.) recorded many exposures of granite, gneiss, and schist on the Benguella plateau. José p’ ANCHIETA (1885), a Portuguese zoologist, briefly described the geology of the neighbourhood of Benguella, and referred to the gneisses and schists of the interior and to the Cretaceous beds of the coast; and asthe called the last Quadersandstein, he had recognised their Cenomanian age. The first important contributions to the geology of the coastal limestones were issued by Meunier (1887) and by Cuorrat and pE Lorton (1888). Meunter described some Cenomanian cephalopods, as well as some foraminifera from Lobito Bay. The studies of CuorraT and DE LorioL were based on collections made by L. MaLHerro, mainly in the Dombe Grande district, and showed the presence of several Middle and Upper Cretaceous horizons.* In this monograph CHorraT summarised what was known regarding the geology of Angola; most of the information referred to Loanda. He showed the presence of Kainozoic rocks both in Loanda and Mossamedes, and the occurrence of Cretaceous rocks at many localities along the coast; he recorded, on the identification of Gomgs, the presence of nepheline-basalt at Dombe Grande. The first important contribution to the petrography of the older rocks of the Benguella plateau was made by J. P. Goms (1898) on specimens collected by R. P. Leconte between Benguella and Catoco, 250 miles to the east-south-east. GomuEs identified granite, biotite-granite, amphibolite, adinole with veins of epidote, mica- schist and other schists, quartzites, diabasic porphyrite, and basic eruptives. In 1895 CHorraT summarised the geology of Angola and gave a classification of | the Cretaceous beds, which range from the Albian to the Senonian. He again reviewed the literature on the country, and described further additions to the Cretaceous fauna in an important memoir in 1905. The widespread distribution of gneiss and granite associated with basalt, quartzite, and diorite in the Bailundo and Huambo districts and of a siliceous limestone at Bailundo was reported by an official mission on the * According to SCHLUMBERGER (1888) the foraminifera indicated a Miocene horizon as well. Se See . ee on are - = ere ne rreeninmirens s — iva CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 497 colonisation of the Benguella plateau in 1909 (Angola, 1910, pp. 41-44). Senhor J. P. NascrmEnro (1912) briefly referred to the geology of the Benguella plateau, and recorded the presence of granite, diorite, basalt, and quartzites, and also the existence of hot springs at Andulo and Chieuca, in what had become known as the volcanic district of Bihé. During the same year Jodo pz AtmErpa (1912) described the Kainozoic rocks along the coast of Mossamedes, and gave further details regard- ing the ancient rocks of the Serra Chella and Humpata plateau, which are the southern continuation of the Benguella plateau. Further references to the geology of Benguella are included in the list (pp. 532-534), which includes a few relating to the coastal rocks of the other provinces of Angola. In 1912 I visited Benguella at the invitation of Mr IsrarL Zancwitt to report on its economic and agricultural suitability for a refugee colony. The special object of the journey involved consideration of the geological structure of the country from the light it would throw on the quality and distribution of the soils. But as geology was only involved in the work of the mission to that extent, I often had to leave points unsettled, if they had no bearing on the economic possibilities of the country. Thanks to the kindness of General Sir J. J. Macnapo, the general manager of the Benguella Railway Co., and Mr EH. Rostns, the chief engineer in Benguella, I had the opportunity of seeing far more than I could have reasonably expected in the time at my disposal. I have also to express my thanks to Senhor ANDRADE, the Minister of State for the Colonies of Portugal, to their Excellencies Senhor Norton, the Governor-General of Angola, and Major Pinto, the Governor of Benguella, and to Professor A. BEN Saupe in Lisbon; also to Senhor Sacramento Monteiro, my kind host at the agricultural station at Quingenge, to Mr Varian and other officers of the Benguella Railway Co., and in fact to all those whom it was my privilege to meet in the country. I must also express my particular indebtedness to the invalu- able help and congenial comradeship of my companion, Professor C. J. Martin, F.R.S., Director of the Lister Institute, who kindly accompanied me in order to report on the healthiness of the country. I have also to express my thanks to the authors of the accompanying reports on the collections, to Mr R. B. Newton on the general Mollusca, Mr G. W. TyRReELt on the igneous rocks, Mrs M. F. Romanes on the fossil algee, and to Mr J. M. Worpre for a provisional list of cephalopods. Mr Woroptr’s departure for the Antarctic pre- vented the completion of this work, which has been kindly undertaken by Mr G. C. Crick. I am also indebted to Professor A. C. Sewarp, F.R.S., for kindly examining some slides of fossil algze, and to Professor hk. A. Berry for analyses* and study of the soil samples collected. It may be convenient to summarise the route followed. We landed at Lobito Bay, the terminus of the Benguella Railway, whence I visited various localities along * These, and an account of the country visited, are published in Gregory, 1913. 498 PROFESSOR J. W. GREGORY. the coast from Hanha to Benguella. Owing to the kindness of Mr Rosins, | accompanied him by motor trolley from Lobito along the railway to Lepi, which was then the railway terminus; and while he was inspecting work along the line I was able to examine many interesting sections and exposures beside it. After organising a caravan at Lepi, Professor Martin and I marched to Huambo, and thence to Bailundo, of which the fertility was so glowingly described by Cameron. Thence we marched north-eastward into the Cutato valley, to the mission station at Ochilesa, which was reported to be a volcanic district. We then went south to the Bihé plateau and the Bulu-Vulu plains, and returned westward to Huambo along, for part of the march, the northern watershed of the Zambesi. I spent a few days in the neighbourhood of Quingenge, while Professor Martin marched to Cubal, whence we returned to Lobito Bay. Il. GEOLOGICAL OBSERVATIONS. A. The Coast Zone and the Cretaceous. The coast of Benguella consists chiefly of Cretaceous rocks and of an alluvial coastal plain. North of Lobito Bay the Cretaceous rocks form a line of cliffs which are washed by the sea; but from that bay to the south of Benguella the Cretaceous plateau is separated from the shore by a low plain composed of alluvium, with some sheets of sand and lines of sand dunes. The coastal plain is separated in places from the Cretaceous beds by a very irregular boundary; but elsewhere, as north of Catumbella, it ends inland at the foot of a straight steep hill face, which has the characteristics of a fault scarp. This scarp is the front of a limestone plateau, which begins west of Lobito at a height of about 300 feet and rises gradually inland to about 500 feet. It ends against another scarp, by which the Cretaceous rocks rise to the height of from 800 to 1000 feet; and behind that scarp is a third plateau, 1500 feet above sea-level. The Cretaceous rocks form a comparatively narrow band along the coast. Their range inland is limited by the great plateau of ancient rocks that form the main foundation of the province of Benguella. The Cretaceous rocks are best known by the work of CHorrat (1888, 1896, 1905), whose studies were based mainly on collections from Dombe Grande, to the south-east of Benguella, by Matyerro, and from the “ Schloenbachia” inflata * beds of Lobito Bay. Little, however, was known of the general succession and distribution of the Cretaceous rocks near Lobito; their sequence is shown in the great sections on the gorge of the Catumbella River. Mr Rosmns gave us the privilege of accompanying him on a railway trolley up the valley to the Catumbella dam; but this hasty visit only gave opportunity to observe the most obvious facts and collect a few specimens from the hillside above the new reservoir. My visits to the Cretaceous rocks were so hurried and the rocks themselves are * Mr Crick tells me that this ammonite is now included in the genus Mortoniceras. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. © 499 so disturbed that I was not able fully to satisfy myself as to the succession. Un- fortunately, the fossils are usually in the form of casts, so many of those collected have proved indeterminable. The list of fossils collected by Matuerro during his long residence at Dombe Grande contains so large a proportion in which the genus only has been determined, that the poor condition of the fossils appears to be general along this coast. (1) The Catumbella Section.—Immediately south of the railway bridge at Catum- bella is a cliff of gypsiferous, false-bedded limestone, which yielded a large Tylostoma globosum*; and close beside this section was an exposure of marls containing the characteristic ammonite “ Schloenbachia” inflata. Further east, along the southern bank of the Catumbella River, the laminated inflata marls dip to the west and are succeeded by underlying conglomerates which contain large boulders of gneiss and granite. This conglomerate series has been bent into many gentle folds. It is succeeded to the east by a thick bed of clay containing some thin bands of limestone. The clay at one place forms most of a cliff about 200 feet high, the upper part of which consists of the bedded inflata limestone. Further east the valley of the Catumbella widens, as the land on both sides is formed of clay ; and where the limestone cap has been removed the clay has been worn by the rain into many spurred ridges. The beds on both sides of the river are repeatedly folded and faulted. The thick clays disappear eastward under hard white limestones, which on the northern side of the river, just below the dam, form a synclinal traversed by several faults. The limestones occur on the floor of the valley beside the new reservoir. At the base of the cliff there is a cream-coloured nodular limestone erowded with concentric alge, which Mrs Romanegs has described as composed of two new species, Gurvanella minima and Lithothammum angolense. About 80 feet above the algal limestone is a white earthy limestone which yielded an Epvaster catumbellensis ; the limestone on a terrace at the height of about 150 feet yielded a Panopea cf. plicata, numerous gastropod casts, and a fragment of an ammonite which Mr Crick has determined as “8S.” Jenzi. This bed is therefore Vraconnian (Lower Ceno- manian). Above this shelly limestone, at the height of about 240 feet, is a band con- taining many hematite nodules, and amongst other fossils a new species of Neithea (N. angoliensis). At the level of about 850 feet the limestone contains abundant nodules of chert and numerous undetermined gastropods. I was unable to go further up the hill; but Mr Rosrns kindly gave me two ammonites, which came from the top of the plateau, and they are excellent specimens of “S.” inflata. The thickness of the limestones beside the Catumbella dam between the Lithothamniwm bed below and the inflata marls must be about 1000 feet. Further up the valley the inflata beds are said to rest directly upon the older crystalline rocks. The beds along the Catumbella River have been disturbed by a complicated series of parallel faults, of which I noticed fifteen between the railway bridge and half a * For the names of the gastropods and lamellibranchs I am indebted to Mr R. B. Newton. 2000' 000° 4000" 2000' 4000 $ 2 bd 38 : ox Benguetia a‘ Lengwe x < ing LC i* t < [Sommie SS Z Catenpue < < ¢ < 3 < (y) i 8 < & < ? < ry) ene: is Cuba? < < < < < lad < < < x fo) = 3 Quingenge S Cuma t SF g y Lepr Ca < < Hvaméo < < See oars < < < d us CutimahalaR Fig. 1. ° °°. ie} S Fic. 2. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 501 mile above the dam. Many of them, such as a 50-foot overthrust fault on the south bank to the west of the dam, four parallel faults on the right bank just below the dam, and others on the north bank near Catumbella, are unmistakable. But others, of greater stratigraphical importance, [| had not time to confirm. The Cretaceous succession in this section is therefore not free from doubt, but it is apparently as given on the table on page 521. (2) South-east of Lobito (fig. 2).—The inflata beds are most conveniently examined on the south-eastern shore of Lobito Bay. They occur near the landing-place at Old Lobito, and most of the fossils that have been sent to Europe from Lobito were probably collected at that locality.’ I found there “8.” inflata, “S.” elobiensis, “'S.” ef. lenzi, “S.” * candollianus, and two species which Mr Worpte regards as new. Seen from the bay the cliffs on the south-east side appear as a long straight scarp, which cuts across the old valleys, and leaves some of them as hanging valleys. This cliff presents the aspect of a recent fault scarp. I was accordingly not surprised on landing to find the rocks broken by numerous faults which are parallel to the coast. Some of the smaller faults are overthrusts, with the downthrow on the south-western side. The wnflata beds consist of well-bedded nodular limestones and marls. The fossils are mostly in the form of casts. I collected one Epiaster catumbellensis a few feet above sea-level in the inflata marls. The sequence inland is shown in the eanyon of the usually dry river which ends at Old Lobito. In the examination of this section from the shore to the hills 5 miles inland, | had the valuable help of Professor Martin. Both walls of the gorge at first consist of the inflata marls, dipping gently to the north-west. They rise to the edge of the plateau about 300 feet above sea-level. These beds are cut off inland by two parallel faults that bring up intensely false-bedded limestone, of which the prevalent dip is at first to the south- east. After passing more of the marls, with fragments of “ Schloenbachia” inflata and of an ammonite which Mr Worpix tells me is a new species of “ Schloenbachia,” the beds resume their-normal dip to the north-west at 15°. Then follows a series of limestones containing casts of Mollusca; an anticline then brings up the cream- coloured limestone with concentric alge, which, according to Mrs Romangs, are identical with those at the Catumbella dam. Near where the river enters the gorge, at the level of 380 feet, we found fragments of “S.” elobiensis in beds dipping inland. _We climbed on to the plateau at the level of about 500 feet and crossed it to the foot of the next hill face. A little to the north-east of the exact line of section the algal bed occurs on the plateau at the foot of the scarp; this low-lying bed has probably been brought to the surface by the eastern limb of the synclinal, of which the other limb occurs where we had left the gorge. At about 650 feet the first plateau ends at the foot of a scarp, the upper edge of which is at the level of about 810 feet. The scarp consists mostly of hard shelly limestone dipping about 7° to the north-west. Above this limestone the country extends back in an irregular plateau which rises to another scarp. At the height of 850 feet, a short distance 502 PROFESSOR J. W. GREGORY. from the north-western edge of the second plateau, is an exposure of a marl contain- ing Salenia dombiensis, var. triangularis, “ S.” inflata, Epiaster catumbellensis, and some molluses which have been identified by Mr R. B. Newton as Newthea xquicostata, Lima ef. itieriana, Trigonarca diceras, Mrhilaia er. nerine formas. Mr Newton compares these fossils with Turonian, Cenomanian, and Albian species, the affinities being strongest with the Cenomanian. This conclusion is supported by the ammonites, which prove that the bed is on the wflata horizon and therefore Lower Cenomanian in age. The only satisfactory interpretation of the section to the south-east of Lobito is that the high level of the inflata beds here is due to faulting, and that the face of the hills between the second and coastal plateaus is a fault scarp. The smaller faults between the 800-foot scarp and the cliff at Old Lobito are probably due to the beds having been somewhat compressed between the two major step faults. (3) From Lolito to Hanha.—The coast cliffs from Lobito north-eastward to the mouth of the Hanha River give good sections of the Cretaceous beds. The rocks along the shore of Lobito Bay consist chiefly of the inflata marls, but exposures of the hard shelly limestone rise. from below them near the second and third gullies from Old Lobito. The first part of the coast section consists of a straight» scarp ; the plateau surface is even except where it is notched by valleys, most of which are hanging valleys, though a few have been cut down to sea-level. There are some overthrust faults with the downthrow on the south-western side. Further on the beds show repeated changes in dip and are broken through by groups of parallel faults. I landed about an hour’s sail west of the Hanha River, on a beach whence Lobito Lighthouse bears 226°. The beds there dip 25° seaward, and include a hard limestone and a bed containing numerous small fossils converted into ironstone: one of the ammonites Mr Worptg has provisionally identified as “S.” gracillima, Kosmat, from the lower Utatur beds (Cenomanian) of Southern India. A brachiopod has been determined by Mr R. B. Newton as Terebratula depressa. The first plateau is here about 150 feet high. Behind it is a second plateau rising from about 400 to 540 feet; it ends at the foot of a line of hills. Along the path leading south-westward to Lobito the algal limestone is exposed at the foot of this upper scarp. ‘To the east-north-east towards Hanha a road cutting has exposed beds containing many Hpiaster catumbellensis, which is associated with “S.” inflata. — A road section at the edge of the plateau overlooking the deep valley of the Hanha exposes the following in descending order :—The inflata beds with Akera gregoryi ; a chalky limestone containing chert in nodules and veins, shell casts, and Astrocenia ef. konincki, Kd. and H.; a massive white limestone containing a few shell casts: it is cut through by a fault with a downthrow of 3 feet to the west. Below this white limestone is a sandy limestone. ‘The section is there crossed by a crushed white lime- stone, to the east of which is a massive-jointed limestone dipping 4° westward; this is cut off to the east by a broad band of crushed rock due to a fault which hades west- CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 503 ward. On the eastern side of the fault the beds near the surface are covered by talus, beneath which are stratified yellow clays and marls with some layers of chert. From this point an extensive development of these clays is seen in the Hanha valley, and they probably belong to the same series as those of the lower part of the Catumbella section. | From this road cutting a valley, which was obviously at one time the main outlet of the Hanha River, descends to the north-east. On its flat swampy floor, at the height of about 130 feet above sea-level, is an outcrop of the algal limestone. This old valley joins the Hanha River (fig. 3) where it emerges from a picturesque canyon, the walls of which give excellent sections of the limestone series. The beds are gently folded, and a major anticlinal crosses the gorge near the first main bend below the Hanha basin; this upfold brings up a sandy limestone and some conglomerates containing boulders of gneiss, one of which was 9 by 5 by 5 inches in diameter. The size of these boulders indicate that the beds are near the base of the Cretaceous Series. The boulder bed is covered by a quartzose sandy 230! Ww clay wibh le Qyrsum Inflate Beas Sea Level Fie. 3: limestone containing blocks of a massive astrean coral. This coral has been so acted upon by water that all the septa have been removed, and they were not even recognisable in a slice cut from the middle of the mass. The coral with its narrow, long, crowded prismatic corallites, and their union by the walls, agrees with Astrocema ; and of the species described it would agree best with A. konincki, Ed. and H., from the Turonian of Gosau; but the corallites are smaller, the average in places being less than 2 mm, in diameter.* This coral-bearing limestone is succeeded by beds of sandy and sometimes false- bedded marls, one bed of which is 3 feet thick. Above these beds is the great mass of stratified limestone, dipping south-eastward. At the south-eastern end of the canyon the boulder beds reappear with large blocks of gneiss and bands of the algal limestone. I searched here for ammonites, but was unable to find any ; nor could I learn that any specimen of “S.” inflata had been found at this locality, though it occurs on the coast at the mouth of the Hanha River. The walls of the Hanha basin expose a thick bed of red clay over white and yellow clays. A hill of gneiss rises above the floor of the basin beside the manager’s house; the foliation strikes to 50°. This rock is covered at the height of 230 feet above the floor of the valley by gypsiferous marls and limestone containing bands of * Sections of limestone from the road cutting (p. 502) retain the internal structure ; the corallites are about 2-5 mm. in diameter, TRANS. ROY, SOC. EDIN., VOL. LI, PART III (NO. 18). 73 504 PROFESSOR J. W. GREGORY. selenite. I could find no fossils in this gypsiferous bed, but saw no evidence why it should not be included in the lower part of the Cretaceous Series. The correlation of the Cretaceous beds is considered on pp. 521, 522. B. Benguella to the Upper Catumbella River. At the city of Benguella the railway turns inland and goes south-south-eastward over a wide alluvial plain in the valley of the Cavoca River. The Cretaceous rocks are seen in the distance on both sides of this valley. At Bimbas Road Station two wells were sunk through the beds on the floor of the valley to the depth of 56 and 76 feet*; they passed only through sand. South of Bimbas Road a cliff of conglomerate can be seen to the east of the railway, which passes between hills of gravel and coarse conglomerate; the latter contains blocks of gneiss, granite, and schist. The boulders are as much as 4 by 33 by 2 feet in diameter, and they are well rounded and embedded in coarse sand. These beds were clearly deposited as torrential fans at the foot of the gneiss hills to the east. This boulder bed passes gradwally upward into sandstone. Mr Rosrns had seen no fossils in these rocks, and their age is uncertain. They may belong to the lowest part of the Cretaceous Series or may represent part of the Dombe Sandstone, which Cuorrar includes in the Cretaceous, but has been also assigned to the Lower Mesozoic. The conglomerate and sandstone end inland against hills of gneiss, which is well exposed in the Lengwe Gorge. Although the edge of the plateau has been cut away in the Lengwe valley, even there the slope is so steep that the railway ascends by rack work, which begins at 380 feet (at km. 51°8) ¢ and ends at 790 feet (km. 54). The gneiss is well seen on the banks of the gorge and in the cuttings along the railway. ‘The most characteristic rock is a coarse, well-foliated biotite-gneiss, which is mostly gray or dark bluish in colour. Pink bands also occur. Some of the rocks show an augen structure and wisps and clots of biotite.. The foliation is often con- torted. Some of the layers are hornblendic gneiss. Some varieties contain biotite in excess and pass into coarse biotite schists. The gneiss is traversed by quartz- felspar veins, some of which are coarse pegmatites with large biotites. In others there is no biotite, and but little felspar ; so they pass into quartz veins, one of the largest of which is beside the second viaduct. Some of the intrusive veins show well- developed graphic structure. The gneiss is often broken by faults, of which the most typical are small parallel thrust planes that are nearly horizontal. In one case, at km. 504, four such faults occur in a height of 12 feet; the upper side has been moved to the north-west. * One of them was at first a flowing well. Both at first gave good water, but after heavy rains all the water from the deeper well became too charged with magnesia to be of use ; and though this well was several times pumped dry, the water never recovered its original good quality. + The kilometre distances are those along the railway from Lobito, CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 505 The general character of this series of gneisses is remarkably similar to that of the biotite-gneisses in the Lewisian of North-Western Scotland. The strike is generally between 55° and 60° (about N.E. by E.). Above the Lengwe Gorge the railway continues to rise steeply past a series of cliffs of gypsiferous sandstone. This bed rests on an eroded surface of gneiss, as at km. 52°1, at the height of 413 feet (fig. 4). The sandstones are here irregularly bedded, and contain alternations of soft sandstone or sand-rock with layers of loam. Some of these bands are cut through by small slip faults. A band of conglomerate is interbedded in this sandstone. Some of the sandstone bands end off abruptly against vertical steps in the gneiss. One band contains some obscure impressions which may be organic in origin, but I found no determinable fossils. The best clue to the age of this rock is that some of the bands are full of wind-rounded grains. The material is very different lithologically from that of the Cretaceous rocks, and Fie. 4. - it was probably deposited under different climatic conditions. These subaerial sandstones are probably Mesozoic but of pre-Cretaceous age. The gneiss series is again exposed east of the station of San Pedro. The first notable railway cutting is at km. 55, and the rocks there are biotite-schist and a regularly banded gneiss. The foliation is vertical, and the rock is traversed by pegmatite veins and horizontal faults. The strike is to east-north-east. Some of this rock, by its even banding, low proportion of felspar, and arrangement of the biotite, resembles the Moine Gneiss of Scotland, though owing to its vertical foliation and conditions of weathering the characteristic flaggy structure of the Moine Gneiss is not developed. Three kilometres beyond this exposure and shortly beyond Monte Sahoa occur the first outcrops of the granites, which are the predominant rock from 58 km. to the bridge across the upper Catumbella River at km. 272. The granites form a belt of desert country which formed the great obstacle to communications between the coast and the better-watered highlands of the interior. The railway enters this desert on a course to the south-south-east. The level gradually rises from 790 feet at km. 54 to 2970 feet at km. 95. The chief features in the scenery are the numerous 506 PROFESSOR J. W. GREGORY. granite bosses and tors. The rock is exposed in many railway cuttings, and the soil is all a granitic wash. The most widespread variety is a massive biotite-granite. The granite in places is coarsely porphyritic. After passing across the second plateau (the Cretaceous being the first), the railway at km. 724 begins the ascent on to the third plateau. The level of 2972 feet is reached at the Portella Coroteva at km. 95. Occasional exposures of gneissoid rocks occur, as at the cuttings at km.77 and km..90. Portella Coroteva cuts through a ridge of granitoid biotite-gneiss, intersected by numerous pegmatite veins, and interbedded with some biotite-schist.. The numerous tors included in the magnificent view from the Portella Coroteva shows that the granite extends far on both sides of the railway and eastward over the basin of the Catengue to the hills around Cubal. . The railway descends this ridge, passing massive porphyritie granite at km. 1044 and also at km. 112. The granite is intruded by occasional altered basic dykes, as at km. 108 and Portella Solo. At km. 114 the granite is light gray and porphyritic and is associated with decomposed biotite-schist or biotite-gneiss traversed by many quartz veins. ; Large exposures of foliated white granite occur at Catengue, km. 121, and the brown soft decomposed biotite-gneiss reappears at km. 124; and it is here broken by bosses of white granite weathering in large exfoliation surfaces. From Catengue the railway rises to the Portella Solo, km. 1343, 2972 feet. At this pass the granite is in contact with a compact selvage of hornfels, which Mr Tyrreti has determined as a recrystallised basalt hornfels or augitite. One and a quarter km. to the east of the pass the granite becomes more basic; at km. 143 the rock beside the line is a white granite, which probably forms the picturesque tors to the north and the larger granite hills to the south. Near Bombola, at km. 157, is another occurrence of compact hornfels. Thence to the Cubal River (station at km. 197, 2976 feet) the line continues through granitic country between numerous rugged and picturesque tors. On the return journey Professor Martin followed the course of the Cubal for some miles to the north, and found that the ordinary granites extend along the river and are intruded by dykes of augitite or basic basalt similar to that of the Portella Solo. Before reaching Cubal the general condition of the country improves owing to increasing rainfall. After crossing the Cubal River there is a long level tract with no rocks exposed beside the line; and the improvement in the condition of the country is shown by the large baobab at km. 205}, being the last beside the line. The granite continues for km. 70 east of the Cubal, and it rises in high tors and bosses. The granite at km. 238 is hornblendic ; but most of the specimens examined consist of the ordinary gray biotite-granite, though some is more basic and there are occasional exposures of granodiorites, CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 507 C. Oendolongo Mountains. A change in the structure of the country at length appears at km. 275. The Oendolongo Mountains to the west have a stratified aspect and the railway is ballasted with ferruginous sandstone. From the crossing of the Catumbella River at km. 275 to Quingenge, at km. 297 (4460 feet), the line slowly rises up the valley of the Cuati, from the head of which it passes to that of the Cuiva River. The railway is near the junction of granitic country to the north, with the stratified and voleanic rocks which form the Oendolongo Hills to the south. Granite hills are conspicuous to the north, and numerous granite bosses occur close to the railway. ee An excursion from Quingenge to the Cuati showed that the granites outcrop immediately on the northern side of the railway near Cruz’s. A mile further north the white granite has been replaced by bluish-gray dioritic granite. A quarter of a mile further north the ground is strewn with blocks of rhyolitic tuff resting on granite, which continues as far north as the bank of the Cuiva River, where the granite is especially siliceous. Across the river the granite forms the Hala (or Uhilala range) to the north-west, and the Binda range to the north. South of the Cuati River, to the north of Quingenge Station, the rock is gneissoid and is seamed with epidote veins. Continuing eastward along the railway, the sandstones recur after crossing the Cambuacambula stream at kunt. 298$; but after km. 300 the railway is again on granite, of which there are numerous outcrops along this part of the Cuiva valley. Some of this granite is fine-grained, like that to the east of the Agricultural Station at Quingenge ; granite continues along the line to Cuma (km. 317, 4708 feet) and beyond it. At km. 3274 the railway crosses a felspathic sandstone, which forms a spur from the southern hills, and a synclinal in the hill face to the south of km. 323 is probably in rocks of the same series. The granite reappears at km. 328 from beneath red sandstones, which form the hills to the south and probably also the table- topped hills to the north. The railway crosses the Cuiva and thence follows its tributary, the Chicanda River, which is crossed at km. 344 near a granite tor. The railway shortly afterwards reaches the sedimentary rocks that form the mountainous country around Lepi (km. 367, 5371 feet), but granite outcrops continue to the three tors south of Lepi, which, being the most easterly in the district, form an important landmark. The Oendolongo Mountains (fig. 5), which rise to the south of the railway, form - the first break from the monotony of granite to the east of Mount Sahoa. The mountains extend from the Upper Catumbella River on the south-west to the Chicanda River on the south-east. They present a bold scarp face to the north-west, parallel to the railway line, and from their high north-western summit of Kanomba (7250 feet) the mountain face turns abruptly to the south-east. The western corner of this mountain mass consists of well-stratified quartzites, which are well exposed about Babaera, where they are exposed around the station of Senhor Ferreira and 508 PROFESSOR J. W. GREGORY. beside a spring in the valley of Bandionhime. The water of this spring has been bottled and sold in Benguella as a mineral water; but its value appears to depend on its remarkable purity. The spring flows from the quartzites, which there strike to 290° and dip from 15° to 20° northward. The quartzites extend from Ferreira’s to Babaera, where they are well exposed beside the waterfall. The rocks there dip 35° to the north of north-west. To the west of Senhor Piré’s farm the rocks dip 40° west and strike to nearly due south. The rocks are coarse-grained quartzite grits, containing many fragments of felspar ; and the rock has obviously been formed from the waste of the granitic rocks. These quartzites are considerably altered, for they contain authigenous muscovite (Pl. II, fig. 2), lines of authigenous epidote, and inclusions have been developed in the quartz along lines which are continuous across adjacent grains. These rocks are associated with banded cherts, which under the microscope prove to be silicified rhyolites. To the east of these quartzites, and & N Oendortongo Mts Quingenge B Rly Phyoe wte aebris cay Fie. 5. apparently overlying them, are volcanic rocks: which form most of the central part of the Oendolongo Hills. A sheet of columnar rock above the third stream west of Cruz’s can be recognised from the railway, but I had not time to reach it. A sheet of rhyolite tuff is exposed at intervals from just east of the Agricultural Station of Quingenge, westward along the foot of the hills; it rests there upon the granites, which outcrop over the plain to the north of the hills and can be seen in some of the gullies. The tuffs east of Quingenge Agricultural Station consist of fairly coarse fragments; but further west the rock is very fine-grained and compact, so that in hand specimens I was doubtful how much of it was tuff, for some of it might have been a felspar-porphyrite. Mr TyrrELu’s microscopic examination indicates that the whole sheet is a rhyolite tuff. The rock in some places, as close to the Agricultural Station, has been silicified and thus acquired a greasy or horny lustre similar to that of some keratophyre. To the east of Quingenge a mass of pinkish fine-grained granite rises from beneath the rhyolite tuffs. This rock can be followed for a couple of miles east of Quingenge, and rises to a considerable height on the hills. This fine-grained granite is succeeded eastward by the sandstones of the eastern part of the Oendolongo Mountains. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 509 South of the Agricultural Station of Quingenge the Cambuacambula flows for about 3 miles over the fine-grained granite, which is exposed owing to the denudation of the tuffs. As the floor of the valley rises more rapidly up the hills the rhyolite tuffs appear in blocks scattered over the granite. A mile further on the granites are covered by a black rhyolite. I could not find the actual contact, but saw the rhyolite within a foot or two of the junction; and the rock showed no trace of contact alteration and there were no tongues of granite entering it. Hence the rhyolites appear to be younger than the granite. The path then rises steeply, showing numerous exposures of rhyolite, some of which is vitreous and some has a glomero-porphyritic structure. On weathered surfaces the banded structure is often clearly displayed. ‘These rocks continue to the top of the pass at the height of about 6500 feet ; and on the top of the hills on the eastern side of the pass (6670 feet) it is clearly a banded rhyolite with conspicuous phenocrysts of quartz. The red sandstones also occur near the top of the pass, which commands a fine view southward over a large basin ; the outlet to the Catumbella River is through a gorge at the south-western end. A kopje, obviously of granite, occurs on the floor of the basin at 212° from the pass. The basin is bounded to the west by hills which appear to be the continuation of the quartzites of Babaera. The central part of the northern face of the Oendolongo Hills, therefore, consists of a.thick series of rhyolites resting on a rhyolitic tuff, which is itself underlain by the normal granites of the district. The rhyolite series is overlain by a thick series of red and white sandstones. The western part of the hills is composed of white altered quartzites, with some authigenous micas, associated with cherty banded rhyolites. The fine-grained pinkish granite east of Quingenge may perhaps be younger than the main series of granites, but I saw no evidence to connect it with the rhyolites. West of Cruz’s store an olivine-basalt dyke cuts through the granite and probably continues into the tuffs. D. From Lepi to Huambo. Near Lepi (km. 367, 5371 feet) the railway enters a new type of country different from any seen between that station and the coast. Between Lepi and the granite hill to the west at Old Lepi (km. 355, 4783 feet) is an area of deep chocolate-red soil which is due to a sheet of olivine-dolerite that is exposed in the railway cuttings. This rock is doubtless intrusive, since a coarse quartzose sandstone beside it is extremely decomposed. The sandstone belongs to a series of old sedimentary rocks which extend eastward to the Kunhungamua valley; they may be grouped and named after their most characteristic rock, the Lepi graywackes. The railway on leaving Lepi climbs by a very sinuous route up the Chicanda valley on to the summit of one of the highest of the Benguella plateaus. The rocks are well exposed in the numerous railway sections, and they include graywackes, quartzite, cherts and slates, 510 PROFESSOR J. W. GREGORY. and some interbedded rhyolite tuffs which are blue-hearted when fresh but turn gray when weathered. The volcanic rocks are less abundant than in the Oendolongo Mountains, and I saw only tuffs and none of the rhyolites themselves. The most typical rocks of this series are a black, fine-grained cherty graywacke, which in the field I recorded ‘as chert. One variety (No. 130), from a spur east of Lepi and about a mile west of the locality known as Homer’s Camp, appears in a hand specimen to be a banded cherty quartzite ; under the microscope it proves to be a fine-grained graywacke ; the larger grains of quartz range from about a tenth to a quarter ofa millimetre in diameter ; the grains are angular, and include a little felspar, some of which is oligoclase, and white mica. The rock shows an imperfect cleavage, the particles having been to some extent rearranged by pressure. A graywacke (No. 131) on the spur to the east of the last specimen is also fine-grained, but the rock is more massive ; it is not cleaved, and the banding is faint ; to the west of the cherty graywacke (No. 130) the rock may be described as an indurated siliceous mudstone. The series includes some slates. The prevalent dip, in spite of some folds, is to the east, so that the rocks nearer Homer's Camp and in the steep ascent to Calenga are higher in the series than the eraywackes nearer Lepi. ‘These higher rocks are mainly quartzites; im fresh specimens the fracture passes through the grains as readily as around them, and specimens in which this does not happen (7.e. the sandstone beside the olivine-dolerite west of Lepi) are probably decomposed quartzites. The graywackes near Lepi are only imperfectly cleaved; but further east the rocks are more altered, and some show an incipient foliation. This feature is well. marked in a greenish quartzite (No. 135) from near Homer’s Camp. The rock was identified in the field as an epidotic sandstone ; under the microscope it proves to be an altered graywacke ; the original quartz grains show undulose extinction, and the borders have been occasionally altered to a granular mosaic; the base has been recrystallised as secondary quartz, abundant granular epidote, and prisms of clinozoisite. The epidote has been formed along irregular lines, and has thus given rise to an incipient foliation. The Lepi Series is intersected by intrusions of dolerite. Some of the dykes and the slates which they intersect are intensely decomposed, and at about km. 379% they have been altered to a china clay, probably by pneumatolytic action. Close beside the china clay are exposures of decomposed pink clay. Two kilometres further are exposures of blue cherty graywacke and variegated slates, which have been bent into occasional shallow folds. This varied series forms the great dissected plateau which includes the mountain of Holocasa (7848 feet) to the north of the railway, and the range known as the Serra Jorvo Andrade, which extends from the railway south-south-westward across Benguella to the province of Mossamedes. The sedimentary structure of this country is well seen in many views near and east of Calenga. Thus the footpath CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. Bil from Lepi to Calenga, after leaving Homer’s Camp, ascends over weathered gray- wacke and reaches the plateau near a 60-foot kopje of subquartzitic grit, which weathers gray and consists of grains of about one-eighth inch in diameter. Further along the line at km. 387 the white weathered banded cherts are covered by a yellow soil. One and a half kilometres further east a view down the upper valley of the Calai, the head stream of the Cunene, is bounded to the east by a long regular escarpment of the Calenga quartzite, which dips 10° eastward. East of Calenga the granitic rocks soon reappear from beneath the Lepi graywackes, and picturesque tors are seen in the distance; but the sedimentary series continues along the high ground to km. 395. Blue-hearted cherty graywackes are exposed beside the railway at km. 391, and at km. 3924 is a dyke of coarse dark-blue epidiorite. Just east of the locality known as Varian’s Camp, at km. 3938, isa deep gully, which exposes the blue cherty graywackes. They are somewhat coarser in grain than those near Lepi, and hand specimens that show large fragments of felspar resemble porphyritic felsites; the rock contains pyrites, secondary quartz, and granules of epidote, but it lacks the incipient foliation of the epidotic quartzite (No. 135). The railway cuttings on the eastern side of this ravine, at about km. 3943, are of sandstone interbedded with slate and graywacke and dipping 15° eastward. From this point there is a long tract of level ground with no sections. Bold tors to the north and south indicate that the lower ground on both sides of the ridge followed by the railway consists of granite. The ridge itself is probably capped by a thin layer of the graywackes, as just before reaching the Kunhungamua bridge the ground is littered with felspathic graywacke, but I could see none i svtu. Immediately to the east of the railway bridge is a hill of porphyritic granite. The rock has a compact, fine-grained selvage which, as Mr Tyrrevi has shown, contains eassiterite. After leaving these granite exposures we saw no rock i svtu before reaching “ Huambo Crossing,” 5 km. further east. The railway cuttings are in buff rock-sand, which might be formed either from decomposed granite or from dis- integrated arkose. From Huambo Crossing, which is on granite, we could see granite tors both north beyond Huambo Fort and southward, bearing 147°, in the valley of the Cunene. E. Huambo to Ochilesa. From Mount Sahoa, near the head of the Lengwe Gorge, to Huambo, and, as we found later, extending 25 miles further east to Candumbo, is a long belt of granite, which is only hidden from view where it is covered by the sandstones and rhyolites of the Oendolongo Mountains and by the graywackes and quartzites of the Serra Corvo Andrade between Lepi and Kunhungamua. These sedimentary rocks are mainly composed of the waste of the underlying granites, so they give rise to soils of similar quality which are of only second-rate value. The uniformity of the conditions was probably due to our route having followed along the grain of the TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 13). 74 512 PROFESSOR J. W. GREGORY. country and not, as I had expected, across it. So at Huambo Crossing we left the route of the railway, which was there still under construction, hopmg by a traverse northward to reach a more varied series of rocks and better soils. The Commander of the Huambo Fort kindly engaged for us extra porters in order to visit Bailundo and the head streams of the Cutato River, a tributary of the Cuanza.* A mile and a half 18° east of north from Huambo Crossing is an exposure of a coarse-grained white quartzite, and beside it lay fragments of decomposed slate and quartzite. A little over a mile further a stream gully exposes a coarse white granite especially rich in quartz. Thence to Huambo Fort the plateau is littered with fragments of sandstone, while the gullies reach quartzose granite. Near the fort the granite is exposed on the surface of the plateau. The fort stands on a rise of porphyritic granite, which forms part of the watershed between the rivers discharg- ing south-westward through the Cunene and Mossamedes, and those which go north- pees Schistose le Quartzite N : 2 Dolemite 9 6000 oe - > > g oN 4S 8 . 1S & % < Sup id RS es Cars 3 ‘ S =] res ~ 5, ° AQ 3% es a A) ° Gin a Somer 2 a ' >») v ic 5000 Fig. 6. north-westward through the Queve and Culele to the Cuivo and thus reach the Atlantic north of Novo Redondo. The track to Bailundo starts from Huambo on an approximately north-north- easterly course, across the head streams of the Queve and Culele. The country consists of granite for the first 8 miles north of the fort. The oranite rises in tors and monoliths, one of which is known as Cleopatra’s Needle. The granite is generally biotitic, and sometimes porphyritic, with large blebs of biotite, and on the margins of some of the granite bosses are bands of a tough granite porphyry. At the Landula t River this last rock occurs in veins in the normal granite. This river exposes a green, massive quartzite, which is so compact that the grains are barely recognisable except on weathered surfaces. It belongs to the group referred to as the Huambo Quartzites. Approaching the Queve River, the granite is succeeded by decomposed quartzite, and at that river are exposures of a quartz-schist with large clastic grains. The rock is holocrystalline, and the original cement has recrystal- lised in lines of fine-grained quartz-mosaic. They represent a fine-grained, purely siliceous variety of the Bailundo Schistose Quartzites. They ditfer from the Huambo Quartzites by their well-developed foliation. * There is another Cutato in the district, which is a tributary to the Zambesi. + Thus spelt by A. pe ANDRADE; spelt Landulo by Vartan, and also as Londulo. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 5138 North of the Queve River, between it and the track, are hills of schistose quartzite, which is intruded by a boss of diorite-porphyry in which the base has been altered to a granular, hornfelsed mosaic. The edge of this rock is a blue hornfelsed quartz-porphyry. From this point the route continues over a quartzite gravel. On the northern bank of the Lumbanda or Lumbwambwa River are the Ambussa Hills, which trend to a little south of east; they consist of schistose quartzite with a well- developed flaser structure and numerous quartz veins. Some bands are schistose- conglomerates and perhaps crush-conglomerates. The rock strikes to 87°. Near the western end of this ridge are the ruins of a prehistoric stone fort, whence the view southward across the plain includes the hill of Bavi or Kubave, the northern face of which has a fine cirque-like hollow. Just north of the western end of the Ambussa Hills is an exposure of granite, unusually rich in biotite. This rock disappears beneath the schistose quartzites, which extend thence to Bailundo; they are well exposed about the Culele River, where the rock is coarse and saccharoidal (No. 151); and as similar rocks are exposed where we crossed the line of the Lumbanganda Mountains, they are probably also composed of the schistose and saccharoidal quartzites. The hill beside the fort at Bailundo consists of coarse pink gneiss (No. 152) of which the strike is to a little north of east. The rock, no doubt, belongs to the same series of gneisses as those of the Lengwe Gorge; but it represents the pinkish granitoid gneisses of the Laxford type of the Lewisian Series, and not the common biotite-gneiss. From Bailundo we turned north-eastward towards Ochilesa and soon left the Bailundo eneiss. It is succeeded by a granulitic quartzite that weathers into sand- stone, and is often covered by a litter of ironstone conglomerate. The route passed to the west of the Ueia Mountains, some spurs of which have the rough craggy weathering of the schistose quartzites. About 3 miles from Bailundo is a lateritic ironstone conglomerate, including blocks of granulitic quartzite. The quartzite beneath this conglomerate contains angular fragments of felspar, and resembles some of the non-schistose Dalradian Quartzites of the Scottish Highlands. From 4 to 5 miles north-east of Bailundo the track passes over an area of a rich, red chocolate soil, which continues in patches for about 3 miles. They are succeeded by the ordinary yellow loamy soil. Where the rock that gives rise to this chocolate soil could be found it always proved to be igneous, and generally an olivine-dolerite. Coarse biotite-gneiss reappears about 10 miles north-east of Bailundo, near one of the last streams which pass through the Culele to the Cuivo. This gneiss is of the Lewisian type; it contains bands 1 inch thick, which are pink owing to the abundant felspar, and 4-inch veins of pegmatite. Fourteen miles from Bailundo the higher ground is covered by a gray, light, loamy soil due to the decomposition of the schistose quartzites ; but on the floor of the gullies and on the lower land to the south and east of our camp were bosses of 514 PROFESSOR J. W. GREGORY. gneiss. The strike of the foliation is to 68°: and this rock is interstratified with quartzites, quartz-schists, and silliimanite-schist, in all of which the foliation is nearly vertical. The sillimanite is abundant in some layers, and the crystals are large enough to be seen by the naked eye. Lower down this gully the gneiss is invaded by a hornfelsed tonalite-porphyry (No. 159). The exposures at this locality show, as I had suspected from their general character, that the schistose quartzites from the Ambussa Hills northward are a much older series than the sandstones of the Oendolongo Mountains and near Kalenga; the quartzites of the Bailundo district are quartz-schists and holocrystalline quartzites, and are interstratified in the lower Archean foundation. The route next day lay along the ridge that forms the watershed between the Cuivo and the Lovule—the first tributary to the Cuanza. The chief rock seen is a saccharoidal quartzite, which is often capped by an efflorescent ironstone that is . sometimes botryoidal. Some bands of a red chocolate soil indicate the deposition of some basic rock. In the valley of the Lovule, 8 miles from our previous camp, are bosses of dark bluish-gray biotite-gneiss, which encloses large blebs of quartz an inch square, and is traversed by contorted quartz veins. Nearly a mile further on are exposures of gneiss and felspathic quartzite ; further along the route are frequent exposures of a dark-grey biotite-gneiss; and the first main branch of the Lovule River, a little west of the station of Catotlo, flows over sub-granulitic biotite-gneiss. Around this station are further exposures of crystalline quartzites. Between Senhor Leite’s two stations of Catotlo and Chieuca is the upper part of the Lovule basin ; the ground consists of series of flat-topped ridges between the valleys of the successive tributaries. Exposures on the first part of the route are scanty, but the country probably consists of schistose quartzite. At the bridge, about 12 miles north-east of Catotlo, are outcrops of dark-blue gneiss. The higher ground to the east of this stream is littered with quartzite gravel, and about 4 miles before reaching Chieuca is an exposure of the schistose quartz- ites; the foliation is nearly vertical, but with a shght inclination to the north, and strikes from 50° to 56°. The station of Senhor Rodriguez Leite at Chieuca is on a high ridge of volcanic rocks, which weather into a very rich red soil. The summit of the ridge to the north of the station is composed of volcanic tuff and agglomerate. The general aspect of these rocks and their freshness indicate that they are much younger than most of the rocks of the Benguella plateau. Many of them show no signs of pressure, and some of the lavas are still glassy. Mr Tyrre x has identified these rocks as shonkinite, sodalite-syenite, solysberai™ tuff, and vitrophyre. Chieuca is a volcanic neck composed of alkaline lavas. From Chieuca the ground descends steeply to a ford over the Cutato River, the gravels of which consist of felspathic quartzite and gneiss. The rocks were not present in situ, but the boulders on the stream bed are so angular that they cannot have travelled CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. BLS far. Six miles from the Cutato the track crosses the Canji River, beside which are -- outcrops of schistose quartzites and a coarse biotite-gneiss striking almost due east- ward. Some of the gneiss about this locality contains but little quartz. The banks of a rushing stream, about 4 miles north-north-east of the Canji, consist of a white granitoid gneiss in which the foliation is indefinite, but long inclusions show that the direction is to 83°. About 2 miles further on are exposures of a gray gneiss; this rock continues “to the Okanusi Pass, by which the route leads down to the American Mission Station at Ochilesa. Half an-hour before reaching this village I noticed the first exposure of a light-coloured pinkish gneiss very rich in felspar. Mr Tyrre.u’s determination of the ferro-magnesian constituents in this rock as hypersthene shows that the rock 4 is a charnockite. , 3 ; F. Ochilesa to Saccanjimba. q Ochilesa is a locality of especial geological interest, owing to its sodalite-syenites and alkaline dykes ; but we had only time to spend a day and a half there, when we enjoyed the kind hospitality of Dr Woopstpr and Mr Niepp. Along the banks of the stream, the Quime, are a series of warm alkaline springs of which the maximum temperature is reported as 114° F. To the south of Ochilesa is a flat-floored basin, the Changa, which is about 8 miles long and 4 miles wide and surrounded by an elliptical ring of hills. The proximity of this oval depression to the hot springs had led to the view that the basin is a volcanic crater, and the neighbourhood of Ochilesa has been described as the volcanic district of Bihé. I was unable, however, to find here any evidence of recent volcanic action, though shonkinite and orrachitite dykes occur on the banks of the Quime. The Changa basin is a cauldron due to subsidence, and the hot springs arise along the northern continuation of the fault that passes along the western side of this basin. The alkaline nature of the water is doubtless explained by the richness in alkalies of the adjacent rocks. As the Quime is fed by these springs, it deposits along its course a series of beautiful tufa terraces built up by the constructive water- falls which connect the successive basins. The following analysis of the tufa, by Mr W. C. Scorr of Sandusky, Ohio, was shown me by Mr Niepp :— Silica. : : : 2.0 sper cen, Tron and alunaina . : : : A et B38) F Carbonate of lime. ; : é . 89°64 - Magnesia. : : : me a 97°96 A These springs are similar to the group of Katanga hot springs described by Marutev (1913, p. 124), in which the water has a temperature of from 70° to 100° F., and deposits calcareous tufa. The thermal springs in Katanga discharge at the 516 PROFESSOR J. W. GREGORY. contact of the older quartzites with granitic rocks, and those at Ochilesa are probably due to water rising along a deep fault. The warmth of the water has aided the growth along the upper part of the Quime of the most tropical vegetation we saw on the plateau, including palms 60 feet high, a dense undergrowth of ferns, and numerous lianas. To the south of the Changa are some iron lodes which are the most extensive ore deposits yet discovered on the Benguella plateau. I was unable to visit the main ironstone lode, but examined some of the smaller ore masses to the west of it, at. some old native workings at Quimbundu. The lodes here strike on an average from 150° to 155°, and they are bounded by decomposed felspathic gneiss or schist con- taining but little quartz. The ore occurs in lenticles, of which one that 1 measured is 13 yards wide by 30 yards long, and its strike is to 28°. The country rock is a grayish schistose felspar porphyry, and the strike of its foliation, where I observed it, varied between 150° and 165°. The country is cut through by north and south faults by which the rocks have been greatly crushed. The leaching of the country rock along these crushed bands has led to the segregation of the iron into lodes, lenticular ore bodies, thin veins, and small scattered nodules of ore. On the bank of the Quime, near the mission station at Ochilesa, a light-Gray gneiss is interbedded with a blue non-foliated rock, which in the field I recorded as diorite. Mr TyRRELL’S miscroscopic examination shows it to be a dark-coloured hornblende- hyperite and to be a member of the Charnockite Series. This rock includes many inclusions of hornfels, which are no doubt altered fragments of sedimentary rocks ; so that the Charnockite Series is apparently intrusive into the gneiss and schistose quartzites. Dykes of shonkinite occur close by this exposure, but the vegetation there concealed their relations. From Ochilesa we turned southward through the uppermost part of the valley of the Cutato (Cuanza) to the watershed between the northern drainage and that to the Zambesi. We began this march across the basin of the Changa and over the pass to the west of the Iron Mountain. About 3 miles to the south of the pass, beside a village whence Iron Mountain bears 10°, are good exposures of a white felspathic schist interbedded with bands of a schistose quartzite composed wholly of interlocking quartz. Beside this village were some deserted native iron workings. About a mile and a half to the south-south-east, on the banks of the Canji River, is a light-coloured gneiss injected with granite veins; and these intrusions may help _ to explain the unusual strike of the rocks, which is to 153°. After crossing this river we passed some large slag heaps of old iron workings and reached the Bui Aiver, beside which are extensive exposures of a gray gneiss containing biotite in patches one inch long by half an inch thick. The strike here is 47°, with a steep dip to the north-west. South of the Bui River is a plateau littered with schistose quartzites, including =e CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. Dili fine-grained, coarse-grained, and compact glassy varieties. Some of the more massive varieties are similar to the Huambo Quartzite. On the edge of the plateau south of the Chimboia River are large blocks of ironstone conglomerates containing quartzite, and the soils indicate that the quartzites continue for 3 miles south of the river, where the gneiss reappears with a coarsely speckled variety. Quartzite occurs again on the high ground between this locality and the next stream, which is about 24 miles to the south, where half a mile to the east of the track is a boss of gneiss. Hrom this point the gneiss extends for a great distance to the south. It is seen in all the stream beds as well as occasional large exposures on the intervening ridges. On the Warwar River the gneiss is a dark-blue variety, rich in biotite ; it strikes to 78°, and is penetrated by coarse veins of biotite-granite. At the large village of Hmbala Andulo the foliation of the gneiss is very irregular, different measurements in the same exposure giving the strike as 8°, 18°, 43°, and 48°. The variations are due to disturbances occasioned by numerous faults and granitic veins. Most of these veins and also some faulted quartz veins trend to the north-west, while the faults trend to the north-east. The eneiss here is especially quartzose, and con- tains pink felspars and many pegmatite veins. From Embala Andulo the track descends steeply to the Unyamba River, on the south side of which is a coarse-grained gneissoid granite with a very irregular foliation ranging from 48° to 110°. This gneissoid granite is invaded by an amphibolite dyke, which Mr Tyrretu has determined as a hornblende-hyperite belonging to the Charnockite Series. About 3 miles south of the Unyamba and half way to the village of Etunda the gneiss is of a dark-blue dioritic type. Mr Tyrrenu has identified it as a rock belonging to a granulitic intermediate division of the charnockites. The more basic character of the rocks in this neighbourhood was indicated by the red soils upon the plateau. About a mile and a half south of the charnockite is a red sandstone, and quartzite fragments become conspicuous in the soil. Quartzite is seen im sitw just north of Ktunda and for at least 3 miles to the south of that village. Then follows a long interval with no exposures. To the north of Saccanjimba the track descends into the valley of one of the uppermost tributaries of the Cutato, along which are extensive exposures of gneiss of which the strike is at first to 52°. A mile further south, on the floor of the valley to the north of the mission station at Saccanjimba, the rock is a hornblendic or diorite gneiss with irregular, obscure foliation, its direction being in places to 112°. The proximity of intrusive rocks is indicated by the existence of a large pegmatite dyke beside the ford. This dioritic gneiss is succeeded higher up the slope leading to the abandoned buildings of the Saccanjimba mission by white saccharoidal quartzite and a pink quartzite which weathers into a white sandstone. Its strike is to 58° and it dips 70° south, but this may be exaggerated by slip down the hillside. 518 PROFESSOR J. W. GREGORY. G. The Bulu-Vulu to Huambo. South of the old mission-house at Saccanjimba is a plateau of about 2 square miles, covered by black soil, but I was unable to find any exposure of the rock which gave rise to.it. On the slope southward to the river the first rock observed was a gneiss with a strike of 102°, and up the slope on the southern side the strike is 122°, At the next stream, half an hour to the north of Gordon’s Store, the rocks were quartz-schist and saccharoidal quartzite, striking 127° and dipping south-west. (1) The Bulu-Vulu.—Gordon’s Store is on the track from Katanga to Benguella, and on the north-western edge of a great tract of level, grass-covered plains known as the Bulu-Vulu. They extend eastward beyond Belmont, the chief settlement in Bihé, which was established by the famous Portuguese trader Silva Porto. I made an excursion off our main route to examine the nature of this Bulu-Vulu. It consists of black-soil plains. The soil is composed of quartz grains blackened by organic matter. Gullies on the side of this plateau show that the sand is about S.w. Bulu Wulu NE, Cambenge R. Umbaz “er INCL Hf 4 feet in thickness, and beneath it is a layer of yellow clay containing loose fragments of a white quartzite. The yellow clay serves as an excellent paint, and is probably due to the leaching of iron from the quartzite. Beneath this clay is often a gravel, which is composed of quartzite and ironstone, and, on the north-western edge of the Bulu-Vulu, rests on gneiss. The Bulu-Vulu may result from the decomposition of a widespread sheet of quartzite ; but it is probably due toa sheet of younger sandstone— the Bihé sandstone, which is the local representative of the Lubilash beds. Coal is said to occur under the Bulu-Vulu; I could learn, however, of no adequate grounds for this report, which may be based only on the blackness of the soil and on occasional occurrences of a peaty material; but the existence of coal would strengthen the reference of the Bulu-Vulu sands to the Lubilash beds. The relations of the Bulu-Vulu and of the adjacent rocks are shown in the section (fig. 7). (2) Cambenge and the Umbal.—Descending from the Bulu-Vulu we passed first a quartzite and ironstone gravel, which probably represents the decomposed outcrop of the Bulu-Vulu sandstones. This layer rests upon gneiss. The quartzites occur beside the stream, which is one of the upper tributaries of the Umbal. On the plateau near Cambenge Station is a handsome pink, granulitic, quartz-felspar gneiss ; it is succeeded to the north of Correo’s Stores at Cambenge by a blue diorite-gneiss i ig, i, i, eh i CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 919 which strikes to 132°. Journeying thence north-westward the track crosses a hill of quartzite and descends to the deep valley of the Umbal, the banks of which consist of massive gneiss with obscure foliation. This rock is bounded on the western bank of the river by a hornfels-like rock which Mr Tyrretu has determined as a granulitic amphibolite. The strike of the gneiss is here almost due north and south, and on: the margin of the gneissoid rock are many basic segregations or inclusions. The gneissoid rock is then probably intrusive into the amphibolites; but I failed to find any indications of contact alteration in the sandstones, which overlie the crystalline rocks on the western slope of the valley. It appears probable that this sandstone is a younger rock resting on the gneiss and schistose quartzites. (3) The Umbal to Huambo.—From our camp on the Umbal we marched in a _ westerly direction to the railway, which had now been extended as far as Huambo. The route for the first 6 miles was westward over quartzites and a plateau which had part of the character of the Bulu-Vulu, though the grassy plains were narrow and were bordered by woods. After marching 6 miles the track turned south, to keep along the high ground around the head of the Cutato valley, and after about 4 miles in this direction we came on exposures of cherty, coarse, white quartzites. After another 2 miles we saw some bosses, obviously of gneiss, down the valley to the east-north-east; but the route still lay over quartzites, and we continued: on them for about 6 miles, after reaching a waggon road which had been opened from Belmont to the railway head. Blocks of gneiss appeared from beneath the quartzites, and about a mile further on, in a shallow valley to the north of the road, is a water-hole due to a bar of hornfelsed felspar porphyry with abundant scapolite (No. 199). I had not time to visit the Chingwari Hills, which may consist of either quartzite or of hornfelsed porphyry, which occurs to the north as well as to the east of them. At the northern foot of the Chingwari Hills is an extensive plain known as the Little Bulu-Vulu, which appears to be due to a sheet of gray marls and sandstones, of which exposures occur due north of the summit of Chingwari. We continued over this Little Bulu-Vulu for about 6 miles, an occasional valley showing exposures of white quartzite and pebbles of sandstone. To the west of the Bulu-Vulu plain are numerous exposures of quartzite containing grains from one-eighth to one-sixteenth of an inch in diameter. At the head, however, of the Cubango, one of the north- western tributaries of the Zambesi, we again entered an area of gneiss. Most of it is the normal gray biotite-gneiss, but it includes some dark gneiss with obscure foliation, and some intrusions of a felspar porphyry. Travelling along the watershed to the west of the head streams of the Cubango we passed again for about 3 miles over some coarse white quartzite, and then a large boss of a fine-grained gray biotite-gneiss with a foliation of 48°. Close beside our camp, in a picturesque group of gneiss crags, we crossed a bar of schistose quartzite, which strikes to 43°; and on the higher ground above it was a considerable exposure of a coarse granular. sedimentary TRANS, ROY. SOC. EDIN., VOL. LI, PART III (NO. 18). 75 520 PROFESSOR J. W. GREGORY. quartzite (No. 227). The quartzites in this district appear to belong to two different — categories. The older quartzites which have a high dip and are interstratified with the gneisses belong to the Bailundo Schistose Quartzites, while the higher ground along the divide is formed of the Huambo Quartzites. After passing the head of the Cubango the route lay westward over biotite-gneiss with irregular foliation ; and the appearance of a bold tor at Candumbo, about 11 miles to the west, warned us of our approach to the granite belt. About 7 miles before reaching Candumbo, at the head of a deep tributary to the Queve, we passed a band of cordierite-hornfels with its foliation striking to the north. The country between Candumbo and Huambo is composed mainly of granite; there are numerous exposures beside the track and in high tors to the south and north. One of. them we called Chichester Cathedral, from the resemblance of an isolated monolith beside it to a detached belfry ; this mass consists of a coarse-grained biotite-granite. Heaméo . Colymehala Vailey Fic. 8. Near Huambo the granite extends on both sides of the track, which crosses a long strip of a white friable sandstone composed of quartz grains in a felspathic base; and the relations of these rocks beside the Culimahala River show that the granites were once covered by a widespread sheet of arkose due to their decom- position (fig. 8). The streams that flow northward have deep rocky gorges, while those to the south lie on the floor of wide shallow valleys; this contrast shows that the northern streams, which discharge to the Atlantic, are extending inland and thus encroaching on the Zambesi basin. Ill. Tae CorRELATION OF THE BENGUELLA SERIES. The rocks seen in Benguella may be divided up into five groups : (1) Pleistocene deposits, including drifts, alluvial deposits, sand dunes, lake beds, the sands on the Bulu-Vulu, the calcareous tufas of Ochilesa, laterites, etc. (2) The Mesozoic rocks of the coast zone. (3) A series of old sedimentary rocks widely spread on the surface of the plateau. (4) Gneisses, schists, and granitic rocks, which form the foundation of the whole country. (5) Various lavas, and intrusive rocks of undetermined age. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 521 A. The Mesozoic Rocks. Their chief representatives are the Cretaceous beds, of which Cuorrat’s classifica- tion may be summarised as follows :— Senonian Sandstone with Cardita, Rondatreia, ete. Limestone with Inoceramus langi. Turonian White coral limestone with Nerinza capellor. Cenomanian | Marly arenaceous limestone with Epiaster and Cyprina wens. Vraconnian Marly limestone with ‘“‘Schloenbachia” inflata, Epraster, ete. Albian Marls with “ Acanthoceras” * mammuillare, Salenia dombeensis, Alo, ete. The Cretaceous beds seen near Lobito appear to represent only the Albian and the Vraconnian (Lower Cenomanian). the fossils. For instance, Salenia dombeensis, which Cuorrat records at Dombe Grande in the Albian mammillare marls, I found only in almost the highest fossil- iferous bed above sea-level; it was there associated with “S.” inflata and Epiaster catumbellensis, which according to CHorrat characterise a higher horizon. Such extensions of the range of the fossils is naturally to be expected with further collect- ing. The correlation of the beds near Lobito from Hanha to Catumbella appears to be as follows :— A few anomalies appear in the distribution of Age. Catumbella Gorge. S.E. of Lobito. Hanha. Inflatu mars Inflata marls with Salenia | Inflata beds Cherty, shelly limestones dombeensis, var. triangu- | Epiaster beds Hematite nodules and laris,and Epiastercatum- | Chalky limestone with Neithea angolensis bellensis chert nodules Vraconnian Limestones with Panopea | Limestones with gastropod | Massive shelly limestone plicata and Pholodomya casts, Trachycardium | Sandy marls vugnest syriacum, Trigonia, | Limestone with Astrocenia Epiaster mar|s crinoid ossicles, anne- | lids, ete. Algal limestone Algal limestone | Algal limestone and sandy Clays, and marls with bands | limestone Albian of torrential gravel | Clays with torrential wash Gypsiferous marl resting on gneiss The general relations of the fauna, as remarked by Cuorrat (1895, p. 89), are with the Cenomanian of the Mediterranean basin, though some of the fossils occur also in Southern India and Zululand. The ammonites “8.” elobiensis and “S.” inflata range northward along the West * According to current nomenclature Douvilleiceras. 522 PROFESSOR J. W. GREGORY. African coast, as in the French Congo near the mouth of the Gaboon. The Angola Cretaceous beds were probably deposited in an arm of the Tethys, which ran down the western coast of Africa. The evidence for a connection with India around South Africa appears as yet inadequate. ; A lower Mesozoic horizon may be represented by the gypsiferous sandstones with wind-rounded grains from the head of the Lengwe Gorge. The sandstone of Dombe has been repeatedly referred to the Trias (see, e.g., CHOFFAT, 1888, p. 41).* CHOFFAT (1888, p. 45, and 1905, p. 56) suggests that it is the same age as the coal-bearing sandstone of Dondo and of the Dande valley. These three localities are in the north-western corner of Angola, and Cuorrat (1888, p. 45) considers that this horizon is also represented further south at Novo Redondo and at Dombe Grande. The sandstone at the head of the Lengwe Gorge probably belongs to the same group, as its lithological characters indicate that it was laid down under different climatic conditions from those of the Cretaceous, and it is not improbably of Lower Mesozoic age. Some of the deposits at the base of the Cretaceous Series, such as the gyp- siferous beds resting on the gneiss at Hanha or the sandstones or conglomerates at the foot of the Lengwe Gorge, are perhaps also pre-Cretaceous. B. The Older Sedimentary Rocks on the Plateau. Upon the ancient gneisses which form the foundation of the Benguella plateau rest disconnected areas of sedimentary rocks, which have hitherto yielded no fossils. Their correlation may therefore appear impracticable. The wide distribution of old unfossiliferous sediments is one of the most striking features in the geology of South and Central Africa, and fortunately the geologists who have worked in ~ those regions have had the courage to attempt the correlation of these deposits by their lithological characteristics and the extent to which they have undergone dis- location. These rocks clearly belong to various ages, and it is recognised that they range from the pre-Cambrian to the Carboniferous. The older sedimentary rocks which I saw on the Benguella plateau (excluding those near Lengwe) may be divided into five groups— (1) The Bihé Sandstone. (2) The Oendolongo Series. (3) The Lepi Graywackes. (4) The Huambo Quartzites. (5) The Bailundo Schistose Quartzites. In the absence of paleontological evidence the correlation of these rocks depends on their resemblances in structure and succession to those in South and Central Africa. I[t is natural to compare them first with the long and varied succession of * Mr A. Hormus (1915, p. 231) also refers to the Dondo coal-bearing shales and grits of North-Western Angola as of Lower Karroo age. Corner (1894, p. 195) says it is pre-Jurassic. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 523 older sediments in the Belgian Congo, which have been described in a considerable literature, and especially in the memoirs by Cornur.* Recently F. E. Srupr (1914) in a valuable memoir has traced the general distribution of the older sedimentary rocks in southern Africa, has shown that the same general succession occurs through- out, and has correlated the beds of the Congo and Northern Rhodesia with the geological systems of the Transvaal and the Cape. ; The correlation of sedimentary rocks by their lithological characters is necessarily somewhat uncertain, except in the case of beds with exceptional characteristics ; and there are still important differences of opinion as to the relations of the unfossiliferous pre-Karroo deposits of South Africa. These differences will, how- ever, probably disappear with fuller knowledge, and the points of agreement are sufficient to justify the attempt to determine the position of the various Benguella rocks in the general South African succession. Cornet has based a recent classification of the Congo beds (1912, No. 1, p. 9) on the extent to which the rocks have been dislocated. The rocks of his oldest division consist of the Archean and Primary rocks, all of which are more or less metamorphosed and all greatly dislocated. Those of the second division, including his Kundelungu System, are but slightly dislocated. His uppermost division, including, in ascending order, his Lualaba, Lubilash, and Bussira Systems, are not dislocated. Srupt has based his correlation of the rocks of southern Africa upon their general lithological characters, and he attempts to trace the representatives of the South African Systems through Northern Rhodesia into Katanga and the Lower Congo. He considers that representatives of the Karroo, Waterberg, Transvaal, Swazi, and Archean Systems of South Africa can all be recognised in these northern territories ; the only systems missing are the Ventersdorp and Witwatersrand Systems, and their absence leaves a great gap between the Transvaal and Swazi Systems. (1) The Bihé Sandstones.—The long succession of unfossiliferous beds ended with the Lualaba beds, which are succeeded by the Lubilash beds. It is agreed that the Lualaba beds are of Lower Karroo age, for they contain coal-seams in North-Western Katanga, and some entomostraca from these beds have been described by LERICHE as Triassic (LERICHE, 1913, pp. 167, 168). According to Corner (1912, No. 1, p. 4) they are probably Triassic. The position of the Lubilash (or Lubilache) beds is less certain. According to Cornet, who founded it asa “system,” it is later than the Lualaba System (cf. Cornet, 1912, No. 1, p. 9); but Srupr represents it as earlier than the Lualaba beds. The fossils obtained by Batt and SHatER have been examined by H. O. Utricr (Batu and SHatsr, 1910, pp. 687, 688), who regards them as Triassic or Jurassic, and thus supports CoRNET’s view. I am not aware of the existence of any Lualaba beds in Benguella, but the Lubilash beds may be widely represented. ‘The typical Lubilash beds are sheets of soft, easily disintegrated sandstones. Cornet called this rock “Gres tendres,” which * Of. ConNET, 1894, 1 and 2, 1896, and a later synopsis of his classification, 1912, 1. 524 PROFESSOR J. W. GREGORY. Batt and SHALER report as “the common name for the Lubilache formation” (BALL and SHALER, 1910, p. 688). These beds are probably represented in eastern Benguella by the Bihé Sandstones. The Bihé Sandstones are a series of soft sandstones, which weather into beds of loose white quartz sand; they give rise to the wide plains of the Bulu-Vulu, and apparently also of the Hungry Country to the east of Bihé and Fort Silva Porto. I only reached the western margin of this area, but from the descriptions of the country to the east of Bihé, it is probable that the wide extent of poor woodland in that part of Benguella is due to the disintegration of the Bihé Sandstones.* The paleontological evidence, though meagre, is in favour of the Karroo and probably Triassic age of the Lubilash beds, for Bart and SHatEr (1910, p. 687) definitely correlate them with the Stormberg beds. Sruprt, however, correlates them with the Waterberg Sandstones. South African opinion appears to be steadily growing in favour of the view that the Waterberg System is Devonian and is the Transvaal representative of the Table Mountain Sandstone. This opinion is not universal, for Professor Scuwarz has advanced (1912, pp. 187-188) arguments in favour of the pre-Devonian age of the Waterberg System; he has referred to the similarity of the Waterberg Sandstones to the Torridonian of Scotland, though he accompanies this suggestion by the warning that the two formations are too far apart for the lithological correlation to be convincing. The Bihé Sandstone and other Lubilash beds present some important differences from the Waterberg System of the Transvaal, which includes an upper series of well- cemented red sandstones and a lower series of volcanic rocks. It seems to me there- fore that the Bihé Sandstones may be provisionally correlated with the Lubilash beds, and are younger than the Waterberg System; and from the evidence in the Congo they are probably of Triassic age. (2) The Oendolongo Series.—This series consists of firmly coherent sandstones, quartzites and shales, and of rhyolites and rhyolite tuffs some of which have been silicified into banded cherts. The rocks are neither foliated nor cleaved. The sand- stones and quartzites are composed chiefly of quartz, but contain abundant grains of felspars (PI. II, fig. 1). They agree in character with the “Grés durs fels-. pathiques” of Corner (1897, No. 1, p. 29); he identified these rocks in 1896 as his Kundelungu System, and subsequently (1912, p. 9) included them in his slightly dislocated division. The Oendolongo beds, apart from the presence of rhyolites and tuffs, agree best with Corner’s description of the lower part of the Kundelungu System ; this subdivision he has called the ‘‘ Syst¢me de la Mpioka,” and described it (1897, No. 1, p. 29) as composed of red, micaceous, argillaceous beds which pass into “psammite,” and alternate with medium or fine-grained sandstones which are very coherent, often felspathic, and red, gray, or dark in colour. They are slightly * This Hungry Country is said by Captain Boyp CunnincHamE (1904, p. 155), who knows it well, to have a good soil, and he regards its poverty as unexplained. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 525 undulating, and contain veins of quartz. This Kundelungu System of Corner has been assigned to the Carboniferous or Permian. Ban and SHater (1914, p. 609) remark, e.g., that the Kundelungu is the “ Permian of most geologists” ; but according to Srupt (1914, p. 71) it represents the Transvaal System, and he regards it as there- fore Devono-Silurian or Silurian in age. The Oendolongo Series has one significant feature in common with the Waterberg System of the Transvaal, in which the upper part consists of red sandstones and the Fig, 9. lower, in the eastern part of the formation, of volcanic rocks and tufts (e.g. ScHwarz, 1912, p. 187). Batt and SHauer have, it is true, remarked (1914, p. 612) that igneous activity, except for some diabase intrusions, had ceased before the deposition of the Kundelungu beds; but according to Srupr (1914, p. 56) the Kundelungu beds are associated with basic lavas as well as having been intruded by a variety of igneous rocks. The Oendolongo Series is probably the Benguella representative of the Kundelungus, and I should be disposed to regard it as equivalent to the Water- berg System of the Transvaal. If so, there is much to be said for the reference of the Oendolongo beds to the Devonian ; but when examining these rocks I felt that they might be much older and possibly be one of those unfoliated and comparatively 526 PROFESSOR J. W. GREGORY. unaltered red sandstones which occur in so many parts of the world between the base of the Cambrian and the great Eparchean unconformity. The considerable re- crystallisation of some of these quartzites, with the development of large flakes of authigenous muscovite (Pl. II, fig. 2) and lines of epidote, is consistent with the pre-Cambrian age of these rocks. (3) The Lepi Graywackes.—The Lepi beds are nearly related to the Oendalenam Series, but they are more disturbed and altered. They include a series of gray- wackes, some of which are coarse-grained and others very fine-grained. Some of the fine-grained varieties have been silicified, and I recorded them in my field notes as cherts. The beds include rhyolite tuffs similar to those in the Oendolongo Hills, but I saw no rhyolites. Kast of Lepi the beds show an incipient foliation, and the felspathic material along these developing fohation planes has been altered into lines of epidote. The Lepi beds appear to correspond with the lower part of the Oendo- longo Series, and contain the same rhyolite tuffs, but they were deposited beyond the range of the lavas. In North-Western Katanga the Kundelungu beds are underlain conformably by the Lubudi beds, which are cleaved and include an extensive series of cherts and oolitic hmestones ; and though I did not find any limestones in my hurried traverse across the Lepi Series, it agrees in so many respects with the Lubudi beds that these groups may be provisionally correlated. I had no opportunity of determining whether the Lepi graywackes are conformable to the Oendolongo Sandstones, for I nowhere saw them in contact; but the Lepi beds are the more steeply inclined and more disturbed. They correspond in petrographic characters to the rocks of the Transvaal System, the age of which is uncertain. It may be Lower Paleozoic, an opinion which has been strengthened by Hermann’s discovery of a badly preserved Orthoceras in the Otavi dolomites of Damaraland (HeRMaNN, 1908, pp. 265, 266), which have been correlated with the dolomites of the Transvaal System. The evidence for this correlation is, however, not convincing; and Hermann (2bid., p. 270) regards the Otavi dolomite as the equivalent of the Bokkeveld beds, and if so they are much later than any part of the Transvaal System. The correlation of these fossiliferous dolomites of German South-West Africa with the Transvaal dolomites may suffer the fate of several previous suggestions as to the identity of marginal marine beds in South Africa with rocks on the high plateaus of the interior. (4) The Huambo Quartzites.—The Lepi beds are bounded to the east by a wide tract of granite, which further east beyond Huambo is covered by granular holo- crystalline quartzites. ‘These rocks must have been coarse sandstones originally ; their cement has recrystallised into small grains, so that the larger grains are fretted together and show strain effects. They show no definite foliation. Specimens were collected on the Lundula River, also east of Candumbo, and north of Gordon’s Store near Saccanjimba (No. 187). Similar quartzites occur further north near Cinjamba and the Canji River. I did not see them clearly associated with the gneisses, and od CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 527 they may be younger than the schistose quartzites that are found to the north of them. It is, however, possible that they belong to the same division as the schistose quartzite and gneiss, and represent granular quartzites due to the alteration of pure siliceous sandstones. These Huambo quartzites appear to agree with the Kafubu quartzites of South-Hastern Katanga and also with the Nzilo beds of North-Western Katanga. The Nzilo beds are described by Srupr (1914, p. 53) as compact to glassy white and red quartzites which are granular and often show false bedding. This description would apply well to the Huambo Quartzites, and Srupr has suggested that the Kafubu beds are of Cambrian age ; but I should incline, from their crystalline character, to the view that the Huambo Quartzites are pre-Cambrian and belong to the lower division of the Kozoic. (5) Bailundo Schistose Quartzites.—To the north and north-east of the Huambo Quartzites are large outcrops of a schistose quartzite. These rocks are coarsely foliated. They are traversed by planes of authigenous micas and pass by increase in the new felspar into coarse gneisses. They are also interstratified with cordierite and sillimanite schists. The schistose quartzites of the Ambussa Hills to the south of Bailundo are typical examples of this group, which occupies a wide extent of country to the north and north-east of Bailundo. The rocks have a very high dip, are interstratified with the coarse gneisses, and belong to the highly crystalline foliated Hozoic foundation. Weathered hand specimens of the older quartzites may be indistinguishable from the quartzites of the Oendolongo Series; but fresh material from the two series is very different in character, and the quartzites which are interstratified with the gneisses can be readily separated in the field by their higher dip. A limestone is reported to occur near Bailundo, but I was unable to visit the locality. It is probably a crystalline limestone belonging to the Kozoic Series. CoRRELATION OF THE OLDER SEDIMENTS. The suggested classification and relation of these rocks may therefore be sum- marised as follows :— Equivalents in Geological Age. Congo. South Africa. Bihé Sandstones P ; Lubilash beds Lower Karroo Trias : r perhaps Oendolongo Series’. . | Kundelungu beds Waterberg System | Devonian a “ae ee 5 Saat 3 pas : a pper Lepi Graywacke Series ; Lubudi beds Transvaal System | Lower Paleozvic | Hone Huambo ()uartzites. : Nzailo beds and ] Kafubu Quartzites | / Swazi System Lower Eozoic Bailundo Schistose Quartzites j TRANS. ROY. SOC. EDIN., VOL. Li, PART IL] (NO. 13). 76 528 PROFESSOR J. W. GREGORY. The various conclusions as to the age of the Waterberg and Transvaal Systems are matters of impression rather than of direct evidence. One current of opinion is in favour of their Lower Palaeozoic age, the absence of fossils being explained by “the imperfection of the geological record” and the still inadequate search. Accord- — ing to the opposite view, the beds are lithologically similar to the unfoliated Upper Kozoic rocks, and the fact that this vast series of sediments has not yet yielded a single fossil, affords strong presumptive evidence that the beds are pre-Paleeozoie. The balance of opinion in South Africa is in favour of the former view; my own tendency, [ must admit, is towards the latter. CU. The Hozoic Gneisses. The schistose quartzites of the Bailundo Series are interstratified with coarse gneisses which are exposed at many localities in the Bailundo and Bihé districts. These rocks so closely resemble the Laurentian and Lewisian gneisses that they are probably foliated plutonic rocks of Lower EKozoic age. Similar rocks are also well exposed at the Lenewe Gorge, where they are not associated with quartzites ; but the gneisses are so similar that they probably belong to the same division, and as the strike of their foliation is roughly east and west, they are probably connected, at least below the surface, north of the Benguella railway from Lengwe to Bailundo. The Benguella gneiss series is intruded by granites which occupy most of the country for 250 miles between Mount Sahoa and Candumbo to the east of the Huambo. These granites have clearly been intruded into the older Archean rocks, which have been altered at the contact into hornfels, while their foliation is often very disturbed. The granites are older than the Oendolongo Series, which is largely composed of their debris. The most typical of the Benguella gneisses resemble those of the Laurentian rocks of Canada and the Lewisian of Scotland; and their associated granites are connected by the charnockites of the Ochilesa district with the ancient rocks of Southern India and with the Ivory Coast of Western Africa, where, amongst granites and gneisses similar to those of Benguella, the existence of a Charnockitic Series has been deter- mined by Professor Lacrorx (1910). D: The Voleame Rocks and Intrusive Rocks of Undetermined Age. The widest sheet of voleanic rocks is that of the rhyolite tuffs of the Oendolongo Hills ; and the occurrence of similar tufts in the beds east of Lepi is a point in favour of the Lepi beds belonging to the same series as the Oendolongo Quartzites. Near Lepi occur some of the best exposures of the widely spread olivine-dolerite. It occurs in dykes in many localities, and apparently in sills near Lepi, and in sills or flows to the north-east of Bailundo. The dolerite is apparently younger than the Oendolongo Series, though | had not time to trace a dyke of this rock west of Cruz’s into the sandstones. CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 929 The volcanic neck of Chieuca may be approximately of the same age as these dolerites, for some of its rocks are fresh and show no signs of pressure. I was in the field disposed to regard the alkali series of Chieuca and Ochilesa as probably belong- ing to the same:age as the alkaline volcanic rocks of British East Africa and of the Atlantic islands and as probably of early Kainozoic age. They are certainly much younger than the Oendolongo Series, though there is no direct stratigraphical evidence to establish this conclusion. The Charnockite Series of Ochilesa is probably much older than the alkaline lavas, and it belongs tothe underlying foundation of older rocks. The charnockites have been intruded into the gneisses, and their age is probably the same as that of the great series of granites along the railway route from Mount Sahoa to Candumbo. Near the charnockites at Ochilesa are considerable occurrences of a sodalite- syenite and shonkinite which doubtless belong to an intrusion of the same age as the Chieuca neck. A sodalite-eleolite-syenite has been described from Zenza do Itombe, in the interior of Loanda (Brra, 1903, p. 359), and a nepheline-basalt from Dombe’ Grande has been identified by Gomes and recorded by Cuorrat (1888, p. 29). The latter rock is doubtless Kainozoic, and is one of the basalts of the coast zone of southern Benguella and Mossamedes ; but the eleolite-syenite was probably intrusive into the biotite-gneisses of Loanda, and may be of any post-Hozoic age. M. PERErRA DE Sousa (1913, pp. 1451-2) has announced the discovery in North-Western Angola of riebeckite-zeeyrine-leptynite, nordmarkite, nepheline-syenite, nepheline-phonolite, and tineuaite.* IV. Tectonic GEOLOGY. The tectonic geology of Benguella appears to be comparatively simple. The problem of chief importance is the origin of the steps by which this part of West Africa rises from the sea to the high plateaus of the interior. This structure was clearly expressed by Livrnesrone, the value of whose observations in Angola is usually underrated, as they are judged from the remarks in his Missionary Travels instead of from his technical paper and section in the Jowrnal of the Geographical Society (1855, p. 232). Livrinesronn’s Ideal Section of Angola was an important contribution to its geology and topography. It shows the trough-like form of the valley of Cassange and the abrupt descent from the main plateau in the Golungo- Alto district ; and it truly represents the coast-lands as formed of marine fossiliferous marls, which are modern near the coast and include older beds inland. The coast zone between Hanha and Benguella shows the step and plateau structure on a smaller scale. The land rises to the height of 1000 feet by three steps; and that they are due to parallel step faults appears conclusively proved by the geological evidence. The inflata beds occur at sea-level on the coast at Old * A description of these and other alkaline igneous rocks from this area by Mr A. Houmes is now in course of publication (Hommes, 1915). 530 PROFESSOR J. W. GREGORY. ~- Lobito, and at the height of 800 feet 5 miles inland. This difference of level might be attributed to a westward dip; but at the step by which the land rises from 500 to 800 feet the algal limestone occurs near the foot of the step only 300 feet lower than the inflata beds, and their relations can only be explained by the step being a fault scarp. ‘The faults in the Cretaceous beds, as remarked on pp. 499, 501, are clearly seen at Lobito and Hanha and especially in the Catumbella valley. A recent subsidence of the whole coastal area has been frequently asserted from the famous submarine canyon off the mouth of the Congo. That submarine trench has been advanced (K. Hutt, 1912, p. 13) as proof of the subsidence of the whole country to the depth of even 6000 feet. There seems, however, to be no evidence of any such lowering of Angola. The existence of this canyon was first announced by Sraturprass (1887, p. 391). It has also been described by Dr J. Y. Bucnanan (1887, pp. 222, 223). Itis 1450 feet deep within the Congo estuary, off Banana Creek ; 35 miles off the mouth of the river it is 3428 feet deep; and it is over 6600 feet deep where it notches the 750-fathom contour 80 miles from the coast. Dr J. Y. BucHaNnaN explained this canyon as having “been built up and not hollowed out.” He regards it as a canyon of accumulation due to the deposition of the Congo silts on either side of the river mouth, like the deep channel off the entrance of the Rhone into Lake of Geneva. This view has also been adopted by the Portuguese geographer VasconceLtitos. The observations by Admiral Purry-Cust in 1899 (pp. 180, 181) show, however, that, instead of “the sea water running up the gully at the bottom and returning in the upper layers, mixed with the river water” (BucHANAN, 1887, p. 223), and thus keeping the canyon empty by a vertical circula- tion, the trench, even within the estuary, is occupied by almost or quite stagnant sea water, and the bottom is covered by deep soft mud containing vegetable matter. Hence the trench is being filled, though perhaps very slowly, by the Congo silts. The most probable alternative to Dr Bucnanan’s theory is that a long stretch of land off West Africa has foundered beneath the Atlantic, and owing to the recent date of this subsidence the continuation of the Congo valley across the sunken belt has not yet been wholly filled. Neither this explanation nor Dr BucHanan’s involves any great subsidence of the existing coast-land, and any such movement is disproved by the physiographic evidence on the mainland. The shore of Angola, with its long even lines, is quite unlike a drowned coast. A recent uplift of the coast near Lobito to the extent of 500 feet has been suggested on the ground that modern marine shells occur at that height on the plateau to the east of Lobito Bay. The aspect of the country, as seen from the sea, certainly resembles a plain of marine denudation. I was, however, unable to find on it any clear evidence of recent uplift. Sea-shells are often carried inland by birds or by natives, and my informant promised to obtain some of these shells while I was in the interior, They were, however, not forthcoming, and it is quite possible Dt hd —————— aS @&«a errr rere eee CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 531 that the report may be based on Cretaceous shells. [ saw no raised beaches or other indications of marine action except upon the low coast plain between Lobito and Benguella. The evidence of the wells near Bimbas Road (p. 504) is not in favour of the marine origin of the recent deposits in the valley of the Cavoea. The age of the coast faults is post-Cretaceous, and it is probably Pleistocene. The cliff on the south-east shore of Lobito Bay appears modern, for some of the valleys which notch it are still in the condition of hanging valleys; and as the rocks on their beds are soft, these valleys must be quite young. Earthquakes frequently happen along this coast, and they may be due to continued movements on these recent coast faults. A more difficult problem is whether the great steps by which the country rises over 6000 feet above sea-level to the Bihé plateau are also fault scarps. The section across the Western Congo by Cornet represents the country as built of a succession of plateaus with steep faces toward the west, and according to his interpretation (Cornet 1897, No. 1, p. 22) the faces are escarpments and the plateaus are long dip- ¢ slopes to the east. In one part of his section, however, near the “créte de Bonza Mantéka,” the plateau cuts across the bedding and its surface is therefore due to denudation. In Benguella the general succession of the plateaus to the west of the coastal belt appears also to be due more to denudation than to faulting. The arrangement of the north-western and south-western faces of the Oendolongo Hills is, however, suggestive of their having been determined by faults. The edges of the main plateaus are so intensely dissected that even if the steps were due to faulting their most conspicuous present features are due to denudation. The successive plateaus are not sharply separated in the section showing the relief of the railway line, for the railway naturally follows the gulleys and sometimes adopts a sinuous course in order to increase the distance and thus lessen the grade. If the railway had adopted a shorter course, rack sections would have been necessary, and they were proposed at two localities where they have been avoided by lengthen- ing the line. The plateau fronts are therefore not well seen from the railway, and the determination of their nature would require long traverses north and south from the railway. If plateaus were originally bounded by faults the original scarps have been destroyed except near the coast, and in the interior the fault lines are buried to the west of the steps. The explanation of the steps by denudation alone would be difficult. The terraces might be regarded as successive pene-planes due to halts in the uplift of the land ; but though that explanation would explain the surface of the plateaus, it does not appear consistent with the nature of the steps. The step-like structure of this country appears to require either a series of monoclinal folds or of step faults; as these two structures are tectonically equivalent, the choice between them is com- paratively unimportant. I saw, however, ne evidence of monoclinals, while faults are numerous. 532 PROFESSOR J. W. GREGORY. s r Benguella presents an interesting tectonic contrast with East Africa. The rock sequence of the two regions has many points in common, and both have been sub- jected to Kainozoic meridional dislocations. Those of Angola are simple coastal step faults, which have disturbed the Cretaceous limestones. Whether the steps by which the country rises to the plateaus of the interior are due to earth movements of the same age is uncertain. Hence the coastal faults are the only proved representative in Angola of the meridional faults which in Kast Africa formed the creat system of rift valleys. REFERENCES. 1912. Atmetpa, Joo pe, Sul @Angola: Relatorio de wm Governo de Distrito (1908-1910), Lisbon, xix + 648 pp., tables and maps. 1885. Ancuieta, José v’, “Tragos geologicos da Africa, occidental portugueza Benguella,” reprinted in Bol. Soc. Geogr. Lisboa, vol. v. : 1910. Ancona, Derr. or GovERNOR-GuNERAL, Lelatorio da Missdo de Colonisagéo no Planalto de Benguella ‘em 1909, 157 pp., 12 tables, 5 diagrams, 4 maps. 1910. Batu, S. H., and Suaner, M. K., ‘A Central African Glacier of Triassic. Age,” Journ, Geol. (Chicago), vol. xvii, pp. 681-701. ss “Economic Geology of the Belgian Congo, Central Africa,” Econ. Geol., vol. ix, pp. 605-— 663, pl. xv. 1895. Barrat, M., “Sur la Scolds du Congo frangais,” Ann. Mines, Paris, sér. 9, vol. vii, Mém., 132 pp. (379-510), pls. xil—xiv. 1903. Bere, G., “Gesteine von Angola, Sad Thomé und St Helena,” Vsch. Min. 1. Petrog. Mitt., vol. xxii, pp. 357-362. 1887. Bucuanan, J. Y, “On the Land Slopes separating Continents and Ocean Basins, especially those on the West Coast of Africa,” Scot. Geog. May., vol. ii, pp. 217-238, plate. 1877. Camron, V. Lovett, Across Africa, vol. i, xvi+ 389 pp., 15 pls., map ; vol. 11, xi1+ 366 pp., 18 pls. 1882. CareLio, HermMEnEciLpo, and Ivens, Koparro, from LBenguella to the Territory of Yacca: Description of a Journey into Central and West Africa (translated by A. Etwes), vol. i, i+ 395 pp., map ; vol. il, xv +350 pp., map. 1886. —— De Angola & Contra-Costa, vol. 1, xxvii+448 pp., 3 maps; vol. i, xiii 4-490 pp., 3 maps. 1887. Cuorrat, P., “Note préliminaire sur les fossiles recueillis par M. Lourengo Malheiro, dans la provinee @Angola,” Bull. Soc. Giéol. France, sér, 3, vol. xxv, pp. 154-157. . 1914. 1887. —— “Dos Terrenos Sedimentares da Africa Portugueza e Consideragdes sobre ‘a Geologia d’Este Continente,” Bol. Soc.*Geog. Lisboa, ser. 7*, pp. 143-150. ’ 1896. —— “Coup d’ceil sur la géologie de la province d’Angola,” Communic. Dir. Trab, Géol. Portugal, vol. ili, f. i, pp. 84-91. 1900-1901. _“ Hehantillons de roches du district de Mossamedes,” Communic. Dir. cee Géol. Portugal, vol. iv, pp. 190-194. 5 “Contributions & la connaissance géologique des colonies portugaises d'Afrique: II, Nouvelles 1905. données sur la zone littorale d’Angola,” Commiss. Serv. Géol. Portugal, pp. 31-78, pls. i-iv. 1888. Cuorrat, P., and pe Lortot-Lerort, P., ‘ Matériaux pour l’étude stratigraphique et paléontologique de la province @’Angola,” Mém. Soc. Phys. Hist. Nat. Geneve, vol. xxx, No. 2, 116 pp., 8 pls. 1894. Corner, J. (1) “ La géologie de la partie sud-est du bassin du Congo et les gisements métalliféres du Katanga,” Rev. Univ. Mines, vol. xxviii, scr. 3, pp. 217-294, pls. vi-vii. 1894. —— (2) ‘Les formations post-primaires du bassin du Congo,” Ann. Soc. Géol. Belg. “rol: xxi, Mém., pp. 193-279, pl. v. 1897. —— (1) “Observations sur la géologie du Congo occidental,” Bull. Soc. Belg. Géol., vol. . pp. 21-30. : : tbs cae | ee la - 1897. 1906. 1912. 1912. 1908. 1904. 1901. 1Sit. 1898. 1898. 1913. 1908. 1915. 1912. 1910. 1878, 1913. 1913. 1855. 1888. 1913. 1887. 1875. 1898. 1912. 1881. 1907. - CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 533 Cornet, J., (2) “Observations sur les terrains anciens du Katanga faites au cours de l’Expédition Bia-Franequi (1891-93),” Ann. Soc. Géol. Belg., vol. xxiv, Mém., pp. 25-190, pl. i. —— “Sur la distribution des sources thermales au Katanga,” Ann. Soc. Géol. Bely., Mém., vol. xxxiii, pp. 41-48, ; (1) “Sur la possibilité de existence de gisements de petrole au Congo,” Ann. Soc. Géol. Belg., vol, xxxviil, Annex. Congo, pp. 9-15. —— (2) “Sur Page des couches du Lualaba,” Ann. Soc. Géol. Belg. vol. xxxviii, Annex. Congo, pp. 3-4. Costa SEN, J. C. va, “A Riquesa petrolifera d’Angola,” Soc. Geogr. Lisboa, 1908, pp. 1-15. Cunincuame, B. A., “A Pioneer Journey in Angola,” Geogr. Journ , vol. xxiv, pp. 153-168, map. Dencavo, J. F. N., “Quelques mots sur la collection d’échantillons collectés par le Rev. Autunes in Angola,” Communic. Serv. Géol. Portugal, vol. iv, pt. 2, pp. 195-201. Freire pe Anprabe, A., “Sur la Position de Senilia senilis dans le. tertiaire de Loanda,” Communic. Commiss. Serv. Géol. Portugal, vol. viii, pp. 87-89. Gomes, J. P., “Echantillons de roches recueillis entre Benguella et Catoco,” Communic. Dir. Trab. Géol. Portugal, vol. iii, fase. ii, pp. 239-243. — “O betune do Libollo (provincia d’Angola),” Communic. Dir. T'vrab. Géol. Portugal, vol. iii, fase. ii, pp. 244-250. ; Gregory, J. W., Report on the Work of the Commission sent out by the Jewish Territorial Organisa- tion under the Auspices of the Portuguese Government to examine the Territory proposed for the purpose of a Jewish Settlement in Angola, xiii+ 50 pp., 2 maps, illustrations. Hermann, P., “Beitrag zur Geologie von Deutsch-Siidwestafrika,” Zeit. deut. geol. Gresell., vol. lx, pp- 259-270. Hommes, A., “A Contribution to the Petrology of North-Western Angola,” Geol. Mag., dec. vi, vol. il, pp. 228-232 ; (continued), pl. ix. Huu, E., Monograph on the Suboceanic Physiography of the North Atlantic Ocean, fol., viii+ 41 pp., 11 pls. Lacroix, A., “Sur Pexistence a la Cote d’Ivoire d’une série petrographique comparable a celle de la charnockite,” Compt. Rend. Acad. Sei. Paris, vol. cl, pp. 18-22. Lenz, O., ‘“‘Geologische Mittheilungen aus West-Africa,” Verh. k. k. geol, Reichs., 1878, No. 7, p. 148. ; LericuE, M., “Les Entomostracés des couches du Lualaba (Congo belge),” Bull. Soc. Belg. Géol., vol. xxvii, pp. 167-168. | “Les Entomostracés des couches du Lualaba (Congo belge),” Revwe Zool. Afric., vol, ii, pp. 1-11, pls. 1-111. Livinestong, D., ‘On the Province of Angola,” Journ. Roy. Geog. Soc., vol. xxv, pp. 229-235. Loriot-Lerorr, P. px, “ Notes sur la géologie de la province d’Angola,” Arch. Set. Phys. et Nat. Geneve, vol. xix, p. 67. See also CHorrar. Maruigu, FV. F., “ Les sources thermales du Bas-Katanga,” Ann. Soc. G'éol. Belg., vol. xl, Aunex. ‘Congo, pp. 103-125, pls. iii-iv. Meunier, S., “Contribution 4 la géologie de Afrique occidentale,” Bull. Soc. Géol. France, sér. 3, vol. xvi, pp. 61-68, pl. 1. Monteiro, J. J., Angola and the River Congo, vol. i, ix +305 pp., map; vol. 1, v+ 340 pp. Nascimenro, J. Persrra po, Haploragdo Geographica e Mineralogica no districto de Mossamedes em 1894-1895, Lisbon, 110 pp., 4 pls., 1 map. . Nascimento, J. Perera po, and Marvos, A. ALEXANDRE DE, A Colonisagéo de Angola, Lisbon, 163 pp., 1 map, 14 pls. Pinto, Serpa, How I crossed Africa: from the Atlantic to the Indian Ocean, through Unknown Countries ; Discovery of the Great Zambesi Afjluents, etc. (translated by A. Euwus), vol. i, xxx + 377 pp., 8 maps; vol. ii, vii+ 388 pp., map. Prizm, F., “‘ Poissons tertiaires des possessions africaines du Portugal,” Comm. Commiss. Serv. Géol. Portugal, vii, pp. 74-79, pls. i and il. 534 PROFESSOR J. W. GREGORY. 1910. Purgy-Cusz, H. E., in Africa Pilot, pt. ii, sixth edition, pp. xxvi, 448. 1912. Roserr, Maurice, “ La stratigraphie du systeme du Kundelungu au Katanga,” Ann. Soc. Géol. Bely., vol. xxxix, Annex. Congo, pp. 5-8. 1888. ScutumbBerasn, C., “‘ Note sur les Foraminiféres fossiles de la province d’ Angola,” Bull. Soc. Géol. France, sér. 3, vol. xvi, pp. 402-404. 1912. Scuwarz, E. H. L., South African Geology, 200 pp., 54 figs. 1913. Sousa, Pererra ve, “ Contribution a Pétude pétrographique du nord d’Angola,” Compt. Rend. Acad. Sct. Paris, vol. clvii, pp. 1450-1452. 1887. Sravuiprass, E., “ Deep-sea Sounding in connection with Submarine Telegraphy,” Journ. Soc. Teleg. Eng., vol. xvi, pp. 479-511, pl. C. 1913-1914. Srupz, F. E., “The Geology of Katanga and Northern Rhodesia: An Outline of the Geology of South Central Africa,” V’rans. Geol. Suc. South Africa, vol. xvi, pp. 44-106, pls. v—xv. 1884. Szasnocua, L., ‘‘Zur Kenntniss der mittelcretacischen Cephalopoden-Fauna der Inseln Elobi an der Westkiiste Afrika’s,” Denk. Akad. Wiss. Wien, vol. xlix, Wien, pp. 231-238, pls. i-iv. 1913. Vincent, Em., “Contribution a la paléontologie des falaises de Landana (Bas-Congo): Mollusques,” Bruxelles Ann. Mus. Conyo, Géol. Pal., etc., section iii, 1, pp. 1-46, pls. i-vi. 1889. We.wirscH, “ Quelques notes sur la géologie d’Angola coordonnées et annotées par Paul Choffat,” Communic. Comm. Trab. Géol. Portugal, vol. u, fase. i, pp. 27-44, pls. i-iv. EXPLANATION OF PLATES. Puate I. Geological Map of part of Benguella. vv Alluvium. cce Cretaceous. Trias, % Lualaba beds. Dombe Sandstones. Lubilash beds. Bihé Sandstones. Palzeozoic Oendolongo Series (sandstones, quartzites, rhyolites, tuffs). or Upper Kozoie. Lepi Series (graywackes, cherts, slates, rhyolite-tuffs, etc.). (Transvaal Systew.) Huambo Quartzites, > oa > eye > he ° — , _— - Lower Eozoiec. Bailundo Schistose Quartzite. (Swazi System.) aes Gueiss, aaa| Kozoic Intrusives. Granites. [sss Shonkinite Series. Alkaline Lavas and dykes. (Probably Kainozoic) Puate II. The older Sedimentary Rocks of the Benguella Plateau. Fig. 1. Oendolongo Sandstone. Pass at head of Cambuacambula River, south of Quingenge Agricultural Station. The large central grain is felspar. (No. 256.) CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 535 Fig. 2. Oendolongo Quartzite. Hill west of Senhor Piré’s at Babaera. With authigenous muscovite (m). (Crossed Nicols.) (No. 241.) Fig. 3. Epidotic Quartzite. East of Lepi. The dark granules are mainly epidote. (No. 135.) Fig. 4. Micaceous Schistose Quartzite with interlocking grains. Bailundo Series. West bank of Lovule River, east of Catotlo. (Crossed Nicols.) (No. 161.) Fig. 5. Schistose Quartzite. Bailundo Series. Ambussa Hills, near old stone fort. (Crossed Nicols.) (No. 149.) Fig. 6. Sillimanite Schist. Bailundo Series. South-east of first camp, about ten miles north-east of Bailundo. (No. 158.) All x 24 diameter. EXPLANATION OF ILLUSTRATIONS IN TEXT. Fig. 1. Geological section along the Benguella Railway from Lobito Bay to Candumbo. The hills outlined in broken lines are situated south of the railway. (Scale: Vertical, 1” =30 miles; horizontal, 1 = about 30 miles.) Cretaceous and alluvium. Q (@) Sandstones at Culimahala (age undetermined). ° fo) Oendolongo Series. [-) ° ° Lepi graywackes. Gneiss. Granite, etc. BIBIAIZ Dolerite. Fig. 2. Section across the Cretaceous beds south-east of Lobito Bay. (Scale: Vertical, 1”=1000 feet ; horizontal, 1” = 1 mile.) 1=inflata beds. a= Algal limestone. * = False-bedded inflata beds. Fig. 3. Section along the Hanha River. (Scale as in fig. 2.) c= Astrocenia bed. a = Algal limestone. tees Boulder bed. 1 Gneiss. Fig. 4.—The sandstone above the gneiss at the head of the Lengwe Gorge. Fig. 5. Section across the Oendolongo Mountains and through the Quingenge Agricultural Station to the Cuiva River. -(Scale: Vertical, 1” = 2000 feet ; horizontal, 1” =2 miles.) s = Sandstone. + = Rhyolite. ¢=Tufis, Q = Quartzite. A = Fine-grained granite. A =Coarse-grained granite. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 13). 77 586 CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. Fig. 6. Section from Huambo to Ochilesa. Approximate average direction south-west to north-east. Seale: Vertical, 1” =1250 feet; horizontal, 1” =15 miles.) — Shonkinite, ete. (Lavas and dykes.) Dolerite?. Charnockite. Diorite. Granite. SSS} Schistose quartzite. of EAE IE Gneiss. ig. 7. Section from the Bulu-Vulu to the Umbal near Cambenge. (Scale: Vertical, 1”=1000 feet ; horizontal, 1” =1 mile.) Bihé Sandstones. a iQ ald Schistose quartzites. LU] Gneiss. g. 8. Section along the track to Candumbo, east of Huambo. (Scale: Vertical, 1”=1000 feet; horizontal, 1”=3 miles.) _ : _s Fi Sandstones (age undetermined), Granite. Fig. 9. Geological sketch-map of part of Angola. Kainozoic. Cretaceous. Dombe Sandstones 'Trias. Lualaba beds. Bihé Sandstone. ?Trias. Lubilash beds, Oendolongo and Lepi Series in Benguella Lower Palzozoic or Upper Hozoic, and Cuanza Series in Loanda. Transvaal System. = Gneiss, schist, schistose and granular Lower Eozdic, Swazi System. quartzites. Granites intrusive in the Swazi System. Alkaline igneous rocks. Basalt. Probably Kainozoic. Trachyte. yoo} Ul SIqSIOF nvaie{q Ply, jo spy ulsjsa~y JO asinog o}eulxoiddy Nevajye[q Puoses jo aspx] UAAIS9A\ JO aSINOD ayeutxoiddy ‘eSOLIYIOQ WIJ puL 0} OquIeNT]Y Woy JIOYINYy ay} Jo aynoy --------— - ‘PLOY UOSSEM\ I9OG eorccecece "eZUeND PUL OF2{ND 9Y} U2aMjeq PeYsieiVA 24) Suoje uvieA “FH I JO amoy .—.—.—.—.-. ABMPVEY BLTONSUS preter ‘sajtpy v71y GOASOIL) uNIGad ‘NOS 8 aIHOLIN Vv Oeeb a yrnw) 14 4404 ~ oannivg 7 \ ‘a | : i Ov. tI IVINVILV Os.e1 ‘T Td—nvalvTd WITsanNONag aHL dO LYVd dO dVW HOLEAWS TvdIDOTOaN « AMODRYND —y 536 oe Fig. 6. Section f (Seale : NING LEAS EAI EINE) Fig. 7. Section fr horizont cw cn # Fig. 8. Section al horizontal > > Fig. 9. Geologi 0g 3 8 2 Se SE) HW EB Roy. Soc. EDIN. Jd. W. GREGORY: SEDIMENTARY ROCKS OF BENGUELLA PLATEAU.—PL. II, Vom Ly (337) XIV.—A Contribution to the Petrography of Benguella, based on a Rock Collection made by Professor J. W. Gregory. By G. W. Tyrrell, A.R.C.Sc., F.GS., Lecturer in Geology, Glasgow University. Communicated by Professor J. W. Grecory, F.R.S. (With One Plate.) (MS. received June 2, 1915, Read June 28, 1915. Issued separately August 1], 1916.) Very little is known of the petrography of the Portuguese West African colony of Angola, and of its three provinces that of Benguella is probably least well known in this respect. Granites, gneisses, schists, limestones, and red sandstones have been recognised by the earlier observers of the geology of the region, and detailed descrip- tions of rocks from Angola, and especially of its northern province, Loanda, have been given by Berc* and Hotmes.t From Benguella itself J. P. Gomes t has described the collection of R. P. Leconte, which contained granites, amphibolite, adinole with epidote veins, various schists, quartzites, diabase-porphyrite, and basic eruptives. Senhor J. P. pe Nasctamento§ in 1912 recorded granite, diorite, basalt, and quartzite on the Benguella plateau. During his recent journey in Benguella, Professor Grecory collected about a hundred and fifty specimens of igneous rocks or their metamorphic derivatives, and these offer a most interesting and varied study. This paper is concerned only with their petrographic character; for their geological relations the reader is referred to the preceding paper by Professor GREGory. For descriptive purposes the classification given below has been found to be most natural and convenient. The order in which the rocks are arranged also closely follows their sequence in time :— I. Basement gneisses and schists. IJ. Charnockite series. Ill. A series of hornfelsed porphyries and other hornfels. IV. Granites, granodiorites, and associated rocks. V. Rhyolite (dellenite). VI. An alkaline series, including nepheline-sodalite-syenite and other syenites, shonkinite, solvsbergite, and ouachitite. VII. Basic intrusions. j I am much indebted to A. Scott, M.A., B.Sc., for two excellent chemical analyses of rhyolite and shonkinite respectively. * Bera, G., “Gesteine von Angola, Sao Thome, und St Helena,” Tscher. min. u. petr. Mutth., xxii (1903), pp. 357-362. + Hoxmzs, A., “Contribution to the Petrology of North-western Angola,” Geol. Mag., dec. vi, vol. ii, pp. 228-232 ; 267-272 ; 322-328 ; 366-370, 1915. + Gomzs, J. P., “ Echantillons de roches recueillis entre Benguella et Catoco,” Comm. Dir. Trab. Géol. Portugal, vol. iii, fase. 11, pp. 289-243, 1898. § NascIaMENTO, J. P. DE, and Mattos, A. A. pz, A Colonisagdo de Angola, Lisbon, 163 pp., 1912. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 14). 78 538 MR G. W. TYRRELL, I. THE GNEISSES AND SCHISTS. The gneisses are predominantly quartzo-felspathic varieties with biotite as the principal ferro-magnesian constituent. They are medium- to coarse-grained, pink or grey, well-foliated rocks, with the biotite aggregated into thick clots or folia, or less often evenly spangling the rock. Hornblendic gneisses occur, but are much less common than the biotite-bearing varieties. In thin section the biotite-gneisses are coarse-grained aggregates of quartz and potash-felspar with folia of biotite, and show marked dynamo-metamorphic effects. The larger quartz crystals are broken up into a comparatively coarse granular mass of angular fragments; but the felspars have only suffered a fine peripheral granula- tion, the products of which, mingled with quartz, form a kind of groundmass in which the larger, more or less rounded, “eyed” crystals of felspar are set. The felspar is prevailingly microcline. Orthoclase is subordinate, and a little oligoclase is occasionally present. The felspars show the effects of crushing in the irregularity, curvature, and fracture of their twin striations. The biotite is of the common yellow variety, and occurs in large clots or folia consisting of small flakes felted together with minute grains of quartz, epidote, and occasionally a little muscovite. Felspar is absent from these aggregates. ' The above description refers particularly to a rock from the Bui River, S. of Ochilesa (185),* which may be taken as typical of the predominant biotite-gneisses. Other rocks of the same type occurring at Bailundo (152) are very rich in “ gitter” microcline. Rocks from the Lengwe Gorge (105, 106, 109) are decidedly richer in soda-lime felspars (oligoclase or albite-oligoclase), and show a development of minute, colourless, euhedral crystals within the felspars. These appear to be of two or three different kinds. Some are clearly small zircons, but others are referred to clinozoisite and epidote. In some types (e.g. (156), W. branch of Lovule River), the foliation is not so definite as in the above-described rocks. The texture is evenly granular and the biotite is uniformly distributed. The granulitic texture becomes more pronounced in a rock from the Benguella Railway, 55 kilometres, E. of San Pedro (114). In this specimen the biotite occurs as thin leaves which appear as narrow bands of biotite flakes in thin section, and produce a bacillar type of foliation. This rock has some resemblance to the Moine gneisses of the Scottish Highlands. Biotite becomes very scarce and the cataclastic texture very prominent, in a pink, granulitic, quartzo- - felspathic gneiss from Kambengi (195). The rocks described above are all orthogneisses derived by dynamo-metamorphism from ordinary biotite-granites. These rocks grade into biotite-schists, e.g. at 55 kilo- metres, Benguella Railway (115). The texture becomes more even and finer-grained, the amount of quartz and biotite increases, whilst that of felspar diminishes, in com- parison with the gneisses described above. Another mode of transition is into a * The numbers refer to the registration numbers of the rocks in Professor GREaoRY’s collection. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 539 group of gneisses with very pronounced cataclastic textures, showing all gradations to mylonite. In the first stage the pseudo-porphyritic “eyed” felspar becomes highly rounded, and the quartz is broken into coarse-grained aggregates of angular frag- ments. Both minerals are set in a minutely crystalline paste of crushed quartz and felspar (190, N. of Saccanjimba). In succeeding stages all the quartz is reduced to mylonite, and the crushed groundmass is thus increased at the expense of the intact erystals of felspar. The latter are much reduced in size, show angular outlines, and are traversed by lines of mylonised material (173, W. of Ochilesa). The ferro- magnesian constituents, such as biotite, have been crushed out of recognisable existence, and appear now as aggregates of chlorite and limonite. A few large erystals of anthophyllite, now in process of alteration to talc, have been formed in the Ochilesa rock. The three hornblendic gneisses in the collection differ considerably from those described above. They occur towards the lower end of the Lengwe Gorge (105b), N. of the Saccanjimba Mission (189), and in the river W. of Kambengi (197). In thin section they consist of a coarse foliated aggregate of hornblende and plagioclase felspar, with subordinate quartz, orthoclase, and biotite. The ferro-magnesian minerals bulk more largely in these rocks than in the biotite-gneisses. The horn- blende is a strongly pleochroic variety, of strong absorption, and with colour extremes of bluish-green and yellowish-brown. Small felts of biotite and epidote cluster with the larger hornblende crystals. The felspars mostly belong to andesine, and contain colourless needles similar to those described above in the biotite-gneisses of the Lengwe Gorge. Cataclastic structures are also frequent. These rocks are typical diorite-gneisses. Only one basic hornblende rock occurs in the collection. This comes from the river W. of Kambengi (195). It is a fine-grained, dark-green, non-foliated rock, con- sisting, in thin section, of an evenly granular mass of hornblende crystals of the usual bluish-green colour. The small and sparse interspaces are filled with quartz and decomposed felspathic material. The rock is a granulitic amphibolite. Il. Rocks oF THE CHARNOCKITE SERIES. These are dark-greenish or greyish fine-grained rocks, in which none of the minerals, save an occasional flake of biotite, are macroscopically identifiable. The outstanding microscopical characters of these rocks are their thoroughly granulitic texture, their pyroxenic composition, and their ideal freshness. The mineral con- stituents, occurring in varying proportions throughout the series, are a pale-green monoclinic pyroxene, hypersthene, plagioclase, orthoclase, quartz, hornblende, biotite, and magnetite. This assemblage is identical with that of the charnockite series of Southern India.* The invariable presence of hypersthene, the granulitic texture, * Hoxuanp, “The Charnockite Series,” Mem. Geol. Surv. India, vol. xxviii, part ii, p. 124, 1900. 540 MR G. W. TYRRELL. and the almost exact correspondence of petrographic types, make the identity of the two series even more certain. , Rocks referable to three of the four divisions into which Hotianp divides the charnockite series are to be found in this collection, as well as a rock which may be described as a charnockite-porphyry. A rock comparable to charnockite in the narrow sense is found to the W. of Ochilesa (177). It consists mainly of quartz, oligoclase, orthoclase, with subordinate biotite, hypersthene, and monoclinic pyroxene. Both quartz and orthoclase show strain-shadows, and the twinning of the oligoclase is often discontinuous, wedging out in short distances and then reappearing. «The soda-lime felspar is rather more abundant than in the Indian charnockite, and considerably ex- ceeds the orthoclase in quantity (see Table I, 1). The biotite occurs in large ragged flakes, and is of a dark-brown, strongly pleochroic variety. Hypersthene is uniformly distributed in typical irregular, prismatic, cross-fractured crystals, with distinct pleochroism and faint lamellar twinning. There is also a pale-green monoclinic pyroxene. This mineral appears in all stages of alteration to a fibrous amphibole which has a distinct pleochroism from yellowish-green to a glaucophane-lke blue. The amphibole occasionally forms a narrow zone grown around the hypersthene crystals. The texture of this rock is not so thoroughly granulitic as in the more basic rocks of the series. It may be described as intermediate between granitic and eranulitic. In the hand specimen this rock has a distinct gneissic banding, which is, however, not apparent in thin section. . A rock from the N. of Etunda (193) is referable to HoLLann’s intermediate — division of the charnockite series. Compared with the foregoing, quartz and ortho- clase have so diminished in quantity as to become merely interstitial to the abundant plagioclase, whilst the amount of ferro-magnesian minerals remains stationary. Magnetite shows a large increase (see Table I, 2). Monoclinic pyroxene does not appear in this rock. The hypersthene is thoroughly anhedral, and has an intense pleochroism from pink to bluish-green. The rock also contains a little pink garnet which forms the centres of irregular aggregates of hypersthene and biotite. There is a tendency for the ferro-magnesian minerals to segregate into narrow linear areas, giving rise to a rude banding.* The water-clear plagioclase is richer in the anorthite molecule than in the charnockite proper, and is an andesine of composition AbgAnsg. The texture of this rock is thoroughly granulitic. Its normal eruptive equivalent would probably be called quartz-mica-hypersthene-diorite or quartz-mica-norite. From the district 8. of Andulo (194a@), and from Ochilesa (176b), come rocks strictly comparable with those called “ hornblende-augite-norite” by Ho.ianp.t These are ideally fresh rocks consisting of a thoroughly granulitic aggregate of water- clear plagioclase (Ab,An,) with pale-green augite, hypersthene, hornblende, iron-ores, and a little biotite. In the rock from Andulo (194q@) the hornblende is a brownish- — * Of, “Charnockite of the Ivory Coast,” Lacrorx, Comptes Rendus, cl (1910), p. 20. + Houvanp, op. cit., p. 157. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 541 ereen, strongly pleochroic variety, and is disposed about the augite in such a way as to indicate that it is due to molecular instability of the pyroxene during or after erystallisation. The hornblende of the Ochilesa rock (176b) is of a reddish-brown tint. Some of the hypersthene shows a close lamellar twinning and a small amount of schillerisation. The rounded grains of iron-ore are embedded in the pyroxenes, and in the augite they form centres from which amphibolisation has started. These rocks are hornblende-hyperites approaching melanocratism, in which the amount of plagioclase and quartz is estimated to be only half that of the ferro-magnesian minerals (see Table I, 3, 4). They belong to Hotiann’s basic division of the char- nockite series. A similar rock (192) occurs near Andulo, and is distinguishable only by its finer grain. Ho.iannp’s ultrabasic division of the charnockite series is not represented in this collection, but a thoroughly melanocratie hornblende-hyperite from Ochilesa (176a) forms an approach to that division. The proportion of ferro-magnesian minerals to plagioclase in this rock is estimated at three to one (see Table I, 5); otherwise the rock differs little from the hyperite described above. This completes the description of the series comparable with the charnockites of India. One other rock undoubtedly belongs to this series, but has no Indian equiva- lent. This rock is from the W. of Chieuca (164), and is interstratified with well- foliated gneiss. It may be described as a charnockite-porphyry. It consists of small phenocrysts of quartz, oligoclase, orthoclase, augite, hypersthene, and biotite, pro- fusely scattered over a minute, evenly granular, microcrystalline groundmass com- posed mainly of quartz grains with subordinate water-clear felspar. The augite is well schillerised, and has fringes of greenish-brown hornblende. Biotite occurs in large clots consisting of small flakes felted with granules of augite and magnetite. It has exactly the appearance of the new-formed flakes in a hornfelsed rock, and the whole aspect of the rock suggests that it is secondary. Oligoclase forms the largest and most abundant set of phenocrysts. The rock appears to be the hypabyssal equivalent of the Ochilesa charnockite. The granulitic texture, the extreme fresh- ness, and the presence of abundant new-formed biotite, indicate that this rock has suffered some thermal metamorphism. It is linked by these characters, as well as others, to the hornfelsed porphyries described in the next paragraph. The quantitative mineralogical composition or mode of this series has been obtained by the Rosiwal method of measurement on five of the rock types, with the results shown in Table I. [TaBLE 542 MR G. W. TYRRELL. TaB_e I, : 1 2 3. 4 5 Quartz . ‘ : ; 39°2 19°8 2°5 tri 2°3 Orthoclase . : : 8°6 4°3 Oligoclase (Ab,An,) —. 28°0 ee Andesine (Ab,An,) ; ar 49-0 bes | ne wee Labradorite (Ab,An,) . e ts 304 | 32°4 22°8 Biotite . , ; : 10°7 86 coe / i See Pyroxenes . : 4 12°6 118 49:1 34°5 39°5 Hornblende . ‘ e fhe Se 16°6 29°5 28°6 Magnetite : 6 5°9 14 wer 6°8 Apatite : . : 3 6 aa af : : ' 758 731 32°9 33:93" 25°1 Felsic/Mafic ratio . : { ve 0 al pa 76 Summation in each case, 100. . Charnockite (177), Ochilesa. . Charnockite, intermediate variety (193), N. of Etunda. . Hornblende-hyperite (194a), 8. of Unyamba River, Andulo. . Hornblende-hyperite (1760), Ochilesa. . Melanocratic hornblende-hyperite (176a), Ochilesa. mm oF de By calculating the ratios of the felsic and mafic constituents,* it is seen that the series falls into two well-marked groups, one dofelsic, consisting of the charnockites proper and the intermediate varieties ; and other domafic, consisting of the hyperites. No rocks intermediate between these groups occur in the collection. The first group is mineralogically distinguished by the abundance of quartz and the presence of orthoclase and biotite. In the domafic group, quartz, while still present, is in ex- tremely small quantity, orthoclase and biotite have disappeared, and pyroxenes and hornblende constitute the bulk of the rocks. Whilst belonging to the ancient basement of the Benguella plateau, the char- * The terms felsic and mafic (Cross, Ippinas, Pirsson, and WASHINGTON, Journ. (feol., xx, 1912, p. 560) are used to denote two main groups of rock-forming minerals, one containing quartz, felspars, and felspathoids, and the other pyroxenes, amphiboles, micas, olivine, iron-ores, etc. Variations in the relative proportions of these groups give rise to the important variations in igneous rocks denoted qualitatively by BréaauR’s terms, leucocratic and melano- cratic. A quantitative meaning may be imparted to the terms felsic and mafic by attaching the prefixes do- and per-, with exactly the same connotation in respect to the mode (actual mineralogical composition) as the analogous terms persalic, dosalic, etc., have in respect to the norm (standard mineral composition) of the American Quantitative Classification. ‘ Felsic constituents Perfelsic Wako consuensenn u : Dofelsic % <= ; > : : Mafelsic ay < 2 > 3. Domafic “3 < 2 S . Permafic 5 < : A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 5438 nockite series is unquestionably intrusive into the gneisses and schists of the basement. : It is interesting to record that Lacrorx has described a very similar series of char- nockites from the Ivory Coast.* These rocks cover a great area in a region on the frontiers of French Guinea, the Ivory Coast, and Liberia. They form a series rang- ing from hypersthene-granite very poor in ferro-magnesian minerals, through quartz- norites, to norites witli 50 per cent. of hypersthene. These rocks are identical with those described above with the single exception that hornblende does not occur to any extent in the more basic types. The country rock is stated to be granitic. Ill. Hornretsep PoRPHYRIES AND OTHER HORNFELS. _ In this section is described a curious and puzzling group of rocks, which, though unmistakably igneous rocks of an acid character, appear to have suffered consider- able thermal metamorphism. The evidence for this lies in their perfect freshness, the recrystallisation of the groundmass into an evenly granular, water-clear mosaic, and the presence of abundant new-formed biotite, characters superposed upon their original igneous characters. The rock described as charnockite-porphyry, although of somewhat different mineral composition, has exactly the same characters as these rocks, and must be included in their discussion. It might also be remarked that the rocks of the charnockite series themselves present features, such as granularity, ideal freshness, and presence of biotite, which might be interpreted as affording evidence of thermal metamorphism. On the other hand, the mineral composition of these porphyries, especially the common occurrence of magnetite-cassiterite intergrowths, links them to the granites described later. Finally, associated with these porphyries, and indubitably of thermometamorphic origin, are rocks determined as scapolite- epidote-hornfels and cordierite-biotite-hornfels. In hand specimens these rocks may be described as “dark porphyries.” They generally show a fine-grained, but obviously crystalline groundmass of a dark-bluish colour, with the peculiar saccharoidal texture of hornfels and granulites, and contain dark phenocrysts of felspar. One or two examples are light-coloured, and are aplitic or halleflintas-like in appearance. The most typical example of these rocks is a specimen from N.E. of Bailundo (159). In thin section this shows phenocrysts of felspar together with peculiar clots or felts of hornblende, biotite, magnetite, and cassiterite, embedded in a microcrystal- line, granulitic mosaic consisting mostly of quartz, and flecked with anhedral shreds of biotite. A plagioclase felspar forms by far the most abundant set of phenocrysts. It is extremely zonal and is characterised by minute, indefinite, bent, and discon- tinuous twin-lamelle, which renders the exact determination of its composition * A, Laororx, “Sur existence 4 la Céte d’Ivorie d’un séries pétrog. comparable a celle de la charnockite,” Comptes Rendus, cl (1910), pp. 18-22. 544 MR G.’ W. TY REEEL, difficult. From the imperfect determinations possible it appears that the range of composition is from AbgsAns5 to Ab7z5Ang5, but the bulk of the felspar is probably near the latter composition. In thin section the felspars have a bluish tinge owing to the abundance of minute, dust-like inclusions. They appear to have suffered magmatic corrosion, as their angles are rounded off, and they have been worn into curious irregular shapes. . The next most abundant phenocryst is a deep-green, strongly pleochroic horn- blende forming irregular plates which poikilitically enclose numerous rounded grains of quartz. Frequently the crystals contain a core consisting of irregular grains of quartz and magnetite, devoid of the green hornblende. The plates frequently retain optical continuity, but occasionally they have been broken up into individual rounded grains intermixed with the quartz mosaic. Next to the hornblende in abundance are the magnetite-cassiterite intergrowths. These generally consist of a highly irregular core of magnetite invested by a thin sheath of cassiterite. The latter mineral is easily recognised by its uniaxial character, its well-marked twin-lamellation, and decided pleochroism from yellow to reddish-brown. Biotite is evenly spread as minute anhedral shreds amidst the quartz mosaic, but is more prominent as a constituent of curious clots or felts of the coloured minerals, which are a feature of the rock. These clots consist of the magnetite-cassiterite intergrowths, with numerous flakes of biotite, and the poikilitic hornblende, which occasionally envelops the other constituents. The groundmass consists of an even mosaic of quartz grains with subordinate water-clear untwinned felspar. The whole aspect of this rock, as described above, is suggestive of recrystallisation at a not very high temperature. It is clearly a hornfelsed porphyry belonging to the granodiorites, although quartz does not occur among the phenocrysts. A rock from N. of Huambo (148) differs from the above in several particulars. Amongst the phenocrysts quartz is abundant, and orthoclase is at least as abundant as oligoclase. Hornblende is absent, but large flakes of muscovite enter as a con- stituent of very closely felted clots consisting mostly of biotite flakes. The ground- mass is of the same nature as that of the Bailundo rock, but is of much denser texture. This rock is clearly a granite-porphyry. In another specimen from the same locality (145) the secondary biotite is very abundant, and epidote enters the biotite felts as a prominent constituent. From 3074 kilometres on the Benguella Railway (124) comes a specimen of the same nature, but the biotite forms wavy, sub-parallel streaks, which, in thin section, give a semi-foliated aspect to the rock. A rock from N. of Cruz’s (233) clearly belongs to the same group as the above. It contains, however, no quartz phenocrysts, and only a few of felspar. The latter are highly altered and are now little more than masses of granular epidote. The new-formed biotite is not aggregated into clots, but is scattered evenly over the - quartzose groundmass. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 545 Scapolite-epidote-hornfels.—This rock, which occurs E. of Chingwari (199), is a dark porphyry of appearance similar to that of the rocks described above. It consists mainly of a mosaic of quartz, biotite, and epidote, in which the first-named is perhaps the most abundant constituent. Magnetite and blue tourmaline form subordinate members of the mosaic. HKmbedded in this groundmass are occasional phenocrysts of andesine, witnessing to the original igneous character of the rock. These crystals frequently enclose the biotite and epidote of the groundmass. There are also numerous irregular areas of scapolite which are crowded with grains of the mosaic groundmass. ‘The mineral is easily recognised by its bright second-order polarisation tints, its cleavage, straight extinction, and uniaxial character. Whilst the poikilitic areas frequently retain optical continuity, they are occasionally broken up into grains with different orientations. The abundant biotite presents no unusual optical features, and is very closely intergrown with epidote. This rock is clearly the pro- duct of combined contact-metamorphism and pneumatolysis. Cordierite-biotite-hornfels—This rock occurs seven miles E. of Candumbo (230) and in hand specimen is a dark-bluish, finely crystalline granulite. In thin section it is seen to consist of a thoroughly granulitic groundmass of cordierite, quartz, together with abundant biotite, muscovite, and magnetite. The cordierite forms rounded sections, occasionally showing lamellar twinning, and the frequent development of a strong yellow pleochroism around minute inclusions. It is undergoing an incipient alteration to small flakes of white mica. The biotite isintergrown with muscovite. The magnetite is evenly distributed in minute subhedral grains throughout the rock. Occasional areas of turbid material with traces of multiple twinning suggest the presence of lime-soda felspar. The original nature of this rock is not so evident as in the foregoing. It was probably an impure sandstone or greywacke, and was almost certainly not of igneous origin, as it is devoid of the felspar phenocrysts characteristic of the types above described. The rocks described in this section appear to form part of the ancient basement of the Benguella plateau, into which the batholiths of tin-bearing granite described later were injected. They are interstratified with and injected into gneiss, and into the quartzites, quartz-schists, and sillimanite-schists of the Bailundo series. They formed a series of acid igneous rocks associated with various sediments. The presence of minerals such as cassiterite, tourmaline, and scapolite, suggest that the alteration they have undergone was due to pneumatolysis as much as to contact-metamorphism. The cassiterite-magnetite intergrowths of the Bailundo rock are identical with those occurring in the Benguella granites (p. 547), suggesting that this mineral association, together with scapolite and tourmaline, were pneumatolytic introductions derived from the Benguella granite batholiths, and that the peculiar modifications these rocks have undergoné were consequent upon the intrusion of these great masses of molten material. The characters of this group of rocks suggest comparison with the “leptites’ TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO, 14). 79 ? 546 MR G W. TYRRELL. and granulites of Finland and Sweden, which include rocks derived from various porphyries, along with cordierite, andalusite, and scapolite-bearing rocks.* Their characters are supposed to be due to “immersion” in a great granite batholith. It may also be noted that scapolite rocks occur along with the charnockite series in Peninsular India. t IV. GRANITES, GRANODIORITES, AND ASSOCIATED Rocks. These are fine- to coarse-grained rocks of white, pink, or grey colour, and of comparatively uniform texture. In only one specimen is there any tendency for felspar to assume a porphyritic development. Hornblende is macroscopically prominent in some of the granodiorite specimens. Under the microscope two quite distinct types may be distinguished. The most abundant group, that of the granites proper, are distinguished by an early generation of euhedral striped felspar approaching albite in composition, and usually much altered with the development of white mica. These crystals are prominent as white Opaque areas in hand specimens and slides. Rocks of this type occur on the Benguella Railway at Kunhungamua (138), Portella Coroteva (116), and Caimbango (123), in the district of Quingenge (248, 249, 252), on the Cuiva River (242), and W. of Chingwari (231). They consist of a coarse hypidiomorphic aggregate of fresh microcline and quartz, enclosing the above-mentioned early crystals of albite. The quartz also occurs as rounded blebs embedded in the microcline. The only ferro- magnesian constituents are biotite, partially altered to chlorite and epidote, and a little magnetite. The quantitative mineral composition of the Quingenge rock, as estimated by the Rosiwal method, is given in Table II. It may stand as a good average sample of the Benguella granites, which may therefore be said to consist of about 50 per cent. of microcline, 10 of albite, 35 of quartz, and 5 per cent. of biotite and iron-ores. The textures are typically granitic and vary in grain-size from coarse (Benguella Railway) to fine (Chingwari and the Cuiva River). The second type is a coarse granitoid rock indistinguishable in general appearance from those described above. In thin section, however, they are seen to contain no albite, the place of which is taken by abundant fresh plagioclase felspar of average composition Ab7Ang (acid andesine). They are further distinguished by the presence of hornblende along with biotite (although the latter remains the dominant coloured constituent), and by a larger proportion of the ferro-magnesian constituents. These rocks correspond to the Continental tonalites and to the American granodiorites. They occur at kilometres 135%, 223, and 238 on the Benguella Railway (119, 121, 123), and at a locality between Quingenge and the Cuiva River (244). The rock from — kilometres 223 on the Benguella Railway is a fresh typical sample of this type. It consists mainly of quartz and oligoclase-andesine in hypidiomorphic relations, with * P. Eskona, “ Petrology of the Orijarvi Region in South-western Finland,” Bull. Comm. Geol. de Finlande, No. 40, 1914, pp. 14, 15, 131. + HOLLAND, op. cit., p. 232. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 547 subordinate alkali felspar, mostly microcline, with a little orthoclase. The ferro- magnesian constituents are biotite, and a green hornblende, which, in the position of greatest absorption, shows a distinct tinge of blue. Magnetite, cassiterite, and apatite occur sparingly, and in one place are associated in a small intergrowth like those in the hornfelsed porphyry from Bailundo (see p. 544). Cassiterite also occurs in the rock from kilometres 238, Benguella Railway, where it is associated with little clots of biotite. Epidote and chlorite occur sparingly as secondary products derived from the alteration of biotite. The quantitative relations of the minerals, as determined by the Rosiwal method, are set out in Table II. A typical Benguella granodiorite therefore contains about 25 per cent. of quartz, 15 per cent. of alkali felspar, 50 per cent. of oligoclase-andesine, and 10 per cent. biotite, hornblende, and iron-ores. The typical American granodiorites are somewhat richer in the mafic constituents.”* Table II shows the contrast between the two types of granitic rocks in Benguella. The granites proper are rich in microcline and albite, whereas the granodiorites are poor in alkali felspar, and contain nearly 50 per cent. of oligoclase-andesine. Along with this they are decidedly poorer in quartz, and richer in the mafic constituents. The granodiorites contain cassiterite, but this mineral is in too small quantity to come under the Rosiwal measurement, and its amount must therefore be considerably less than 1 per cent. TaBLeE II. 1 - Df @uartz 4 - : : 359 26°5 Microcline and orthoclase . 47°3 15°4 Albite (Ab). P ; 10:5 Ax Andesine (Ab, An.) . : Site 48:4 Biotite and chlorite . : 3°9 5°5 Hornblende . : : bee 16 Magnetite : ‘ E 2°4 15 Epidote Ss et : 93°7 90°3 Fel fi { wen 90°3 elsic/Mafic 63 7 1. Biotite-granite, E. of Quingenge, Benguella (248). 2. Biotite-granodiorite, kilometres 238, Benguella Railway (123). Of especial interest in this connection is the occurrence of a stanniferous selvage to the Kunhungamua granite. This rock (139) is an even-grained microgranite con- sisting of quartz and orthoclase, with the latter distinctly in greater quantity. The orthoclase frequently has the mottled and moiré appearance which indicates a con- siderable content of the albite molecule. The dominant ferro-magnesian constituent is a greenish-brown biotite, but there are also a few large ragged patches of a bleached * “Geology and Ore Deposits of Butte, Mont.,” U.S. Geol. Surv., Prof. Paper No. 74, 1912, p. 36. 548 MR G. W. TYRRELL. green hornblende. Cassiterite occurs rather abundantly in spongy anhedral crystals which enclose quartz and felspar. It is yellowish-brown in colour, distinctly pleochroiec, and shows a fine lamellar twinning. It is commonly associated with small grains of magnetite. No other mineral of the tin association is found, except, perhaps, one faint purplish section doubtfully referred to axinite. The only specimen which may be regarded as the hypabyssal equivalent of the © | granites is a coarsely porphyritic rock from E. of Lepi (224), which shows numerous large white felspars in a dense greyish groundmass. In thin section the felspar phenocrysts are seen to form at least half the rock. They are much altered, with the formation of white mica and epidote. Hence they belong to an acid soda-lime felspar (probably oligoclase), a conclusion strengthened by the occasional retention of multiple twinning. These phenocrysts frequently have a ragged external zone of orthoclase which interpenetrates with the groundmass. The only other phenocrysts are large, partially chloritised crystals of hornblende. This mineral also appears in the dense groundmass, which, in the main, is a fine micrographic intergrowth of quartz and alkali felspar. Magnetite and large apatite crystals form rather abundant acces- sories. This rock is a granodiorite-porphyry, and is clearly to be associated with the Benguella granodiorites. It differs entirely in appearance from the hornfelsed and granulitic porphyries described in a previous section. V. RuyorrtE (DELLENITE). An extensive series of lava-flows belonging to this group occurs in the neigh- bourhood of Lepi and Quingenge, In hand specimens they are dense flinty rocks, generally light-grey in colour, occasionally dark-grey to black, with numerous, small, inconspicuous phenocrysts of quartz and felspar. In thin section they show numerous small phenocrysts of nae, orthoclase, oligoclase, and biotite, in a dense, cryptocrystalline groundmass. The quartz ranges from 1 to 3 mms. in diameter, and is subhedral, rounded, and embayed by corrosion. The felspars are euhedral, averaging 2 mms. in diameter, and mostly belong to ortho- clase. In some of the rocks, however, oligoclase becomes a prominent constituent, and nearly equals the orthoclase in quantity. Its composition ranges from oligoclase (AbsAnj) to oligoclase-andesine (AbsAns3). The biotite is bleached and decomposed, and is often represented by chloritic decomposition-products and epidote (257, 260, 261, Quingenge). The groundmass is dense, uniform, cryptocrystalline to glassy, and is besprinkled with minute unidentifiable microlites. A flow-orientation is indi- cated by faint differences of colour and texture in different parts of the groundmass. The proportion of phenocrysts to groundmass is estimated at one to two, and quartz is about equal to the felspars amongst the phenocrysts. In some rocks (245, Quin-— genge) the groundmass has apparently undergone silicification, which has also involved most of the felspars, leaving “ghosts” of the latter and much diffused sericitic matter. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 549 Some of the specimens from Lepi are compacted rhyolite-tuffs (225, E. of Lepi). These contain numerous fragments of quartz, oligoclase, orthoclase, and microcline, along with numerous angular fragments of microcrystalline quartz-porphyry, in a dense, cryptocrystalline groundmass. A rock generally similar to the rhyolites described above, from W. of Cruz’s, Quingenge (234), is much poorer in phenocrysts, and is devoid of phenocrystic quartz. A chemical analysis of one of these rocks (258, Quingenge) has been made for me by A. Scorr, M.A., B.Sc., with the result set out in Table II, 1. apie, Lol ] 2 3: 4 Si0, , a 68°98 72°60 66°91 69°48 TiO; : ’ 23 30 333} ve AO. : 15°24 13°88 16°62 13°88 HesOr . ; 2°83 1:43 2-44 2°67 FeO : ; 43 82 133 1:53 MnO . ’ 05 cit) 04 “SE MgO’... F 1:08 38 1°22 777 Il CaO ‘ 5 3°01 1:32 3:27 2 39 (Ba, Sr)O. tr. sine oe ‘8 Na,O_. : 3°88 3°54 4°13 3°74 K,O ; : 3°53 4-03 2°50 4:44 H,O+ . 53 | : : : 0) 39 | 52 1:13 1:19 BAO ; 09 06 08 CO, ; é nt. fd. Je aks 100:27 100-00 100-00 100°18 1. Rhyolite (dellenite), Toscanose-Lassenose (I’, 4, 2’, (3) 4), Quingenge, Benguella (258). Anal., A. Scorr. 2. Average rhyolite, quoted from Daty, Igneous Rocks and their Origin, 1914, p. 19. 3. Average dacite, ibid., p. 25. ; 4. Dellenite, Toscanose (I, ‘4, 2, 3), Dellen, Helsingland, Sweden. Quoted from Ippines, Zgneous Rocks, vol. ii, 1913, p. 128. Consideration of this analysis, combined with what is known of the mineral composi- tion, shows that the Quingenge rocks are the lavaform equivalents of the grano-diorite rather than of the granite family, and belong therefore to the group called dellenite by Bréccrr.* The toscanites of Italy are chemically, but not mineralogically, similar to the Benguella type. The averages of sixty-four analyses of rhyolite, and of thirty dacites, as calculated by Daty,j are given for comparison in Table III. The average rhyolite is richer in silica, poorer in lime, and has a slightly greater content of alkalies in which potash is predominant, than the Quingenge rock. On the other * See Ippinas, Igneous Rocks, vol. 1i, 1913, p, 104. + Igneous Rocks and their Origin, 1914, pp. 19, 25. 550 MR G. W. TYRRELL. hand, the average dacite is poorer in silica, and has a decided predominance of soda over potash. In most respects, therefore, the Quingenge rhyolite is intermediate between average rhyolite and dacite. The analysis of the type dellenite from Sweden (Table III, 4) is very similar to that of the Quingenge rock. It is, however, richer in potash and poorer in lime. The Swedish rock contains hypersthene, and was originally called hypersthene-andesite ; but the analysis shows that this identifica- tion was completely astray. It is worth while comparing the norms of these rocks (Table IV). TaBLE IV. 1 2. Quartz . : 5 ‘ 25°4 25°6 Orthoclase 2 ; ; 20°6 26°1 Albite . : ; ‘ 33:0 30°9 Anorthite : : 3 13°6 8:3 Femic minerals, water, etc. 74 gel! 1. Rhyolite (dellenite), Quingenge, Benguella. 2. Dellenite, Dellen, Helsingland, Sweden. The norms correspond very closely, especially in regard to the salic constituents, to the modes that would have been developed had the rocks been holocrystalline. The dellenite is shown to be somewhat richer in orthoclase and poorer in oligoclase than the Quingenge rock, but the figures are sufficiently close to justify the identi- fication of the latter as dellenite. I have not been able to find records of ancient African rhyolites similar to those of Benguella, but some of the Tertiary rhyolites of Algeria approach them closely in chemical and mineral composition. In the American Quantitative Classification the dellenite of Quingenge falls into the subrang lassenose, but is transitional to toscanose. The type dellenite of Sweden falls into toscanose. VI. Tue ALKALINE SERIES. An interesting series of alkaline rocks occurs in Benguella in the district of Ochilesa, the extreme north-eastern point reached by Professor Grecory. At Chieuca these rocks are found within a volcanic vent, associated with tuffs and igneous breccias composed of the same materials; but there are also intrusive masses within the district which are not in actuai connection with the vent. The principal alkaline types found in the collection are nepheline-sodalite-syenite and other types of syenite, shonkinite, solvsbergite, and ouachitite, which are treated in order below. 1. Syenites. A. Nepheline-sodalite-syenite.—This type occurs as an intrusion at Ochilesa (209). It is a thoroughly leucocratie or perfelsic, coarse-grained rock, consisting mw ainly of A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. sor broad, simply-twinned anhedrons of soda-orthoclase, which are irregularly mottled in polarised light, together with subordinate nepheline and sodalite.. The nepheline is distinguished by its lower interference tints, and total absence of the mottling which characterises the orthoclase, as well as by refractive index and interference figure tests. The sodalite has occasionally a faint bluish tinge, and is crowded with minute bubbles and other inclusions. It is identical with a mineral which occurs in the shonkinite described later, and which has been determined by chemical means. The ferro-magnesian constituents consist of ragged anhedrons of a_pleochroic greenish-brown hornblende, which envelops small tablets of biotite, and occasionally contains a small core of almost colourless pyroxene. Irregular grains of egirine are frequently inserted around the margins of the crystals. Apatite and magnetite are also abundant as inclusions in the amphibole sections. The pleochroism of the mineral is from pale yellowish-green to a light brown-red which has a tinge of violet. There is always an irregular marginal zone in the crystals in which the colour becomes bluish-green. The extinction in most sections is in the neighbourhood of 17°, but may range up to 31°. The mineral is probably a soda-amphibole belonging to the katoforite-arfvedsonite series, or possibly to keersutite. girine also occurs as small anhedrons embedded in the felspars. An estimate of the relative propor- tions of the constituents is: soda-orthoclase, 80 per cent.; nepheline and sodalite, 10 per cent. ; hornblende, augite, biotite, etc., 10 per cent. B. Sodalite-syenite.—A second variety of syenite occurs in the volcanic vent of Chieuca (163). It differs from the above in its much finer grain, in the substitution of a pale-purplish titanaugite for hornblende as its principal mafic constituent, and in a smaller proportion of nepheline and sodalite. The titanaugite encloses biotite tablets, apatite, and magnetite ; and is in all stages of alteration to granular egirine- augite. The mass of the rock, however, is soda-orthoclase, with a very subordinate amount of euhedral nepheline and sodalite. The latter mineral penetrates the felspars in channels directed by the cleavages, as though it had made its way by corrosion. Sphene is an important accessory constituent. Decomposed specimens of this type, in which the pyroxene has completely passed over to aggregates of fibrous amphibole, and the felspar completely “ kaolinised,” occur in the Ochilesa district (208a, 209e, 219/). C. Akerite.—An alkali-syenite containing neither nepheline nor sodalite occurs to the W. of Ochilesa (1726). It consists mainly of anhedrons of mottled soda- orthoclase, with oligoclase (AbsAn) in little inferior quantity. Buiotite forms the principal ferro-magnesian constituent, but is accompanied by a little brown hornblende similar to that of the nepheline-sodalite-syenite of Ochilesa described above. These erystals may contain a core of nearly colourless pyroxene. Crystals of apatite and magnetite are common enclosures within the ferro-magnesian minerals. This rock has a close affinity with the akerite of Norway, although the principal ferro-magnesian constituent of the latter is augite. 552 MR G. W. TYRRELL. Nepheline-syenites appear to be widely distributed in Central Africa. Bera * has described a nepheline-sodalite-syenite from Senze do Itombe, and Hotmzsf has recently described a nepheline-syenite from the same locality in north-western Loanda, the northern province of Angola. This rock contains zegirine-augite as the principal ferro- magnesian constituent, but this mineral is frequently bordered by amphiboles which are identified with hastingsite or a mineral of the katoforite-arfvedsonite series. 2. Shonkinite. This remarkable rock occurs in the Chieuca volcanic vent (170b, 170g), and at Ochilesa (172a, 2086, 208c). It appears to be the dominant type in this volcanic district. It is a fine-grained rock of bluish-grey colour and doleritic appearance, and is besprinkled with numerous small black crystals of hornblende ranging up to ¢ inch in diameter. In thin section the rock is ideally fresh, and consists mainly of orthoclase with interstitial sodalite, and a comparatively large quantity of mafic minerals which include titanaugite, barkevikite, and magnetite. Apatite is an important accessory. The texture of the rock is thoroughly granitoid or hypidiomorphic. Many of the augite and hornblende crystals are larger than those of the orthoclase composing the mass of the rock, but their porphyritic character is not so apparent in thin section as in the hand specimen. ‘The felsic constituents poikilitically enclose numerous small crystals of titanaugite, magnetite, and apatite, which doubtless gives the rock its dark colour. The orthoclase forms elongated prisms which are simply twinned, or it builds more equidimensional subhedral crystals which contain numerous, evenly spaced, parallel rows of very minute bubbles, apparently of liquid. These rows are perpendicular to a direction of cleavage within the mineral, and are also parallel to a direction of extinction. A pale-bluish sodalite fills. up many interspaces, and is frequently crowded with minute bubbles and inclusions. The presence of sodalite was confirmed by obtaining a silver chloride precipitate in a nitric acid digest of the rock, after removing apatite by means of a heavy liquid. The felsic minerals form a sort of coarse groundmass to numerous, pseudo-porphy- ritic, perfectly euhedral crystals of barkevikite and titanaugite, and also poikilitically enclose a swarm of small grains of the same minerals along with apatite and magnetite. The barkevikite and titanaugite are in roughly equal amount. The former occurs in fine, euhedral, red-brown crystals, occasionally with simple twinning, and narrow darker-tinted borders. In some of the rocks considerable resorption of the hornblende has taken place, the final appearance being of an irregular mass of magnetite mixed with granules of a greenish amphibole, the whole being surrounded by a dense corona of magnetite granules. The pyroxene has occasionally undergone a similar alteration, but to a lesser extent. The augite is of the purplish tint which is held to indicate a * Tscherm. min, petr. Mitth., vol. xxii (1903), p. 359. t Geol, Mag., dec. vi, vol. ii (1915), p. 323. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 553 high content of soda and titania. It is zonal, with hour-glass structure, faint pleo- chroism, and strong dispersion of the bisectrices. In some slides the’ smaller crystals, and the borders of the larger ones, have a greenish tinge, indicating some admixture of the egirine molecule. Apatite occurs abundantly in large dusty erystals with rounded corners. In some, an addition of fresh material has erown around them with sharp crystalline boundaries, in a manner which recalls the quartz erystals of the Penrith sandstone. A little sphene occurs in the usual wedge-shaped erystals ; and there is frequently a border of granular sphene fringing irregular masses of titaniferous iron-ore. The affinities of this rock are clearly with the type described as shonkinite by Presson. In his latest work Rosensuscn™* defines shonkinite as a hypidiomorphic granular rock of the alkaline series, characterised by the mineral combination potash- felspar and nepheline, and by the predominance of coloured constituents. This definition would make it the melanocratic facies of the nepheline-syenite group. Ippines { defines the rock as a phanerite characterised by alkali-felspar and equal or nearly equal amounts of mafic minerals, with small but notable amounts of fels- pathoids such as nepheline or sodalite; and he treats it as the mesocratic facies of nepheline-syenites. In the original shonkinite of Square Butte, Montana, nepheline, along with sodalite, was regarded as a rare accessory mineral, and was, indeéd, only detected by chemical means.{ The rock was thus regarded as the mesocratic or melanocratic facies of alkali-syenites. The later work of Prrsson showed, however, that his type shonkinite contained 5 per cent. of nepheline and 1 per cent. of sodalite.§ The term shonkinite, in fact, has been used to cover rocks which belong to the meso- eratic and melanocratic divisions of both alkali-syenites and nepheline-syenites. Recognising this, Lacrorx has proposed to restrict the term to the melanocratic types of the alkali-syenite family,|| and to use another term for the equivalent rocks of the nepheline-syenite family.1 In his latest description RosenspuscH himself speaks of “the typical, nepheline-poor shonkinite of Montana.”** The Ochilesa rock falls most naturally under the original definition of the term. The quantitative mineral composition of the Ochilesa shonkinite has been determined by the Rosiwal method, with the result shown in Table V, 1. It is * Gesteinlehre, 3rd ed., 1910, p. 199. + Igneous Rocks, vol. ii, 1913, p. 230. { Weep and Pirsson, Bull. Geol. Soc. America, vol. vi, 1895, p. 415. § Presson, “Geol. and Petr. of Highwood Mountains, Mont.,” U.S. Geol. Surv., Bull. 237, 1905, p. 104, || Lacrorx, Nouv. Arch. du Mus., 1902, p. 179. ‘| Lacrorx proposes to use the term theralite in this sense. This, however, ignores RosenBuscu’s plain intention of using this term for the plagioclase-nepheline rocks of plutonic habit. It is true that the original “theralite” of the Crazy Mountains, Montana, on which Rosmnsuscu erected the group, was erroneously described by WouFF as carrying plagioclase, and is in fact a shonkinite in the sense of RosenpuscH and Ippinas. LacRorx’s usage of the term thera- lite, however, merely perpetuates the original mistake in the diagnosis of the Crazy Mountain rocks. When RosENBUSCH discovered the mistake, he promptly withdrew these rocks from the theralite group, and placed them under shonkinite. He then created as his type theralite the unquestionable plagioclase-nepheline rock of Duppau in Bohemia, thus preserving his term theralite with its original connotation. ** Loe, cit., p. 208. TRANS. ROY. SOU. EDIN., VOL. LI, PART III (NO. 14). 80 554 MR G. W. TYRRELL, there compared with the type shonkinite of Square Butte, Montana, and with a shonkinitic syenite from Middle Peak in the same district. TaBLE V. 1 2. 3 Alkali-felspar* . 61:0 20 53:2 Nepheline : oS, 5 = Sodalite . : 2-0 1 SH Biotite . : : aoe 8 4:0 Titanaugite : 14:1 46 32°3 Barkevikite .. : 13°4 oa 2 Olivine : : be 10 Jeg Tron-ores + é ] 165 6 7:9 Sphene : ‘ 6 os Apatite. : ‘ 1-4 4 9 1. Shonkinite, Ochilesa, Benguella. 2. Shonkinite, Square Butte, Montana. Quoted from Pirsson, U.S. Geol. Surv., Bull. 237, 1905, p. 104. 3. Shonkinitic syenite, Middle Peak, Highwood Mountains, Montana ; zbid., p. 93. The Square Butte rock is much richer in the mafic constituents than that of Ochilesa, which resembles the Middle Peak shonkinitic syenite much more closely. As is shown in the discussion of the chemical analyses, the type shonkinite of Square Butte tends now to become an extreme type as compared with the average of the rocks which have been grouped as shonkinites; and the Ochilesa rock is very close to the average shonkinite. It differs from all previously described shonkinites in containing abundant hornblende. It compares well with the Montana types in con- taining a little sodalite, and in being associated with sodalite-syenites. The chemical composition of the Ochilesa shonkinite has been kindly determined for me by A. Scorr, M.A., B.Sc. The figures are set out in Table VI, 1; and are there compared with the type shonkinite of Square Butte, and with the average shonkinite. The composition of the latter has been derived from the average of — twenty-one analyses of shonkinitic rocks found in Ippines’ Igneous Rocks, vol. ii (1913), and in Rosensuscn’s Gesteinlehre, 3rd ed., 1910. In the collection is included the shonkinites of Montana, Arkansas, Celebes, and the Katzenbuckel, Odenwald ; also the malignites (nepheline-pyroxene-garnet-shonkinites) of Ontario and British Columbia ; and the marosites (biotite-augite-shonkinites) of Celebes. The range of rocks covered by this average corresponds to that covered by Rosenpuscy’s defini- tion and descriptions, and consequently includes nepheline-bearing types. Certain “ microshonkinites” described by Lacrorx from the Los Islands, West Africa, are excluded, since they are far too rich in nepheline to come under the heading of _— shonkinite. * Mostly soda-orthoclase. + Magnetite and ilmenite. -A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 555 TABLE VI. 1 2. 3 SiOu eal 4741 46-73 47°46 MMOs? b: ; 1°86 78 1:14 ATOR 12°30 10:05 14:28 Fe,0, . 6:99 3°53 4°16 Reo. 5:03 8:20 4:97 MnO a‘ : ‘14 ‘28 19 MeO i 6-82 9:27 6-64 CaO — . ry 9°57 13°22 9°42 | (Ba, Sr)O Shite sip ste Ol. Ne Na,O . 3:06 IES 7, 3°86 Oni (ac Mi 4°05 376 | 4°65 ee Ene | 1:24 1:24 | Os = Bi no Foe Leen BOs.” 1°62 1°51 1:02 | cl Ce oe Eee é a 21 ape 79 CO, ; : | nt. fd. ho ( | 100-68 10056 100-17 Less 0=Cl, F. -10 04 100°58 100°52 1. Shonkinite, Zamarose (IIT, 5’, 2, 3), Ochilesa, Benguella. 2, Shonkinite, Shonkinose (ITT’, 6, 2, ‘3), Square Butte, Montana. Quoted from Prrsson, U.S. Geol. Surv., Bull. 237, 1905, p. 102. 3. Average shonkinite, Shonkinose (III, 6, 2, 3). The Square Butte shonkinite is richer in lime and magnesia, and poorer in alkalis, than the Ochilesa rock or the average shonkinite, thereby revealing its more mafic character. The analysis of the average shonkinite, however, is extraordinarily like that of the Ochilesa type. It is somewhat richer in alkalies and alumina, but the correspondence of the other constituents is very close. A notable feature of most analyses of shonkinite, including the Ochilesa rock, is their abundance in phosphorus pentoxide, which constituent commonly reaches 1°5 per cent., indicating about 33 per cent. of apatite. The Ochilesa shonkinite falls into the subrang ILI, 5, 2, 3 (lamarose) of the American Quantitative Classification, but is intermediate towards the next subrang, III, 6, 2, 3 (shonkinose), in which the typical shonkinites are found. The average shonkinite falls almost centrally into the subrang shonkinose. Light-coloured veins of alkali-syenite-aplite (208c) occur in the shonkinite of Ochilesa. On the other hand, a boulder of sodalite-syenite from the volcanic vent of Chieuca shows a vein of microshonkinite or shonkinite-aplite. It is unnecessary to describe these rocks in detail, as they are merely fine-grained modifications of the rocks above described. 556 MR G. W. TYRRELL. 3. Solusbergite. The voleanic vent of Chieuca is filled with tuff and agglomerate consisting of boulders, pebbles, and comminuted débris of shonkinite, sodalite-syenite, and grey or green compact rocks identified as solvsbergite. It is possible that the latter rock also occurs as small intrusive masses or as lava-flows. The principal rocks of the vent may be described as shonkinite-tuff (170h), and as solvsbergite-tuff (170c, 170/). Many specimens of the solvsbergite were collected (170, 170a, 170e, 162b); and amongst these three varieties may be distinguished. One contains large phenocrysts of soda-orthoclase, oligoclase (AbsAn,) diopside with mantles of zgirine-augite, and small crystals of egirine-augite, in a dense, felted, trachytic groundmass consisting of minute laths of orthoclase, granules of green pyroxene and magnetite, with some ™ indeterminate material. A few, small, partially resorbed crystals of barkevikite occur. In the second variety the barkevikite constitutes the only phenocrystic mineral apart from small crystals of szgirime-augite, and this type may therefore be called hornblende-solvsbergite. The groundmass here contains abundant egirine and a little nepheline, and the rock is therefore transitional to tinguaite. The barkevikite has suffered considerable resorption ; the crystals are always surrounded by a dense corona of granules of green pyroxene. Frequently the core of barkevikite has entirely disappeared. A third type is characterised by a colourless glassy groundmass crowded with acicular green microlites. It carries small and sparse phenocrysts of euhedral barkevikite, eegirine-augite, orthoclase, and a little nepheline. In some of the slides are small rounded areas of sodalite. The glassy base is probably highly alkaline, and the rocks may be termed solvsbergite- and tinguaite-vitrophyres. These rocks un- doubtedly represent the hypabyssal or volcanic phases of the syenites described above. The lavaform equivalent of the shonkinite does not appear in the collection, although it probably occurs in the field. 4, Ouachitite. A remarkable rock belonging to the alkaline series occurs as dykes in the sodalite- syenite of Ochilesa (209). In hand specimens it is a dark-green, grey, or brown, compact rock, which effervesces freely with acid, and in which the only identifiable constituent is biotite. It is characterised by an irregular banding due to the alterna- tion of strips differing slightly in texture, colour, or mineral composition. In thin section the bulk of the rock is seen to consist of biotite and a grass-green pyroxene, embedded in a groundmass consisting of allotriomorphic plates of calcite. Anatase derived from ilmenite, and apatite, are important accessories ; whilst soda-orthoclase, and a mineral identified as melilite, occasionally appear in the slides. The biotite forms large plates of a bleached yellow tint, with much darker borders. Large cross- sections ranging up to 5 mms. in length are frequent, as are also good basal sections. - It is extensively altered to a peculiar yellow mineral, which is usually granular in A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 557 texture, but occasionally becomes massive and then shows aggregate polarisation. This mineral retains the cleavage of a mica, although it is not so perfect as in the biotite. The double refraction is rather lower than that of biotite, but its pleochroism is usually very feeble. In some sections, however, a pleochroism from orange to a pale-green was observed. The mineral is sensibly uniaxial. These characters agree on the whole with those of the penninite variety of chlorite. In the text-books, how- ever, the double refraction of penninite is stated to be extremely low, whereas the mineral in question has a double refraction not much lower than that of biotite itself. The pyroxene occurs in small euhedral crystals, or as granular aggregates em- bedded in the plates of calcite and biotite. It is grass-green in colour, with only a slight pleochroism from green toa yellowish-green. The cross-sections are frequently rectangular, with well-marked cleavages parallel to the rectangular sides. There is occasionally a slight truncation of the corners giving rise to octagonal sections. The extinction diagonally bisects the cleavages and the rectangular sections. The crystals are therefore prismatic with only a feeble development of the pinacoid faces. There is great dispersion of the bisectrices, and some of the sections fail to give perfect extinction in white light. Many of the pyroxene crystals are mere shells, the interiors having been entirely replaced by calcite. The biotite and augite are embedded in a groundmass of large allotriomorphic plates of calcite. Within the latter, or enveloping biotite or augite, are found areas of a colourless mineral with very low double refraction and a dark-blue polarisation tint. There is occasionally some approach to a prismatic form, and on the thin edge of sections a cleavage may be observed parallel to the direction of elongation, which is also the negative or “fast-ray” direction. The extinction is straight in regard to the cleavage, the mineral is optically uniaxial, and has a refractive index considerably higher than that of Canada balsam. These characters agree with those of melilite, but the characteristic peg structure of this mineral has not been observed. There are numerous irregular patches of a black mineral, which in very thin section becomes translucent and of a beautiful blue colour. The identification of this mineral as anatase was put beyond doubt by finding it growing, in small rectangular tablets, out of decomposing ilmenite. Apatite is also very abundant in small euhedral crystals. In one of the slides (209d) one of the bands or schlieren consists of biotite and granular pyroxene in a groundmass consisting entirely of grains of apatite. In another section a considerable amount of soda-orthoclase appears in large anhedral plates, with which the biotite forms peculiar poikilitic inter- growths. Notwithstanding the alteration of the rock, and the rather uncertain identification of melilite, the general character of this rock is similar to that of the alnoite group, as defined by RosenBuscu,* and it is here provisionally classed with that group. The type alnéite of Alno, Sweden, differs from the rock here described only in containing serpentinised olivine, garnet, and perofskite as the titaniferous * ROSENBUSCH, Gesteinlehre, 1910, p. 304. 558 MR G. W. TYRRELL. mineral in place of ilmenite and anatase. It is very rich in apatite, and contains abundant calcite as the result of the alteration of the augite and melilite. In the _ absence of olivine and garnet, the Ochilesa rock has closer affinities with the biotite- monchiquite or ouachitite of Arkansas than to the type alndite. According to RosENBuUSCH* the ouachitite of Hot Springs, Arkansas, contains a small quantity of melilite, and is free from garnet. It is included by him in the alndite group. VII. Basic IntRvsIons. A group of basic rocks including dolerites and basalts is widely spread over the area traversed by Professor Grecory. Their geological relations suggest that they are the youngest set of igneous rocks in the region. t A rock from the railway W. of Lepi (127) is a coarse-grained ophitic olivine- dolerite, consisting of a plexus of broad laths of labradorite (AbyAnq) in ophitic re- lations with irregular plates of brown augite. There is also some anhedral olivine, magnetite-blackened, and partly altered to colourless serpentine ; and titaniferous iron-ore. The rock is mesocratic or mafelsic, 7.e. the felspars are roughly equal in amount to the ferro-magnesian minerals. A similar rock forms a dyke W. of Cruz’s, Quingenge (236). It is, however, finer in grain, and the iron-ore is more abundant, than in the Lepi rock. The augite has a marked purplish tone. Rocks from near Cubal (300c) and Portella Solo (118) are very basic basalts approaching augitite. The Cubal rock contains numerous large phenocrysts of brown augite in a fine-grained groundmass consisting of augite and plagioclase, of which the former is by far the more abundant. The texture is obscurely ophitic, and the augite phenocrysts enclose very small laths of plagioclase. The only remaining consti- tuents are ilmenite passing over to leucoxene, and small green patches which may represent a little original olivine. The rock from Portella Solo differs from the above only in its non-porphyritic character and the slightly greater proportion of the felspar. It occurs at the margin of a granite mass, and appears to have been re- crystallised as a basalt-hornfels, with little or no new mineral formation. Both these rocks are practically devoid of olivine, and in the abundance of augite approach the augitite end-member of the basalt family. EXPLANATION OF PLATE. Fig. 1. Charnockite, Ochilesa. Ordinary light, x 15. A granulitic aggregate of clear quartz, cloudy felspars, with biotite (centre of field) and pyroxenes (below biotite). The latter is mostly hypersthene. _ Fig. 2. Hornblende-hyperite, Ochilesa. Ordinary light, x 15. A granulitic aggregate of felspar (plagioclase) with a little quartz, abundant hypersthene, and hornblende. ,_ The hypersthene is easily dis- tinguishable by its comparatively high relief. * Rospnsusen, Gesteinlehre, 1910, p. 305. A CONTRIBUTION TO THE PETROGRAPHY OF BENGUELLA. 5959 Fig. 3. Hornfelsed porphyry, Bailundo. Ordinary light, x 15. A recrystallised granulitic quartzo- felspathic groundmass, which contains irregular clots of intergrown hornblende, biotite, magnetite, and occasionally cassiterite ; and cloudy phenocrysts of felspar (orthoclase and oligoclase). Fig. 4. Shonkinite, Ochilesa. Ordinary light, x 15. Shows pseudo-porphyritic euhedral crystals of barkevikite (centre), and titanaugite (south-eastern quadrant), in a groundmass consisting of orthoclase, and a little sodalite which is not distinguishable in the photograph. The felsic minerals contain numerous small inclusions of titanaugite and barkevikite, and apatite is enclosed within the coloured minerals. Fig. 5. Solvsbergite-vitrophyre, Chieuca, Ordinary light, x 24. Shows small phenocrysts of egirine. augite and magnetite (centre of field), in colourless glassy groundmass crowded with microlites of egirine. Fig. 6. Ouachitite, Ochilesa. Ordinary light, x 15. Shows numerous small euhedral sections of green augite, and large plates of bleached biotite (centre of field), in groundmass of calcite (colourless areas), and a little melilite (?) which cannot be distinguished in the photograph. The irregular black sections are of blue translucent anatase which has been derived by the alteration of ilmenite. «¥ ANS. Roy. Soc. EDIN. VoL. LI. G. W. TYRRELL: PETROGRAPHY OF BENGUELLA.—PL. I. i Ss ss i CPsel 4) XV.—On some Cretaceous Brachiopoda and Mollusca from Angola, Portuguese West Africa. By R. Bullen Newton, F.G.8., Geological Department, British Museum.* Communicated by Professor J. W. Grecory, F.R.S. (With Two Plates.) (MS. received June 2, 1915. Read June 28,1915. Issued separately August 14, 1916.) CONTENTS. PAGE Introduction . : ; : : : : ‘ : : : ; . 561 Distribution Table 563 Description of Specimens :— Brachiopoda . : ‘ : : : : , ; ; ; . 563 Pelecypoda_ . : : : : : , ; : : : . 564 Gastropoda . : ; : : ‘ : . : : ; . 576 Explanation of Plates : ‘ E : . ‘ ; : : : s > OS) INTRODUCTION. The following notes have been prepared as the result of a study of a small collection of fossils belonging chiefly to the Pelecypoda and Gastropoda, which were obtained by Professor J. W. Grecory, F.R.S., from the Cretaceous rocks of the neighbour- hood of Lobito Bay, situated to the north of Benguella, in the province of Angola. At first sight the specimens appeared to be of too unsatisfactory a character for determination, on account of their fragmentary condition, besides often consisting of natural casts. From a closer examination, however, it has been possible to trace some details of structure which have afforded clues to the identification of certain genera and species, a record of which will assist in extending our present limited knowledge of the Cretaceous conchology of this region of Africa. During the progress of these researches, a few further specimens from the same locality were added to Professor Grecory’s collection, having been specially obtained by Myr EH. Rosins, of the Benguella Railway Company. So far as the literature of the subject is concerned, we appear to be indebted to the memoirs of Professor Paun Cuorrat + of Lisbon for most of our information on the Cretaceous fauna of Angola, although reference should also be made to a small and important paper by M. Srantstas Meunter } issued in 1888, which contains the earliest known figures of * Published by permission of the Trustees of the British Museum. + “Note préliminaire sur les Fossiles recueillis par M. Lourengo Malbeiro dans la province d’Angola,” Bull. Soc. Géol. France, 1887, sér. 3, vol. xv, pp. 154-157 ; “ Matériaux pour l'étude stratigraphique et paléontologique de la province d’Angola,” Mém. Soc. Phys. Hist. Nat. Genéve, 1888, vol. xxx, No. 2; “ Nouvelles données sur la zone littorale d’Angola,” Contrib. Géol. Colonies Portugaises Afrique—Com. Serv. Géol. Portugal, 1905. t “Contribution 4 la géologie de Afrique occidentale,” Dull. Soc. Géol. Hrance, 1888, sér. 3, vol. xvi, pl. 1, figs. 1-5, pp. 61-63. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 15), 81 562 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA Cretaceous mollusca (Cephalopoda) from Angola, consisting of Schloenbachia inflata, J. Sowerby, Desmoceras cuverviller, Stan. Meun., Hamites virgulatus, Brongniart, and H. tropicalis, Stan. Meun. These specimens, obtained from the limestones of Lobito Bay, north of Benguella, were considered to be of Albian age, the forms of S. inflata being regarded as identical with a variety of that species which had been figured by Szasnocua* from the Elobi Islands, situated off the north-west territory of the French Congo. From an examination of the mollusca and other organisms, Cuorrat divided the Cretaceous rocks of Angola into :—Albian (beds with Pholadomya plewromyx- formis); Vraconnian (beds with Schloenbachia inflata); Cenomanian—Turonian (beds with Cyprina invensi, Actxonella, Nerinea); and Senonian (beds with Ostrea olisiponensis, Roudaireia forbesi, Ostrea bayler, Cardita barroneti, and Inoceramus lang). . In the light of these researches, and from a study of the present fossils, it is recognised that the fauna here described belongs to the Vraconnian stage of the Cretaceous series, which includes the zone of Schloenbachia inflata. Strong con- firmation of this horizon is offered by the Cephalopod remains forming part of the collection, and which are in course of description by Mr G. C. Crick, as there are certain Ammonites showing close relationships with Schloenbachia inflata. Whether the zone should be regarded as the lowest Cenomanian or latest Albian need not now be discussed, it being sufficient for the purposes of this paper to acknowledge that it belongs to RENEVIER’S Vraconnian stage as adopted by Cuorrat, and so representing an intermediate or passage series of beds between the Albian and Cenomanian. The Angola fauna undoubtedly bears the European facies, being, besides, related in its details to that characterising North African regions, a few of the species occurring as well in Syria, Socotra, and Zululand. It is connected, likewise, with the southern Indian Cretaceous series on account of the presence there, at the base of the Utatur group, of Schloenbachia inflata. Further, a resemblance may be noted with the fauna of the Sergipe beds of Brazil described by C. A. Wurrn,t and which has already been pointed out by Dr F. Kossmat,{ who, on account particularly of the presence of Ammonites of the group S. inflata, regarded those deposits as of Cenomanian age. Certain specimens of the present collection help, also, to confirm the existence of a Brazilian facies in the Angola fauna, among which may be mentioned :—Neithea xquicostata, N. tricostata, Trigoma crenulata, Tylostoma globosum, and Akera gregoryr. The distribution of the species referred to in this paper is set out in the following table :— * “Zur Kenntniss der mitteleretacischen Cephalopodenfauna der Insel Elobi an der Westkiiste Afrika’s,” Denkschr. k. Akad, Wiss, [Wien], 1884-1885, vol. xlix, p. 231, with 4 plates. ; + “Contributions to the Paleontology of Brazil,” Archiv Mus. Nacion. Brazil, 1888, vol. vii. t“On the Importance of the Cretaceous Rocks of Southern India, etc.,” Records Geol. Surv. India, 1895, vol. xxviii, p. 45. [=English translation of memoir issued in the Jahrb, k.k, Geol. Reichs., 1894, vol. xliv.] AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. 5638 Inst of Cretaceous Mollusca (excluding Cephalopoda), ete., from Angola, collected by Prof. Gregory and Mr EK. Robins. Geographical Distribution. Genera and Species. Ceol ogiea) sles «| le Distribution. Si ecy lesa.) 26 el ES oe ae Oo} 8 || ea). |S] a) 3] |) Sole) Se ee le 1) aie RP icp Ne pee eerene Sh =i Ee 2 ‘ | Z BracHIopopA— * Terebratula depressa, Lamarck . A ptian- Vraconnian x x PELECYPODA— Exrogyra cf. flabellatu, Goldfuss Cenomanian-Turonian | x | x | x | x | x | x x Neithea xquicostata, Lamarck . zy x x pau * 4, angoliensis (sp. nov.) . Vraconnian x * 4, quadricostata, J. Sowerby Albian-Cenomanian x x x » tricostata, Coquand . | Cenomanian 5 aul i x * Lima cf. itiercana, Pictet and Roux. | Albian-Cenomanian x x ll Bx Volsella sp. (= Modiola) . Vraconnian x * Trigonarca cf. ligeriensis, Orbigny Cenomanian x || 36 me x “a 5 ef. diceras, Seguenza on S| S21 Se x * Trigonia crenulata, Lamarck 5 54) Se il x x * Anthonya cf. baudeti, Coquand a x || Seal] x * Trachycardium cf. syriacum, Conrad if : x x * Panopea cf. plicata, J. de C. Sowerby | Aptian-Cenomanian x x x * Pholadomya cf. vignesi, Lartet. Cenomanian-Senomian | x | x | x ales allans Toucasia sp. ‘ Vraconnian x GasTROPODA— * Tylostoma globosum, D. Sharpe Cenomanian-Turonian | x x ellie * Mrhilaia cf. nerineformis, Coquand . | Cenomanian 4 ill SS 1 3 x * Hipponyx sp. . < : Vraconnian x * Akera gregory? (sp. nov.) - x x ” Note.—The asterisks (*) to the left of the above names indicate new records for the Cretaceous of Angola. DESCRIPTION OF THE SPECIMENS. BRACHIOPODA. Fam. TEREBRATULID. Terebratula depressa, Lamarck. (PI. I, fig. 1.) Terebratula depressa, Lamarck: Hist. Nat. Anim. sans Vert., 1819, vol. vi, part 1, p. 249; Davidson: Ann. Mag. Nat. Hist., 1850, ser. 2, vol. v, pl. xiii, fig. 15, p. 435; British Cretaceous Brachiopoda, Mon, Pal. Soc., 1854, pl. ix, figs. 9-24, p. 70; Suppl. 1874, pl. iv, figs. 1-10, p. 40. Remarks.—A single specimen of this species is all that is represented in the collection. It exhibits a good external view of the ventral or pedicle valve, with only the crushed remains of the brachial or dorsal valve, which is too much mutilated for the determination of characters. The ventral valve is very shallow 564 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA and gently arched, with markedly straight and oblique lateral margins diverging from a fairly narrow and tapering summit region to the rounded and expansive base. Numerous concentric growth striations cover the surface, the more prominent of which are more or less equidistantly arranged. On the dorsal aspect of the specimen is seen the remains of an apparently large foramen, the details of which, on account of fracture, are too obscure for description. The microscopically punctate character of the Terebratulidz is well displayed in the test of this specimen. * Dimensions (ventral valve).— Length es a: ; 5 : 3 37 mm. Width 4 : : ; : 34 «Cs, Diameter. : , ‘ : : A es This is a very variable species. The Angola specimen compares favourably with the more elongate examples from the type locality of Tournay in Belgium which occur in the “Tourtia” beds, regarded as of Cenomanian age. Excellent figures have been published of the species, LAMARCK’S type specimen having been drawn by Davinson, while p’Arcutac has issued illustrations and descriptions of examples from Belgium under the name of T. nerviensis. The specimen gives no indication of folds or plications, being perfectly smooth; hence it differs from both T. biplicata and T. dutempleana, which are recorded in the lists of Cretaceous fossils from northern areas of Africa. Distribution.—The later studies of Davipson have demonstrated that the species had its origin in Aptian times, having been found at Faringdon, Upware, Shanklin, and other English localities. In the present case the species may be considered as of older Cenomanian or Vraconnian age, since it is associated with fragmentary Ammonites indicating the zone of Schloenbachia inflata. Occurrence.—In a limestone similar to that associated with Nezthea tricostata, exhibiting minute specks of mica and glauconitic grains. Locality.—No. 295—N.H. of Lobito Bay. PELECYPODA. Fam. OSTREID. Exogyra cf. flabellata, Goldfuss. (Pl. II, fig. 1.) Exogyra flabellata, Goldfuss: Petrifacta Germanix, 1833, vol. 1, pl. xxxvii, fig. 6, p. 32. Ostrea flabella and flabellata, Orbigny: Pal. Francaise: Terr. Crétacés Lamellibvanchia, 1846, vol. iii, pl. 475; p. 71%. : Exogyra flabellata, Pervinquiére: Etudes Pal: Tunisienne: Gast. Lamellib. Crétacés, 1912, pl. xiii, figs, 6-8, p. 189. Remarks.—The collection contains a fragmentary pair of valves in the closed condition which may apparently be referred to an exogyriform oyster probably ee —————eEOEOEe elle AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. 565 related to Hxogyra flabellata of Goupruss, although the umbonal region is entirely absent. The lower valve consists of a fairly well-arched portion of possibly the ventral region, which is furnished with remains of seven obtusely angulate, equidistant radial costz divided by wide suleations which are ornamented with robust, concentric and regularly undulating ridges and striations, these being continued over the summits of the coste. The opposing surface representing part of the upper or flatter valve, which fits closely into the other, exhibits five radial, funnel-shaped, roundly convex, and smooth plications separated by moderately narrow and deep grooves. For an explanation of these structures it is necessary to refer to good figures such as those published by Orpiany of examples from the French Cenomanian. _On that author’s plate (475) the details similar to those im the Angola fossil are more or less well represented. The strongly developed, funnel- shaped costee of the upper valve are well seen in his fig. 1, whereas the more obtusely angulate character of the coste of the lower valve with their wide dividing furrows are to be observed in fig. 2. Under the same name (Ostrea cf. flabellata), but without description or figure, M. Cuorrar* has recognised a probably similar mollusc in the Cenomanian-Turonian beds of Dombe Grande, Angola. It is interesting to note that the double plication of this species of Hxogyra was used by Baye as the significant character for the establishment of his genus Ceratostreon, the true Hxogyra having only its lower valve plicated, while the upper bears a concentric sculpture. This fossil measures 50 by 49 mm., its greatest diameter being 20 mm. Distribution.—This species has a wide geographical range, being found in Europe (France, Italy, Portugal, etc.), Algeria, Egypt, Tunisia, Arabia, Palestine, etc. It is essentially Cenomanian, and chiefly belongs to the upper part of that horizon, as well as occurring in the Lower Turonian rocks of the island of Socotra, according to Kossmat. Occurrence.—The fossil is without matrix, except as an infilling material between. the valves, composed of a hard light-coloured limestone. Locality.—No. 12—Hanha near Lobito. Fam. PECTINIDA. Neithea xquicostata, Lamarck. (PI. I, figs. 2 and 3.) Pecten xquicostatus, Lamarck: Hist. Nat. Anim. sans Vert., 1819, vol. vi, part 1, p. 181. Vola xquicostata, Choffat: Recueil Etudes Pal. Faune Crét. Portugal, 1901-02, pl. iii, fig. 10, p 152. Pecten (Neithea) xquicostalus, Pervinquitre: Etudes Pal. Tunisienne: Gast. Lamellib. Terr. Crétacés, 1912, p- 135 Remarks.—A somewhat small lower valve exhibiting an external view is referred to this species. The length and height are the same, viz. 16mm. It is furnished * Mém. Soc. Phys. Hist. Nat. Geneve, 1888, vol. xxx, No. 2, p. 25; Contrib. Géol. Colonies Portugaises Afrique, 1905, part 2, p. 59. 566 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA with about 20 equal costee and furrows, which are ornamented with a close and trans- versely striated surface. The ears are not visible. PERVINQUIbRE refers to small forms of this species, with a similar number of ribs to the present specimen, which he found in Tunisia, although the presence of a transversely striated sculpture is not mentioned in connection therewith, nor is it a usual character of the shell, which is supposed to bear a smooth surface. Such a sculpture does exist, however, in well-preserved specimens of the species. It should be mentioned that Vola cf. xquicostata has been recognised by M. CHorrat™ as occurring in the Schloenbachia inflata zone of Dombe Grande, Angola; but whether it belongs to the same species as Dr GrEGoRY’s shell it is impossible to say, as the statement is without description or figure. It is of interest, also, to remark that the present Angola specimen shows a resemblance to Neithea sergipensis of C. A. WuiTE + from the Schloenbachia inflata zone of Brazil (Sergipe). Distribution.—This species is characteristic of the Cenomanian rocks of Europe (France, England, Germany, Portugal, ete.). In Portugal it occurs in Turonian beds, » PERVINQUIERE having alluded likewise to its occurrence in the Cenomanian of Tunisia. Occurrence.—The specimen is embedded in a cream-coloured limestone, associated with small examples of a Lama. Locality.—No. 294—from E. of Lobito, plateau 800 feet. Neithea angoliensis, sp. nov. (Pl. I, fig. 4.) Description.—Lower valve small, of arched contour, umbonal region comparatively narrow ; latero-marginal surfaces nearly vertical, depressed, smooth ; pallial region furnished with 6 robust, equidistant primary coste, and intermittent furrows occupied by a single median rib with an occasional accessory rib. Dimensions.— Length : : : t ; ; 10 mm. Height : , : ‘ : ; 15 ” Remarks.—This specimen is quite a small form of Nezthea, and exhibits an external view of a lower valve. Any fine surface sculpture that may have originally existed is entirely obliterated. The ears are not disclosed. Only one probable accessory rib can be traced, this being seen very obscurely near the ventral margin on the posterior corner of the specimen ; otherwise a single secondary rib occupies each of the furrows. In contour the specimen appears to be related to both Janira tricostata of Coquanp{ from the Cenomanian of Algeria and Tunisia, and Janira ficalhoc§ of * Uontrib. Géol. Colonies Portugaises Afrique, 1905, part 2, p. 58. + “Contributions to the Paleontology of Brazil,” Archiv ‘Mus, Nacion. Brazil, 1888, vol. vii, pl. iii, figs, 6, 7, p. 39. + | Géologie et Paléontologie Constantine, 1862, pl. xiii, figs. 3, 4, p. 219. § Mém. Soc, Phys. Hist. Nat. Genéve, 1888, vol. xxx, No. 2, a y, figs. 8-10, p. 89. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. 567 Cuorrat from the Albian of Dombe Grande, Angola. Both these species, however, have more than one secondary rib between the primaries, the first-mentioned having three and the other two. The nearly unicostate character of the present shell is an interesting feature which, so far as can be ascertained, has not previously been observed in this genus. Occurrence.—In a creamy-coloured limestone with ironstone nodules. Locality.—No. 20—80 metres above the Catumbella Dam. Neithea tricostata, Coquand. (PI. I, figs. 5-7.) Janira tricostata, Coquand: Géologiz et paléontologie de la région sud de lu province de Constantine, 1862, pl. xiii, figs. 3-4, p, 219 [non Pecten tricostatus, Bayle]. Pecten coquandi, Peron: Desc. Moll. Foss. Crétacés Tunisie, Explor. Scient. Tunisie, 1891, part 2, p. 224. Pecten (Neithea) shawi, Pervinquitre: Etudes Pal. Tunisienne: Gtastropodes et Lamellibranches des Terrains Crétacés, 1912, pl. ix, figs. 1-3, p. 136. Remarks.—-There are two examples of this species in the collection, the better one being of considerable size and in a fine state of preservation. It consists of a lower valve with the central part of the shell substance removed by fracture. Otherwise the characteristic external sculpture is entire and well displayed. The interspaces dividing the six great primary cost are furnished with three equidistant ribs of unequal size, the central being the most robust, whereas the rib on each side is narrower and less elevated. A few nearly obsolete radial ribs may also be traced on the large lateral areas, the surfaces of which, in common with the remaining parts of the valve, are covered with fine concentric and closely arranged striations. The valve has a greater height than length, and in addition it may be stated that the ventral margin is of strong polygonal contour, while the aliform expansions, although only partially seen, give the idea of extreme smallness of size. Dimensions.— Length : ' : ; . 80 mm. Height : ; : : é CH) 55 Diameter . : ‘ é : : 40 ,, This species has been often confused with J. SowERBy’s Pecten quadricostata, from which it differs in its smaller auricles and in possessing a greater relative height, and therefore not presenting a sub-quadratic contour; while further the intermittent coste are of unequal size instead of being generally of equal strength as in Sowersy’s shell. It is not possible to agree with PERVINQUIERE in regarding SHaw’s* old figure of a North African fossil as representing this species ; it is much more likely to be a form related to, if not identical with, N. quadricostata, on account of its much more nearly square outline, and from the fact that a similar measurement characterises the height and length of the valve. This species offers con- * “ Pectinites,” THomas SHaw: Voyage en Barbarie, 1748, vel. ii, fig. 24, p. 128 (fide PERvINQUIzRE) ; German edition, 1765, pl. xxviii, fig, 24, p. 407. 568 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA siderable resemblance to C. A. Wurte’s* Nevthea quadricostata from the Cretaceous deposits of Brazil (province of Sergipe), and only seems to differ in possessing more prominent costal furrows. It would appear to be quite permissible to retain CoqgUuAND’S specific name of tricostata, to which was attached the generic term Janira, not- withstanding that Baye } had described in previous years another fossil as Pecten tricostatus because the specific term had been applied to two different genera. Distribution.—Hitherto this species appears to have been known only in the Cenomanian beds of Northern Africa, Algeria, and Tunisia. Occurrence.—In a cream-coloured limestone with occasional ferruginous staining, and containing minute specks of glauconite (?) and mica. Locality.—Lobito, Angola. Collector.—Mr E. Rogins. Neithea quadricostata, J. Sowerby. (PI. I, fig. 8.) Pecten quadricostata, J. Sowerby : Mineral Conchology, 1814, vol. i, pl. lvi, figs. 1, 2, p. 121. Neithea quadricostata, R. B. Newton: Trans, Roy. Soc. South Africa, 1909, vol. i, pl. ii, figs. 18-20, p. 55. Remarks.—The collection contains a specimen showing an internal view of a small upper valve in the limestone matrix; it belongs in all probability to Pecten quadricostata, J. Sowerby. Although of considerably less size, it agrees remarkably well with an excellent text-figure in Mr H. Woops’ Monograph of the Cretaceous Lamellibranchia of England, published by the Paleeontographical Society, 1903, vol. i, part 5, fig. 5, p. 214, depicting a similar interior of a specimen from the Upper Greensand of Warminster, which is attributed by some authors to the Cenomanian horizon. In the Angola fossil the length is greater than the height, which is one of the characters of the species; and the shell is furnished with a relatively large posterior ear, the anterior one not being preserved. The surface is covered with some twenty rounded radial coste separated by furrows of rather greater width, to see which a wax squeeze has been necessary. Otherwise the flattened-looking ribs of the fossil represent the broad furrows, as would be seen on the external surface if it were so exposed. The wax impression also proves that the valve was distinctly concave. Dimensions.— Length : ; ; 3 5 : 21 mm. Height : : ; : : : 19 ” Distribution.—According to Mr Woops, the species ranges from the Albian to the Cenomanian in British and European regions; the species has also been recorded from Africa (Zululand, etc.). ; Occurrence.—In a cream-coloured limestone. Locality.—No. 19—Catumbella Dam. * “(Contributions to the Paleontology of Brazil,” Archiv Mus. Nacion. Brazil, 1888, vol. vii, pl. iv, figs. 1, 2, p. 37. + FournEL, Richesse minérale de ? Algérie, 1849, pl. xviii, fig. 30, p. 369. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. 569 Fam. Limip&. Inma cf. itieriana, Pictet and Roux. (PI. I, fig. 9.) Lima parallela, Orbigny: Pal. Francaise: Terr. Crétucés Lamellibranchia, 1846, vol. ii, pl. 416, figs. 11-14, p. 539, non J. Sowerby. Lima ttieriana, Pictet and Roux: Desc. Moll. Foss. Grés Verts, 1852, pl. xl, fig. 5, p. 484. Remarks.—The collection contains two limestone casts of right valves which appear to present affinities with this species. They possess the same number of coste, about 20; no ornamentation can now be traced, nor is there any evidence of a thin rib occupying the grooves, which is a characteristic feature of the shell when properly preserved. A resemblance undoubtedly exists between these specimens and the Lima interlineata of C. A. Wurrr,* which has been recorded from the- Schloenbachia inflata beds of Brazil (Sergipe). The species itself is synonymous with Plagiostoma elongata of J. px C. Sowersy, which has been included by Mr H. Woops in his Lima (Mantellum) gaultina.+ Distribution. —The species ranges in Kurope from the Albian to the Cenomanian, but both Peron and Pervinqutire have recognised it in rocks of Cenomanian age in Tunisia. Occurrence.—Found in a cream-coloured limestone with Neithea xquicostata. Locality.—No. 294—Plateau 800 feet EK. of Lobito. Fam. Mytinipa. Volsella sp. (PI. I, figs. 10 and 11.) Remarks.—Vhis genus is represented by one specimen possessing very tumid valves, a gently excavated ventral margin, a pair of nearly terminal umbones, a pro- minent lanceolate ligament cavity, and a concentrically striated ornamentation. Dimensions.— Length : 5 5 : : é 60 mm. Height : . ‘ ; 5 DBT s, Diameter (with elared valves) : : ; 24 ,, Judging from the lateral view of this specimen, there is much in its contour and general appearance to connect it with Modiola siliqua of Marutron from the “Craie chloritée” or Cenomanian beds of France, { especially Orpicny’s interpretation of that species,§ although a comparison shows that the Angola fossil is furnished with much more inflated valves and is relatively wider at the anterior end. It is probably a new species, although the surface ornamentation is too * “Contributions to the Paleontology of Brazil,” Archiv Mus. Nacton. Brazil, 1888, vol. vii, pl. iii, figs. 3, 4, p. 42. + Mon, Pal. Soc, : Cretaceous Lamellib. England, 1904, vol. 11, pl. v, figs. 16-20, p. 31. t Cat. Méthod. Foss. Bouches-du-Rhéne, 1842, pl. xxviii, figs. 5, 6, p. 178. § Pal. Francaise: Terr. Crétacés Lamellibranchia, 1844, vol. iii, pl. 339, figs. 3, 4, p. 274. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 15). 82 570 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA obscure, through possible erosion, for a more complete description of an otherwise interesting specimen. Occurrence.—A limestone specimen and more or less of a cast; the surface is tinged with a brownish colour. It was found associated with Trigonarca, crinoidal stem ossicles, ete. Locality.—No. 292—Canyon E. of Old Lobito. Fam. ARCIDA. . Trigonarca cf. lgeriensis, Orbigny. (Pl. I, figs. 12 and 13.) Arca ligeriensis, Orbigny: Pal. Francaise: Terr, Crétacés Lamellibranchia, 1844, vol. iii, pl. 517, figs. 1-3, p. 227 (non figs. 4 and 5). Trigonarca ligeriensis, R. B. Newton: Trans. Roy. Soc. South Africa, 1909, vol. i, pl. iv, figs. 13-18, p. 33. Remarks.—A small form of Trigonarca occurs rather plentifully in the cream- coloured limestones, mostly as natural casts. The specimens are rather more quadrate than in the normal condition of this species; otherwise they are closely related, and may even be young examples. One specimen fits into a limestone impression which exhibits the strongly and divergently grooved ligament region— the teeth on the cast itself possessing striations on the lateral surfaces. The same impression shows remains of external sculpture which consists of fine and closely arranged concentric striations crossed by obscure radial lines, which is especially well seen around the umbonal region. Dstribution.—The species is known from the Cenomanian of Europe, Syria, Keypt, and Zululand. Occurrence.—These casts are found in the cream-coloured limestone associated with Trachycardium, ete. | Localities.—Nos. 281, 292—Plateau N. of Lobito (281); Canyon E. of Old Lobito (292). | | Trigonarca cf. diceras, Seguenza. Arca diceras, Seguenza: Studi geol. Cretaceo Italia: Atti Accad. Lincet (Roma), 1882, ser, 3; Mem., vol. xii, pl. xiv, fig. 1, p. 158. Arca (Trigonoarca?) diceras, Pervinquitre: Htudes Pal. Tunisienne: Gast. Lamell. Crétacés, 1912, pl. vii, figs. 23, 25, 26, p. 102. Remarks.—This determination refers to a rough limestone cast exhibiting both valves in the closed condition, and showing a pair of very anterior, distant umbones that are separated by a wide and deeply excavated cardinal region, while the adductor scar impressions are represented by large and prominent pits of more or less trigonal contour. Dimensions.-— Length é 5 ‘ , - : 68 mm. Height : F ; ; : . 42 5, Diameter. ‘ : : x 50 ,, AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. ant The fossil is closely related to Area diceras of SeGuENzA, from the Cenomanian of Italy, which was founded on a similar limestone cast, and chiefly seems to differ in its rather greater length of valve, the dimensions of the type showing a length of 60, height 45, and diameter 50 mm. Dstribution.—This species belongs to the Cenomanian of Italy and Tunisia, and is also found in Algeria, according to PERVINQUIERE. Occurrence.—The specimen was found associated with some fragmentary ammon- ites of the group of Schloenbachia inflata, and some other mollusca. Locality.—No. 294—Plateau 800 feet E. of Lobito. Fam. TRIGONIIDA. Trigonia crenulata, Lamarck. (PI. I, fig. 14.) Trigonia crenulata, Lamarck: Hist. Nat. Anim. sans Vert., 1819, vol. vi, part 1, p. 63; Orbigny, Pal. Frangaise: Terr. Crétacés Lamellibranchia, 1843, vol. iii, pl. 295, p. 151; Coquand, Géol. Pal. Constantine, 1862, p. 290. Remarks.—A small limestone cast with both valves closed together agrees in every particular with specimens of this species in a similar condition from the Cenomanian of France (Rouen), in the British Museum. It shows a well-excavated areal region, while the sides of the valves retain obscure evidence of the obliquely arranged costz. Dimensions.— Length : : ; : 3 ; 25 mm. Height : : : We ; : 205, Diameter. : ‘ : : : Ii Ss; A better specimen of what is apparently the same species is preserved as an impression on a limestone slab, from Catumbella, Angola, belonging to the British Museum (Ansorge collection). A wax squeeze of this yields an externo-lateral view of a fragmentary right valve 1m which the antero-marginal region is not present. It shows, however, a sharply defined posterior carina to which is con- nected anteriorly a regular series of equidistant coarsely crenulated coste, much finer costze of similar but more minute structure being present on an incomplete postero-areal surface. Great resemblance may be noted between this specimen and Trigonia sub- erenulata as figured by C. A. Wuite* from the Schloenbachia inflata zone of Brazil (province of Sergipe), which is much more likely to represent a form of T. crenulata rather than T. subcrenulata described by Orpieny ¢ from the Aptian rocks of Columbia, South America, which has fewer and more distinct costz, besides belonging to an older Cretaceous horizon. Distribution.—This species was originally described from Mans in the north- * “Contributions to the Paleontology of Brazil,” Archiv Mus. Nacion. Brazil, 1888, vol. vii, pl. v, fig. 3, p. 70. + Voyage dans ? Amérique Meéridionale, pt. 4, “ Paléontologie,” 1842, vol. iii, pl. xix, figs. 7-9, p. 87. 572 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA west of France. Coquanp has referred to its having been found in Algeria, and PrRon* mentions casts from Tunisia which he determined as Trigonia cf. crenulata. The age of the species is usually regarded as Cenomanian. Occurrence.—In a hard creamy-coloured limestone. Localities—No. 292—Canyon E. of Old Lobito; Catumbella ..:, British Museum, L. 18,582. ‘ Fam. CRASSATELLITID®. Anthonya cf. baudeti, Coquand. (Pl. I, figs. 15 and 16.) Crassatella baudeti, Coquand: Géol. Pal. Constantine, 1862, pl. xiii, figs. 5-7, p. 198. Crassatella (Anthonya) baudeti, Pervinquitre: Etudes Pal. Tunisienne: Gast. Lamell. Orétacés, 1912, p. 250. Remarks.—It is interesting to be able to determine one of the specimens of this collection as being closely related to Coquann’s Crassatella baudeti from the Cenomanian deposits of Algeria, a species which is also known from rocks of similar age in Tunisia and Italy. Its dentition, however, has never been ascertained, so that the identification hitherto has had to be based on external and cast features only. Quite recently PERVINQUIERE, in referring to its occurrence in Tunisia, has placed the species in Anthonya, a genus founded by Gass for a Crassatelliform shell from the Upper Cretaceous of California,t with which its sculpture and contour show considerable resemblance. But the interest of Anthonya is its dentition, which consists of two strong divergent cardinals in each valve; moreover, the hinge region is without the ligamental fossette so characteristic of Crassatellites (=Crassatella), details which are of sufficient importance for the recognition of Anthonya as a genus and not merely a section as implied by Pervinquikrg, although it must be eranted he was without knowledge of the dental features of the shell. Dr Grecory’s Angola specimen is curiously fractured in the umbonal region, so that the dentition of the left valve is well displayed, and is seen to agree in its main features with that of Anthonya. There is, however, this distinction, that the teeth are laterally and transversely striated, { no mention being made of such a structure by Gass, although it may possibly be present in specimens of thoroughly good preservation. A difference exists also in the sculpture, the Angola fossil showing minute vertical striations between the concentric coste which ornament the valves, a feature not yet recorded in the history of Coquann’s species nor in the type form of Gage’s genus. Although found in another region of Africa, this interesting specimen may be considered as furnishing a clue to the dentition of C. baudets, and therefore connecting it with Gasp’s genus Anthonya, as first suggested by PervinquikzRE, on external characters only. * Moll. Foss, Urétacés Tunisie: Hauplor. Scien. Tunisie, 1891, pt. 2, p. 263. + Geol. Surv. California ; Palxontology, 1864, vol. i, pl. xxx, figs, 236, 236a, p. 182. + A similar structure has been observed by Gass in his Cretaceous genus Remondia from Mexico (Geol. Surv. California: Paleontology, 1869, vol. ii, pl. xxxvi, figs. 17, 17a, p. 270), which, however, is distinct from Anthonya in possessing more teeth, and is besides of a totally different contour. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. | B68) Dimensions.— Length : . : : : : 40 mm. Height ; : ‘ ; é ; Zo. 55 Distribution.—This species is regarded as of Cenomanian age, and has been reported from Algeria, Tunisia, and Italy. Occurrence.—The specimen, embedded in a cream-coloured limestone, consists of a pair of united valves in the closed condition showing the external surface of the right, and only the hinge of the other valve being exposed. In the same limestone are examples of Trigonarca ef. ligeriensis, crinoidal stem ossicles, and Nullipore structures. Locality.—No. 286—Plateau N. of Lobito. Fam. CarpmIp&. Trachycardium cf. syriacum, Conrad. Cardium syriacum, Conrad: Lynch’s Offic. Rept. United States Exped. Dead Sea, 1852, pl. vii, fig. 45, p. 217. Cardium (Acanthocardia) syriacum, Hamlin: Mem. Mus. Comp. Zoology Harvard, 1884, vol. x, No. 3, pl. iii, fig. 7, p. 48. Remarks.—The cream-coloured limestone yielding Trigonarca and other shells contains also some excellent internal casts of Cardium-like valves which closely resemble Conran’s Cardiwm syriacum from the Cenomanian rocks of Palestine. They possess a narrowly elongate and very much incurved umbonal region, while the surface is furnished with numerous fine and closely arranged radial coste which appear to show obscure traces of minute lateral asperities. In contour the valves are subglobose, heart-shaped, somewhat oblique, and of greater height than length. The hinge line is short. Such characters would suggest the inclusion of this species in MOrcu’s genus Trachycardium (type =Cardium isocardia, Linneus) rather than in Acanthocardia (type=C. aculeatum, Linnzeus) as suggested by Hamu, in which the valves are usually longer than high, besides possessing a more extended hinge region. Dimensions.— Length : . : ‘ : ; 25 mm. Height : , : é PA) Diameter (with closed valves) : : : U3 as Distribution.—So far as can be ascertained, this species has been hitherto restricted to the Cenomanian (Buchiceras zone) rocks of Palestine. Occurrence.—Found in the cream-coloured limestone with Trigonarca, crinoidal stem ossicles, and other remains (No. 281); another example occurs with Hipponyx (No. 286); and some isolated specimens are in a similar inatrix (No. 292). Localities.—Nos. 281, 286, 292—Plateau N. of Lobito (281, 286); Canyon E. of Old Lobito (292). . 574 MR Rk. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA Fam. SAXICAVID. Panopexa cf. plicata, J. de C. Sowerby. Mya plicata, J. de C. Sowerby : Mineral Conchology, 1823, vol. v, pl. 419, fig. 3. Panopxa plicata, J. de C. Sowerby, Mineral Concholoyy, 1829, vol. vi, p. 211; Pervinquiére, Eludes Pal. — Tunisienne: Gast, Lamell. Crétacés, 1912, pl. xxi, figs. 1-4, p. 284 (see this work for comprehensive synonymy). Remarks.—A large limestone cast, with united valves, exhibits the contour lines of this species. Its length is roughly about twice the height given in the original deseription. Dimensions.— Length : : ; : : - ) LO mam: Height : . : . ; : GO” ;, Diameter. : ; : : : 47s, Distribution.—Generally regarded as a Lower Cretaceous species, being found in Britain, Europe, and Tunisia. PERVINQUIERE mentions a probable Cenomanian example from Tunisia. | Locality.—No. 11—from 50 metres above Catumbella Dam. Fam. PHOLADOMYID. Pholadomya cf. vignesr, Lartet. Pholadomya vignesi, Lartet, Explor. Géol. Mer Morte, 1877, pl. xi, fig. 9, p. 126; Blanckenhorn, Beier. Geol. Syriens, 1890, pl. v, figs. 14-17, p. 94; Pervinquitre, Htudes Pal. Tunisienne: Gast. Lamell. Crétacés, 1912, pl. xxi, figs. 8, 9, p. 290. . Remarks.—It is interesting to be able to record a small form of Pholadomya which is evidently related to P. vignesi. The specimens consist of two limestone casts representing the subglobose valves of this species, and which are furnished with numerous radial costee—excepting posteriorly, where the surface is smooth— and obscure concentric striations. This sculpture, best seen in the umbonal region, includes also a minutely granulated structure. As remarked in a previous account of the species from Zululand by the present writer,* it is unquestion-— ably related to Cardium lucerna of Forses from the Upper Cretaceous rocks of Southern India,t and would appear to differ chiefly in possessing finer and more numerous coste. Dimensions.— Length i ‘ : : ; : 22 mm. Height : ; : : 5) Diameter (with closed valves) : : ; Lido * Trans. Roy. Soc. South Africa, 1909, vol. i, p. 80. + Trans. Geol.“Soc, London, 1846, ser. 2, vol. vii, pl. xvii, fig. 10, p. 145. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. a7a This form of Pholadomya is not to be confused with either P. pleuromyxfornus, Chotfat, or P. cf. collombi, Coquand, which have been recorded by CHorrat from the Albian rocks of Dombe Grande in Angola.* Distribution.—The species is characteristic of the Cenomanian, having been found in Algeria, Tunisia, Palestine (Buchiceras zone), the island of Semha near Socotra, and Zululand. Occurrence.—The examples consist of isolated limestone casts. Locality.—No. 15—Cliffs beside Catumbella Dam. Fam. DICERATIDA. Toucasia sp. (Pl. I, fig. 17.) Remarks.—The specimen consists of a badly preserved internal cast in a hard calcareous sandstone, which appears to represent a form rather similar to that figured by Orpieny + as Requienca (Caprotina) carantonensis from the Cenomanian of France. In the chief lateral view of this specimen, which is evidently a lower valve, a well-arched umbonal region is observed with remains of a secondary umbonal body beneath it possessing a strong inward curvature at the nucleus, while the opposing lateral surface is much too obscure for definition; the valve throughout is regularly compressed and of moderate width; no indication exists of the upper valve. Dimensions.— Length : é j : : é 55 mm. Depth 4 ; : , : 5 25°" 55 Width (dorsal) : ; : 5 8 : 32) 55 The occurrence of similar fossils in the Cenomanian-Turonian rocks of Dombe Grande, Angola, has already been referred to by Cuorrar { under the name Requienia, but without description or figure, as forming part of the fauna found in the “calcaires coralliens blancs.” It should be mentioned also that the generic name Toucasia was founded by Monrer-Cuatmas § on Marueéron’s. type of Requena carinata from the Aptian rocks of France,|| a species related to the well-known British fossil Diceras lonsdalir of J. pe C. SowErsy, from beds of the same horizon, and which was subsequently recognised as occurring in the Urgo-Aptian beds of Algeria by Coquanpn, under the genus Toucasia.** There seems to be little doubt that Dr Grecory’s specimen, which resembles those species, may be regarded as belonging to this genus. * Mem. Soc. Phys. Hist. Nat. Geneve, 1888, vol. xxx, No. 2, pl. iv, fig. 12, and pl. v, figs. 1-3, pp. 83, 84. + Pal. Frangaise: Terr. Orétacés, 1850, vol. iv, pl. 592, fig. 3, p. 259. + Mém. Soc, Phys. Hist. Nat. Genéve, 1888, vol. xxx, No. 2, p. 25; Contrib. Géol. Colonies Portugatses Afrique, 1905, part 2, p. 59. § Journ. Conchyl., 1873, vol. xxi, p. 74. || Cat. Foss. Bouches-du-Rhéne, 1842, pl. ii, figs. 1-5, p. 104. { Trans. Geol. Soc. London, 1836, ser. 2, vol. iv, pl. xiii, fig. 4, p. 338. ** Etudes Supplém. Paléontologie Algérienne, 1880, p. 191. ° 576 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA Occurrence.—The specimen is associated in the same matrix (calcareous sand- stone of brownish-grey colour) with a compound tubular coral of crystalline and mineralised structure and therefore quite indeterminable. Locality.—No. 273—Hanha Gorge. GASTROPODA. Fam. Naricip&. Tylostoma globosum, D. Sharpe. (PI. I, figs. 18, 19.) Tylostoma globosum, Sharpe: Quart. Journ. Geol. Soc., 1849, vol. v, pl. ix, fig. 5, p. 379. Tylostoma aff. xquiaxis, Peron: Moll. Foss. Crétacés Tunisie: Explor. Scient. Tunisie, 1890, pl. xix, fig. 23, p. 57. Remarks.—This form is represented by two limestone casts of which the largest is much crushed in front and so obliterates all details of the aperture. It is of greater dimensions than the type, although agreeing with it in contour and structure of the spiral region. The second specimen is of more normal measurements, show- ing a well-inflated body whorl, and a narrow and semilunate aperture with indica- tions of a marginal constriction ; the summit is much covered with matrix, although the penultimate whorl is well exposed. Dimensions.— Larger Smaller Example. Example. Length : : : ‘ , : 79 67 mm. Width , ; ‘ . : 3 74 BD: "55 Distribution.—The species was originally described from the Cretaceous of Portugal, and Cuorrar has recognised it as ranging in that country from Cenomanian to Turonian. PERoN quotes it from the Senonian of Tunisia, whereas PERVINQUIERE regards it as belonging to the Lower Turonian of that country. It should be also mentioned that the present Angola examples of this species bear considerable resemblance to C. A. Wurrr’s* 7. globosuwm (?) from the Cretaceous beds of Brazil (Sergipe), which therefore indicates the zone of Schloenbachia inflata. The species is in no way related to Cuorrat’s Tylostoma pechuelt from the Albian of Dombe Grande, Angola,t nor can it be stated to be equivalent to Tylostoma sp. of the same author from the Schloenbachia inflata zone of Dombe Grande and Catumbella, Angola, because the characters of that form are not known, the determination being without figure or description.} | Occurrence.—In a light-coloured limestone. Locality.—No. 100—Gypsum beds by Catumbella Bridge; and Lobito (collected by Mr E. Rostns). * “Contributions to the Paleontology of Brazil,” Archiv Mus. Nacion. Brazil, 1888, vol, vii, pl. xvii, fig. 10, p. 190. + Mém. Soc. Phys. Hist. Nat. Geneve, 1888, vol. xxx, No. 2, pl. iv, figs. 6, 7, p. 80. { Contrib. Géol. Colonies Portugaises Afrique, 1905, No. 2, p. 57. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. OT Fam. NERINAIDA. Mrhilaia cf. nerineformis, Coquand. (PI. I, fig. 20.) Turritella nerineformis, Coquand : Géologie et paléontologie de la région sud de la province de Constantine, 1862, pl. iii, fig. 2, p. 176. Mrhilaia nerinxformis, Pervinquiere: Etudes Paléont, Tunisienne : Gastropodes et Lamellibranches Crétacés, 1912, pl. iii, figs, 34-38, Daoos Remarks.—This Gastropod is represented by a fragmentary limestone cast, showing only the basal whorl and an imperfect aperture with obscure evidence of a columella plication. The whorl has ‘deeply excavated sides, besides bearing an indication of marginal nodulations. Similar and better-preserved casts are among the Angola fossils in the British Museum (Ansorge collection) ; they exhibit a wide and deep sutural excavation, as well as occasional concentric bandings on the concave walls of the volutions, which continue over the basal surface of the body whorl. The existence, however, of the marginal nodulations, not referred to in the original or later descriptions, creates a doubt as to the Angola specimens belonging really to this species. They are, however, regarded as being closely related, and not to be confused with the highly sculptured species Nerinea capelloi which Cuorrat has described and figured from the Dombe Grande area of Angola of a rather later Cretaceous horizon.* Distribution.—The species belongs to the Cenomanian stage, and has been recorded from Algeria, Tunisia, Dead Sea area, and Italy. W. PavuicKke t+ has recognised a similarly determined form in the Albian-Cenomian rocks of Peru. Occurrence.—In a cream-coloured limestone with mica and glauconite grains. Localities.—No. 294—Plateau 800 feet E. of Lobito, associated with Ammonites of the group S. inflata and other mollusca; and Catumbella (British Museum specimens). Fam. Hieronycip. Hipponyx sp. (Pl. I, fig. 21.) Remarks.—A specimen of Hipponyx, consisting of a limestone cast, must have been originally about the size of an adult form of the well-known H. cornucopix from the Eocene (Lutetian) formation of France. ‘The proportions and contour likewise show a considerable resemblance to the Tertiary species. It possesses mostly a smooth surface, with, however, some obscure indications of the presence of longitudinal striations, especially in the region of the basal margin, where they are seen to be of thick and robust character. The fossil shows no concentric coste such as are typical of Calyptrea (?) corrugata of Forses from the Upper Cretaceous rocks (Ariyalur group) of India,{ which, besides, is rounder and of smaller dimensions. * Mém. Soc. Phys, Hist. Nat. Geneve, 1888, vol. xxx, No. 2, pl. uli, figs. 15-18, p. 77. + Neues Jahrb. Beilage—Band xvii, 1903, p. 275. t Trans. Geol. Soc. London, 1846, ser. 2, vol. vii, pl. xii, fig. 11, p. 137. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 15). 83 578 MR R. BULLEN NEWTON ON SOME CRETACEOUS BRACHIOPODA Probably this is a form more or less intermediate in character between the Tertiary and Cretaceous species mentioned, but further material is required for a more accurate determination. Dimensions.— Length : : 5 ; : : 30 mm. Diameter. : 5 . : 22, Occurrence.—Accompanying Cardiwm sp. and other shell remains, crinoidal stem ossicles, etc., in a cream-coloured limestone. . Locality. No. 286—Plateau N. of Lobito. Fam. AKERIDA. Akera gregory, sp. nov. (Pl. I, fig. 2.) Description.—Shell very large, broadly ovate, ventricose; volutions three (probably), slightly projecting above outer whorl, deeply canaliculated at suture ; last whorl widely convolute, subcylindrical in front for half its length, columella broadly excavated anteriorly ; aperture extensive, curved, narrowing posteriorly, widest in front, outer lip free at the summit; sculpture showing distant, longi- tudinally curved growth lines without flexuosity. Dimensions.— Length : ; : ; : : 60 mm. Width (maximum) . : : : ‘ DOME temarks.—-This description apples to a natural limestone cast which probably represents one of the largest forms of Akera yet referred to in literature. Not a vestige remains of the delicate thin shell that once invested the surface, and which is so characteristic of the genus. The specimen is related to Coquann’s* Bulla tevesthensis from the Turonian of Algeria, and to a lower Turonian form from Syria described by WuitrieLp{ as Akera siliciosa; but affinities of possibly greater alliance may be traced with C. A. Wurrr’s{ Akera browns from the Cretaceous of Brazil, a species showing considerable resemblance in general contour, but of smaller dimensions, being only about half the size, besides possessing distinctly sinuous crowth lines, a lesser anterior excavation of the columella, and an outer lip terminat+ ing with a closer attachment to the spire. Unfortunately, the Brazilian type is without definite locality, the species being stated to occur both at Sergipe and Pernambuco, the rocks of which areas, according to Kossmat, are of different Cretaceous horizons, the Sergipe deposits belonging to the Schloenbachia inflata group, whereas those of Pernambuco are of later age and would be recognised as equivalent to the Ariyalur stage of India or the Danian of Europe. * (oi. Pal. S. Constantine, 1862, pl. v, fig. 9, p. 189. + Bull. American Mus. Nat. Hist., 1891, vol. iii, pl. x, figs. 14, 15, p. 441. { “Contributions to the Paleontology of Brazil,” Archiv Mus. Nacion. Brazil, 1888, vol. vii, pl. ix, figs. 6-8, p. 206. AND MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA. a9 The specific name is in honour of Prof. J. W. Grecory, F.R.S., who, from a series of extended studies as well as from personal surveys, has greatly advanced our knowledge of African geology. Occurrence.—The cast is in a cream-coloured limestone, and was found associated with an Ammonite of the Schloenbachia inflata type. Locality.—No. 266—plateau south-west of Hanha. EXPLANATION OF PLATES. (Figures are of the natural size, except when otherwise specified.) Puate I. Terebratula depressa, Lamarck. Fig. 1. An external view of the ventral or pedicle valve, the oblique lateral margins and concentric growth striations being well marked. Localtty.—Hanha, near Lobito (No. 12). Neithea xquicostata (Lamarck). Fig. 2. External aspect of a small lower valve with a similar length and height, exhibiting a series of equal coste. Fig. 3. Magnified view of costz of same specimen, showing an obscure transversely striated sculpture. Locality.—K. of Lobito (No. 294). Neithea angolwensis, sp. nov. Fig. 4. Lateral view of lower valve, showing the six robust primary coste with a secondary rib between each, of apparently shorter length. x 2. Locality.—80 metres above the Catumbella Dam (No. 20). Neithea tricostata (Coquand). Fig. 5. Lateral view of a large lower valve, showing the six primary cost with. the interspaces occupied by the three unequal and nearly equidistant ribs. Fig. 6. Anterior aspect of same specimen, exhibiting the curvature and structure of the costal system. Fig. 7. Magnified details of sculpture, showing the close and undulating concentric striations. Locality.—Lobito ; collected by Mr E. Rosrns. Neithea quadricostata (J. Sowerby). Fig. 8. An internal view of a small upper valve embedded in limestone, showing a large posterior ear, the flattened-looking ribs indicating a series of broad furrows. Locality.—Catumbella Dam (No. 19). Lima cf. itiertana, Pictet and Roux. Fig. 9. Right lateral aspect of a limestone cast, showing about twenty ribs with resemblances to this species. Locality.—E. of Lobito (No. 294). Volsella sp. Fig. 10. Dorsal view of specimen with united valves, showing considerable tumidity. Fig. 11. Left lateral aspect of same specimen concentrically striated. Locality.—Canyon E. of Old Lobito (No. 292). 580 MOLLUSCA FROM ANGOLA, PORTUGUESE WEST AFRICA: Trigonarca cf. ligeriensis (Orbigny). Fig. 12. External aspect of summit region, preserved as a limestone cavity, showing the divergently an grooved ligament region as well as microscopically fine and closely arranged concentric cost crossed by obscure radial striations. From a wax impression. x 2. Fig. 13. Natural limestone cast of another specimen, showing internal characters of a right valve, including muscular impressions and teeth. ; Locality.—N. of Lobito (No, 281), Trigonia crenulata, Lamarck, Fig. 14. External view of a right valve preserved as a limestone cavity, showing the coarsely — crenulated costs and the more minute ornamentation of the post-areal surface. From a wax impression. Locality.—Catumbella. British Museum (Ansorge Coll.), L. 18,582. Anthonya cf. baudett (Coquand). Fig. 15. View RE fragmentary specimen with closed valves, the right being fractured in the summit region and so exposing the dentition of the opposing valve. lig. 16. Enlarged portion of same specimen, showing the two divergent cardinal teeth and minute vertical striations between the closely arranged concentric costa. x 2. ' Locality.—N. of Lobito (No. 286). Toucasia sp. Fig. 17. Natural internal cast, giving a lateral view of a lower valve with an arched umbonal’ region and a secondary umbonal structure beneath it with a similar curvature. Locality.—Hanha Gorge (No, 273). Tylostoma globosum, D. Sharpe. Fig. 18. Medium- aed example illustrating the semilunate aperture of this species. Locality.—Lobito ; collected by Mr E. Roszins. Fig. 19. A giant fort of the same species, showing the characteristic irregularity in the suture of the body whorl. * Locality.—Catumbella Bridge (No. 100). > - Mrhilaia cf. nerineformis (Coquand). Fig. 20. Front view of limestone cast, showing the excavated faces of the volutions, the deep suture and obscure traces of marginal nodulations, Locality.—Catumbella. British Museum (Ansorge Coll.), G. 17, 024. Hipponyx sp. Fig. 21. Lateral view of a limestone cast, showing remains of longitudinal striations at the basal margin. — Locality.—N. of Lobito (No. 286). Nore.—The specimens herewith figured, as well as other portions of this collection, are in course of presentation to the British Museum by Professor Gregory, with the exception of those represented by figs. 14 and 20, which already belong to that institution, PuateE II. This plate bears its own explanations. Mowe. et ;. Roy. Soc. EDIN. R. BULLEN NEWTON: CRETACEOUS BRACHIOPODA AND MoLLusca, ANGOLA.—PL. I. Prd) BY J. GREEN. CRETACEOUS BRACHIOPODA, PELECYPODA AND GASTROPODA from ANGOLA, PORTUGUESE WEST AFRICA. [Except when otherwise marked, the figures are of natural size. | >" - fi * ri ai | 4 ie ¥ be UMN ba . ae ¥ ae ul Trans. Roy. Soc. Edin. Vou. LI. R. Buruen Newron on some Cretaceous Brachiopoda and Mollusca from Angola, Portuguese West Africa.—Puarte II. Fic. 1.—Exogyra cf. flabellata, Goldfuss (figures of nat. size). A, ventral view of lower valve of specimen without umbonal characters, showing the widely distant coste and the undulating concentric ridges both of the furrows and coste. B, ventral aspect of the upper valve of same specimen, showing wide funnel-shaped cost and prominent grooves. Locality.—Hanha, near Lobito (No. 12). Note.—The dotted lines indicate probable contours of the missing umbonal portions of the fossil. Fic. 2.—Akera gregoryi, sp. nov. (figures of nat. size). A, ventral view, showing sub-cylindrical body whorl and the extensive aperture. B, dorsal view of same, showing the slightly projecting penultimate whorl. OC, summit view, showing the canaliculate suture, Locality. —Hanha, near Lobito (No 266). + i t i ( 581 ) XVI.—Note on an Algal Limestone from Angola. By Mrs M. F. Romanes, late Harkness Scholar, Newnham College, Cambridge. Communicated by Professor J. W. Grecory, F.R.S. (With One Plate.) (MS. received June 2, 1915. Read June 28, 1915. Issued separately Aprvl 25, 1916.) The specimens here described were collected by Professor J. W. GREGcory in Angola, and belong to the series described in the paper by him on the geology of that country. The horizon of the beds from which the limestone was obtained is Albian. The specimens are of a yellowish pisolitic limestone. In hand specimens the weathered surface has a warty appearance—the diameter of each knob being about | lem. A fractured surface shows circular forms varying in size and having decided concentric markings. These concentric markings are, however, less evident in microscopic sections, and this is probably due to the effects of preservation. A similar case has been described by Mr H. Yasr,* where the genus Metasolenopora is affected in this way. These circular forms make up the greater part of the rock, and are cemented together by a fine calcareous paste in which are fragments of shell and quartz grains, with occasional grains of other minerals, such as felspar and muscovite. The specimens of limestone examined varied to some extent in texture, some being more compact, and even slightly siliceous, but the contained organisms are present in similar amounts regardless of the texture of the rock. With care, small roughly circular nodules can be extracted from the looser varieties of the limestone. In thin section, under the microscope they are seen to consist of algal growth round a nucleus. The substance of the nucleus varies, and may be the plate of an echinoid, or more commonly is the fragment of a spine. In some cases a group of quartz grains serves as a centre round which growth has taken place. Instances of this have been noted by Mr WrErHERED ¢ among the pisolites of the Jurassic. Some- times also small bodies, cf. calcisphere,{ help to form a nucleus. Some also occur scattered throughout the matrix of the rock. Their possible presence in this lime- stone is interesting in view of the fact that the association of calcispherze with abundant plant remains has already been noted.§ Two different genera of algze have been determined making up the pisolitic nodules. The genus present in the greatest abundance has been identified as Girvanella Nich.| * H. Yass, Sez. Rep. Imp. Tohoku Univ. Sendai, Japan, vol. i, No. 1, 1912, p. 2. + E. WErHERED, Creol. Mag., dec. 3, vol. vi, 1889, pp. 196-200. t E. J. G@arwoop, @.J.G.S., vol. Ixviii, 1912, p. 580, pl. xlvii, “Carboniferous Succession in N.W. England.” § RC. M‘LRan, Proc. Camb. Phil. Soc., vol. xvi, pl. vi, 1912, p. 512, “Group of Rhizopods from Carboniferous || A. NicHonson and R. Eraeriper, A Monograph of the Silurian Fossils of the Girvan District, Edin., 1878, p. 23, pl. ix, fig. 24. TRANS. ROY. SOC. EDIN., VOL. LI, PART 111 (NO. 16). 84 582 MRS M. F. ROMANES ON The tubuli of this organism form the centres of all the nodules, and can be dis- tinguished amongst the fine calcareous material, in association with the quartz grains, ete., which form the nuclei of the nodules. The average diameter of the tubuli is 6p. Their length it is not possible to measure, as they twist and bend, the same cell- thread passing to different levels in the section. Sometimes they appear to branch dichotomously. They form an anastomosing bundle of cell-threads in the centre of the nodules, and have no concentric arrange- ment till towards the outside, and here, owing to the state of preservation, it is frequently impossible to distinguish them. The boundary of each pisolitic grain 1s, however, clearly defined against the cementing matrix. In every respect the tubuli are comparable with the smaller varieties described by Mr WrtHERED ™* in his papers on Girvanella. The dimensions of the tubuli in these species are as follows :— G. minuta, 74; G. intermedia, 10m; G. incrustans, 10m. The first two occur in the Coralline oolite of the Jurassic; and the third, together with G. sinensis, described by H. Yasr,t and having tubuli 10g in diameter, is confined to the Carboniferous. This species from the Albian beds of Angola is, therefore, the smallest so far described, and may be known as Grvanella minima, n. sp. The genus is now accepted as a calcareous alga, although its systematic position has not yet been definitely ascertained. There is reason to believe that it is most nearly related to the Siphonee.{ So far, Girvanella has been described as ranging from the Cambrian to the Jurassic ; but, as Mr CuapmMan remarks, “ others, no doubt allied to fossil forms, exist at the present day.” It is not surprising to find it in Cretaceous rocks which have been deposited under suitable conditions. The second genus has been identified as Lithothamnion Phil. This alga occurs towards the circumference of the nodules, frequently as little excrescences, and can be detected on nodules that have been freed from the matrix. In longitudinal section the cells have a width of from 84-12, and their length is greater than this; but measurement is difficult, as there is reason to believe that the transverse walls have suffered more by replacement than the longitudinal walls. The quadratic form of the cells, characteristic of Lithothamnion,§ is not so clearly developed as in some species, but can be made out, and the lack of clearness is almost certainly due to the state of preservation. The thallus is made up of cells, somewhat radially arranged ; branching takes place at different levels, so that the whole assumes a fan-shaped form growing from a point. The edges, where growth has been last in process, are irregular and lobate. This applies equally to sections cut tangentially near the erowing point. The measurements of the cells in tangential sections vary from 8px 10pm to * E. WETHERED, Q.J.GS., vol. xlvi, 1890, p. 270, “On the Occurrence of Genus Girvanella in Oolitic Rocks” ; also Geol. Mag., dec. 3, vol. vi, 1889, p. 196, “ On the Microscopic Structure of Jurassic Pisolite.” + H. Yass, op. cit. { F. CuapmMan, Rep. Aust. Ass. Adv. Scz., vol. xi, 1907, p. 377. § A. RorupLerz, Zeit. deutsch. Geol, Ges., vol. xliii, 1891, p. 302, “ Calcareous Algz.” AN ALGAL LIMESTONE FROM ANGOLA. 583 12u x16. The outline of the cells in some of the sections is polygonal, often with five sides. Some of the tangential sections show large polygonal markings; these are due to growth having taken place from different adjoining centres, with the resulting compressional structure (Plate). The most characteristic feature of the species is the presence of reproductive organs, which have been observed in several sections, and in sections made from one specimen they were very abundant. The conceptacles have a distinctive shape, as will be seen from the plate. Their dimensions bear a close resemblance to those of species which are mentioned by Dr Rorupietz in his paper on “ Calcareous Alege.” * In L. ramosissumum Reuss. the measurements for the conceptacles are given as 280m long x 80p high, in L. suganum Rothp. as 250u long x 100p high, and in L. torwlosum Giint. as 400p long x 150m high. From the plate it will be seen that owing to the curved shape of the conceptacles measurement is not satisfactory, but an average of well-formed ex- amples, reckoning from edge to edge, gave 350 long x 1254 high. The conceptacles occur most usually towards the edge of the thallus. Although they are usually isolated and are situated most commonly on the same circumference, it seems that sometimes they may be so close together as to coalesce in forming a band of “ con- ceptacle tissue,” with the loss of individual shape. In a single thallus, isolated conceptacles can be seen to pass into the conceptacle band. No spores have been observed in the conceptacles, although they contain a certain amount of structure, suggesting canals. When writing on the classification of the Nelobesiex, Fosuir t distinguishes between those Lathothamma in which the sporangia are grouped in conceptacles, and those in which they are not, placing the former in subgenus Eulithothammion. This subgenus he divided into two sections according to whether the conceptacles remained superficial or not. In the species under dis- cussion the material of the thallus subsequently grows round the conceptacle, so that this species falls into the section Innate Fost., to which section also belong L. ramosissimum, L. suganum, L. torulosum. The conceptacles in the first two are smaller, and in the third larger, than in this species. It resembles those three which are Tertiary most nearly, but is distinguished from them by the characteristic shape of the conceptacles. The specific name of Lithothammion angolense, n. sp., is pro- posed for it. Girvanella forms the largest proportion of the nodules, but the two have been observed in association in all the sections examined. It appears that bundles of Girvanella tubes have first been formed, enclosing fragments of echinoid plates or quartz grains, and that subsequently Lathothamnion has begun to grow on the nodules so produced. Its growth has, however, been arrested at an early stage, for in no instances are large masses found, and the original outlines of the Girvanella nodules are only slightly modified. * A. RoTHPLeErz, op. cit. t H. Fosuin, Kongl. Norske Vidensk. Selsk. Skr., 1900, No. 5, p. 12. 584 AN ALGAL LIMESTONE FROM ANGOLA. Localities.—The limestone is from beds near Lobito Bay in the province of Benguella. The localities from which the specimens were obtained are given below— the numbers refer to Professor GrEGoRY’Ss collecting list. 8 and 10. Catumbella dam. This rock is cream-coloured and very compact. 264. Hanha Gorge. The matrix in this specimen is coarser, and contains much quartz and some felspar, both oligoclase and microcline. The specimen is weathered, and the nodules stand out clearly, some attaining a length of 1°8 em. . Beach, one hour 8.W. of the mouth of the Hanha River. This specimen is greyer in colour than the rest, and is partially polished. The algze bo i | — were here associated with Schloenbachia, n. sp. 290. Coastal Plateau, at the foot of the scarp of the Second Plateau, about 6 miles from Old Lobito, on the track to Hanha. This specimen is compact and deep yellow in colour. The material from here gave the best con- ceptacles. 297. This limestone is less compact, and contains a greater number of shell frag- ments forming nuclei. The specimens are so similar, especially microsopically, that there seems no reason to suppose they do not all belong to one bed. Throughout the observations recorded in this note I have been indebted to Professor GREGORY. EXPLANATION OF PLATE. Fig. 1. Girvanella minima, sp. nov. Section of 264 x 66. Fig. 2. Lithothamnion angolense, sp. nov. Longitudinal section of 290 x 33. - Hie: 3. x Transverse section of 290 x 66. Fig. 4. ‘ . * Transverse section of 290 x 66, showing lines produced by compression, due to growth. Fig. 5. x be PS Group of conceptacles in 290 x 24. Fig. 6. * 3 5 Conceptacles which have coalesced, forming a band, in 290 x 66. This shows structure suggesting canals. _ VOL. LI. Mrs. Romanes—Algal Limestone, Angola. Fic, 3. Fic: 4 Fic. 5. Bemrose, Collo., Derby. (era85)-2) XVII.—On some Cretaceous Echinoidea from the Neighbourhood of Lobito Bay. By Professor J. W. Gregory, F.R.S., D.Sc. (MS. received June 2, 1915. Read June 28, 1915. Issued separately August 21, 1916.) Nine species of Echinoidea were described by pr LorioL* from Benguella and mainly from the neighbourhood of Dombe Grande. 1. One of his species, Epiaster catumbellensis,t is common in the Lobito district. It is recorded by pr Loriot from the Inflata beds at Catumbella, and the younger Cyprina wensi sandstone at Dombe Grande. It is a variable species. According to the dimensions given by DE LorioL, the length varies from 25 to 36 mm., and the breadth varies from 88 to 94 per cent. of the length and the height from 66 to 77 per cent. of the length. His illustrations show that the species has two chief forms: in one the shape is depressed and the posterior margin is nearly vertical ; in the other the posterior interambulacrum is subcarinate, and the posterior margin is more sloping owing to the projection of the lower posterior corner. In DE Lorior’s figure, pl. viii, No. 4, the posterior margin has a slope of only 7° from the vertical ; in his fig. 5 this slope is 15° from the vertical. . I collected specimens of this species 50 feet above the algal bed at the Catumbella dam, where it was below the level of the Inflata beds; from marls with “S” inflata at Old Lobito; from some marly beds, in which it is abundant, in the upper part of the Inflata zone; and also from the Salenza bed and at a slightly higher level (850 feet) above that bed. Many of the specimens are poorly preserved, and some are found only as internal casts. The dimensions are therefore quoted, as they are useful in the determination of the species. It is possible that two of the forms figured by pg Loriox should be separated, but they were both found together in Lobito. SPECIMENS COLLECTED. { No. 292. Gorge near Old Lobito. 1. Good internal cast. 1., 29; b., 26; h., 194; a depressed variety. 2. 1, 29; b., 29; h., 17; very depressed variety. 3. 1, 32; b., 325; h., 25; broad high variety. 4. Specimens with best-preserved granulation. * In Cuorrat and pe Lorton, “Matériaux pour Vétude stratigraphique et paléontologique de la province d’Angola,” Mém. Soc. Phys. Hist. Nat. Geneve, vol. xxx, 1888, 116 pp., 8 plates. + Ibid., pp. 112-114, pl. viii, figs, 3-6, { Dimensions are in mm. :—l, = length; b. = breadth; h, = height. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 17). 85 PROFESSOR J. W. GREGORY ON SOME CRETACEOUS ECHINOIDEA or [9 6) for) No. 269. First Plateau; two specimens, of which one has subcarinate posterior ambulacrum more strongly developed than is shown in DE LorIoL’s fig. 4. 1., 263; b., 245; h.; 20. No. 281. Collection of specimens from the Hpiaster bed on the First Plateau on the road to Hanha, south-west of the steep descent to the Hanha_ basin ; one large specimen of the depressed variety has 1., 33; b., 30; h., 22. No. 288. Summit of Second Plateau; south-west of the Epiaster bed, on road to Hanha. A weathered cast. 1., 32; b., 274; h., 21. No. 294. Salenia bed; Upper or Second Plateau, 800 feet above sea-level; five miles east-south-east of Lobito Bay. 1 305" bs, 28a 235. 2. 1., 264; b., 26; h., 20. 3. 1., 31; b:, 29; h., 24. No. 270. Three poorly preserved specimens; top of Upper or Second Plateau, 850 feet above the sea-level; above the Salenia bed; five miles east-south-east of Lobito. 2. Holaster dombeensis, Lor.; a broken specimen and a fragment from the Inflata marls at Old Lobito agree in the characters available for comparison with this species from the Jnflata beds at Dombe Grande. 3. Salenia dombeensis, Lor., var. triangularis, n. var. Two worn specimens of Salenia were found on the top of the Second Plateau at the height of 800 feet above sea-level, five miles east-south-east of Lobito Bay. One specimen is fairly well preserved, though the granules are only faintly indicated. It agrees very closely with Salenia dombeensis of DE LorioL,* which, however, occurs on a lower horizon. According to DE LorioL, that species is found at Dombe Grande in the mammillare beds, which are Albian, whereas the Salenia bed near Lobito contained “S” inflata, and a series of mollusca which Mr Newron has shown have Cenomanian and even Turonian affinities. The epistroma of this Salena is poorly preserved; but sufficient patches of granules are recognisable to show that the arrangement is practically identical with that of S. dombeensis. The Dombe Grande specimens are a little flatter: the dia- meter recorded by DE Lorion varies from 11 to 28 mm., and the height varies from 64 per cent. to 78 per cent. of the diameter; in the specimen from near Lobito the diameter is 19 mm., the height 153, so that the height is nearly 82 per cent. of the ciameter. The apical dise lacks many of the slits shown in DE Lorior’s figure, but they are very variable, and their presence is not regarded as of specific value. The only definite difference which I can recognise is, that in the Lobito specimens the madre- * Op, cit., pp. 100-102, pl. vi, figs. 2-4, FROM THE NEIGHBOURHOOD OF LOBITO BAY. 587 porite occupies a triangular instead of a nearly circular depression ; it differs markedly from that in the apical dise shown by DE Loriot in his pl. vi, fig. 2d; but in his figure, pl. vi, fig. 4, there are faint indications of a smaller depression below the genital pore. Were it not for the difference in horizon I should have been disposed Salenia dombeensis, Lor., var. triangularis, n. var. to attach little value to this character; but it may be worth while to separate the new Salenia as a distinct variety, named, from the shape of the madreporite, var. triangularis. The apical dise of the Lobito specimen is shown in the figure. 4. Cottaldia, allied to Benettzz, Konig: diameter, 18 mm.; height, 135 mm. Found in the Epiaster bed on the First Plateau, south-west from Hanha. This specimen is too covered with matrix for description. Me 2. & ry i ’ ° — cab 7 ’ -% _ 7 @ i Ay eC eae ‘ : e 4 * MONS OS -aet be Oa ee " =, wu PAT Tae ( 589 ) XVIII.—Contributions towards a Knowledge of the Anatomy of the Lower Dicotyledons. I. The Anatomy of the Stem of the Papaveracee. By R. J. Harvey-Gibson, D.L., M.A., Professor of Botany, University of Liverpool; and Minnie Bradley, M.Sc., Hartley Research Scholar, University of Liverpool. (With Three Plates.) (MS. received December 6, 1915. Read December 6, 1915. Issued separately September 16, 1916.) INTRODUCTORY. The systematic study of the cluster of orders lying at the base of the Dicotyledon phylum has of late years become of fundamental importance in view of the various theories advanced by different authorities as to the origin of the so-called Mono- cotyledons. The anatomical structure of many of these groups has been but little investigated from this aspect, and the present paper is the first of a series which it is hoped to publish with the object of discovering to what extent anatomy supports or contradicts one or other of views held on the phylogeny of flowering plants. One such research is already far advanced, viz. on the anatomy of the Magnoliaceze, and another, the present contribution, deals with the stem structure of the Papaveraceze. Although both researches have been retarded owing to one of the authors being engaged on war service, there would seem no reason for delaying the publication of such observations as have been made. Naturally, no attempt has been made in the present paper to draw any general conclusions outside the genera actually investigated ; such generalisations must be postponed until the completion of other work now in progress. THE ANATOMY OF THE STEM OF THE PAPAVERACES. _ According to Pranri and Kinvice (8), the order Papaveracez includes the genera usually grouped under Fumariaceze, but by them placed in a section Fumarioidee. Twenty-eight genera in all are described in the Pflanzenfamilien, and of these twelve, represented by twenty-six species, have been investigated by us, viz. Papaver, Meconopsis, Argemone, Chelidonium, Glaucium, Bocconia, Eschscholzia, Hypecoum, Dicentra, Adlumia, Romneya, and Corydalis. Some of the remaining sixteen we hope to discuss in a future paper, as material becomes available. Papaver Rheas, L., is placed by Lieer (11) in a subgroup Rheeades, which includes P. Argemone, L., P. dubium, L., and P. hybridum, l., and he selects P. dubium for detailed description, stating, however, that the whole group exhibits great similarity in anatomical structure. In the present instance P. Rhwas has TRANS, ROY. SOC. EDIN., VOL. LI, PART III (NO. 18). 86 590 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON been selected for detailed study and other species are referred to only in so far as they differ from it. ; A transverse section through an internode of P. Rheas (figs. 1, 2) shows, | successively inwards, epidermis, chlorophylliferous cortex, several layers of sclerotic pericycle, vascular bundles arranged in an undulating circle, medullary rays, and parenchymatous medulla. The epidermis has a moderately thick cuticle, and stomata occur in the interfascicular regions. Multicellular hairs are present, as stated by Licer, but a distinct exodermis is absent. The term “exodermis” is applied by Licer to the layer of non-green cells found immediately below the epidermis in many Papaveracee. In P. Rheas the subepidermal layer contains chlorophyll and its thin-walled cells resemble those of the cortex. The cortex consists of three or four layers of thin-walled parenchyma containing chlorophyll, with prominent intercellular spaces. An endodermis is not distinguishable, although Lkcer claims to have identified a fairly well-marked one in, P. dubiwm consisting of non-green cells larger than those of the cortex. He draws attention to the danger of con- - fusing the endodermis with the pericycle owing to sclerosis. In P. Rheas sclerosis has advanced so far as to render it impossible to distinguish between endodermis and pericycle, in the adult at least. The pericycle consists of three or four layers of cells polygonal in transverse section and more elongated than the cortical cells. Their walls are strongly lignified and abundantly pitted. In the interfascicular regions this pericycle merges gradually into thin-walled medullary ray parenchyma, while in the fascicular regions it abuts directly on the hard bast. The vascular bundles are arranged in one or more circles according to the region of the stem under examination. The phloem consists of lignified fibres, unlignified sieve tubes, companion cells, and phloem parenchyma. The fibres are narrower and much longer than those of the pericycle. Pits are present on their walls, but these are not so prominent as those of the pericycle. Lrerr found no pits on the walls of the bast fibres either in P. dubiwm or the present species. As to the sieve tubes, LEGER states that the sieve plates are poorly developed and often unrecognisable, but careful examination, after suitable treatment, shows that the plates are of the normal angiospermic type. The lateral walls are well supplied with sieve fields, broken up into subareas by secondary thickening of the wall as the tube gets older (fig. 3). Later on callus becomes deposited on either side of the membrane. Hutt (2) points out that the connecting filaments react differently to reagents after the formation of callus, and regards them as “slime-strings.” In P. Rheas sieve tubes are by no means infrequent, although owing to the small quantity of callus formed they are not always easy to locate (fig. 4). The companion cells are usually as long as the sieve tubes, and in communication with them by groups of protoplasmic threads (fig. 5), as are also the abundant phloem parenchyma, not mentioned by Lkcrr. When callus is formed in these cases the thickening occurs on the side of the sieve tube only. THE ANATOMY OF THE STEM OF THE PAPAVERACEA. 591 The laticiferous tissue, so characteristic of the Papaveracez, is found in the phloem, and is arranged in an are between the primary and secondary elements (fig. 6). The tubes are articulate, and form a much-branched network (fig. 7). In the developing stem protuberances arise from the vertical tubes, which push their way between adjoining sieve tubes and unite with other later tubes in the vicinity, The course of the tubes may readily be observed if fresh material be boiled for a few seconds in a 10 per cent. solution of KOH and afterwards well washed and sectionised. The xylem is separated from the medulla and medullary rays by conjunctive par- enchyma, and similar cells are interspersed among the xylem elements themselves. The protoxylem elements have single or double spirals followed by annular and later by pitted vessels. The older secondary wood consists of radially arranged pitted trachez and fibres, the latter forming the main mass of the secondary tissue and corresponding to the conjunctive parenchyma of the primary xylem (figs. 8, 9). The secondary phloem consists of narrow elongated elements with a few laticiferous tubes. Only very rarely is anything in the nature of a sieve tube or companion cell recognisable (fig. 4). The narrow elongated cells are, however, in protoplasmic communication with each other through their lateral walls. The secondary latici- ferous tissue is in the form of tubes running side by side and connected by large perforations in contiguous walls, A transverse section through the basal region of the stem shows, as LEGER points out, marked differences from such a region as that already described. The epidermis is deficient in cuticle; the cortex consists of about seven layers of non-green, slightly elongated cells; the pericycle and hard bast are much less lignified, and finally, in regions nearest the base, form with the cortex one homogeneous zone of thin-walled cells. Meanwhile the vascular bundles coalesce and both secondary xylem and phloem are produced abundantly. Medulla and medullary rays are slightly lignified. The internodes are short, owing to the crowding of the radical leaves, and the epidermis and outer layers of the cortex tend to die off and exuviate. At the points of entry of an axillary shoot concentric bundles often ‘appear, the occurrence of which is not noted by Licer (fig. 10, a). The phloem in these cases is central, and has associated with it a complete ring of laticiferous vessels (fig. 11). The secondary xylem in this region consists entirely of lignified fibres with pitted walls. L&cer describes the course of the vascular bundles only in the regions immedi- ately below the flowers. The anatomy of several floral axes was investigated right through the thalamus; some of these flowers had eight, some ten, and others thirteen carpels, but they all agreed in their essential characters, and the following account may be taken to apply equally well to all types examined. The vascular bundles of the pedicel are arranged in a ring, large and small bundles usually alternating. These bundles, as the thalamus is approached, increase in number by repeated divisions until an almost compiete cylinder of vascular tissue has been 592 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON formed. A large number of cords are then given off to supply the sepals and petals ; the ring then re-forms, and almost immediately gives off numerous bundles to supply the stamens. Once more the ring closes, and then finally splits into as many vascular cords as there are placental plates. Between these cords there is at first no vascular tissue, but later on independent groups of tracheids are differen- tiated, branching in all directions, but especially towards the ovarial cavity. Further, the placental cords send innumerable branches to the ovules scattered over the placental plates. It was noted that the number of vascular bundles in the pedicel did not always correspond to the number of placental bundles in the fruit wall. The vascular system as a rule becomes stable just below the flower head and remains unchanged until the insertion of the first leaf-trace, but occasion- ally forking or fusion occurs, resulting either in an increase or a diminution of the number of the bundles from that seen at a higher level. The observations made on the mode of insertion of the leaf-traces agree on the whole with those of LkcEr, though more variations were met with. Just before the entry of the first leaf- trace the bundles of the stem lose their definite arrangement in three concentric circles and form a more or less undulating ring. The bundles of the leaf on approaching its base fuse into three or more cords. Where the system consists of three cords (fig. 12, af) the main central bundle forks and gives off anterior strands, while the adjacent stem bundles fuse laterally. With this compound bundle the anterior strands unite. This composite mass then splits, and the four bundles of the leaf-trace enter by the gap so created. Finally, the two branches, arising by the forking of the median leaf-trace bundle, unite with the lateral bundles of the same leaf-trace and run down the axis as two independent cords. After a time these bundles split into smaller strands and the normal arrangement of the vascular system becomes established. L&cER states that anterior strands are given off from all the leaf-trace bundles. That is not so; further, in some cases no anterior strands are formed at all. Numerous variations are met with. Thus in one imstance where no anterior strands were formed the entire leaf-trace did not enter by the main gap produced by the forking of the compound stem bundle ; in another case a special layer of parenchyma enclosed both the leaf-trace and the vascular system of the axillary branch. The leaf-traces at the base of the stem are simpler, for in some only one bundle enters the foliar gap. The smaller anterior strands, instead of fusing into’ an are and uniting with the axis system, may remain distinct and persist in this condition for a considerable distance. The vascular tissue of the axillary branch is inserted one- half on either side of the leaf-trace, just as the latter enters the stem. The stem anatomy of other species of Papaver need be described only in so far as it shows differences from that of P. Rhaas. Papaver dubium, u.—The layer beneath the epidermis usually forms a well- defined exodermis in LicER’s sense of the term, but not infrequently it is indis- THE ANATOMY OF THE STEM OF THE PAPAVERACE. 593 tinguishable from the general cortex, although Licer describes its presence as a constant character of this species. Papaver somniferum, L.—A transverse section shows numerous vascular bundles arranged in concentric circles, giving the general appearance presented by a mono- cotyledonous stem (fig. 13). The exodermis is distinct, the chlorophyll zone is five to seven layers in thickness, the pericycle is less sclerotic than in P. Rh@as, and there is a gradual transition both in the shape of the cells and in the intensity of sclerosis from the pericycle to the medullary rays. A band of non-sclerotic cells intervenes between the sclerotic pericycle and the hard bast in the fascicular regions, but the hard bast is not well developed in this species. Papaver pilosum, Sibth.—The anatomy of the stem resembles very closely that of P. Rheas. Both exodermis and endodermis are well marked, the pericycle is very sclerotic, and this character is maintained in the medullary rays. The peri- eycle closely surrounds the hard bast, save where the latter abuts on the sieve-tube region. The vascular bundles are more deeply seated than in any of the preceding species. Papaver orientale, L.—The main axis of this species is perennial, and gives off erect annual shoots. The arrangement of bundles is monocotyledonous in char- acter, as in P. somniferum, the smaller outer bundles being more or less imbedded in a slightly sclerotic pericycle. The cortex is composed of several layers of cells, the outer alone containing chlorophyll. There is no distinct exodermis, and the peripheral bundles frequently consist of phloem and laticiferous tubes only. Papaver levigatum, Bieb., resembles in anatomy P. Rheas, and concentric bundles also occur as in that species. Papaver glaucum, Boiss.—The anatomy resembles that of P. levigatum in all essential points. Papaver lateritiwm, Koch, is similar to P. Rhwas, but the hard bast is only feebly developed. Papaver nudicaule, L.—The main stem is extremely short, while the erect axes are correspondingly elongated and well developed. No distinct exodermis is visible, and the chlorophyll zone varies in thickness. There are three or more layers of green cells about half way down the axis, but below and above this level these cells decrease in number and finally disappear altogether. The pericycle is thick and sclerotic, merging gradually into the non-sclerotic rays, but abutting directly on to the hard bast in the fascicular regions. Papaver alpinum, L.—This species on the whole resembles P. nudicaule, but on a reduced scale. The peduncles are sometimes fistular. By some authors P. alpinum is regarded as synonymous with, or merely a variety of, P. nudicaule. Papaver rupifragum, Boiss.—The structure of the stem is in all respects comparable with that of P. Rheas. Here also there is a tendency towards the formation of concentric bundles. 594 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON The stem anatomy of other genera of Papaveraceze may now be considered. Meconopsis cambrica, Vig.—The main axis is short and thick, and from it numerous leaves and flower-stalks arise. The exodermis is indistinct, and the inner chlorophyll-bearing layers are often sclerotic in varying degree. The endodermis is indefinable, the pericyele, five or six layers deep, is sclerotic and merges gradually into the non-sclerotic medullary rays in the interfascicular regions and abuts directly on the hard bast opposite the bundles. The medulla is non-sclerotic and often fistular. The bundles, which are of the normal dicotyledonous type, are arranged in circles, and each is surrounded by one or two clearly defined layers of conjunctive cells. The hard bast is strongly sclerotic, while the sieve tubes are prominent and show sieve areas on the walls adjacent to the phloem parenchyma. These cells also exhibit a tendency to bulge out and meet similar protuberances from neighbouring cells, as do the laticiferous elements in Papaver. Latex tubes also occur and run parallel with the phloem elements, but without any sign of branching. Argemone mexicana, L.—The exodermis in this species is fairly well defined, but no endodermis can be distinguished. The protoxylem elements are scattered amongst an inner mass of parenchyma which is itself bounded by a well-marked V-shaped band of sclerenchyma. Laticiferous tissue occurs in the pericycle, in the conjunctive tissue of the medullary rays, along the borders of the primary phloem and in the secondary phloem, in the xylem parenchyma, and a few elements are found even in the hard bast and in the lignified tissue bordering the inner parenchymatous mass of each bundle. The elements are branched, and have sinuous walls in the phloem, but in other regions resemble more in form the tissues among which they occur (fig. 14). Prominent perforations occur on the walls of the latex tubes. The stem anatomy of Argemone hispida and of Argemone ochroleuca differs in no essential respect from that of Argemone mexicana. Chelidonum majus, L.—The exodermis of this species is fairly well defined, but its cells more nearly resemble those of the epidermis than of the cortex, which consists of a single layer of green cells (fig. 15). There is no obvious endodermis, and the pericycle is represented by about six layers of sclerotic cells, the intensity of sclerosis diminishing from without inwards, so that the pericycle gradually merges into the non-sclerotic medullary rays. Laticiferous tissue occurs at the periphery of the vascular bundles, more rarely in the secondary phloem and medullary rays. Most of it consists of a series of cells placed end to end, showing no lateral branching, but those cells which occur in the medullary rays are less regularly superposed and occasionally show indications of branching. Their walls contrast with those of Papaver in being free from perforations save, infrequently, where the lateral walls are contiguous. Glaucium corniculatum, Curt.—The peripheral region is shown in fig. 16. There is no definite endodermis, but just within the chlorophylliferous zone is a region of sclerotic cells five to ten layers deep which comprises pericycle, hard THE ANATOMY OF THE STEM OF THE PAPAVERACEA. 595 bast, and the outer portions of the medullary rays, the whole forming a continuous cylinder of mechanical tissue. In section the laticiferous tubes which occur at the periphery of the hard bast assist in differentiating pericyle from hard bast owing to their larger lumina. Pranti and Kiwnie state (L.c.) that in Glaucium laticiferous tubes occur only in the roots, but this is not the case, and no doubt these authors failed to recognise them in this situation. A few isolated tubes are also to be found in the pericycle and medullary rays and in the secondary phloem. When old the latex tubes lose their contents, and those external to the vascular bundles become sclerotic. The cells are usually isolated, and show no tendency to form a series. Although, as above stated, there is usually a gradual transition in the interfascicular region from well-lignified pericycle cells to less sclerotic and finally non-sclerotic medullary ray cells, sometimes the whole of the medullary ray as well as the “medulla itself is transformed into mechanical tissue. The xylem and phloem do not call for remark beyond the fact that the sieve tubes are especially well developed. Bocconia microcarpa, Max.—A transverse section through the stem of this species shows a well-marked epidermis of brick-shaped cells whose walls tend to become lignified (fig. 17). Beneath the epidermis lie a double layer of collenchyma and two or three layers of chlorophyll-bearing cells. The collenchymatous cells show traces of chloroplasts, indicating their modification from ordinary cortical cells. Further, traces of collenchymatous thickening appear in the thin-walled, ereen cortical cells, so that the difference between these two zones is one of degree rather than of kind. No definite endodermis is discernible. The pericycle is only slightly sclerotic, and in the interfascicular regions merges gradually into the medullary ray tissue, which, like that of the medulla, is parenchymatous. The vascular bundles are arranged in a single compact circle close to the periphery. The xylem consists of vessels with bordered pits (fig. 18) and wood parenchyma. Laticiferous tubes are found at the periphery of the hard bast, but in the adult they become less obvious owing to lignification. They occur also on the inner margin of the internal parenchymatous mass, and may be recognised, even when lignified, by their large lumina and contents. Eschscholaa californica, Cham.—The stem of this plant is furrowed, the angles being rounded and about ten in number, though varying in different regions of the stem (fig. 19). Stomata occur in the furrows, while the subepidermal tissue opposite the ridges is collenchymatous (fig. 20). The chlorophylliferous tissue lies in the furrows, and merges into thin-walled parenchyma continuous beneath the collenchy- matous bands. Many of the cells show traces of latex, but the latex tends to disappear almost entirely in the adult, affording an explanation of the statement by PrantTL and Kinnie (/.c.) that latex does not occur in the stem of this genus. The pericycle is only feebly sclerotic, and is from one to three cells deep. The latex tubes occur chiefly in the deeper cortex, though occasionally also in the secondary 596 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON phloem and medullary rays. The tissue resembles in all respects that already described for Glaucium. Hypecoum grandiflorum, Benth.—The main stem is very short, most of the aerial portion being formed of long slender leaves and floral axes bearing smaller leaves arranged in a 2/5 spiral. The stem anatomy is as follows: The epidermis is thin-walled and shows no definite cuticle, and beneath lie three layers of cells containing either chlorophyll or red sap (fig. 21). The inner cells abut on an amyliferous layer which is more evident in young plants. The sclerotic pericycle follows next, and many of its cells are laticiferous. The vascular bundles are few in number, only four occurring above the entry of the first leaf-trace. Hach bundle is surrounded by a sheath of collenchyma specially well developed on the inner side. Laticiferous elements are present in all the tissues, and latex occurs even in the xylem vessels (fig. 22). The tubes are unbranched. Dicentra formosa, DC.—The chief axis is obliquely ascending, its lower region being covered with crowded leaf-bases, and its parenchyma is crammed with starch. A transverse section of the floral axis exhibits a fluted outline with the larger bundles occupying the angles. The epidermis is only slightly cutinised, and the subepidermal tissue is poor in chlorophyll but rich in latex. The pericycle varies in thickness and is sclerotic. Laticiferous tubes are present, but latex occurs also in the ordinary cells of the pericycle. Latex tissue is also found in the cortex, phloem, and wood parenchyma, and latex may even occur in the xylem vessels. There is no branching or anastomosing of latex tubes (fig. 23). Adlumia cirrhosa, Raff.—The only available material was some stem fragments showing leaf insertion, and hence the following account of the anatomy of this species must necessarily be very incomplete. The plant is a climber, and has very long slender ribbed stems (fig. 24). Immediately below the epidermis is a layer of non-green cells representing an exodermis. The ridges of the stem are occupied by bands of collenchyma, while the furrows are subtended by loose chlorophylliferous tissue with large intercellular spaces opening on to stomata. A sclerotic pericycle abuts on the collenchyma and chlorophyll-bearing tissue, but ends abruptly inwards. The hard bast is deeply imbedded in the sclerotic pericycle, the whole forming a continuous cylinder of mechanical tissue. The laticiferous tubes are solitary or in short rows, and are found chiefly in the subepidermal layer, the cortex, soft bast, and conjunctive parenchyma, but the latex tends to disappear as the stem gets older. Romneya trichocalyx, Kastw.—The exodermis in this form is fairly clearly defined. The subjacent chlorophylliferous zone is discontinuous, and separated into bands by non-green plates (fig. 25). The inner cortex from which these plates arise consists of ordinary parenchyma ten to twelve layers deep. No endodermis or pericycle is distinguishable. The elements of the xylem are worthy of note. Spiral thickenings occur on their walls in conjunction with more or less bordered pits. The material available was old, and here as in other species the latex tends to dis- 7 e @ THE ANATOMY OF THE STEM OF THE PAPAVERACE. DOT appear as age progresses. The distribution of the elements could not therefore be determined with certainty. Corydalis racemosa, Pers.—The epidermal and subepidermal cells in this form have curiously folded walls (fig. 26): Though furrowed like Adlumia, subepidermal collenchyma is absent. The endodermis is not distinguishable, and the pericycle is only slightly sclerotic and varies in thickness. Cells with folded walls similar to those found in the epidermis and subepidermis also appear in the conjunctive parenchyma surrounding the vascular bundles or scattered among other cells on the inner side of the vascular bundles (fig. 27). Similar cells with folded walls occur in other species of Corydalis and also in species of Argemone, Dicentra, and Eschscholzia. Latex occurs in the subepidermal layer, in the conjunctive parenchyma of the vascular bundles, and also but to a less extent in the xylem and phloem. Some of the medullary ray: and medullary cells also contain latex, but in trifling amount, and in all cases it tends to disappear in the adult. Corydalis glauca, Pursh., agrees in its anatomy with C. racemosa in all essentials, as does also C. capnoides, Wahl. As has been already pointed out under Papaver Rheas, the anatomy of the stem base may differ considerably from that of the aerial portion, and a few notes may be added as to the peculiarities of that region in other species of Papaver and in the genera just discussed. The only point worthy of note in the stem bases of P. dubium and P. somniferum is the feebleness of the lignification, while in P. pilosum the secondary wood even consists of only a few xylem vessels scattered among thin-walled, slightly prosenchymatous cells. In P. orientale LkcER states that the vascular bundles remain isolated even at the base of the stem, and arrange themselves in one or two cycles. This is not the case, however, as may be seen in fig. 28, where the bundles are not isolated though their individuality may still be made out by their protoxylems. LicErR also emphasises the permanence of hard bast in the basal region, in contrast with the condition obtaining in preceding species. It is true that the bast fibres do persist for some time, but eventually they disappear entirely. One point of interest not noticed by L&cER is the occurrence of a secondary zone of cambial activity inside the vascular ring, 7.e. in the medulla, which produces xylem outwards and phloem inwards,(figs. 29, 30). The two xylem zones, the outer formed by the normal the inner by the medullary cambium, are separated by a layer of cells identical with those of the medulla, though in some places the two xylems are practically confluent. Save for variations in the degree of sclerosis in the medulla and medullary rays and in the amount of secondary xylem formed, the stem bases of the remain- ing species of Papaver examined do not call for any special description. The secondary phloem in the stem base of Meconopsis cambrica exhibits abundant latex. tubes arranged in concentric circles. No special features call for mention in the stem bases of Argemone mexicana, A. hispida, A. ochroleuca, Chelidonium majus, TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 18). 87 598 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON Glaucium corniculatum, Bocconia microcarpa, and Eschscholzia californica. The main stem of Hypecoum grandiflorum is short, and corresponds to the stem bases of types already mentioned. In its upper regions its vascular system consists of from ten to twelve isolated bundles which rapidly increase in number as the base is approached owing to the insertion of leaf-traces. Lateral fusion then takes place, and finally the vascular system. forms a complete ring broken only at intervals by the entry of leaf-traces. The xylem vessels are abundant, and the stem contrasts in this respect with the preceding types. The anatomy of the stem base in Dicentra formosa closely resembles that of Hypecoum, as does also that of Romneya trichocalyx. In the genus Corydalis the features are in all respects similar, save for certain un- important variations in the degree of sclerosis of the cortical and medullary cells and in the amount of secondary wood formed and the extent of its lignification. The course taken by the vascular strands in Papaver Rheas has been already described, but a few notes may be added as to conditions obtaining in this respect in other species of Papaver and in other genera of Papaveraceee. So far as the course of the bundles in species of Papaver is concerned, it may be said at once that it differs in no essential from that found in P. Rha@as. In Meconopsis cambrica the capsule contains four placental plates with four placental vascular strands, while the peduncle contains a ring of several isolated bundles which exhibit frequent fusions and splittings. In Argemone mexicana there are also four placental bundles, horseshoe-shaped in section. Hach opens out and fuses laterally with its neighbours so as to form a ring of vascular tissue prior to the entry of the stamen traces. Small leaves are attached close to the top of the peduncle, and their method of entry is shown in fig. 31, a, b, ¢. In Chelidonium majus the flowers are arranged in corymbs, though some authors regard the inflorescence as a true umbel. The vascular tissue, however, shows that the main axis gives off a series of bundles to small bracts each with a flower-stalk in its axil, the whole arranged in a close spiral. The ovary consists of two carpels, and the strands from each of these, together with those from other floral parts, form in the pedicel a circle of four large and three or four smaller bundles which follow independent courses until the base of the pedicel is reached, where they fuse to form a vascular ring. The bract trace consists of a single bundle, and the vascular system of the pedicel at the point of attachment to the main axis splits, one-half being thrown to either side of the bract trace, all three cords entering by a gap formed in the vascular system of the chief axis. The bract traces tend to become multi- fascicular in the lower regions. The course of the bundles in Glauciwm cornculatum does not differ in any essential respect from that described in Papaver Rheas, while that seen in _ Bocconia microcarpa closely resembles the arrangement in Chelidonium majus. In Eschscholzva californica the five vascular cords from the carpels fuse into a ring at the base of the ovary, to which are attached the cords from the stamens and petals. THE ANATOMY OF THE STEM OF THE PAPAVERACEA. agg The vascular strands from the sepals, however, fuse into an outer vascular ring, while the inner ring, at a lower level, opens out once more and its elements unite with the outer ring. The conjoint vascular tissue then splits up into independent strands of various sizes and descends the peduncle in that form. In Hypecoum grandiflorum the main stem is short, and most of the aerial region is composed of long slender leaves and floral axes, each of the latter bearing several flowers in racemes. The vascular tissue of each pedicel consists of four bundles prior to the entry of the first leaf-trace. Hach pedicel (save the terminal one) receives two leaf insertions before becoming attached to the main axis. Lach is unifascicular, and the single strand enters a gap formed by the splitting of the nearest pedicel bundle, the branches fusing with the adjacent axial strands. After the two inser- tions the vascular system of the pedicel thus consists of two large and two small bundles. When, as is sometimes the case, the leaf-traces are trifascicular, each member of the trace maintains its independence and the base of the pedicel then shows the presence of eight bundles. Similarly, the vascular system of the pedicel enters a gap formed for it in the system of the chief axis, one-half of the pedicel system ranging itself on either side of the bract trace. Lateral fusions later on result in the ten to twelve bundles seen in a section of the main axis. The inflorescence in Dicentra formosa is a biparous cyme. The vascular system of the youngest pedicel consists of five bundles, that of the subjacent bract of one or three. The vascular tissue of each pedicel fuses laterally into two masses, one on either side of the incoming bract trace, and these three strands are often distinguish- able for some considerable distance. The course of the vascular system in Romneya trichocalyx does not differ materially from that found in Papaver Rheas. In Corydalis racemosa the vascular tissue ‘of the terminal flower-stalk consists of four independent cords. These fuse in pairs lower down, thus creating two gaps through one of which the vascular system of the next peduncle and its bract enters. The vascular cord of the bract unites with that of the peduncle at the moment of in- sertion. LicEr worked out the course of the vascular system in the inflorescence of C. soda, but the arrangement of cords in C. racemosa, C. glauca, and C. capnoides does not quite correspond to his description. In the last-named species the vascular tissue of the terminal flower is represented by a single cord which splits in two just before the entry of the next lower peduncle and bract. Some notes may now be added on the insertion of leaf-traces in other species of Papaver and in other genera of the order. Papaver dubium.—The leaf-trace entry in this species closely resembles that described for P. Rhwas. Litcer says that at the base of the leaf the numerous leaf bundles unite into three, the middle bundle forks, and its two sections fuse with the lateral bundles. He also states that all the bundles of the leaf-trace give off strands inwards which either arrange themselves into an are whose extremities later on fuse with the nearest stem bundles, or remain free and follow an independent course down 600 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON the stem. Dissection shows, however, that the number of bundles into which the vascular system of the leaf resolves itself depends entirely on the size of the leaf, that the forking of the central cord occurs very late, and that the interior strands are by no means constant, many leaf-traces showing none at all. Fig. 32, a-c, shows the most fully differentiated anterior strands that could be found, and it will be seen that only those of the median bundle are developed. The course of the leaf-trace in P. somniferum follows no fixed rule, several different methods having been noted. In no case was. there any indication of the formation of anterior bundles. Several variations were also met with in P. pilosum, in which species also anterior bundles are absent.- The leaf-trace in P. omentale consists of several strands which finally merge into three groups, the central consist- ing of one bundle and the lateral of several. The anterior bundles are present, but are only feebly developed ; these fuse into three ares. These anterior bundles consist only of phloem accompanied by laticiferous tubes. The leaf-trace enters by three gaps in the axial system. Irregular fusion takes place between the axis bundles, even the innermost series of strands being disturbed. On this point LicEr’s account differs. The leaf-trace in P. levigatum consists generally of seven bundles, and anterior bundles are occasionally found, and even then are feebly developed. Here also variations in the mode of entry are frequent. In P. glaucium the leaf base largely ensheathes the stem (fig. 33, a—c), and the leaf bundles finally unite into six, which give off small anterior strands. Further fusions result in the formation of three main cords which pursue an independent course down the stem. No anterior cords appear in young leaves. P. lateritiwm shows two types of leaf-trace entry with general features in common but differing in detail. When a leaf has an axillary branch the entry is effected as shown in fig. 34, a, b; when there is no axillary branch the method is as shown in fig. 35, a-c. In P. nudicaule the leaf-trace consists of one large median and two large and several smaller lateral bundles. As the trace approaches the stem cylinder the bundles approximate to each other. The median bundle enters first, followed by the lateral bundles, and these gradually fuse in their passage down the stem. In the peduncle, on the other hand, the leaf-trace bundles fuse into one cord before entry. The same is true of the leaf-traces of P. alpinum. The mode of entry of the leaf-trace in P. rupifragum is very similar to that met with in P. Rheas, but as in that species variations occur even in the same plant. When an axillary branch occurs the vascular tissue consists of a ring of isolated bundles. These fuse, but leave a gap opposite the incoming leaf-trace, half the cords of the axillary branch being placed on either side of the leaf-trace. In the leaf- traces entering the stem near its base the anterior bundles are not so prominent nor is the forking of the median bundle so well marked ; indeed, as a rule the two final strands representing the leaf-trace rapidly unite into one broad bundle. Variations, THE ANATOMY OF THE STEM OF THE PAPAVERACEA., 601 as above stated, are frequent ; thus one leaf-trace examined showed no forking of the median strand, but one of the lateral bundles fused with the median one and the leaf- trace was continued down the stem as one large and one small bundle, the former formed by fusion of the median strand with one of the lateral cords. Later on these strands fused also. In another case the complete leaf-trace failed to enter by one gap; the median and one of the lateral bundles entered through the main gap, «while the other lateral bundle inserted itself between two axial bundles, with which it ultimately fused. The compound bundle so formed then behaved like an ordinary lateral leaf-trace bundle, fusing with one of the forks of the median leaf-trace. There are two chief types of leaf-trace insertion in Meconopsis cambrica. In the upper leaves inserted just below the flowers the leaf-trace consists of three strands uniting into one cord which enters the axis system by a gap in that system. In the lower leaves the leaf-trace consists of several strands, united ultimately into one median and two lateral cords. An axillary branch occurs in the axil of each of these leaves, whose vascular system splits into two parts, each half inserting itself on either side of the gap in the stem system. The median strand of the leaf-trace enters the gap, followed by the lateral strands. The median trace splits, and each half fuses with the adjacent lateral cord. These two strands fuse later into one, which merges into the vascular ring of the axis. In Argemone mexicana two types of leaf-trace insertion are also found. In the case of the upper leaves the leaf-trace consists of several bundles showing a tendency to ageregate into three groups. ‘These give off well-marked anterior strands which turn through 180 degrees, thus making the leaf-trace take the form of a circle of bundles. The vascular bundles of the stem separate opposite the leaf insertion, and the leaf-trace throws its anterior strands right and left as it enters. Irregular fusions occur, so that ultimately the number of bundles in the stem is scarcely increased by the leaf-trace entry. In the case of the lower leaves three or five circles of bundles are found. The median series enters first, while the lateral circles follow, as shown in fig. 36. When an axillary branch is present some of its bundles are sometimes intercalated between those of the leaf-trace, but usually all of it passes down the stem, half on either side of the leaf-trace. The tendency for the leaf-trace bundles of the lower leaves to unite into three or five circles is not so marked in A. hispida; the anterior strands are later in appearing, and are not so well developed. The same features are presented by A. ochroleuca. The leaf-traces in Chelidonium majus are multifascicular at their points of in- sertion, and the vascular system of the axillary branch forms two semicircular cords, the gaps left being opposite the main axis and the leaf-trace. The median strand of the leaf-trace enters by the gap opposite to it. A gap is then formed in each of the semicircular cords coming from the axillary branch, and the lateral strands of the leaf-trace enter by these gaps. The whole of these bundles may maintain an 602 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON > independent course through the next internode, but more usually the median bundle only runs distinct, while the lateral bundles fuse with adjacent bundles. “This species shows occasionally two axillary branches in the axil of each leaf. Glaucium corniculatum.—The leaf-trace consists of several strands which follow fairly independent courses down the stem, the vascular tissue from the axillary branch being intercalated between the leaf-trace strands. Fusion then takes place between these bundles and the stem bundles, so that the number of cordsin the stem is not much increased over that present above the leaf entry. In Boccoma microcarpa the leaf-trace usually consists of seven strands which (save the median one) split and again reunite in the basal region of the petiole before entering the stem cylinder (fig. 37). The mode of entry is very gradual. First of all the vascular system of the axillary branch separates and the median bundle from the leaf enters; then the lateral bundles of the axillary branch separate to permit the lateral cords of the leaf-trace to enter between them, but there are no immediate fusions. Further down the bundles of the branch fuse and taper off, becoming much less prominent than the leaf-trace cords intercalated between them. The entry of the basal leaf-traces is effected in a similar manner. In Eschscholaa californica the leaf-trace is formed of one large median and four to six smaller lateral bundles. These latter fuse among themselves, while the median bundle remains distinct. As a rule the median and lateral cords do not enter by the same gap, but the former is separated from the latter by certain stem bundles. The median bundle retains its individuality for a considerable distance, while the lateral strands merge into the stem cylinder. In Hypecoum grandiflorum the leaf-trace is uni- or tri-fascicular. The upper leaf-trace, which is usually unifascicular, is inserted between two segments of a stem bundle, and maintains its independence through an entire internode. Although the phyllotaxis is a 2/5 spiral, the first two leaves arise so close together that the leaf- traces are inserted practically simultaneously. When the leaf-trace is trifascicular the method of entry is identical with that of Hschscholzia californica; many varia- tions, however, occur. In Dicentra formosa the traces are unifascicular and enter without forking, while the vascular tissue of the axillary branch arranges itself half on either side of the entering leaf-trace. The radical leaves have generally three main bundles and several smaller ones, all of which enter by one gap; and, although at first they remain independent for some time, they gradually fuse in groups and ultimately lose their identity in the stem cylinder. The leaf-trace in Adlwmia cirrhosa consists of an are of several independent strands which enter the stem system by one gap. Their number is, however, reduced to five by lateral fusion before actual entry (fig. 38). Two marginal strands, one on either side, fuse with adjacent stem bundles—sometimes simultaneously, sometimes in succession. Three bundles remain, of which the two lateral unite with these com- THE ANATOMY OF THE STEM OF THE PAPAVERACE, 603 pound bundles, while the remaining median bundle pursues an independent course » for some distance before fusing with the stem cylinder. The entry of the leaf-trace in Romneya trichocalyx is effected in a very simple manner. The trace consists of five to seven strands; the three central bundles usually fuse to form one large median bundle. The vascular tissue of the axillary branch distributes itself half on either side of the entering leaf-trace and gradually loses its independence by lateral fusion, although the median bundle maintains its identity for a considerable distance. The traces from the lower leaves do not, however, enter entirely by one gap, although the median bundle, as in the case of the upper leaves, remains distinct. In Corydalis racemosa the leaf-trace at the moment of insertion on the stem consists of three strands, one large median and two smaller lateral. The bundles of the axillary branch fuse laterally to form two vascular masses (fig. 39), between which the median leaf-trace enters, while the two lateral leaf-trace bundles enter by smaller gaps between adjacent stem bundles. The same method is adopted in C. glauca, but in C. capnoides the three bundles enter together. The median cord divides into three just before entry, all five bundles showing variations in degree of fusion amongst each other and with the stem bundles. ANATOMICAL PECULIARITIES IN THE STEM OF THE PAPAVERACE. In the foregoing pages attention has been drawn to certain peculiar anatomical conditions in the structure of the stem of the Papaveracez, of which the following are the more noteworthy :— 1. Secondary Thickening in the Stem Base of Papaver orientale. An extra cambium forms in the medulla, producing xylem outwards and phloem inwards. ‘This is the only case of abnormal secondary thickening that has been met with in the order. The condition is in all respects comparable with that described by Santo (7) as occurring in the stem of Tecoma radicans, and indications of a similar anatomical peculiarity appear in Rumex crispus. 2. The Secondary Thickening in the Walls of the Xylem of Romneya trichocalyz. The pitted vessels and tracheids possess a double spiral thickening such as LicEr found to occur in Dendromecon rigidum, Benth. 3. The Occurrence of Concentric Vascular Bundles. Concentric vascular bundles occur in Papaver Rheas, P. somniferum, P. lergatum, P. alpinum, and Glaucium corniculatum. The phloem invariably 604 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON occupies the centre of the bundle, and the concentric character is found to be due to the insertion of an axillary branch. Such concentric vascular cords also occur at the lower ends of the leaf-trace bundles of many rhizomes of Monocotyledons, e.g. Iris germanica, Cyperus aureus, Carex arenaria, Papyrus, etc. (5, 6). 4. Folding wn Cell Walls. Many of the cortical cells of Corydalis racemosa as well as of the parenchyma surrounding the vascular bundles show marked folding of their walls. Often these cells form a continuous layer, suggestive of an endodermis. Frequently such cells are unequally distributed, e.g. the dorsal and lateral surfaces of the bundles are destitute of them, while the ventral side may have a layer three or four cells thick. This irregularity of distribution, however, scarcely suggests an endodermis. In Corydalis solida the cells with folded walls are often restricted to one layer round each vascular cord, which may have led to the conclusion (3) that this layer was an endodermis and that Corydalis solida was polystelic. Detailed investigation shows that this view is untenable. In Argemone ochroleuca cells with folded walls occur not only round the vascular bundles but also in the epidermis and cortex, while in Dicentra formosa practically the entire medulla consists of such cells, as well as the cells in the more immediate neighbourhood of the vascular bundles. In Adlumia cirrhosa and Eschscholaa californica folds occur on the walls of some of the cortical cells as well as in those near the vascular system. Licer and DE Bary also suggest that these folds are caused by the limiting nature of the tissue in which they are found, but this theory does not account for the folds in the medullary cells of Dicentra formosa nor for those in the cortical cells of Corydalis racemosa. 5. The Occurrence of: Reticulate Secondary Thickening i the , Walls of the Secondary Xylem. In by far the majority of plants the walls of the vessels and tracheids of the secondary xylem are pitted, but-in Papaver there is a preponderance of the reticulate type of secondary thickening, while P. pilosum, P. lateritium, P. rupifragum, P. orientale, P. nudicaule, and P. alpinum show reticulate elements, to the exclusion of all pitted forms. Reticulate elements are also found in Meconopsis cambrica and Chelidomum majus. Similarly, the xylem of the Crassulaceze also shows reticulate thickenings in the secondary wood, though the stems in that order are much more lignified than in the Papaveracez. 6. The Occurrence of Bordered Pits. Bordered pits occur in the xylem elements of Bocconia microcarpa similar to those found in other dicotyledonous woods, e.g. Cheilanthus arboreus, Vitis, ete. THE ANATOMY OF THE STEM OF THE PAPAVERACES, 605. 7. Peculiarities of the Cortex of Romneya trichocalyx. “Arms” of rather thick-walled cells span the gap between the pericycle and epidermis. These appear to be primarily for support, and may be compared with the collenchymatous pads of Eschscholzia californica. SUMMARY. 1. The stem structure throughout the Papaveracez is very uniform, the chief characteristics being (a) A subepidermal band of chlorophyll-bearing tissue. (b) A sclerotic pericycle (except in Argemone mexicana and Romneya trichocalyzx). (c) A massive pith, often fistular, e.g. Hypecowm procumbens. Minor differences occur which are constant for individual genera and thus useful for classification purposes, e.g. (a) The amount of cortical tissue; eg. two layers in Chelidoniwm majus, the outer being a colourless exodermis, the inner only being green, or several layers, as in Papaver orientale and P. somniferum. (b) The distribution of chlorophyll. All the cortical cells may contain chlorophyll, e.g. Papaver Rheas, or only some of them, e.g. P. orventale, Corydalis racemosa, Adlumia cirrhosa, Romneya trichocalyx, and Dicentra formosa. (c) The intensity of sclerosis in the pericycle, which may vary considerably. In Romneya trichocalyx and Argemone mexicana the pericycle is not sclerotic, in Papaver pilosum and P. Rheas it is entirely so, while it it is only feebly sclerotic in Hschscholua californica, Corydalis race- mosa, Papaver somniferum, and P. orentale. (d) The thickness of the pericycle. In the fascicular regions in Meconopsis the pericycle is usually five or six layers deep, while in Hypecoum and Eschscholzia it consists of only one layer. In the interfascicular regions it usually merges gradually into the non-sclerotic medullary rays, but in Adlumia cirrhosa there is an abrupt change from sclerotic to non-sclerotic tissue in this region. (e) The degree of differentiation of the exodermis. The exodermis (of LicEr) is well marked in some genera, e.g. Chelidonium, in others, e.g. Papaver Rheas, scarcely at all. (f) The distribution of the vascular bundles. These are numerous, and arranged in a single compact circle in Argemone mexicana, they are few and far apart in Hypecoum procumbens, numerous and arranged in a monocotyledonous manner in P. orientale and P. somniferum. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 18). 88 606 R. J. HARVEY-GIBSON AND MINNIE BRADLEY ON (7) The intensity of sclerosis in the border of the internal parenchymatous mass. In some forms the border is non-sclerotic, in others feebly so, while in others still, e.g. Argemone mexicana and Boccona microcarpa, the sclerosis is intense. This feature is, however, not constant, for the genus for Argemone hispida has no sclerotic border. (h) The nature of the secondary thickening in the trachee and tracheids is distinctive of Romneya trichocalyx. (7) The sectional outline of the stem and the degree of differentiation reached in the cortex, ridged stems generally having subepidermal pads of collenchyma. The stem bases in the Papaveracez also exhibit a very uniform anatomical structure. In all types examined, save Corydalis capnoides and LEschscholza californica, the vascular bundles as they approach the base of the stem proceed to unite into a cylinder broken only by the entry of the leaf-traces of the radical leaves. The stems are not all equally lignified; thus Bocconia muicrocarpa, Romneya trichocalyx, and Papaver nudicaule have well-lignified stem bases, those of P. Rheas and many others are only moderately lignified, while those of Chelidonium majus, Corydalis capnoides, and C. racemosa are scarcely woody at all, their secondary xylem consisting almost entirely of unlignified parenchyma. The secondary xylem consists of trachez, tracheids, xylem fibres, and parenchyma. In the woody stem bases the fibres predominate, while in the non-woody forms the parenchyma is more abundant. The vessels and tracheids of the secondary xylem are often pitted, but in many species of Papaver the reticulate type is more common, as also in VMeconopsis cambrica and Chelidonium majus. Asa general rule reticulate elements are more common in the secondary xylem of non-woody forms, but Papaver alpinum and P. nudicaule are exceptional in this respect. The secondary phloem consists of cambiform cells amongst which the sieve tubes and companion cells are difficult to locate. 2. The methods of leaf-trace entry, as has been shown, are numerous. Often the same species presents three or four different types. The most common type is that exhibited by Papaver Rheas, P. dubium, P. levigatum, P. laterituwm, and P. rupifragum. Another type of entry is that seen in P. pilosum, P. alpinum, P. nudicaule, and P. sommferum, while P. glawcum and P. orientale show characters common to both, but at the same time with peculiarities of their own. Meconopsis is the only other genus examined showing marked similarity to P. Rheas, but it possesses no anterior strands, a feature exhibited by Argemone only. The leaf-trace insertions of Glaucium and Dicentra closely resemble each other, while those of Eschscholzia, Hypecoum, Romneya, and Corydalis form another group, the remaining genera showing special features of their own. 3. Similarly the vascular system found in the axillary axis varies. The bundles THE ANATOMY OF THE STEM OF THE PAPAVERACEA. 607 composing it may fuse laterally just before their insertion on the vascular ring into two or four vascular cords. In other cases the vascular bundles remain isolated throughout. Again, the vascular tissue of the axillary branch may be inserted half on either side of the leaf-trace or may be partly or wholly intercalated between the leaf-trace strands. 4. Laticiferous tissue occurs in all the Papaveracez examined, but its distribution and histological differentiation vary in different cases. There is a gradual transition from the simple laticiferous tissue of Corydalis, where it occurs chiefly in the non- vascular areas, to the complex system seen in Papaver and Argemone, where it occurs exclusively in the vascular system. Meconopsis and Chelidonium form transitional types, for their laticiferous tissue is found at the periphery of the vascular bundles and consists of articulate tubes which are seldom branched. The discussion of the exact affinities of the different genera must of necessity be postponed until the anatomy of the other organs has been investigated. BIBLIOGRAPHY. (1) Etxan, Tentamen monographiz generis Papaver. Konigsberg, 1839. (2) Hi11, “ Histology of the Sieve Tubes of Angiosperms.” Annals of Botany, xxii, 1908. (3) Van TizeHem, “Un nouvel exemple de tissue plissé.” Journal de Botanique, t. v, 1891. (4) Linx, Icones Anatomico-Botanicxe. Berol, 1837-42. (5) SoLEREDER, Systematic Anatomy of the Dicotyledons. Oxford, 1908. (6) De Bary, Comparative Anatomy of Phanerogams and Ferns. Oxford, 1884. (7) Santo, Botanische Zeitung. 1864. (8) Eneuer and Pranti, Die natirlichen Pflanzenfamilien. Vol. iii, 1891. (9) Literr, “Les laticiféres de Glaucium et de quelques autres Papavéracées.” Bull. Soc. Linn. Normandie, sér. 4, vol. v, 1891. (10) Léger, ‘“ Recherches sur l’appareil végétatif des Papavéracées.” Loe. cit., vol. xviii, 1895. (11) Lies, “Sur la présence des laticiféres chez les Fumariacées.” Compt. Rend. Acad. Sci., cxi, 1890. (12) Mosstus, Ber. deut. bot. Gesell. 1887. (13) Sarton, ‘‘ Rech. exp. sur l’anatomie des pl. affines.” Ann. Sci. Naturelles: Bot., sér. 9, t. ii, 1905. (14) Junes, “‘Beitrage zur Kenntnis des Gefasspflanzen Schleswig-Holsteins.” Jahrb. Hamb. wiss, Anst., xxii, 1904. (15) Eastwoon, Proe. Calif. Acad., ser. 3, Bot. 1. EXPLANATION OF PLATES. (All figures marked ““L” are drawn under a magnification of approximately 80, those marked “H” under a magnification of 450, save where otherwise stated. ) Puate I. Fig. 1. Transverse section of stem of Papaver Rheas (L). », 2, Transverse section of the peripheral region of the stem (H). », 3. Longitudinal section of the stem, showing sieve plates and sieve fields ( x 800). a 1 12, 13. 14. 15. THE ANATOMY OF THE STEM OF THE PAPAVERACE. . Longitudinal section of the phloem, showing sieve tubes with callus and laticiferous cells with perforations (H). . Longitudinal section, showing sieve tubes and companion cells (H). . Transverse section of vascular bundle (H). . Longitudinal section of phloem, showing laticiferous tubes (H). . Transverse section of primary xylem, showing conjunctive parenchyma (H). . Transverse section of secondary xylem, showing wood fibres replacing conjunctive parenchyma (H). 10. Transverse section of vascular system at the base of the stem, showing concentric bundles (L). (on Plate II), Transverse section, concentric bundle from stem base (H). a-—f. Successive transverse sections, showing the mode of entry of the leaf-trace (L). Transverse section of stem of Papaver somniferum (L). Laticiferous cells of Argemone mexicanu (H). Superficial layers of the stem of Chelidonium majus (H). 15a. Transverse section of stem of Chelidonium majus, diagrammatic (L). 16. re 37. 38. 39. Superficial layers of the stem of Glaucium corniculatum (H). Superficial layers of the stem of Bocconia microcarpa (H). Puate IL. . Bordered pits in surface and section from the xylem of Bocconia microcarpa (H). . Transverse section of stem of Eschscholzia californica (L). . Peripheral layers of the stem (H). . Transverse section of peripheral layers and vascular bundle of Hypecowm grandiflorum (H). . Xylem elements containing latex (H). . Transverse section of peripheral layers and vascular bundle of Dicentra formosa (H). . Transverse section of stem of Adlumia cirrhosa (L). Prate IIf. Cortical region of the stem of Romneya trichocalyx (H). . Epidermis and subepidermis of Corydalis racemosa (H). . Cells with folded walls on the inner margin of the vascular bundles (H). . Fusion of the vascular bundles in the stem base of Papaver orientale (L). . Medullary cambium in the stem base (L). . Formation of secondary xylem and phloem from the medullary cambium (L). a, b, c. Successive stages in the entry of the leaf-trace of Argemone mexicana (L). a, b, c. Successive stages in the entry of the leaf-trace of Papaver dubium (L). a, b, c. Successive stages in the entry of the leaf-trace of Papaver glauciwm (L). 35. Types of leaf-trace entry in Papaver lateritium (L). a, b, c. Entry of leaf-traces in Argemone mexicana (L). Entry of leaf-trace in Bocconia microcarpa (L). Entry of leaf-trace in Adlumia cirrhosa (L). Entry of leaf-trace in Corydalis racemosa (L). a ee a ee Von cot Trans. Roy. Soc. Edin* R. J. Harvey Gisson anp M. BRADLEY: ANATOMY OF STEM OF PAPAVERACEZ—PLATE I. M. Bradley, del. M‘Farlane & Erskine, Lith., Edin. > ee oe ” “= Sa ae a P as i ee > .« |Trans. Roy. Soc. Edin‘ Vout. LI. R. J. Harvey GissoN AND M. BrapLey: ANATOMY OF STEM OF PAPAVERACEZ—PLATE IT, = oom 2 oe eee tee ne pO eae fie —— Sa aa i ae 20. OIE os chl Vtg he SAE spas Ss soibresemnasestecsc ce LAL GTO ee ee Alga sies tie rs V. C4 ES aaa AM ATI

SAU EEK RSE DESIGHI-NSE A Tf \ \A ROE VOSGES Ss @re (RSeacee SOI On ees BRUTE GL UALS j ‘ aS S N ——— i RE Chis CEG Ta iM. Bradley, del. M‘Farlane & Erskine, Lith., Edin. : Niro il: trans. Roy. Soc. Edin‘ OL R. J. Harvey GIBSON AND M. BrapLtey: ANATOMY OF STEM OF PAPAVERACEZ—PLATE III. »P0%a, M‘Farlane & Erskine, Lith., Edin. c— ( 609 ) XIX.—Apractocleidus teretipes: A new Oxfordian Plesiosaur in the Hunterian Museum, Glasgow University. By William R. Smellie, M.A., B.Sc. (With One Plate.) Communicated by Professor J. W. GRecory, IDS, elhase Edin. and Lond. (MS. received January 15,1916. Read March 20, 1916, Issued separately August 29, 1916.) This Plesiosaur was collected from the Oxford Clay near Peterborough by A.N. Leeps, Esq., F.G.S., and was acquired for the Hunterian Museum, Glasgow University, by Professor J. W. Grucory. The major portion of the skeleton is present, and the bones are in an excellent state of preservation. The specimen has many striking resemblances to Cryptocleidus oxoniensis, but detailed examination showed that it could not belong to that species, and the differences existing in the paddle, shoulder- girdle, number of vertebrze, and conditions of ossification are such that it cannot be retained in that genus as now defined. Under the circumstances it seems advisable to create a new genus. A short preliminary account of the specimen has already been published (Smetrie, 1915), and from that account the following diagnosis of the genus has been taken :— APRACTOCLEIDUS, gen. nov.* Plesiosaurs in which the neck is composed of about twenty-nine vertebree, of which the centra are wider than high and slightly higher than long. The oval articular ends are concave in the centre but convex near the margin. There are three pectoral vertebrae and twenty-three dorsals, or two pectorals and twenty-four dorsals. The shoulder-girdle is of HElasmosaurian type, and so ossified as to form a very rigid structure. The coracoids are exceptionally broad, and the postero-lateral processes are greatly produced. The dorsal rami of the scapule are widely extended in a ge, and in the adult the anterior portion has grown forward beyond the clavicles, which are very thin films similar fashion. The ventral ramus of the scapula is lar of bone lying wholly within the visceral surface of the scapule. A rudimentary interclavicle may be present. In the mid-ventral line the scapule and the anterior parts of the coracoids impart a slightly carinate appearance to the shoulder-girdle. The humerus is greatly expanded distally, and articulates with four elements. The pelvis is of great breadth, and the wide spread of the antero-external angles of the pubes, taken in conjunction with the breadth of the shoulder-girdle, indicates a genus of exceptionally broad build. The femur is slightly smaller and more slender than the humerus, and is not greatly expanded distally. * The name Apractocleidus, meaning idle or functionless*clavicle, has been adopted to accord with the names of the related genera T’ricleidus and Cryptocleidus. For the suggestion of the name I am indebted to the Rev. Gavin Warnock, B.D. : TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 19). 89 610 WILLIAM R. SMELLIE ON The genus is represented by one species, Apractocleidus teretupes, n. sp., the specific name (from teves=rubbed, well-chiselled, elegant, and pes=a foot) being descriptive of the symmetrical outline of the paddle. . In the type specimen the parts preserved form a large portion of a fully adult individual, and the state of preservation of the bones is remarkably perfect even for Oxford Clay fossils. The head, unfortunately, is absent, but the vertebral column, ribs, shoulder and pelvic girdles, and fore and hind limbs are all well represented. Vertebral Column. The bones present comprise the atlas and axis and other twenty-seven cervical vertebree—three pectorals, twenty-three dorsals, and one sacral. All posterior to the | >. Hh ian PRS | ZS Ul ‘\ } INNS q Pi Va Wy, “ WSs; y oie \ af 4 \ We) Mie 1 Ii) <0 . aS ss N In] [ii mil “S \\ od. I ca a -fY —\\\ Wipe Ny ZA Dl (SS — FE a.w.b Trext-Fic. 1. Atlas and axis of Apractocleidus teretipes: A, from front ; B, from right side ; C, from behind. (Type specimen V. 1091, nat. size.) at.a., neural arch of atlas; a.w.b., anterior wedge-bone ; a#.a., neural arch of axis; aa.c., centrum of axis; hy.7., hyapophysial ridge; 2.c., neural canal ; n.sp., neural spine ; od., odontoid ; p.z., posterior zygapophysis; 7.,, rib of atlas ; 7.2, rib of axis, first sacral are missing. From the atlas to the sacral the series seems complete, and, while it is not impossible that the neck lacks its full complement, it is certain that pectoral and dorsal regions show no missing vertebra,.and it seems a sound assump- tion to consider the series complete as far as it goes. The atlas and aais (text-fig. 1, A, B, C).—The various bones composing the atlas and axis are so completely fused in the specimen that it is with the greatest difficulty — and uncertainty that the sutures can be even partially traced. In the atlantal cup the sutures between the anterior wedge-bone (a.w.b.) and the odontoid (od.), and between the odontoid and the lateral pieces of the neural arch of the atlas (at.a.), are almost obliterated. The cup itself is deeply excavated and smooth, and, in section, APRACTOCLEIDUS TERETIPES. 611 shows considerable flattening on the top. The anteridr wedge-bone forms the lower third of the cup, and the rim extends downwards to the base of the prominent hyapophysial ridge (hy.r.). The lateral pieces of the neural arch of the atlas (at.a.) are broken, but, from what remains, it is clear that they did not unite above but continued back and fused with the anterior zygapophyses of the axis, thus enclosing a small opening on each side between the arch of the axis (aa.a.) and the arch of the atlas. In the axis there is a low neural spine (n.sp.) with an indifferently developed ridge at its base. The posterior end of this spine is 1 em. thick, and contains two deep, semicircular facets separated by a vertical ridge (text-fig. 1,C). The neural canal (n.¢c.), in section, is an ellipse with the long axis vertical, and at the AXIS end this elongation is more pronounced. The rib of the atlas (7.,) has evidently Trext-Fia. 2. Anterior cervical vertebra (sixth) of Apractocleidus teretipes: A, from left side ; B, from behind. (Type specimen V. 1091, 4 nat. size.) a.a., anterior angle of rib; a@.z., anterior zygapophysis; c.7., cervical rib; n.c., neural canal ; n.s., neural spine ; p.z., posterior zygapophysis. been considerably smaller than the rib of the axis (r.,), although both are now incomplete. The facets for those ribs are not confluent, but distinctly, though slightly, separated. The posterior face of the centrum (az.c.) is much wider than high, and the upper side, forming the floor of the neural canal, is quite flat. The central portion is gently concave, but between this part and the circumference there is a convex portion and the rim is turned back, the reflection of the under part being most pronounced. Besides the atlas and axis there are other twenty-seven cervical vertebre (text- fig. 2; Plate, fig. 3). In the measurements of the. centra, the proportions are not found to differ in any essential from those of Cryptocleidus owoniensis. The ends of the centra are transversely oval, the upper part, forming the floor of the neural canal (n.c.), being flat or slightly concave. The ends are decidedly concave in the centre, but convex and reflexed at the margins. The ventral side, between 612 WILLIAM R. SMELLIE ON the ribs, is concave both longitudinally and transversely except for the longitudinal ridge between the nutritive foramina. This ridge is inconspicuous in anterior cervical regions, but, towards the posterior, the nutritive foramina become more widely separated and the consequent increase in the width of the dividing ridge causes the ventral side of the centra to become actually convex from side to side. By the time pectoral and dorsal vertebra are reached this process has caused the centra to assume a much more circular outline. The sides of the cervical centra above the rib facets are gently concave. ‘The centra in middle cervical regions have rugosities and tubercles situated both on the sides and under surfaces, close to the posterior ends. These are less well marked in the anterior ends. The rugosities persist into dorsal regions and form definite plications. Throughout the neck the neural arches are fused to the centra, but the sutures can be traced, most easily in the anterior region, as concave lines coming well down the sides of the centra. The neural arch occupies the whole length of the centrum, and bears strong anterior and posterior zygapophyses (a.z. and p.z.), the latter pro- truding a greater distance over the end of the centrum. In the anterior part of the neck the anterior and posterior zygapophyses are connected by a sharp ridge, concave upwards, which persists, although becoming less sharp, in the posterior part of the neck. The neural spines (n.s.) are fairly perfect in cervical regions. The spine on the first cervical is very similar to that on the axis, and half way between the tip and the line of the zygapophyses there is a pronounced rough ridge sloping downwards and forwards. This ridge persists in the other anterior cervicals, but is less apparent, and is represented finally by a few tubercles half way up the side of the spine. Both anterior and posterior borders of the spine are very sharp. The cervical ribs (¢.r.) are noteworthy for their variation. They are greatly compressed from above downwards, especially in middle cervical regions. Some might be described as straight, flat rods of bone with shght rounding at the anterior end of the extremity. Others show the anterior angle (a.a.), relic of the double head, almost as prominently as the cervical ribs of Pzcrocleidus. These last occur chiefly in the anterior part of the neck, where the ribs are short. Towards the posterior the ribs become long and straight and lack the terminal expansion present in the others (text-fig. 6, A). Hach rib broadens at its junction with the centrum, and occupies practically the whole length of the centrum in anterior cervical regions ; but the posterior cervicals have a distinct space between the anterior edge of the rib facet and the end of each centrum. Going from the posterior end of the neck forwards, the cervical ribs show a progressive degree of fusion. Thus, on the left side, from the first to the nineteenth cervical, the ribs are closely fused to the centra (text-fig. 2) and the suture closed and hard to trace. The next three ribs are fused imperfectly, leaving the suture gaping. The next has been slightly fused and has separated along the suture, while the remainder are quite separate and show no APRACTOCLEIDUS TERETIPES. 613 trace of ever having been fused (text-fig. 6, A). They have large, rough convex heads (h.), which are accommodated by corresponding concavities in the centra. The free cervical ribs are much less compressed from above downward (cf. text-figs. 2, c.r., and 6, A), and about the twenty-seventh a section through the neck of the rib is an equilateral triangle. It is noteworthy that the neural arches of the posterior cervical vertebre are closely fused to the centra, the sutures being even less visible than in the anterior cervicals, so that the freedom of the ribs is not matched by any corresponding looseness in the neural arch of the same vertebra. The freedom and lack of breadth of the ribs in the posterior of the neck must have served to mitigate Text-Fie. 3. Pectoral vertebre of Apractocleidus teretipes : from left side. (Type specimen VY. 1091, 4 nat. size.) «.z., anterior zygapophysis ; 7.s., neural spine ; p.%, posterior zygapophysis ; 7./., rib facet. the excessive rigidity which, as insisted on by WrLListon (1913), must have existed in the thick portion of the neck of Plesiosaurs. Pectoral vertebrx (text-fig. 3).—The sutures between neural arch and centrum are so indistinct in posterior cervical and pectoral regions that nothing definite can be deduced from them. The rib facets begin to rise on the centra about the twenty-fifth cervical vertebra, and after the twenty-seventh a sudden change occurs— instead of the rib articulating in a rough depression in the side of the centrum, a short process with smooth termination receives the rib, and this process is situated higher up on the centrum. The next two vertebre are very similar, but show the process steadily ascending on the side of the centrum. The smooth terminations of the processes (rf) of the pectoral vertebrae show slight sinuosities on their other- wise flat surfaces. In the next vertebra (the first dorsal) the transverse process is borne wholly on the neural arch, is considerably longer, and has a decidedly convex oval termination whose anterior dorsal margin is reflexed. In this region 614 WILLIAM R. SMELLIE ON of the back the centra present a more circular outline, but still preserve the flattening under the neural canal. Dorsal vertebrex (text-fig. 4, A, B)—In the dorsal vertebrze the neural spines (n.s.) and transverse processes (¢.p.) have suffered much from fracture and crushing. In the anterior the transverse processes continue to rise on the neural arch and to increase in length. They terminate in expanded oval facets (rf), the long axis of the oval sloping downwards and backwards. These rib facets are strongly convex and have the upper anterior lip reflexed. This is matched by a similar structure on the concave articular surfaces of the heads of the dorsal ribs (text-fig. 6, B, C, D). AUisp-ayelaites 2b Middle dorsal vertebra (39th) of Apractocleidus teretipes: A, from right side; B, from front. (Type specimen VY. 1091, 4 nat. size.) a.z., anterior zygapophysis ; ”.c., neural canal ; n.8., neural spine; p.z., posterior zygapophysis; 7.f., facet for rib ; ¢.p., transverse process. In dorsal regions the nutritive foramina, following the rise of the transverse processes, rise on to the sides of the centra. Towards the posterior end of the back the transverse process becomes shorter, more slender, and lower set on the neural arch, and the articular ends of the centra become wholly concave and lack the convexity and reflexion at the margins. The sacral vertebra (text-fig. 5, A, B) shows a wide, rough, excavated facet (rf.) for articulation with the sacral rib (C), borne partly on the centrum and partly on the neural arch. The sacral rib has not been fused to the vertebra, and the head of the rib bears a rough articular surface corresponding to the rib facet (rf) on the vertebra. This surface is formed of two surfaces meeting at an angle of about 100 degrees, the upper portion, doubtless, corresponding to that part of the facet APRACTOCLEIDUS TERETIPES. 615 contributed by the neural arch, and the lower to the portion contributed by the centrum. The vertebre posterior to the first sacral have not been preserved. The ribs (text-fig. 6) in cervical regions have been referred to above. In anterior dorsal regions the ribs are short (B, C), but increase rapidly in length (D) towards middle dorsal regions. The heads (h.) of the ribs are smooth and concave, and fit the rib facet on the corresponding transverse process with great accuracy. The upper anterior margin of the articular surface is broken by a small lip sharply folded back. In middle dorsal regions this lip becomes less abrupt and the articular surface of the head becomes almost flat. The head (h.) bears numerous roughened pits and tubercles (s.) for muscle attachment, and these are most pronounced on the upper and posterior portions of the head. The rugosities continue down the anterior side Trxt-Fia. 5. Sacral vertebra and rib of Apractocleidus teretipes: A, from right side; B, from behind ; C, sacral rib. (Type specimen We 1091, 4 nat. size.) a.z., anterior zygapophysis ; m.c., neural canal; 1.s., neural spine ; p.z., posterior zygapo- physis (broken) ; 7./., rib facet ; s.f., facet for ilium. of the rib and converge towards the smooth ridge (r.), which is most marked in the ribs of anterior dorsal regions. In the long ribs (D) of middle dorsal regions this ridge is a much less prominent feature: These long heavy ribs contain a well-defined pit (p.) a short distance from the end, on the upper anterior side. This pit probably served for the insertion of a strong ligament for the support of the long and heavy rib. The lower portions of the long ribs, below the ridge (r.), remain of almost uniform diameter and oval to circular in section. At the distal end there is a shght expansion, and the bone ends in a concave, oval surface, having been tipped with cartilage in life. The sacral ribs (text-fig. 5, C) are short strong rods of bone some- what expanded at the facets at their ends. The facet (s.f.) for the attachment to the ilium is an irregularly convex surface with well-defined margins but no prominent tubercles. Numerous ventral ribs and rib fragments are present, and doubtless these went to the formation of a plastron of ventral ribs (text-fig. 9) arranged as in Cryptoclerdus (ANDREWS, 1910, p. 175, text-fig. 86) 616 WILLIAM R. SMELLIE ON Text-Fie, 6. Ribs of Apractocleidus teretipes (type specimen V. 1091, } nat, size): A, posterior cervical rib; B and C, anterior dorsal (33rd) rib, left side ; D, middle dorsal rib, left side. h., head; p., pit for insertion of ligament ; 7., crest of ridge ; s., roughened surface for muscle attachment. The shoulder-girdle (text-fig. 7, A; Plate, fig. 1) consists of fused coracoids (cor.), fused scapulee (sc.), the anterior parts of the left missing, and right clavicle (cl.), the left having been lost with the anterior part of the scapula. The scapulx are triradiate bones each consisting of a dorsal ramus (d.sc.) directed upward and out- ward, a heavy bar directed backward and carrying the articular surface for the APRACTOCLEIDUS TERETIPES. 617 humerus (g/.) and the surface of union with the coracoid, and the ventral ramus (v.sc.), Which meets the ventral ramus of the other scapula in the mid-ventral line, the two sending back the posteriorly directed process which passes imperceptibly into the anterior median extension of the coracoids. The dorsal rami (d.se.) are chiefly remarkable for their length and wide spread, and this gives the first intimation that the specimen is an animal of distinctly Trext-Fic, 7. A, Shoulder-girdle of Apractlocleidus teretipes, from above. (Type specimen, V. 1091, about + nat. size.) B, Shoulder-girdle of Cryptocleidus oxoniensis, from above. (R. 2616, about 4 nat. size. After ANDREWS, 1910, text-fig. 87.) cl., clavicle; cor., coracoid; d.sc., dorsal ramus of scapula ; for., foramen between coracoids; g/., glenoid cavity ; p.c.p., postero- external process of coracoid ; sc., scapula; v.sc., ventral ramus of scapula. broader build than Cryptocleidus, a fact which is further emphasised by both coracoids and pubes. Each dorsal ramus has suffered a curious fracture just where it leaves the main body of the scapula, so that the end has been slightly displaced and rejoined, while in neither case is the tip complete enough to show how the bone terminated. Just anterior to the above-mentioned fracture, and about 2 cm. from the outer edge of the bone, a roughened ridge rises and runs forward and inward on the ventral side of the bone almost parallel to the margin. This ridge termin- ates close to the edge of the bone near the point reached by the postero-external angle of the clavicle (cl.), but a continuation of the line would lead along the TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 19). 90 618 WILLIAM R. SMELLIE ON smooth concave part of the margin and terminate in the strong outwardly directed tubercle. The posterior bar of the scapula is triangular in section, and terminates in two faces making an angle with each other distinctly greater than a right angle. The one face, the glenoid surface (gl.), forms half an ellipse, is comparatively smooth and gently concave ; it measures 5°5 cm. by 6 em., the line of junction with the coracoid being the smaller. The other face, the surface of union with the coracoid, is a triangle 8 cm. high and almost right-angled in the outer (ventral) angle. This surface and the corresponding one on the coracoid are very rough, but there has been no extensive symphysis, for, in places, the separation of the bones has left pieces of the one adhering to the other, indicating a close union and at least partial fusion during life. About 3 cm. from the glenoid surface the lower border rises into a rough ridge which runs in a straight line towards the anterior extremity of the scapula, but flattens out before it passes the well-marked tubercle on the anterior lateral border. Between this ridge and the similar one on the dorsal ramus there is a well-marked angular hollow. The inner edge of the posterior bar of the scapula is thin and sharp; it forms the outer border of the coraco-scapular opening. The ventral ramus (v.sc.) is partly covered on its inner (visceral) side by the clavicle (cl.), and the covered part shows a sudden and appreciable reduction in thickness. The anterior lateral border is composed of two concavities separated by the prominent tubercle referred to above. The border approaches the median line at an angle of 45 degrees, but, when it is about 3°5 cm. distant from the median line, it turns abruptly inwards and the extreme anterior portion bends up protectively in front of the clavicle and has a roughened unfinished edge as if the bone had a cartilaginous continuation in life. The median border of the scapula is fused to that of its fellow, and, as they have met after making a sharp bend downwards, a strong median ridge has resulted. ‘The posterior prolongation of the united scapulee passes imperceptibly into the anterior prolongation of the fused coracoids. The bones have been separated by fracture, and there is no trace of symphysis or suture to indicate that they were separated in life. The median ridge above mentioned continues into the cora- coids, and only flattens out when crossing a line joming the outer angles of the glenoids. This ridge is very rough, and at its highest part finishes with a concave roughened tip indicating a cartilagimous termination. The ventro-dorsal depth of this bar is 7°5 cm., while its diameter between the coraco-scapular fenestra is only 18 cm. A section through the corresponding part of Cryptocleidus is nearly circular. The inner (visceral) side of the ventral rami (v.sc.) shows considerable thickening between the anterior ends of the coraco-scapular foramina, but forward from this the median line is occupied by a hollow corresponding to the ridge on the under side. Only the right clavicle (text-fig. 7, A, cl.) is present. It lies very closely on the inner surface of the scapula, and does not protrude beyond the border of the scapula APRACTOCLEIDUS TERETIPES. 619 except for a small part of its extreme postero-lateral border. The posterior border of the clavicle is deeply concave, and its outer angle bears a very rough facet which fits a corresponding depression in the scapula, the edge of which is roughened in a similar fashion posteriorly to this point. The outer border of the clavicle follows closely the lines of the scapula, and bears a small protuberance, corresponding to the tubercle on the scapula, with gentle concavities on each side. As the anterior of the clavicle is approached the border is found to retreat further within the border of the scapula till the tip is fully 1 cm. behind the tip of the scapula. The median border is straight, and follows the downward curve of the scapula closely, although the median hollow is not conspicuous owing to the thickness of the clavicle here. The anterior end of the clavicle is much roughened at the point where the margin joins the median line. Except for the median border the whole clavicle is of very thin bone. The coracoids (cor.) are fused into one massive bone, no trace of symphysis or suture remaining. The combined bone might be described as built on a framework of four thick bars—one stretching between the glenoids; another, which thins near its middle point, occupying the median line; and one each side stretching from the glenoid to the postero-lateral process (p.ep.); while the intervening spaces are occupied by bone, thin in proportion to its distance from any of those massive bars. Thus at a point midway between the postero-lateral process and the centre of the bone the thickness is less than 2 mm. The keel-like anterior prolongation of the coracoids has already been described in reference to the scapula, and the glenoid and surface of union with the scapula are very similar to the corresponding surfaces in that bone, but make an angle of 135 degrees with each other. Between the glenoids the bone is very massive and almost flat on its inner (visceral) side. Behind this flat portion is a saucer-shaped depression, in the middle of which the bone is very thin and pierced by a foramen (for.) which is situated practically in the middle of the combined bone but slightly to the right of the median line. Continuing backwards from this point along the median line the bone thickens again, and is 1°5 cm. thick at a point only 4 cm. from the margin. The postero-external processes (p.e.p.) are chiefly remarkable for their wide spread, the distance between them being slightly greater than the length of the whole shoulder-girdle. The tips are rough and deeply concave, indicating a cartilaginous termination. This condition is continued along the posterior border of the left process for a distance of 9 em., while the right presents a smooth, rounded, finished edge at this part. Other two prominences, each situated about mid-way between the postero-external angle and the median line, terminate with thickened concave borders; otherwise the posterior border of the coracoids terminates with a fine edge of bone only about 1 mm. thick. The lateral borders, from the glenoid to the postero-external angles, are concave, the bone being thick and smooth until 14 cm. off the postero-external processes, when for 5 cm. the edge of the bone is raised into a sharp ridge with rough tubercles, 620 WILLIAM R. SMELLIE ON The figures (text-fig. 7, A and B) and measurements of the shoulder-girdle show some of the points of difference between this species and Cryptocleadus oxomensis. Notable points of difference are seen in the relative proportions of the shoulder- girdle, Apractocleidus being much shorter and relatively broader; processes are longer, ossification is much more advanced, and the mid-ventral portion of the girdle is carinate between the coraco-scapular fenestra. The increase in the ventral ramus of the scapula has brought it forward till it forms the anterior end of the shoulder- girdle, leaving the retracted clavicle lying functionless behind the upturned tip. In the scapula the angle between the glenoid surface and the surface of union with the coracoid is 110 degrees, as compared with 90 degrees in Cryptocleidus. This change better enables the glenoid to receive a more directly forward thrust from the head of the humerus at each drive of the paddle, and this makes for increased speed. It is uncertain whether the shoulder-girdle contained an interclavicle or not. The clavicle is obscurely notched on the median margin about a third of its length from ‘the forward end, so that it is possible that a rudimentary interclavicle may have been present and either lost or never ossified. The bones of the fore limb are in a very perfect state, have suffered no crushing or distortion, and exhibit details of articulation and muscle attachment as clearly as freshly macerated bones. Comparison of the figures (text-fig. 8, A, B, C) shows the striking manner in which the fore limb combines the characteristics of Tricleidus and Oryptocleidus. The humerus (B, hum.) is very similar to that of Cryptocleidus (C, hum.), but can be easily distinguished from it by the facets on the distal end. The head (h.) is almost circular in outline, and bears a convex roughened surface ; the convexity and roughness are greatest in the centre, while the rim at the cireum- ference is comparatively flat and smooth. The edge is sharp and cleanly cut, and the state of the whole head is such that the cartilage cap might have been removed only recently. The surface of the head is continued across a narrow (3 cm.) concave passage into that of the tuberosity (tw.) which forms an irregular quadrate prominence situated on the outside of the head and slightly to the posterior. The tuberosity has also been capped with cartilage, but it is convex only near the outside and is deeply concave between that and its junction with the head of the humerus. The prominent anterior and posterior borders of the tuberosity are soon lost amid the deep rugosities (m.r.) which cover the -outside of the upper surface, and shade inconspicuously into the general contour of the oval neck: The short, strong shaft of the bone appears slender by comparison with the massive head and wide distal expansion. Both margins are concave, the posterior being more so. The rugosities on the posterior margin of the great tuberosity are continued to the posterior border and inner side of the shaft, where they become very pronounced. The anterior margin is sharply rounded and smooth except for a few rugosities on the upper part of the shaft which are a continuation of those on the anterior of the great tuberosity. The distal expansion is very wide (24°6 cm.), but only 4 cm. deep at its thickest. It APRACTOCLEIDUS TERETIPES. 621 is covered with rugosities and plications which rise from the smooth shaft and spread fanwise over the expansion. These terminate in a line of small depressions situated about 1 em. from the margin, and a comparatively smooth strip is left on the -distal and post-axial borders. The distal expansion bears four distinct facets for the articulation of four bones—radius (r.), ulna (w.), pisiform (p.), and post-axial accessory ossicle (a.). The facet for the radius is the largest, and it is slightly concave. The facets for the ulna and pisiform are differentiated by a slight salient, and almost at right angles to the line of the last two there is a convex facet, about Ui, Puy ) i & S) Cm) iS) a aa zi ies = a ® | z 5 = Fy < Pies eae i ae | al | Length of centrum in| 44, 2°38) 32 3°6 3°7 40) 39 | 38 4°3 46 | 47| 46 mid-ventral line Width of posterior| 2°9) 36/ 4:2 4-7 52 | 58] 6:0 61 6°2 64 | 6:3] 61 end of centrum Approx. | | approx. Height of posterior | 2-4| 2:9) 3:3 3° Ae AG oe 4:8 5-1 53 4-8 end of centrum | approx. approx, approx. | approx. Height to top of | 55| 74) 86 |10-4 ... {13:5 | 14:0 | neural spine | “approx. | Shoulder-Girdle compared with Cryptocleidus. V. 1091. R. 2616.* Greatest length of combined scapula and coracoid 60-0 66°5 Coracoid—Length from middle of glenoid cavity to tip of postero- lateral process . 31°5 Hoe Width of united bones at hinder angle of glenoid cavity . 40°6 42:2 Pr . narrowest “point 31:7 33°U - between outer ends of postero-external angles 62°5 y7/Ol Greatest length (median) 39°5 approx. 46°7 Scapula—Greatest length (middle of glenoid to anterior ‘extremity) 33°2 32°2 Length of median border 20°5 approx. 18°8 Length i in a straight line from median border to top of dorsal ramus : 26°0 24-5 Length of glenoid surface 55 6°2 = surface for coracoid 8:0 | 8:5 Antero-posterior diameter of coraco- -scapular opening 13:1 | 12:1 Lateral a 5s 9°6 11:0 Clavicle—Length of median border 12°5 1271 Width (median border to outer angle) 127 on Length of anterior border : 175 18:0 * ANDREWS, 1910, p, 195, 628 WILLIAM R. SMELLIE ON Fore Limb compared with Cryptoclerdus. ’ Vi. LOOM: R. 2412.* = ~ OS (Sv) Humerus—Length Diameter of head Greatest width of upper end Width of shaft at narrowest Width of distal expansion Radius—Length of pre-axial border y humeral _,, - ulnar 3 Ulna—Greatest length : Length of humeral border Pisiform—Greatest length Length of humeral border Length of ulnar border = j= em mee bo Na hes CECE CB Tees SSIS) (S219) ADSDSCOUSCKABROUGS He ho Hey Ope OSA St RONCO ICS EON MOMAODN —_ ae — * ANDREWS, 1910, p. 194. Pelnic Girdle compared with Cryptocleidus. Ve, LOS. | R. 2860.*|R. 2616.* Left Right Side. Side. Thum— Length 18:8 17:2 20:0 Width of upper end . 7'8 73 6:2 lower end . ; 64 56 8-0 - middle of shaft . aye a ae ects Pubis—Length 20°8 20°9 25°3 25°8 Width (from antero- external angle to symphysial border 35'0 B45 30°5 34°2 ,, of acetabular surface 4°3 5'1 10°0 8-7 Depth z : 74 6°9 558 5°9 Width between antero- ‘external angles of two pubes : 66 55°5 66°6 Ischium—Greatest length of symphysial expansion f 21°6 22°6 18-4 21-9 Width (from acetabular surface to aN) 20°0 20°1 18-1 22°3 », of articular head 6-1 onl 88 10°7 3 Jofmeck . 56 6:4 5:2 67 2 Antero-posterior diameter of obturator foramen 11:5 11:5 10-2 © Lateral - BiH) 13:5 * Hind Limb compared with Cryptocleidus. Ls Weprooie R. 2860,* Femur—Length 30°1 27:0 Greatest width of head 9-1 ee) Width of upper end with trochanter ae 95 ” shaft at narrowest. 6:1 58 " distal expansion 19:5 16-0 ad * ANDREWS, 1910, pp. 191 and 196, APRACTOCLEIDUS TERETIPES, 629 EXPLANATION OF PLATE. Apractocleidus teretipes, gen. et sp. nov. 1, Shoulder-girdle from below (about 4} nat. size), 2. Pelvis from above (about } nat. size).* , 3. Atlas and axis and other thirteen anterior cervical vertebrz, from the left (about } nat, size). 4. Outer side of left fore paddle (about } nat. size). 5. Outer side of left femur (about } nat. size). REFERENCES. AnprREws, C. W.— 1909. “On some new Plesiosauria from the Oxford Clay of Peterborough,” Ann. and Mag. Nat. Hist. [8], vol. iv, pp. 418-429. 1910. B.M. Cat., Marine Reptiles of the Oxford Clay, part i. 1915. “Note on a Mounted Skeleton of Ophthalmosaurus icenicus, Seeley,” Geol. Mag. [vi], vol. ii, p. 145. Sretzy, H. G.— 1874. “On some Generic Modifications of the Plesiosaurian Pectoral Arch,” Quart. Journ. Geol. Soc., vol. xxx, pp. 436-449. 1892. “The Shoulder-Girdle and Clavicular Arch in Sauropterygia,” Proc. Roy. Soc., vol. li, pp. 119-151. Sme.iiz, W. R.— 1915. ‘On a new Plesiosaur from the Oxford Clay,” Geol. Mag. [vi], vol. ii, pp. 341-343, Witurston, S. W.— 1914. Water Reptiles of the Past and Present, pp. 80 and 91. * The right ilium is a plaster model made by Mr KineHorn, the Hunterian Museum. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 19). 92 : 7 ax . O ae > bd ; : : : fi ‘ ; 4. De a 5 ne A vi eed 5 : : : , : ‘. p ao - : f 3 << vP gral tin 2s eee ke ty Ju, have) Oral pay ota - < : I : . i aa rely Lae el ee ae = a! 7s ‘ , : ‘ fs , ov) (Hb 1G Se : : - . } ¥ iat The et a (ee : é re fb: : P ae aed + F ononers. we J 20 optic Qa icapeg\ i ALI % 5 a = [ a “ a af bi iF . q ea ane 1? a . as , : yan ; es xO Met ti iy hi ahha! pe ‘ a 2 i c) - — a » As ? yted ; ' i Pass E 3 "2. '* ‘ % ‘7 Af ut 7 on he — 2 — _ ” ; - Trans. Roy. Soc. Edin. Vou. LI. Wituam R. Suuviie: Apractocleidus teretipes. Fic. 1.—Shoulder-Girdle. Fic. 2.—Pelvis. Fic. 3.—Cervical Vertebre. Fic. 5.—Femur, Fic. 4.—Fore Paddle, BS © weal) 3) XX.—The Anatomy and Affinity of Platyzoma microphyllum, R. Br. By John M‘Lean Thompson, M.A., B.Sc., Chief Assistant in Botany and Robert Donaldson Research Scholar, Glasgow University. Communicated by Professor Bower, F.R.S. (With Four Plates.) (MS. received February 7,1916. Read February 7, 1916. Issued separately November 6, 1916.) In the year 1810 there appeared in Ropert Brown’s Prodromus Flore Nove Hollandiz et Insule Van Diemen, p. 160, a brief description of a rare Australian Fern to which the generic name Platyzoma was given. The habit of stem and leaf, and in general the form and position of the sporangia, led Brown to the conclusion that Platyzoma was of Gleicheniaceous affinity. It was noted that the plant was heterophyllic; for not only did it possess pinnate leaves suggestive of those of certain Gleichenioid forms, but also small, compressed, and filiform leaves devoid of pinne. To the form of the larger and pinnate leaves Brown paid particular attention, noting chiefly that the leaves of Platyzoma were unbranched, whereas a branched condition was typical of the leaves of Gleichenias. This unbranched condition of the leaf seemed to Brown a suffici- ently distinctive feature to justify the foundation of a new Gleicheniaceous genus —the monotypic genus Platyzoma; and he named the single species Platyzoma macrophyllum. Brown’s opinion was, then, that Platyzoma was allied to Gileichenia, but was, at the same time, worthy of recognition as a distinct genus. It would appear, however, that at a later date Brown doubted the value of the character on which he had laid most emphasis in founding the genus Platyzoma, for in 1814* he wrote: “Among the Australian Ferns there is no genus absolutely confined to that country, except Platyzoma, but this, perhaps, ought not to be separated from Gleichena.” Still later, in 1838, while describing Polypodiuwm (Dipteris) Horsfieldii,t Brown referred to the axis of Platyzoma, and indicated that it differed anatomically from the axis of all Gleschenias he had examined. The pinnate leaves and sporangia of Platyzoma were figured in 1842 by Baver in Sir Wm. Hooker’s Genera Filicum, and here it was affirmed that “ Platyzoma differs from Gileichenia more by its undivided leaves, than by any differences in their sporangia.” In the Synopsis Filicum, 1874, 1883 (Hooker and Baker), Platyzoma was placed as the first genus of the Gileicheniacex, and a note was appended affirming that Platyzoma is “a genus which has been too * General Remarks, Geographical and Systematic, on the Botany of Terra Line ORAS Misc. Bot. Works of Robert Brown, vol. i, p. 59, Ray Society Publications, 1866. + Plante Javanice Rariores, p. 2. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 20). 93 632 MR JOHN M‘LEAN THOMPSON ON nearly allied to Gletchenia.” F. Mi.uer, however, rejected the opinion expressed in the Synopsis Filicum, and maintained that Platyzoma should be considered a species of Gleichenia; accordingly, he named it Gleichenia platyzoma. In 1878, BreNTHEM endorsed the opinion of Hooker and Baker, and, writing in Flora Australiensis, vol. vil, p. 696, he held that Platyzoma “is a genus limited to the single species endemic in Australia. The great difference in ‘fructification’ as well as in habit appears to preclude its union with Gileschenia as proposed by F. Miuurr.” In 1897, H. Curist* sank Platyzoma in Giletchenia, and named it Gleichenia microphylla, while in 1902 Diets f followed a similar course. And lastly, in 1906, Cart CHRISTENSEN, in his Index Filicum, placed Platyzoma in very close relationship to Gleichenia, but at the same time was not disposed to reduce Platyzoma to the rank of a species of Gleichenia. It will be noted, then, that the systematists, without exception, have followed Brown’s example in placing Platyzoma in the Gleicheniacex, and that the only point upon which opinions have differed is the rank which Platyzoma merits within this group. The following characters of Platyzoma have been recorded by the systematists :— The axis is a short, horizontal rhizome beset with a thick covering of long, glistening, golden-brown hairs. The branching and fibrous roots spring from the lower surface, and occasionally from the sides of the stem. Regarding the heterophylly there is but a single statement, namely, that which is found in Brown’s original description of Platyzoma. It is remarkable that later systematists have not even referred to Brown’s remarks on this feature, for he stated clearly that Platyzoma possesses not only undivided pinnate leaves, but also leaves of a compressed and filiform character. The pinnate leaves range from about 6 inches to 1 foot in length. They are densely crowded upon the upper surface and sides of the rhizome; they are dark and sclerotic, and their bases are surrounded by clumps of firm hairs. They are simply-pinnate and unbranched, and both petiole and rachis are smooth and slender. The pinne are small and numerous, and seldom exceed 2 mm. in length. They are ovate-orbicular in outline, and their margins are so revolute that the stomata—which are restricted to the lower surface—are in a nearly closed cavity. The upper epidermis is smooth and strongly cuticularised, and much of the mesophyll is composed of water- storage tissue. The margins of the pinnz bear numerous minute irregular hairs, and similar hairs are distributed upon the lower surfaces of the pinne. The venation is a short sympodium. There are but a few sporangia in any fertile pinna. They are disposed terminally upon the lateral branches of the sympodium, and are at maturity globular and almost sessile. They are hidden beneath the broad revolute margin of the pinna; they are irregular in form, and their dehiscence is. variable. : BentHeM has noted (loc. cit.) the localities from which Platyzoma has been * Ine Farnkraiiter der Erde, p. 339. 7 Ena, and PRL, 1, 4, p. 355. THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 633 recorded. In N. Australia both Brown and GuLuiver collected this plant by the Gulf of Carpentaria, while Méiuer found it by the Fitzmaurice River and at Providence Hill. Later on, Martin encountered Platyzoma by the Glenelg River on the north-west coast. In Queensland, both Brown and Hitt have found Platy- zoma at Facing Island, while other collectors have discovered it at York Peninsula, Rockingham Bay, and on the downs of the interior. In the Synopsis Filicum it is referred to as inhabiting the tropical and subtropical regions of Australia. In 1901, Boopte* laid emphasis on the xerophytic nature of Platyzoma; and in 1910, H. Curist+ figured a heterophyllic specimen, and referred to Platyzoma as a xerophytic plant from the desert zones of Australia. Regarding the anatomy of Platyzoma very little is known. In 1893, Porravuir{ briefly described the rhizome, and referred to its curious type of stele. It was here noted that there is a central sclerotic pith, completely surrounded by an endodermis, outside which is a zone of parenchyma surrounded by a broad ring of parenchy- matous xylem. It was further noted that at the periphery of the xylem there isa narrow zone of phloém, which is in turn surrounded by a large-celled pericycle and an endodermis. Special reference was made to the shortening of the internodes, and the crowding of the leaf-traces in the cortex. No leaf-gaps were recorded by Porrautt. Nevertheless, he considered this anomalous stelar state to have arisen by obliteration of the leaf-gaps from an original solenostele. We are indebted to Boopie§ for fuller information regarding the rhizome. He discussed in greater detail the medullate stele, with its annular xylem and internal endodermis, and described the origin of a leaf-trace by the nipping out of a group of phloém and xylem elements from the periphery of the stele, and indicated the collateral nature of the leaf-trace on its way out through the cortex of the rhizome. He regarded Platyzoma as “a xerophytically reduced form in which the leaf-traces have become small and crowded.” He also indicated the possibility of Platyzoma having been derived from a solenostelic form by obliteration of the leaf-gaps and disappearance of the internal phloém of the stele. But, on the other hand, he emphasised the possibility of Platyzoma having been derived from a protostelic Gleichenia, by the formation de novo of a pith and internal endodermis. On the one hand it was noted that the stele in Gleichena is typically protostelic, and on the other that a widely different type of structure—a solenostelic structure—exists in one species, Gleichenia pectinata. Platyzoma might then be regarded as occupying—as far as stelar con- struction is concerned—a position intermediate between Gileichenia pectinata and all other known Gleichenias. To the view that the present stelar condition of Platyzoma has arisen by trans- formation of a solenostele both JEFFREY and Tansey have lent their support, the * “Anatomy of the Gleicheniace,” Ann. Bot., xv. + Die Geographie der Farne, p. 180. { “Recherches sur les Cryptogames vasculaires,” Ann. des Sciences Nat. Bot., 7¢ sér., tom. xviii. § “Anatomy of the Gleicheniacex,” Ann. Bot., xv, 1901. 634 MR JOHN M‘LEAN THOMPSON ON latter stating definitely * that “ Platyzoma, with its extreme xerophily and anomalous leaf-traces, may almost certainly be regarded as reduced from the solenostelic type.” Bower and others have inclined to the belief that the original condition of Platyzoma was protostely. But in any case, BoopiE considered that the construction of the stele of Platyzoma was sufficiently individualistic to justify the separation of Platy- zoma from Gleichenia, as proposed by Hooker and Baker. It is then evident that the facts advanced regarding the stele have led neither to the removal of Platyzoma from the Gleicheniaceex nor to a definite settlement of its position within this group. A brief consideration of the characters which have been employed in determining Platyzoma’s systematic position may not be out of place at this point. The habit of stem and leaf, and in general the form and position of the sporangia, led Brown to place Platyzoma in the Gleichenracex. These same general features have likewise been the main guides of later systematists. But at the present day, when convergent. evolution is being widely recognised among Ferns, it is necessary, for systematic determinations, to be provided with detailed knowledge of the anatomy of stem, leaf, and root, and of the original position and form of the sporangia. The exact position and form of the annulus and the form and output of spores per sporangium are of much importance. The nature of the dermal appendages should also be determined. In fact, it has been found that while general habit has frequently proved a true index in phyletic groupings, at other times—when taken by itself— it has led to gross error. In the case of Platyzoma no definite information has hitherto been advanced regarding the anatomy of leaf and root, nor has any accurate account been given of the form and development of the sporangia, the character of the annulus, and the spore form and output. Both PorrauLt and Boopie have contributed to our knowledge of the stele of Platyzoma. Nevertheless, a wider knowledge on this subject is necessary, and the facts already advanced are far from sufficient to prove or disprove the proposed Gleicheniaceous affinity. Further, one would have expected that the heterophylly noted by Brown, and shown in Curist’s figure, was a character of too great im- portance to be neglected by the systematists. Boopie has informed me that in five of the sheets of specimens of Platyzoma in Kew Herbarium the heterophylly is clearly shown; while in a beautiful specimen, collected in 1902 by Professor BaLpWwIN SPENCER at Boorroololoo, N.T., and kindly supplied to us by the authorities of the National Herbarium of New South Wales, Sydney, this heterophylly is a conspicuous feature (figs. 6 and 8). As to the original distinction drawn by Brown between the branched leaves of Gleichenia and the unbranched leaves of Platyzoma a few remarks are necessary. It was noted by Bowsr in 1899 + that the pinnate leaf of Platyzoma is occasionally forked, as is seen in specimens of a “forma furcata” sent by the late Baron Sir F. V. * “Lectures on the Evolution of the Filicean Vascular System,” New Phytologist, 1908. + “Studies in the Morphology of Spore-producing Members,” Phil. Trans. Royal Soc. London, iv. _ a i THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 635 Mutuer to the Kew Herbarium. Regarding this branching I am indebted to Boopie for fuller information. Forking of the leaf is shown in two cases in which the leaf branches once. In the first (text-fig. 6) the leaf is attached to the rhizome, the branching takes place at some distance from the base, and pinne are distributed on the rachis below the forking. The terminal portions of this leaf had been damaged. In the second specimen (text-fig. 7) pinnee are again developed on the rachis below the point of branching; and while it is impossible to determine the length of the EE & TExt-Fic, 1. TEXT-FIG. 2. TEXT-FIG. 3. TEXT-FIG. 4. Trxr-Fic. 5. Trext- TEXxtT- TEXxT- TEXT-FIG. 9. FIG. 6. FIG. 7. FIG. 8. TrEXxtT-FIG. 10. unbranched portion of the rachis—since the leaf is detached,—the branching is suggestive of a dichotomy. And further, in a specimen in the herbarium of Glasgow University a pinnate leaf was found showing perfect dichotomy (text-fig. 8). In this case the branching occurs close to the base of the leaf, and there is no pinnation below the point of forking. It is worthy of note that both simple and branched leaves are found in Gleichenia spelunce, Br. (GI. circinata, Sw.); while in Gleichenia simplex, Hk., is found a species characterised by simply-pinnate leaves. From con- sideration of these points it is apparent that the chief distinction drawn by Brown between Platyzoma and Gleichenia has broken down, and the question remains 636 MR JOHN M‘LEAN THOMPSON ON whether or not there are other characters which will justify their separation. And finally, a brief consideration of the facts above detailed regarding Platyzoma will suffice to show that the Gleicheniaceous affinity has not been proved, and that, even supposing Platyzoma be admitted to this group, its nearer relationships are in much doubt. In this paper are set forth some additional facts gleaned from the specimen supplied to us from Sydney Herbarium. It is still impossible to speak with assur- ance of the affinity of Platyzoma, but the facts here detailed may serve as a basis for comparisons at a later date, when the development of the sporangium will be described. The axis is a slender, horizontal rhizome measuring some 2% inches in length (fig. 6). It is closely invested by a dense covering of beautiful golden-brown un- branched hairs (figs. 2 and 8), which spring from all sides and cluster around the leaf-bases. The roots are inserted on the lower surface and occasionally on the sides of the rhizome. They are strong and fibrous and are frequently branched. The root hairs are remarkably long and persistent, and their walls are thick and brown. The heterophylly is well shown. The large pinnate leaves are about 11 inches long, while the small filiform leaves vary from about 23 inches to 4 inch in length. The petiole of the pinnate leaves measures about 1 inch. ‘The leaves are closely crowded upon the condensed and apparently slow-growing rhizome, and are definitely zoned, so that, if the leaves of a single year’s growth be examined, they are found to divide roughly into two groups. Of these the first are composed almost exclusively of long pinnate leaves, the second of delicate filiform types. At first sight it might be con- cluded that the pinnate leaves definitely precede the filiform leaves in order of development. Further, the majority of the pinnate leaves spring directly from the upper surface of the rhizome, and only a few encroach slightly upon its sides. No pinnate leaves arise from the lower surface of the stem. The filiform leaves are in- serted upon the sides and lower surface of the rhizome, and, curving forwards and upwards, pass out in a fairly definite group in front of the pinnate leaves. There might be a slight tendency towards a succession from pinnate to filiform leaves in a season’s growth, but this is not the chief cause of the zoning. It is due primarily to a definite tendency in this rhizomatous plant to initiate the larger leaves on or towards the upper surface, and the smaller leaves upon the sides and lower surface. The length of the pinnate leaves varies from about 10 inches to 12 inches. They are either sterile or fertile, but stature is of little value as a guide to their sterility or fertility. The filiform leaves are invariably sterile. The pinnate leaf is typically unbranched, but, as has already been noted, a dichotomous branching may occur. Only slight differences exist between the pinne of sterile and fertile leaves, and, further, a leaf may be partly fertile and partly sterile. The fertile pinne are, if any- thing, more robust than the sterile pinne, and their margins are strongly revolute (compare text-figs. 1 to 5 and 9). The vascular supply is in both cases a sympodium THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 637 which varies in detail from pinna to pinna. Irregularities in the numbers of the lateral branches of the sympodium, and the tendency to establish sympodial dichotomies, are the features worthy of note (text-figs. 1 to 5). The filiform leaves varied much in both length and form. The longest measured some 2% inches, while the smallest were minute and slender structures of about 4 inch in length. The larger filiform leaves were inserted chiefly upon the sides of the rhizome; the smallest sprang, almost invariably, from the lower surface. In fig. 7 a filiform leaf, which measured about 14 inch, is represented as seen from its lower surface. Three distinct regions—a basal, middle, and terminal region—are recognisable. The basal region measures about 4 inch. Its superficial tissues are dark brown and sclerotic, and on its upper surface it widens out on both sides into a distinct flange. Beneath each of these flanges is a shallow lateral yroove. The flanges die out as the middle region of the leaf is approached, but the grooves persist, and, having passed to the lower surface, they deepen and run parallel until the terminal region is reached. The middle region constitutes the greater part of the leaf, and variations in its length are mainly responsible for variations in the length of the larger filiform leaves. It is a more slender and delicate region than the leaf-base. Its superficial tissues are not sclerotic, but its epidermis is strongly cuticularised. As has been indicated, it is deeply grooved on both sides of its lower surface. On its upper surface are found a pair of grooves, similar to those already noted on the lower surface. ‘These upper-surface grooves likewise continue to the terminal portion of the leaf. The terminal region is short and bulbous. In it, both the upper and lower pairs of grooves become rapidly shallow, and finally die out just before the leaf-tip is reached. A leaf of the smaller type taken from the lower surface of the rhizome, and which had remained almost hidden among the hairs, is represented in fig. 3. It measured about 4 inch, and its tip was distinctly circinate. But in this case no division into basal, middle, and terminal regions could be made, and no distinct grooves were recognised. ‘The leaf became gradually slender, and the superficial tissues passed gently from a sclerotic to a strongly cuticularised condition, as one passed forward. The great majority of the filiform leaves were of the first type, and only a very few resembled the second type described. Four remarkable leaves formed a reduction series, illustrating a transition from the typically pinnate to a filiform condition. The first stage in this reduction is shown in fig. 5, which illustrates a leaf found among the broken petioles of dis- carded leaves. It measured about 19 inch in length, and possessed about 20 pinne of very variable size and irregular distribution. The lower third was entirely devoid of pinne, and might therefore be considered petiole. The remainder of the leaf was mainly pinnate, and was therefore undoubtedly rachis. Here, then, as in the typical pinnate leaf, the rachis bulked more largely than the petiole. The superficial tissues of both petiole and rachis were sclerotic, as in the typical pinnate leaf, and neither 638 MR JOHN M‘LEAN THOMPSON ON basal flanges nor grooves were present. Despite the fact that the pinnz varied greatly in size, their venation was invariably a sympodium. This leaf was inserted upon the upper surface of the rhizome, and might be regarded as a pinnate leaf which had suffered extensive reduction of its rachis. The second step in leaf reduction is shown in fig. 9. It represents a leaf which measured roughly 1 inch in length. The numerous pinnz were crowded upon a condensed rachis which measured little more than 4 inch in length. The venation of the majority of the pinnz was a simple midrib, but no vascular supply was dis- covered in several of the smaller terminal pinne. At the base of the rachis an abortive branch arose as a short and slender projection. The nature of this branch- ing will be discussed later. In this leaf, then, is seen a further reduction of the rachis, and a simplification, and even total loss, of the venation. A third step in this reduction process was shown by a leaf which is depicted in fig. 2. Of the 2 inch which it measured, the greater part consisted of petiole of the smaller type of filiform leaf already described. The pinnate portion of this leaf had been damaged, but the vascular supply to each of the four persisting pinne was a simple unbranched vein. As will be shown later, when the anatomy of these reduced leaves is being described, the rachis of this leaf had been very short, and the number of pinnze which had been destroyed was probably very small. There is here, in fact, a.leaf in which the rachis has been reduced to a small and condensed - column, bearing a few small pinnze with the simplest possible venation. The fourth step in leaf reduction remains to be considered. It is illustrated by a small leaf, + inch long, which was found among the hairs on the side of the rhizome (fig. 12). Its tip was hooked, suggestive of an original circinate vernation, and it bore almost terminally two small pinne, to which the vascular supplies were simple unbranched veins. There is here the minimum of pinnation, and one step further in reduction would lead to the smaller type of filiform leaf. It is a far cry from the elaborate pinnate leaf to this smaller type of filiform leaf, but the reduction series here briefly outlined can bridge the gap between them. It may be asked what is the position of the larger filiform leaves in this varied aggregation of leaf types. It is my purpose to show in succeeding paragraphs that the larger and smaller filiform leaves alike are reductions from, or replace- ments of, larger pinnate types. It is believed, indeed, that the leaves of Platyzoma have undergone drastic morphological changes, resultant from the present rhizo- matous habit and extreme xerophytic conditions. But the small filiform leaves, and the pinnate leaves showing reduction, were rare occurrences In my specimen, while, on the other hand, the larger filiform leaves were numerous. I am of the opinion that, when abundant material has been examined, it will be found that the larger filiform leaf is the typical reduction leaf of the present day, just as the unbranched pinnate leaf, and not the branched pinnate leaf, is typical of Platyzoma in its present condition. The forked pinnate leaf, the reduced simply- THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 639 pinnate leaf, and the smaller filiform leaf, may be regarded as harkings back to steps taken earlier in an evolution of reduction, from which the unbranched pinnate leaf and the larger filiform leaf have emerged as dominant types. The anatomy of the various leaf types recorded can be considered when we have before us the main facts concerning the anatomy of the rhizome. It will be conyenient to delay the consideration of the sporangia until the end, and accordingly the examination of the chief anatomical features of Platyzoma will now be made. The axis of Platyzoma possesses a broad cortex, developed peripherally as a strong zone of sclerenchyma, and more centrally as a mass of storage parenchyma (fig. 1). The stele is centrally placed, and from it depart a close succession of leaf-traces and roots, which pass out rapidly through the cortex and emerge in close succession upon the surface. The leaf-traces, on their way out through the cortex, are oval or almost circular in transverse section, and vary in size from large traces, originating mainly from the upper surface of the stele, to smaller traces, which arise from the sides and lower surface of the stele. The majority of the roots come off from the lower surface of the stele, but occasionally they have a lateral origin. There is a bulky central sclerenchymatous pith. Its cell walls are not, however, strongly thickened, and centrally they are of a light brown colour, and enclose mucilaginous protoplasts. The pith is surrounded by an un- interrupted endodermis, which is immediately followed by two or three layers of large-celled parenchyma.. There is no inner phloém, and the large-celled parenchyma is followed immediately to the outside by a remarkable ring of xylem. The xylem-ring varies in thickness as one passes from the upper to the lower surface of the stele. On its upper surface the ring is invariably thin, while on its lower surface it is relatively thick. The thickness of the xylem-cylinder increases gradually during the passage from the upper to the lower sides of the stele. At some points along the course of the rhizome this difference in the thickness of the upper and lower portions of the stele is only slightly marked (figs. 4 and 10), while at other points it is a very conspicuous feature (figs. 1, 13, and 63). The bulk of the sclerenchymatous pith does not, however, vary to any great extent from point to point, for increase in the thickness of the lower side of the xylem is not attained so much by encroachment centrally, as by a centrifugal thickening. The xylem-cylinder is composed of scalariform tracheides together with chains and groups of parenchymatous cells, which follow irregular courses between the tracheides, and which fuse and branch repeatedly. No protoxylem elements can be recognised, but the xylem-cylinder can be divided roughly into two concentric cylinders according to the characters of the tracheides and the distribution of the parenchyma. The inner cylinder is composed of wide storage tracheides associated with numerous chains and masses of parenchyma. It forms, essentially, TRANS. ROY. SOC, EDIN., VOL. LI, PART III (NO. 20). 94 640 MR JOHN M‘LEAN THOMPSON ON the parenchymatous part of the xylem. The outer cylinder is composed of small and narrow tracheides which are more closely packed than are the tracheides of the inner cylinder. Parenchyma does not bulk largely in the organisation of this outer cylinder, but the chains and groups are short and narrow, and the elements composing them are few in number. This outer cylinder is alone involved in the process of supplying the traces of the leaves gnd roots. It has been noted that the xylem is thin on the upper surface of the stele, while on the lower surface it is usually comparatively thick. It may be stated at once that the difference in xylem thickness, between the lower and upper sides of the stele, is not primarily due to a progressive thinning of the xylem on the upper surface as a whole. It is mainly due to an increase in the bulk of the parenchymatous storage xylem towards the lower surface of the stele (figs. 1, 4, 10, 13, 63). The outer xylem-cylinder does not vary markedly as one passes from the upper to the lower surface of the stele. It is true that it is of irregular thickness from point to point, but it is maintained consistently on all sides of the stele, whereas the inner xylem- cylinder may be locally massive towards the under surface, and at the same time reduced almost to the vanishing point on the upper side of the stele. It is therefore evident that it is the parenchymatous inner portion of the xylem- cylinder which is most liable to fluctuations, either in the direction of increase or reduction (figs. 4 and 63). Further, the great increase in the parenchymatous xylem just noted is subject to local accentuation ; for, while this type of xylem is certainly conspicuous at all points along the lower surface of the stele, it becomes very massive in those regions where the numerous and closely crowded traces are being liberated (figs. 13 and 63). On the other hand, it is usually less dominant in those portions of the stele where the leaf-traces are less numerous and crowded. As to the absence of definite protoxylem groups a word may be said. It is generally recognised that the presence or absence of definite protoxylems depends upon the degree and rate of extension which take place in the xylem. There can be no doubt that the axis of Platyzoma is either slow-growing or is subject to little or no extension as maturity is approached. To this the close succession of the leaves and the condensed nature of the rhizome bear ample witness, and accordingly the absence of protoxylems need be cause for no surprise. The absence of protoxylems may then be regarded as indicative of a slow rate of growth or of but slight exten- sion of the stele. It is an index of the peculiar physiological conditions of Platyzoma, but as a systematic character it is unreliable. The phloém follows almost immediately upon the outer surface of the xylem. It is separated from the xylem only locally by a single layer of small parenchymatous cells; it is from three to five layers broad, and is composed entirely of typical fern- sieve tubes. Itis only interrupted at points where leaf- and root-traces are departing. The phloém is followed externally by a large-celled pericycle of three to four paren- THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 641 chymatous layers, and beyond this is an outer endodermis separating the stele from the cortex. This curious stele has been interpreted by various observers as a reduced soleno- stele which has lost its inner phloém. But while the leaves are numerous, and many of them are relatively large and are provided with strong leaf-traces from the stele, no leaf-gaps involving a breaking of the continuity of the stele occur at any point. As will be shown directly, the outer endodermis is in no case even temporarily incomplete when a trace is being liberated. A break in the endodermis does cer- tainly arise, but the cortex cannot invade the stele. Further, the xylem of the stele may be so depleted by the demands of a leaf-trace that at the point of departure of the trace the xylem-ring may be broken or traversed by a sheet or mass of parenchyma. But at no point along the entire length of stele examined was the inner endodermis found to be defective. It was neither broken nor did it show any tendency to break, opposite any leaf-trace or at any other point. PorravLr has attempted to explain this stele as a reduction from a solenostele from which leaf-gaps have been obliterated. But it may be remarked that not only are leaf-gaps non-existent in the stele of Platyzoma, but no anatomical evidence can be advanced in support of the suggestion that they have existed at any time. There is no connection whatever, either by leaf- gaps or perforations, between the cortex and the pith, and the pith is accordingly recognised as a completely isolated mass of sclerenchyma, bounded on all sides by an unbroken endodermis. But if no evidence can be advanced in support of PorravLt’s suggestion of a solenostelic origin for the stele of Platyzoma, a serious objection has presented itself to Jmrrrey’s theory of the origin of the pith in Pteridophytes. JEFFREY has maintained that the pith is, 7 all cases, of cortical origin, and that it has arisen through an invasion of the stele by the cortical parenchyma at the points of departure of leaf-traces, or at other points. It seems more than difficult to explain the origin of the bulky pith of Platyzoma by such a theory, since neither leaf-gaps nor perforations, nor any tendency towards the formation of such gaps and perfora- tions, have been found in the stele. And further, even supposing the inner endo- dermis could be shown to be locally imperfect or interrupted—so that parenchymatous connections could exist between the cortex and pith—the stele of Platyzoma would still be open to interpretation as of directly protostelic origin. For if the Gleicheni- aceous affinity of Platyzoma could be definitely established, its stele might be interpreted as a transformed protostele, in which sclerenchymatous replacement of the xylem core has taken place, and an inner endodermis has arisen, de novo, around the pith. With one exception—namely, Gileichenia pectinata—the genus Gleichenia is protostelic. The core of the typical Gleicheniaceous protostele is a parenchymatous mass of large storage tracheides; and if the stele of Platyzoma has been originally protostelic, the inner parenchymatous xylem-cylinder—which has been described as the most variable part of the stele—might be considered the last remnant of a solid xylem core. Lane has shown recently that endodermis can 642 MR JOHN M‘LEAN THOMPSON ON arise de novo,* and it will be shown immediately that, in the liberating of the leaf-traces of Platyzoma, endodermis is formed where and when required. It is then not impossible that the inner endodermis of Platyzoma is of relatively recent origin, and has been at no point in descent in connection with the outer endodermis, either through leaf-gaps or perforations. If this be so, the pith of Platyzoma cannot have been of cortical origin, and the stele cannot be interpreted as a reduced solenostele. We come now to the origin and nature of the leaf-traces. There are both large and small leaf-traces (figs. 1, 4, and 13). The largest traces arise typically from the upper surface of the stele, but occasionally their points of origin are at a slightly lower level. The smallest traces originate from the lower surface of the stele, and between the largest and smallest types traces of intermediate size exist. These latter are inserted typically upon the sides of the stele. There is, then, a gradation in the sizes of the leaf-traces as one passes from the upper to the lower surface of the stele. Slight differences of detail occur from trace to trace, but one general method is adopted throughout in the formation of the leaf-traces. It will suffice to describe, in detail, the departure of a large leaf-trace, and to indicate briefly such points as are of interest in the smaller types. The first indication of the formation of a leaf-trace is the marking out and pro- . jection of a peripheral, tangential band of tracheides from the outer xylem-cylinder (fic. 41). Beneath this band of tracheides a definite mound of parenchyma has formed, while above it the phloém, pericycle, and endodermis have been raised in a gentle curve. As the band of xylem becomes more distinctly curved and projecting, the band of phloém, displayed upon its convex side, becomes thinner. Very soon . the xylem band begins to free itself from the xylem of the stele, and to pass out as the crescent-shaped. xylem of a leaf-trace (fig. 42). It may sever its connections with the stelar xylem simultaneously at both sides, or it may become free on one side more rapidly than on the other. But in any case, while the phloém which clothes the outer surface of the leaf-trace xylem has become thinner, no diminution of phloém has taken place at the points where the leaf-trace xylem has freed itself. In fact, no sooner has an edge of the leaf-trace xylem become free, than phloém begins to appear in the gap which separates it from the stelar xylem, and, as the gap widens, a sheet of phloém appears in the parenchymatous mass lying beneath the leaf-trace xylem (compare figs. 42 and 43). This sheet of phloém is an inward con- tinuation of the stelar phloém beneath the developing leaf-trace. It is destined to meet a similar phloém sheet projecting inwards from the gap which separates the leaf-trace xylem from the stele at the other edge; and when these phloém sheets have met and fused, a continuous layer of phloém will be spread once more over the surface of the stelar xylem. Two stages in this process are represented in figs. 43 and 44. Meantime the leaf-trace xylem and its accompanying crescent of phloém > have continued their outward course, and very soon the phloém of the leaf-trace * Ann. Bot., 1913, pp. 236-239. THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 643 becomes free from the phloém of the stele (compare figs. 44 and 45). The endo- dermis and pericycle have moved outwards in order to accommodate the xylem and phloém of the leaf-trace, and when the limit of this accommodation has been reached the leaf-trace is a crescent of solid xylem, with phloém on its convex surface and around its margin (fig. 45). When any gap which had arisen in the peripheral cylinder of stelar xylem has been repaired, and when the phloém covers the stele once more as a continuous sheet, endodermal cells appear on both sides of the leaf-trace (fig. 45). These new endodermal cells spread obliquely inwards from points upon the outer endodermis, and, having passed rapidly inwards, unite to form a continuous endodermis which completely shuts off the leaf-trace from the stele (fig. 46). But further, no sooner has this endodermal formation been completed, than the curved portion of the endodermis, which has accompanied the leaf-trace on its outward course, becomes free as a crescent-shaped band, following the general outline of the leaf-trace. The endodermis of the leaf-trace is then, at this point, of an imperfect character, being maintained upon the outer side of the trace, but absent from the inner side. This condition is not, however, long maintained, for an inner band of endodermis soon appears, repairing the gap (fig. 46) and determining the contour of the leaf-trace as it passes out through the cortex. It is here a solid though narrow band of xylem, with a thin layer of phloém upon its convex surface, and a well-developed pericycle (fig. 47). It is well to note that in this process of leaf-trace formation the inner endodermis is in no way involved, and, further, that there exists no point where the . stele is not entirely enclosed in outer endodermis. Not only has it been established that the cortex and pith are connected at no point, but it will now become evident that no connection exists between the cortex and the pericycle at the point of final liberation of the leaf-trace into the cortex. The possibility of such a connection has been removed by the local doubling of the outer endodermis as described above. It frequently occurs, however, that when the liberation of a leaf-trace imposes a large drain on the tracheides of the outer stelar cylinder, a parenchymatous bridge appears passing athwart the xylem-ring (fig. 63). But many chains of parenchyma traverse | the xylem from side to side—though their courses may be irregular and narrow,— and in no case do these bridges disturb the inner or outer endodermis. It happens occasionally that the xylem of the leaf-trace is, at the point of departure of the trace, a compact mass of tracheides which appears almost circular in section; but as the trace is followed outwards, the xylem becomes extended laterally, and on sectioned shows the typical crescent arrangement described above. The smallest leaf-traces may now be considered. Two stages in the liberation of a small trace from the lower surface of the cylinder are represented in figs. 48 and 49. It will be evident that they reproduce, in miniature, the method of trace- departure described above for the largest leaves. As they pass outwards into the cortex the steps described for the large leaf-traces repeat themselves, and. when the 644 MR JOHN M‘LEAN THOMPSON ON traces are free within the cortex, they appear, in transverse section, as represented in fig. 51. They merely reproduce, in miniature, the characters of the largest leaf-traces. A leaf-trace of the transitional type, and springing from the side of the stele, is represented in transverse section in fig. 50. It shows the re-formation of the layer of stelar phloém below the departing leaf-trace, and the liberation of the leaf-trace phloém from the stelar phloém. I am not aware of any leaf-trace departure which is exactly comparable with the one described in these paragraphs, and for the present we may consider it an unique type. The origin of the roots presents no points of special interest. The roots are restricted to the lower surface and lower portions of the sides of the stele. They originate, as do the leaf-traces, by a protrusion of the outer xylem-cylinder (figs. 1 and 52). When they have entered the cortex they appear in transverse section as a solid mass of xylem, with small peripheral and larger central tracheides. Around this xylem core is a narrow zone of phloém, a well-developed pericycle, and an endodermis (fig. 53). A typical root-structure is not established until the cortex has been traversed. When free from the cortex, the root has typically a diarch stele (fig. 54). The pericycle and phloém are well developed, and the endodermis is of large and delicate cells. A two-layered zone of sclerenchyma gives the fibrous character to the mature roots, and the outer part of the cortex consists of two or three layers of large parenchymatous cells, with which are associated numerous, long, and often persistent root-hairs. There is nothing remarkable in the roots except the persistency of the long root-hairs and the cortex. This persistency was so marked a feature that perfect sections could be obtained from old roots of the specimen which I had at my disposal. The largest leaf-traces pass into the large pinnate leaves and the smallest traces typically supply the filiform leaves. The traces of intermediate size divide roughly into two groups. The first group includes traces of moderate size which spring from the upper portions of the sides of the stele, and, as a rule, supply pinnate leaves. The second group is composed of traces springing from the lower portions of the sides. Almost invariably they supply filiform leaves. The leaf-zoning has there- fore found its explanation in the dispositions of the large and small leaf-traces upon the stele. Passing now to the anatomy of the leaves, the outstanding features of the typical simply-pinnate leaf may be considered first. The petiole is circular or oval in section (fig. 14). The epidermis and the peripheral zone of cortex are markedly sclerotic, but the inner cortex is composed mainly of large-celled parenchyma with thin walls, and mucilaginous protoplasts. The protoplasts of the cortical sclerenchyma are also highly mucilaginous, and the general condition of the cortex is indicative of extremely xerophytic conditions. The trace is roughly oval in section, and the endodermis is slightly indented on the upper surface. There is practically no THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 645 pericycle, and the phloém is a narrow band displayed over the lower convex surface of the xylem and around its hooked margins. The xylem is an extended narrow strap, which is composed of two types of elements. The lower, central, and more massive portion is composed mainly of large tracheides. The incurved margins of the strand are of numerous, very small, and relatively thin-walled tracheides. This marked difference in size between the elements of the central and marginal portions of the xylem is a conspicuous feature throughout the greater portion of the leaf. The incurved xylem-strap encloses a parenchymatous ground tissue, and if protoxylems are at all developed they are involved in the small-celled and arched margins of the strap. As the rachis is approached the petiole becomes somewhat flattened dorsiventrally, and this is accompanied by a lateral extension of the trace. Within the rachis itself little or no change takes place in the epidermal and cortical tissues. but the leaf-trace has become extended laterally, and the margins of the xylem-strap have become thinner and more attenuated (fig. 15). The pinna- traces are small, and of extra-marginal origin. They are given off from the convexity of the incurved leaf-trace, their xylems being contributed by the sharply eurved parts of the small-celled and incurved margins of the xylem-strap. The pinna-traces pass quickly through the cortex to the almost sessile pinne. The venation of the pinne has already been indicated. Towards its distal end the rachis becomes slender and laterally compressed. The xerophytic characters of the superficial and cortical tissues already noted are still dominant, and the leaf- trace, though now greatly reduced in bulk, retains the general features noted in the lower regions of the leaf (fig. 16). The pinna-traces are still of extra- marginal origin, but now the drain made by the pinne upon the sharply curved portion of the xylem-strap leads frequently to a depletion of the small tracheides ; so that, when seen in section, the margins of the xylem-strap appear fragmented, and separated from the main lower and central portion of the strap. The margin of the xylem-strap is, in fact, in process of reduction. Close to the apex of the leaf the rachis has become very slender. Its cortex is greatly reduced, and the leaf-trace, though now very small, is still clearly defined (fig. 17). The infolded margins of the xylem-strap have been entirely reduced at a lower level, and the strap has become a thin, irregular, and disintegrating assemblage of tracheides, from which the last few small pinne cannot fail to be supplied in a marginal manner. From what has been recorded, it will be evident that in general form of leaf-trace, and in the dominantly extra-marginal origin of the pinna-traces, the pinnate leaf of Platyzoma falls easily into line with most Gleichenioid types. From the section of a small pinna shown in fig. 17, it will be clear that the epidermis is well developed, and that the mesophyll of the pinna is partly lacunar. The upper epidermis is strongly cuticularised, and is devoid of stomata. The stomata are restricted to the lower surface, and lie hidden within the almost 646 MR JOHN M‘LEAN THOMPSON ON closed cavity formed by the infolding of the membranous leaf-margins. Hairs arise from the leaf-margins and along the course of the veins, and assist still further in the protection of the stomata. The anatomy of the pinnate leaf has revealed characters expressive of extremely xerophytic life conditions. It will be convenient to record, at this point, the anatomical features of the forked pinnate leaf which I had at my disposal (text-fig. 8). It forked near to its base, no pinnee appeared below the point of forking, and the leaf-branches were of equal length. The pinnate portions of this leaf were of no special interest, for they agreed in general plan with the rachial part of the simply-pinnate leaf. The branching petiole was, however, of considerable interest, for its trace was found to undergo a perfect dichotomy. The leaf-trace—in transverse section,—as seen in the leaf base, is shown in fig. 28. There is not so marked a difference in size of element between the central and marginal portions of the xylem-strap as existed in the simply-pinnate leaf, but the general plan of construction of leaf-trace is the same in both cases. The leaf-trace of the dichotomising leaf was larger than that of the simply-pinnate leaf, and its xylem wasmore extended. As the petiole of the forked leaf was ascended, a small group of tracheides appeared in the parenchymatous bay enclosed in the curved xylem-strap (fig. 29). At a still higher level, the trace had become extended laterally, and the island of tracheides in the bay of parenchyma had grown considerably (fig. 30). Very soon a junction was established between the island and the middle portion of the much extended xylem-strap (fig. 31), and two promontories had developed—one to right and the other to left of the free end of the original island,—so that two parenchymatous bays were now formed on the upper or adaxial side of the leaf-trace. A split then developed in the middle region of the xylem-strap (fig. 32). It traversed the strap from its upper to its. lower surface, and divided it into two similar portions. Further on, when the endodermis, pericycle, and phloém had become nipped across along the same direction-line, there were formed two similar leaf-traces (fig. 33) which supplied the two limbs of the dichotomised leaf. It would be of interest to know what takes place in the branch- ing of the leaf in the two specimens already referred to as belonging to Kew Herbarium. It seems probable that they too are dichotomous branchings, but their further interest lies in the facts that they are pinnate below the points of branching, and that they branch at points considerably removed from their bases. It is doubt- ful whether these bifurcations of the leaf of Platyzoma will bear any phyletic inter- pretation. The bifurcations are seen to occur at various levels, and seem to have no particular relation to the pimne. They are probably of the same nature as those bifurcations found in many other Fern leaves, e.g. Polypodium vulgare, Scolopendrium, Nephrodium molle, etc., and are akin also to those distal forkings which give rise in varying degree to the “forma furcata” varieties. The remarkable point is that the forking should occur in such a Fern as Platyzoma, for it is more usual in advanced Leptosporangiates. But nevertheless it is seen in Osmunda. rye THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 647 The typical filiform leaf is now to be described. At its point of insertion upon the axis it is compressed laterally and its epidermis is thickened (fig. 18). The cortex is composed of mucilaginous cells which form a bulky water-storage tissue _towards the lower side. The leaf-trace is small and compressed, and has altered very little from the condition which it showed on its way through the cortex of the rhizome. A slight indication is already present of the formation of the flanges which are so conspicuous in the basal region. Fig. 19 represents a section at a slightly higher level. The trace has become slightly more bulky, and the xylem-strap has become thinner and broader. The flanges are clearly established, and the lateral grooves are apparent. In fig. 20 a typical section of the basal region is depicted. The flanges are well developed, and the cortex is lacunar. The trace has grown con- siderably, and, while the xylem-strap has extended, no very marked differences can be recorded between the elements of its middle and marginal portions. The passage from the basal to the middle region is characterised by the possession of two pairs of grooves—one pair on the upper surface, and the other on the lower surface. The grooves on the lower surface of this middle region are the continuations of those grooves which exist in the basal region, and which have passed to the lower surface at the junction of the basal and middle leaf-regions. The grooves on the upper surface are new grooves which have commenced at the base of the middle region. A typical section of the lower half of the middle leaf-region is shown in fig. 21. The epidermis is strongly cuticularised, but not thickened as is the epidermis of the basal region. Stomata are occasionally found in the grooves on the lower surface. The xylem of the trace is of particular interest. It has developed into a broad strap with long, thin, curved margins. The margins are composed of small tracheides, while the middle is a zone of large, thick-walled tracheides. It will be apparent that in this region the trace thus resembles the trace of the simply-pinnate leaf in the rachial portion particularly. But this condition of the xylem-strap is not always found. In a number of filiform leaves the condition represented in fig. 22 is found. It is clear that the margin of the xylem is not characterised by small tracheides, and, in fact, short storage tracheides are irregularly distributed in both the middle and marginal portions. Passing forward in this middle region the grooves on the upper surface deepen, -and the xylem of the trace undergoes remarkable modifications. As is shown in fig. 23, the middle zone of the xylem remains typically large-celled, but the margins have become extended until they have met to form a practically closed tube of xylem around a core of parenchyma. Here again there are variations, for in several leaves the condition found at this region is as shown in fig. 24. The xylem is more massive than in the former case, the parenchymatous core is almost obliterated, and there is not so marked a difference in size between the tracheides of the middle zone of xylem and those of the revolute and now fused margins. Sections through the tips of the terminal lobes of two filiform leaves are repre- TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 20). 95 648 MR JOHN M‘LEAN THOMPSON ON sented in figs. 25 and 26. In the first of these the lower surface grooves had died away before the upper grooves, while in the second the upper surface grooves had been lost first. In both cases the infolded margins of the xylem-strap are being reduced, and storage tracheides are appearing in the middle xylem-zone. 4 The anatomy of these remarkable filiform leaves has then shown that the trace is of an elaborate type, comparable in form with the trace of the pinnate leaves ; and that, while the trace is admirably constructed for the supplying of extra-marginal pinna-traces, similar to those which feed the ordinary pinne, no pinna-traces are initiated in the filiform leaves. It may be that the ridges of tissue which run along the margins of the filiform leaves mark the middle zone as of rachial nature. But until definite evidence is advanced, showing that those ridges represent coalesced and aborted pinnee, no definite opinion can be formed regarding the interpretations of the three regions of the filiform leaves. That these leaves are reduction forms seems undeniable ; but whether they are useless abortions or serviceable structures cannot be determined from an herbarium specimen. The anatomy of the transitional leaf-forms remains to be considered. Regarding the small filiform leaf with the circinate tip, represented in fig. 3, it will suffice to note that for the greater part of its length its leaf-trace remained of the type shown by the basal region of the longer filiform leaves, and that, as the leaf-tip was approached, the margins of the xylem-strap were reduced and the trace died out. The anatomy of the first of the reduction series of four leaves (fig. 5) showed no essential points of difference from an ordinary simply-pinnate leaf. It might be con- sidered a pinnate leaf in miniature. The anatomy of the second leaf of the reduction series (fig. 9) showed two points of interest. In its basal portion it resembled the foregoing leaf, but at a point near to the rachis the petiole became distended. At this point the stele showed a peculiar condition. The xylem-strap was, as usual, divisible into the marginal and middle regions, but a parenchymatous replacement of the tracheides was taking place in the centre of the strap (fig. 27). This led, finally, to what appeared to be an equal division of the stele into two parts. Of these, the one part passed into the pinnate branch, but the other rapidly degenerated, and only an irregular group of tortuous tracheides and a few softer’ elements entered the aborted branch. I am of the opinion that this leaf might have developed a true dichotomy, but that it became malformed during development. No pinna-trace was found supplying the minute pinna-like protuberances which stood distally on the pinnate branch. The terminal portion of the small leaf which provided the third step in the reduction series is worthy of remark (fig. 11). At least five pinnee were supplied with traces. The supplies for the basal pair of pinnee are represented in fig. 34 on their way through the cortex. In fig. 35 the trace to one of the second pair has already been liberated, while the second trace is in process of separation. It is noteworthy that the second pair of traces are here much more bulky than are the first or lower THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 649 pair. This seems to be due mainly to a thickening of the extraordinarily extended leaf-trace margins, at the points where the infolding of the margins takes place. This thickening of the margin is more pronounced in fig. 36, which shows the extra-marginal departure of the fifth pinna-trace. When this last pinna-trace had departed, the xylem became suddenly a solid mass of tracheides (fig. 37). It is probable that this marked the approach of the leaf-tip, and that few if any more pinne were developed on this leaf. And lastly, in the pinnate portion of the small leaf which is represented in fig. 12—and which showed the simplest possible pinnate condition—the supply to the pinne is of extra-marginal origin. Three stages in the preparation of the pinna-traces of this leaf are shown in figs. 38, 39, and 40. The condition typical of the trace in the rachis of the pinnate leaf, and in the middle region of the larger filiform leaf, was found to dominate the stalk. It is depicted in fig. 38. Fig. 39 shows the lateral extension of the trace as the pinna-traces begin to form, and in fig. 40 one of the two pinna-traces is almost liberated. The facts regarding those various leaf-types have been recorded in some detail ; for not only was it necessary to recognise the transitions from pinnate to filiform leaves by external form, but a knowledge of the anatomical features could provide a valuable check on any conclusions which might be drawn. On the ground of external form, it seemed almost certain that the large filiform leaves might justifiably be regarded as reductions from pinnate leaves, and the evidence which has now been advanced—both regarding these filiform leaves themselves and also the “transitional” types—seems to lend strong support to this opinion. It is not to be supposed that the “transitional” types are considered illustrations of definite transitional forms, lying between the departure from the pinnate leaves and the arrival of the filiform leaves, but im themselves they are proofs of leaf-reductions in Platyzoma, and the filiform leaves may be resultants of a like reduction process. The sporangia remain to be considered. At present I am unable to give an account of their development, but there are certain outstanding features which ‘may be noted here. They are developed either terminally upon the main lateral branches of the sympodium, or, if more than one sporangium is developed upon a branch of the sympodium, the distal portion of the branch is forked and the sporangia are again terminal (text-fig. 9). The bundle-endings which thus supply the isolated sporangia are worthy of note. They are composed of long tracheides with spiral and reticulate markings, and a sheath of long, narrow, and delicate cells which persist to the very tip of the bundle (text-fig. 10). I am not aware of any other Leptosporangiate Fern in which the condition just described exists. It may be most nearly matched by some early Kusporangiates. The sporangia in any one pinna vary in size, but no definite arrangement of the sporangia according to size can be followed. There are large and small sporangia, 650 MR JOHN M‘LEAN THOMPSON ON and, between these, transitional forms exist (text-fig. 9, and figs. 55-62, 67-68). The large sporangia contain small numbers of large spores, the small sporangia larger numbers of small spores (figs. 64, 65, and 66). The small sporangia are invariably short-stalked, but in form they are very variable. In some cases the capsule is globular, in others it is ovoid or almost pear-shaped, while frequently it is very lopsided (figs. 55-59, 61, and 62). The annulus is variable in position and form. It is seldom, if ever, perfectly regular, and is frequently very irregular. In a few cases it was found to be almost horizontal, passing round the capsule as an irregular and broken ring, and having no organised stomium (figs. 61 and 62). But in the majority of the sporangia examined the annulus was almost vertical, though typically irregular or broken. These sporangia possessed a definite stomium (figs. 55-58). Figs. 55-58 show a single sporangium from above (fig. 57), below (fig. 58), back (fig. 55), and front (fig. 56). From these it is apparent that the annulus begins at the stalk, passes obliquely and with irregularities upwards, and, having passed over the top of the capsule, descends to the stalk once more as an irregularly developed stomium. The annulus is, then, interrupted by the stalk. A more regular—though also more oblique—annulus is represented in back view in fig. 59, while in fig. 60 the wall of a sporangium which has been opened out is represented. It is apparent that in this case the annulus began close to the stalk, and passed almost vertically over the summit of the capsule. But here an irregularity arose. This led toa displacement of the stomium, so that it did not descend directly to meet the stalk, but, having passed to one side of it, came into close contact with the end of the annulus which had met the stalk. The number and the size of the spores which were found in those smaller sporangia varied. The largest number of spores found in any small sporangium was 32, and such numbers as 30, 29, 28, and 26 were frequently counted. The spores are tetrahedral, and the lines of junction of the tetrads are well marked (figs. 64 and 65). The exospore is beautifully sculptured, there being numerous small areas enclosed by wavy borders, ‘and possessed of little mounds which give an added roughness to the spore coat. The smallest type of spore found in a small sporangium is shown in fig. 65. It was taken from a sporangium containing 32 similar spores. The largest type of spore found in a small sporangium is shown in fig. 64. It was taken from a sporangium containing 29 similar spores. One of the largest sporangia found is represented in figs. 67 and 68. It is apparent from the back view (fig. 68) that the annulus stops short at the stalk and passes irregularly over the top of the capsule. The stomium is shown in fig. 67. It is separated from the annulus, and passes downwards obliquely to the stalk. The largest number of spores found in a large sporangium was 16; 14 was an average count, and 12 was a rare number. A typical large spore is shown in fig. 66. It resembles the smaller spores in form and external markings, but differs from them remarkably in size. Two large sporangia contained both large and —_ i SS ae. THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 651 small spores. In the first of these there were 14 spores of the largest type and two of the intermediate size; in the second instance there were 14 large spores and one small spore. It is impossible to decide at this point whether there is here a true heterospory, or the variation in spore size is merely the result of peculiar local nutritive conditions. I have not yet examined the contents of these remarkable spores, and accordingly can give no indication from this standpoint of the significance of the variation in spore size and output per sporangium. It is hoped to throw more light on the nature of these spores when the development of the sporangium has been studied and the spores can be germinated. But while in general form of sporangium, and obliquity and irregularity of the annulus, Platyzoma somewhat resembles some species of Gleichenia, the differences of sporangial and spore size and the small spore output are not characteristic of any known Gleichenia. The chief facts obtained from the specimen of Platyzoma have now been considered. The habit of the plant, upon which the systematists chiefly relied, has sunk in importance before the remarkable characters of the stele, the leaf- traces, the leaves, and the sporangia. This plant has proved anomalous at almost all points, and in fact it would be difficult to indicate any known Fern with which Platyzoma can be reasonably compared. One is forced to the conclusion that it is not a Gleichenia. The development of the sporangia and the true nature of the spores may throw a fuller light upon its true affinity; and until information on these points is forthcoming, it would be foolish to separate Platyzoma from a relationship which may prove to be its nearest affinity. But although one cannot speak with any degree of confidence regarding the actual place of Platyzoma in the phyletic system, the study of its anatomy has laid bare facts regarding the stele, reduction processes in the leaf, and sporangial characters which may be of value in later discussions. The material available for the investigation was an herbarium specimen, and accordingly it was impossible to do more than collect the most important facts of mature anatomy. The present memoir is to be regarded as an instalment of the facts relating to this remarkable Fern, and the statement is not to be considered a final one. This being so, it has been thought well not to enter upon detailed comparisons, though the remarkable facts relating to Platyzoma provoke them. It has likewise been deemed well to abstain from any very definite statement as to the systematic position of Platyzoma until the sporangial conditions have been more fully examinéd. ‘These matters can be better dealt with when properly preserved materials are available. I am indebted to the Director of Sydney Botanic Garden for the specimen upon which these observations have been made; to Professor Bowrer for much valuable guidance and criticism ; to Mr BoopxeE for information regarding the Kew specimens of Platyzoma, and to the Carnegie Trust for their assistance in the production of the illustrations. 652 MR JOHN M‘LEAN THOMPSON ON SUMMARY. The chief facts obtained from the single herbarium specimen available of Platyzoma have now been described. At some points the account has been less complete than would have been the case if sufficient material, properly preserved, had been at hand. Development has been necessarily omitted. But the fuller facts relating to mature structure which have been acquired may help towards establish- ing the position which this plant should hold in relation to other ferns. The habit of the plant is chiefly notable in relation to the heterophyllous character of its xerophytic foliage. The horizontal rhizome bears with some degree of regularity zones of smaller—probably reduced—foliage leaves with pinnz vestigial or absent, and these alternate with zones of larger pinnate leaves which are mostly fertile. The rhizomatous condition is probably derivative from one.in which the axis had an upright habit, and the heterophylly has probably followed upon the change of habit. Occasional cases of bifurcation of the leaves have been seen. It is probable that these are of the same nature as those branchings seen in the “forma furcata” of many Ferns. The crowded grouping of the leaves is a feature in strong antithesis to the isolation of the leaves prevalent in the Gleichenias, with which Platyzoma has habitually been grouped. The presence of hairs and absence of flattened scales accords with what is seen in Gil. linearis and pectinata, and is again a point marking Platyzoma off from many of the Gileichenias; while it is in itself a character of relatively primitive Ferns. But external habit, upon which the systematists have hitherto placed their chief reliance, has sunk in importance before the remarkable characters of the stele, the leaf-trace, and the sporangia. The stele is little removed from the protostelic state. It consists of a continuous ring, which has never been seen to be broken by leaf-gaps or perforations. Definite protoxylem has not been recognised. But the xylem is differentiated into two zones, an outer consisting of smaller tracheides with occasional parenchymatous groups and chains, the inner of large storage-tracheides intermixed with larger groups and chains of parenchymatous cells. The inner zone appears to be primarily a storage zone. Centrally lies a bulky pith which is sclerotic and mucilaginous. It is surrounded by a continuous inner endodermis which is throughout distinct and separate from the outer endodermis. Between the inner endodermis and the xylem- ring lies a narrow zone of parenchyma, but no phloém. LHxternally to the xylem-ring lies a continuous narrow zone of phloém, a pericycle, and external endodermis. The leaf-traces depart without any leaf-gap. They vary in size according to the dimensions of the leaves which they serve, but even those of the largest fertile leaves, though they may interrupt the xylem to its inner limit, do not disturb the inner parenchymatous zone or the inner endodermis: thus there is no leaf-gap. The leaf- trace draws off from the outer zone of xylem as a crescentic mass of tracheides, accompanied by an outer are of phloém. There is no definitely constituted proto- THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 653 xylem. As this simple leaf-trace separates from the stelar cylinder, the stelar phloém and xylem right and left of it close behind the outgoing leaf-trace. Mean- while the outer endodermis is unbroken—a state which is maintained until the stelar ring has been reconstituted. Only then does the are of outer endodermis accompany- ing the departing leaf-trace break its connection with the remainder of the endodermis. This it does at two points right and left. An outer separate arc of endodermis is thus liberated, and it accompanies the outgoing leaf-trace. Meantime the continuity of the outer endodermis of the stele has been maintained by the appearance of tracts of endodermis, right and left, which finally meet beneath the leaf-trace before the are of leaf-trace endodermis has become detached. Following the leaf-trace outwards, its peripheral arc of endodermis is soon extended on the adaxial side, and completes the enclosure, in which state the trace enters the leaf. The zonation recognised externally is seen anatomically to be apparent rather than real. For it is found that the leaf-traces directed downwards are smaller and supply the smaller vegetative leaves; those towards the upper side of the rhizome are larger and supply the larger and fertile leaves. The apparent zonation depends upon the separate grouping of the larger and smaller leaves in their forward and upward course. The smaller and larger leaf-traces originate from the stele in the same way, and differ only in the extent to which they affect the stele on their departure. This accords with the facts of abortive pinnation of the vegetative leaves—an indication that they are of reduced type. The leaf-trace remains undivided throughout, and the pinnation is always of extra-marginal type, except towards the tip of the leaf, where a reduced condition exists. The sporangia are distributed solitarily upon the lower surface of the pinnee, and terminally on the veins. They do not constitute definite sori. A well-developed bundle ending runs to the base of each sporangium. There is no indusium, but hairs are distributed along the veins and the revolute margins of the pinne. The sporangial stalk is short and consists of three rows of cells; the head is globular and lopsided. The annulus is typically oblique, but is irregular in position and in constitution. The number of cells of the annulus is variable, and their succession is apt to be irregular and interrupted. The annulus is always interrupted at the stalk. The stomium is variable in form and position, and where the annulus is almost horizontal no organised stomium is found. This indicates a transitional state. Roughly, the sporangia are differentiated into two types—the larger sporangia with few large spores (16, 14, and 12), the smaller with more numerous small spores (32-26). In actual dimensions a large spore may be many times the size of a small spore. The facts above detailed would still allow of a Gleicheniaceous relation for Platyzoma. But their divergence from the Gleichenia characters is so pronounced that it seems quite impossible to retain a view of close affinity. The sporangial characters are specially divergent, and indicate a probable derivative position for Platyzoma from some Gleicheniaceous source. Meanwhile the retention of hairs and 654 MR JOHN M‘LEAN THOMPSON ON the anatomical state show a greater conservatism. In fact, Platyzoma is again a type somewhat anomalous in its characters. Like Cheiroplewria it shows sporangial advance with a relatively primitive vegetative condition. DESCRIPTION OF FIGURES IN THE TEXT. Figs. 1-3. Diagrams of 3 sterile pinne of Platyzoma, seen from the upper surface. They show the chief variations in the sympodial venation. Figs. 4, 5. Diagrams of 2 sterile pinne of Platyzoma, seen from below. They show further variations in venation, The strongly revolute margins of the pinne bear fringes of small teeth-like hairs. Fig. 6. Diagram of a forked pinnate leaf of Platyzoma in Kew Herbarium. The leaf is attached to the rhizome, and bears pinnz both above and below the point of branching. The terminal portions of this leaf are damaged. Fig. 7. Diagram of a forked, and probably dichotomised, pinnate leaf of Platyzoma in Kew Herbarium, The leaf was detached from the rhizome. It bears pinne both above and below the point of branching. ' Fig. 8. Diagram of a perfectly dichotomised leaf of Plutyzoma, taken from an herbarium specimen in Glasgow University. The leaf dichotomised at a point near to its base, and no pinnz were developed below the point of branching. Fig. 9. Diagram of a fertile pinna of Platyzoma, seen from below. The margins are very revolute. The sporangia are of variable size, but are invariably inserted terminally upon the main lateral veins or upon branches of these. Fig. 10. Bundle-ending of a vein, supplying a sporangium of Platyzoma. It consists of a conspicuous group of long reticulated tracheides, and a sheath of more delicate and narrow cells which persist to the tip of the bundle. DESCRIPTION OF FIGURES IN THE PLATES. _ Puate I, Fig. 1. Transverse section of the rhizome of Platyzoma. Four large leaf-traces are seen on their way out through the cortex; a fifth is being liberated from the upper surface of the stele. On the right a small leaf-trace is separating from the stele, and on the left a root-trace is departing. (x 12.) Fig. 2. A multicellular unbranched hair taken from the rhizome. (x 10.) Fig. 3. A filiform leaf with circinate tip, and measuring about + inch. It was taken from the lower surface of the rhizome. (x 6.) Fig. 4. Transverse section of the stele of Platyzoma, showing a large leaf-trace about to be liberated on the upper surface. Small leaf-traces are being separated from the sides and lower surface. The xylem of the stele is clearly differentiated into an outer zone of small tracheides, and an inner zone of broad tracheides intermixed with many parenchymatous groups. (x 24.) Towards the upper side of the stele the inner xylem-zone is greatly reduced. Fig. 5. A reduced pinnate leaf, taken from the upper surface of the rhizome. It was about 14 inch long. The pinne are variable in size, but their venation is invariably a sympodium. (x 6.) Fig. 6. A photograph of the specimen of Platyzoma upon which this paper is founded. (x 4.) Fig. 7, A typical filiform leaf measuring about 14 inch, seen from its lower surface. (x 6.) It is differentiated into a broad basal region, a slender middle, and a bulbous terminal region. . Fig. 8. A photograph of part of the specimen upon which this paper is founded. The heterophylly is clearly shown, and the crowding and grouping of pinnate and filiform leaves is demonstrated. (Natural size.) Fig. 9. A “reduction” leaf which measured about 1 inch, The venation of the majority of the pinne THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. 655 is a simple midrib, and no vascular supply is seen in the small terminal pinne. At the base of the rachis is a small aborted branch devoid of pinne. (x 6.) Fig. 10. Transverse section of the stele of Platyzoma. The sclerenchymatous pith is breaking from the stelar ring along the inner endodermis. In this section the difference in thickness between the upper and lower sides of the xylem-cylinder are not so pronounced as in figs. 1, 4, 13, and 63. (x 36.) Fig. 11. An incomplete “reduction” leaf, 2 inch long. Only a few terminally aggregated reduced pinne are developed. (x 6.) — Fig. 12. A “reduction” leaf, + inch long, taken from the side of the rhizome. Towards its hooked tip are two small pinne with simple unbranched veins. ( 6.) Fig. 13. Transverse section of the stele of Platyzoma, showing 3 large leaf-traces departing from the upper surface ; a trace of intermediate size being liberated on the upper portion of the right side; and small traces departing on right and left towards the lower surface. (x 36.) Puate II. Fig. 14. Transverse section at the base of the petiole of a typical pinnate leaf of Platyzoma. (» 15.) Fig. 15. Transverse section of the lower part of the rachis of a typical pinnate leaf. (x 15.) Fig. 16. Transverse section of the wpper part of the rachis of a typical pinnate leaf. (~ 15.) Fig. 17. Transverse section of the terminal part of the rachis of a typical pinnate leaf. (x 15.) Vig. 18. Transverse section of the very base of a typical filiform leaf. (x 15.) Fig. 19. Transverse section of the same leaf at a slightly higher level in the leaf-base and showing the establishment of the lateral grooves. (x 15.) Fig. 20. Transverse section of the same leaf-base at a still higher level. The xylem of the leaf-trace has widened laterally, the water-storage tissue is established towards the lower surface, and the flanges and grooves are fully developed. (x 15.) ; Fig. 21. Transverse section at the base of the middle zone of a typical filiform leaf. (x 15.) Fig. 22. Transverse section at the base of the middle zone of another typical filiform leaf. (x 15.) The xylem of this leaf is much more massive than in the leaf seen in section in fig. 21. Fig. 23. Transverse section of the upper part of the middle zone of a filiform leaf. (x 15). Fig. 24. Transverse section of a similar portion of another filiform leaf. The xylem is here much more massive than in fig. 23. (x 15.) Figs. 25, 26. Transverse sections through the bulbous terminal portions of two typical filiform leaves. (x 15.) Fig. 27. Transverse section of the petiole of the reduced leaf depicted in fig. 9, and showing the prepara- tion for the splitting of the xylem prior to supplying the two branches of the rachis. ( 15.) ‘Figs. 28-35. An ascending series of transverse sections of the stele in the basal portion of the dichotomised pinnate leaf represented in text-figure 8. The series illustrates the chief steps in this dichotomy. (x 15.) Puate III. Figs. 34-37. An ascending series of transverse sections of the rachial portion of the small leaf repre- sented in fig. 11. The series illustrates the extra-marginal pinna-trace supply, and the massing and strengthening of the xylem towards the tip of the leaf. (x 15.) Fig. 38-40. An ascending series of transverse sections through the pinnate portion of the small leaf shown in fig. 12. It shows the preparation for and liberation of a pinna-trace. (x 15.) Figs. 41-47. A series of sections showing the origin and liberation of a large leaf-trace from the upper surface of the stele. In fig. 47 the trace is as seen in transverse section on its way out through the cortex. (x 15.) Figs. 48-51. A series of sections showing the origin and liberation of a small leaf-trace from the lower surface of the stele. In fig. 51 the trace-is as seen in transverse section on its way out through the cortex. (x 15.) TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 20). 96 656 THE ANATOMY AND AFFINITY OF PLATYZOMA MICROPHYLLUM. Fig. 52. A section of the stele of Platyzoma, showing the origin of a root-trace from the outer xylem- cylinder, (x 15.) Fig. 53. Transverse section of a root-trace on its way out through the cortex of the rhizome. (x 15.) Fig. 54, Transverse section of a root of Platyzoma. (x 15.) Prats LV. Figs. 55-58. A small sporangium of Platyzoma seen from behind, in front, above, and below. The irregularities in the nearly vertical annulus, and the interruption of the annulus by the stalk, are shown. (x 36.) Fig. 59. A sporangium with almost globular capsule, and an oblique—though almost regular— annulus. (x 36.) Fig. 60. The wall of a sporangium opened out so as to show the irregular and broken annulus. (x 36.) Figs. 61, 62. Two views of a rarer type of small sporangium of Platyzoma. The annulus is almost horizontal, but is irregular and broken, and no clearly defined stomium is developed. (x 36.) Fig. 63. Transverse section of the stele of Platyzoma, at a point where the xylem towards the lower surface is broad. The groups of tracheides appear white in the photograph, and the chains and groups of xylem parenchyma are black. (x 36.) Fig. 64. A spore of the intermediate size. (x 90.) Fig. 65. A spore of the smallest size. (x 90.) Fig. 66. A spore of the largest size. (x 90.) Figs. 67, 68. A front and a back view of one of the largest sporangia of Platyzoma. (x 36.) SI = ~ ‘a ae a¥ © & U = me = P2 oS N 3 at a A. Zz. HOMPSO fe MM. JOHN Collo., Derby rv fh o a 2 L. — = = =, 8 ; ww > i = ¢ i : ay 1 = : { if aH | ¢! ea Q/ eo XY 1) se = | fae S ww ae a ~ ~ &% iS) =| rs < — gy lon — N Fal sl z : _— ay Zz © Nn QO — = © a —! | — <. ne cq ~ — Trans. Roy. Soc. Edin* Ss 2 a : 3 rans. Roy. Soc. Edin‘ Vol. LI. Joun M.THompson: PLATYZOMA MICROPHYLLUM. PLATE. III. “a ————————————— s—“— vans. Roy. Soc. Edin: oll. Joun M.THompson: PLATYZOMA MICROPHYLLU\ PLATE IV. Bemrese, Collo, Derby. 9 g whe } * 7 A aan ; 1 * Ar nes7 es} XXI.—On Leaf-Architecture as illuminated by a Study of Pteridophyta. By F. O. Bower, D.Sc., F.R.S., Regius Professor of Botany in the University of Glasgow. (With One Plate.) (MS. received March 6, 1916. Read March 6, 1916. Issued separately November 8, 1916.) ‘ The expression “architecture” as applied to the leaf was introduced by PRantL in his monograph on the Hymenophyllacee.* It may be adopted as connoting the sum of the facts of construction of leaves; together with those principles or methods deduced from them, upon which we find the leaf to be built up. The varieties of size, form, and complexity of leaves appear infinite; but simi- larities in the scheme of their construction are obvious. It cannot be assumed that where similarities occur they are necessarily due to immediate community of descent. They may or may not be. Parallel development under similar conditions may be, and probably has been often, the source of such similarity. But even so it may be possible to connect the simpler and the more complex within the several lines of nearer relationship, and a study of several such lines may be expected to disclose certain underlying principies or methods which have ruled in the construc- tion of foliar organs at large. The recognition of these, in their evolutionary aspect, is the proper basis for a scientific knowledge of leaf-architecture. There are three chief avenues which may lead to such knowledge :— (1) A comparative study may be made of adult leaves, in the mature state or in the course of their individual development, in a large number of different types. (2) A study may be made of the juvenile leaves of the individual, and the gradual steps be traced up to the adult form. Comparison may then be made of the results in various forms related or systematically apart, and especially in the case of those types that are believed to be relatively primitive. (3) The results of such comparisons may be placed in relation to the fossil record, and any conclusions thus obtained as to phyletic progression may be checked accordingly. All of these avenues should be pursued in order to arrive ata scientific knowledge of the “architecture” of leaves in Vascular Plants. Naturally the first of these took priority in the history of the science; and unfortunately it was first practised for the most part upon the Higher Flowering Plants. This often led to the interpretation of the lower in terms of the higher, and to the consequent failure to recognise the full significance of the facts observed. In some measure this has been corrected by the pursuit of the third avenue of study, viz. by the comparison of the fossils. The flattened form of the leaf-blade, which has led to its very perfect preservation * Leipzig, 1875, p. 7. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21). 97 658 PROFESSOR F. 0. BOWER ON in the state of impressions, has supplied important features. The second avenue has, however, been somewhat neglected, and especially in the case of those plants which should serve as the basis for such comparisons—the Pteridophyta. I am not aware of any methodical synthesis having been given of the facts relating to the juvenile leaves of the Pteridophyta, so as to make a general comparison of them possible. In many of the more important primitive types the facts are still very imperfectly known. The attempt has therefore been made here to supply the material for pursuing this second avenue of study, and to place the results in relation to those from other sources. It is only when the facts have been ascertained and the underlying principles have been laid down, in some such way as this, that a beginning can properly be made for the discussion of such larger questions as the phyletic origin of that member designated as the “leaf,” and its probable relation to the axis which bears it. But before the detailed description of the juvenile leaves is entered upon, it will be well to summarise the views which have been expressed, and the con- clusions attained by others, on the subject of leaf-architecture, in so far as they have been based upon the elaborate. leaves of the Filicales. Hormrister™ states that “the formation of the pinne of the frond in the species of Pteris, as in the rest of the Polypodiacez, is the result of a true bifurcation of the punctum vegetations . . each of the new shoots is alternately more strongly developed, thus changing the direction of the bifurcation to the right or to the left. The weaker one is pushed on one side so as to appear to be lateral.” The continual change in the direction of the less vigorous bifurcations causes the feather-like form of the frond, whose seg- ments (as is well known) are in no species exactly opposite to one another. This description, based on a special example, includes facts which will be found to be of wide application. It is doubtless this passage to which Sacus refers in his Teat- book,t where he introduced the terminology which has since been generally adopted for branching. He noted the apparently true dichotomy in Platycervum alcicorne, and that “according to an older statement of HormErIsTER it appears that branching of Fern-leaves generally commences dichotomously, although mature leaves mostly resemble a monopodium.” Further, he notes that “since the branches or lobes are apparently always alternate and not opposite, and the terminal lobes of the leaves are frequently developed as equally strong bifurcations, leaves of this kind may be considered, according to Hormristrer’s hypothesis, as dichotomies developed in a sympodial manner . . . the midrib representing the sympodium, and the apparent lateral branchlets the weaker branches. . .. A similar interpretation may perhaps be permitted for the simply pinnate leaves of the Cycadacez.” The observations of SapEBEcK { translated these conclusions into terms of cellular construction, thus actually demonstrating the dichotomous character of the ultimate * Higher Cryptogamia, p. 209. + First English edition, p. 161. { Friedrichs-Realschule Jahresbericht, Berlin, 1874. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 659 branchings of the lamina in the juvenile leaf of Aspleniwm. They were followed closely, in 1875, by Kny’s monograph on the Parkeriacee. Here the demonstration was given that where the single initial cell is still active in leaf-forming pinne, these do not necessarily coincide with segments from the initial cell.* In the same year Prantt’s monograph on the Hymenophyllacee appeared. The general conclusion from his detailed developmental and comparative study was that in them the leaves, simple or complicated, were the result of purely dichotomous branching, and usually developed sympodially. He suggests that this conclusion is applicable also to other Ferns.t In 1881 he published similar observations on the Schizeeaceze. The result was that he was able to refer the branching in all of them ultimately to dichotomy. He for the first time followed carefully the steps of departure to the sympodial development, both in the juvenile leaves and in the pinne of the adult leaf: and he constructed diagrams illustrating that progression for Lygodium circinatum.t The term “dichotomy” has appeared repeatedly in the preceding paragraphs. It must not be assumed in all cases where it has been used, or where it will be used in this memoir, that there has been the equal division of an apical cell. The observations for or against such a conclusion are mostly wanting. Still, a near approach to such a theoretical perfection of dichotomy, if not an actual demonstra- tion of it, has been given by SaDEBECK in the seedling leaf of Asplenium. From such a condition, which seems actually to exist im some cases, the gentlest gradations lead to what is clearly monopodial branching. What is often seen is then “apparent dichotomy,” that is, the “continuation of one shoot by two equally strong ones.”§ Cases where the resulting shoots are not equal, lead to what in our descriptions are designated “sympodial developments of dichotomy.” It is believed that the fundamental construction in leaves was the true dichotomy, and that these are derivatives from it along lines which are clearly suggested in the juvenile leaves of many Ferns. It will be superfluous here to abstract the scattered literature giving details of further isolated examples, whether in adult or in juvenile leaves, where the demon- stration of dichotomous branching has been given. It must suffice to say that in very many cases the dichotomy is obvious enough. Accordingly, we may pass directly to the general statements of GorBEL on the subject of leaf-construction in Ferns, given in his Organographie, as representing the outlook of present morphology on this matter.|| He noted the similarity of juvenile leaves, even in Ferns where the adult leaves are divergent in character. These may be either dichotomous or sympodial in their development. The latter is the case where the growth is apical (Scheitelzellwachsthum), but apical growth merges into marginal (Randzellen- * P. 41, Taf. vii. fobs 14, {eRe § Lane, Address to Sec. K, Brit. Assn., 1915, p. 8. || 2te Auflage, 1913, p. 337 ; see also Engl. ed., part ii, pp. 313-321. 660 PROFESSOR F. 0. BOWER ON wachsthum). Where the growth is apical the origin of the pinnee may be mono- podial, as already shown in Ceratopteris by Kny, and exemplified by GorBEL in Adiantum Edgeworthi.* Where the marginal growth has supervened the branch- ing becomes dichotomous. This has been summed up by GorBet in the following words: “In all Ferns the lateral primordia of pinnz appear on the primordium of the leaf, and if the leaf is a greatly elongated one the lateral parts are laid down in rapid succession ; but where surface-growth predominates, then there is dichotomous branching, and there is no formation of a strong leaf-spindle or midrib.” Such facts suggest a phyletic meaning. JBearing on this question GOEBEL recognises three stages of leaf-development in Ferns: (i) marginal growth and dichotomous branching; (ii) apical growth, with monopodial branching, but later marginal growth and dichotomous branching ; (iii) persistent apical growth, forming long-growths and short-growths, the latter limited, as in Gleichenia and Lygodium. He remarks that “whether this series forms an ascending or a descending one, or whether we must recognise it as simply one construction, cannot certainly be determined. What is certain is that the configuration of the primordium of. the leaf is connected with that of the mature condition in the manner that has been indicated above.” The facts and comparisons to be brought forward in this memoir will help to resolve this question. It may here be briefly stated that they are believed to indicate that the marginal growth and dichotomous branching represent the primitive foliar development ; that the stronger apical growth with prevalent monopodial branching is a derivative state, which is absent from many juvenile leaves, and, though commonly present in the adult, it lapses into the primitive dichotomy in the distal region. Already there is much evidence from anatomical sources which indicates that in Fern leaves the apex and base show more primitive characters than the middle region. It will be important to keep this conception in mind when examining the leaf-constructions to be detailed below. GOEBEL describes the juvenile leaves as arrested members (Hemmungsbildungen).} But if the comparative evidence leads to the conclusion above stated, this would appear to be an inversion of the true evolutionary conception of the leaf, as exemplified in the ontogeny. A more probable thesis would be that the juvenile leaf is not arrested, but is something similar to the primitive and fundamental type ; while the adult leaf is a promoted condition, and its higher development is consequent upon a phyletic advance in complexity comparable to that seen in the ontogeny. But the later and more elaborate leaves may be subject to arrest due to various causes, which throws them back to such a state as is seen in the juvenile leaves. This is the theoretical position which will be entertained in this memoir, and it will be seen later how far it is in accord with the facts to be described. * Organography, Engl. ed., ti, p. 317, fig. 20. + E. W. Sinnott, Ann. of Bot., 1911, p. 167 ; and other writers. | Flora, 1889, p. 28; Vergleichende Lntwickelungsgeschichte, p. 254 ; Organographie, 2te Auflage, p. 378. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 661 While the statements of Gorse, in his Organographie present the purely morphological position, Poroni had been approaching questions of leaf-architecture from the avenue of paleontological comparison.* He regarded dichotomy as the primitive type of branching, and monopodial (pinnate) as the derivative. He notes the greater prevalence of the former in Paleozoic than in more modern times, and that it still exists, though less frequently, in the Higher Plants, being represented atavistically in their cotyledons. He discusses why the pinnate should have replaced the dichotomous, and sees the explanation in the point that a dichotomous blade would result in a circular surface of leaf-expanse, while a pinnate blade approaches the oval, which has mechanical-physiological superiority, and is better fitted to resist leverages of weight and wind-pressure. He passes from Ferns to Vascular Plants generally, and asserts (provisionally) that all monopodial branchings origi- nated phylogenetically from dichotomies. This question will come up for discussion towards the end of this memoir. Turning more especially to the leaves of Ferns, Potonis } discusses their archi- tecture in the ancient fossils, as regards both venation and branching. He notes as some of the chief features in early Ferns: (1) the preponderance of dichotomy in them ; (2) their frequently unsymmetrical development by interchange of forking and pinnation, and transitions from one to the other; (3) the frequent occurrence of lobes of the ultimate order between those of the penultimate order; and (4) the frequently large size of the katadromic pinnule. In addition to these features he remarks also that most Paleozoic fronds have katadromic construction, while in recent types the anadromic greatly preponderates. Also that in the earliest a midrib is absent in the ultimate pinnules, while in later forms it is present. Further, that reticulation becomes more prevalent from lower to higher horizons. He notes that Srur in his Culm-Flora does not represent a single case of reticulation: this is initiated in the Middle Coal Measures, and becomes more prevalent upwards, while the highest type of reticulation, with smaller meshes within larger areol, occurs first in the Mesozoic period. Such statements as these, based upon a wide knowledge of fossils, supply a foundation upon which to work. They serve as checks to comparative conclusions ; and it is a very reassuring fact that the results already attained by comparison of living forms are in substantial accord with them. A real advance has thus been made along the two avenues of study on the adult leaves and on the fossil types. But a knowledge of the actual facts as to the structure of the primordial leaves is still very incomplete. It may be a question how far their characters will bear an atavistic interpretation. But on this it is impossible to approach any opinion until the facts are before us. These will now be described so far as they are available. * “Die Beziehungen zwischen dem echt-gabeligen und dem fiederigen Wedel-Aufbau der Farne,’ Ber. d. d. Bot., Ges, xiii, 1895, p. 244 ; also Lehrbuch der Pflanzen-Palaeontologie, Berlin, 1899, p. 110, ete. + Lehrbuch, p. 110, ete. 662 PROFESSOR F. 0. BOWER ON The method of preparation has been a very simple one. The leaves are made transparent by eau de Javelle; stained with safranin; cleared in bulk in oil of cloves; and mounted in Canada balsam. The result is a transparency, with the vascular tissue so stained that the connections, even those at the base of the blade, can be readily followed. The relation between branching of the leaf itself and branching of the veins is obviously a close one. It suggests that where the branch-veins run parallel the venation represents a latent branching of outline, and that originally distinct lobes have been “ webbed,” or laterally merged together. This seems highly probable in simple cases, but it may be doubted whether the branching of veins is always the equivalent of a webbed branching of outline. In complex cases where reticulation occurs there can be no question but that vascular strands have been formed de novo. At the moment it is difficult to make any general statement on this point. It is best to be fully aware of the difficulty, and to recognise provisionally that a close relation subsists between branching and venation. When a large number of examples have been considered in detail it may be possible to draw some more definite conclusion. Comparative study of the leaf-architecture of the Pteridophyta is the natural foundation for a similar inquiry for the Spermophyta. The actual lines of evolu- tion may be obscure, but there can be no doubt that the one will illuminate the other. This was the view of Hormersrer, Sacus, PRANTL, and other morphologists. Accordingly we*may be prepared to trace at least analogies, if not some nearer degree of resemblance, between the construction of the leaves of the more primitive Pteridophyta and that of the relatively advanced Seed-bearing Plants. No special observations relating to the latter will be described here, and it may be that further inquiry will be necessary before full advantage can be taken of the basis for com- parison which the Pteridophyta can supply. But at the close of this memoir some attempt will be made to carry onwards to the Flowering Plants the applica- tion of the results acquired, which may help to elucidate the architecture of their -leaves also. Description of the Architecture of the Leaves of the Pterrdophyta, with special reference to their Primordial Leaves. . LYCOPODIALES. In all of these the leaf is relatively small, and its construction is always simple both in form and venation. It was, however, of considerable actual size in the Lepidodendracez and Sigillariacez, as it is also in the living Jsoetes. It is traversed by a single unbranched vascular strand. The only known exception to this among living or fossil Lycopods is in certain Sigillarias, where in the middle region of the leaf there are two equal strands.* These fuse towards the apex, and the whole * Scort, Studies in Fossil Botany, p. 230, fig. 95, of Sigillaria sulcata. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 663 structure suggests not a dichotomy but a local segregation of the vascular tissue of the single strand in the middle region of the leaf, where the blade is broadest. The analogy with what is seen in certain Conifers is obvious. Notwithstanding this constant simplicity which the leaves of the Lycopodiales show, they do not differ essentially in mode of origin from the leaves of other Vascular Plants. Indeed, they may be matched very nearly in point of size and complexity by the leaves of the Hydropteridez, while the primordial leaves of Pilularia and Marsilia resemble in many of their features the submerged leaves of Jsoetes. It may remain an open question whether the simplicity of the Lycopod leaf is primitive, or the result of reduction. A circumstance which favours the former alternative is the ereat uniformity which they show, not only in the living forms, but also in the fossils. ‘The juvenile leaves in these plants do not show characters of comparative interest, beyond the fact that they are often more succulent than the leaves of the adult. I find myself unable to follow Lienier in the distinction which he draws between “phylloides,” by which term he designates the leaves of the “ Prohépatiques” and Lycopods, and the leaves of the Filicinee. As regards their relation to the axis, their acropetal succession, and their general characters of form and structure, the foliar appendages of the Lycopodiales fall into the category of “leaves” just as much as those of other Vascular Plants. It is not, however, necessary to regard them as all derived along a single ancestral line.* EQUISETALES. The leaves of Hquisetum are mere webbed teeth, each provided with a single vascular strand, and united into a sheath. They are probably reduced representa- tives of an ancestry with larger leaves, less fully webbed, or even free. Thus in Annularia and Phyllotheca the leaves are slightly webbed: but in Asterophyllites, which is traced back to the Devonian Period, the leaves appear quite separate in widely divergent whorls. These leaves were simple, though of larger dimensions than the leaf-teeth of Hquwisetum; and more effective than these as assimilating organs. But among the earliest forms, such as Asterocalamites (Schimper) from the Culm, the separate leaves were themselves branched in repeated dichotomies, as shown in the well-known drawing of Stur.t These facts indicate that in the earlier Equisetales the leaves were separate assimilating organs, and were sometimes susceptible of dichotomous branching. That they were liable to cohesive webbing. That with this in some cases went a reduction, so that they became mere protective sheaths to the assimilating axis, asin Hgwsetum. In others they became enlarged, and the sheaths split into broad * Compare Licnier, “ Equisétales et Sphénophyllales: leur origine Filicinéenne commune,” Bull. de la Soc. Linn, de Normandie, 50 série, 7° vol., Caen, 1903. t See Land Flora, fig. 199 ; or Zu1LLER, Palzxobotanique, fig. 112. 664 PROFESSOR F. 0. BOWER ON pseudo-leaves, each with several veins, which however do not appear to have them- selves branched. This is seen in Schizoneura.* For our comparisons the most important fact is the existence of equal dichotomy of the linear leaves within the Kquisetales (Asterocalamites), though in most of them the leaf is unbranched. In this connection must be mentioned two organisms of great antiquity but of rather uncertain affinity. Pseudobornia, which was described by NatrHorst from the Upper Devonian of Bear Island, had jointed stems and whorled leaves, according in this with the Equisetales and Sphenophyllales. But the leaves are palmately branched with repeated dichotomies, the segments having serrate edges and the whole presenting a fern-like habit, with a more complex leaf-structure than is found in the Equisetales or Sphenophyllales.| Another early type with elaborate leaves borne in whorls on a jointed axis is Chevwrostrobus, of which only the fertile cone is known. Each sporophyll is composed of three equal sterile lobes and three sporangiophores. In so compact a branching as it shows it is difficult to be sure — of the method of branching. It suggests trichotomy. ‘But there is no doubt that the distal end of each of the sterile lobes shows a dichotomous branching. It thus appears that, though in both cases the knowledge is limited and development unknown, still there is clear evidence of dichotomy in the construction of their foliar organs.{ SPHENOPHYLLALES AND PSILOTACEA. But dichotomy is much more prevalent in the leaves of the Sphenophyllales and their living correlatives the Psilotaceze. The latter bear in the vegetative region simple leaves with an unbranched vascular strand; but their sporophylls show normally a single dichotomy associated with the production of a sporangiophore. In abnormal cases the dichotomy may be repeated. A possible interpretation of these facts would be that the vegetative leaves are simple by reduction, while the forked sporophylls represent the more primitive state. This would bring them into line with the fossil Sphenophylls, in which branching exists both in the vegetative leaves and in the sporophylls, and is carried further than in the Psilotaceze. This is particularly well seen in Sphenophyllum majus, where the sporophyll may show two successive, equal dichotomies; but these are repeated still further in S. cuneifoliwm, and especially in the relatively large-leaved Trizygua. There is a great variety in the state of “webbing” in this family. All grades of it can be illustrated, from narrow linear dichotomising segments with a single vein, as in S, tenerrimum, to webbed segments with a marginal tooth at the distal end of each vein of the dichotomy, as seen in S. cunerfolium. The most advanced state of webbing is seen in those with dichotomous venation, but entire margin, as * See FeistMaNTHEL’s drawing reproduced in ENeLER and PRANTL, i, 4, fig. 345. + Natnorst, K. Svensk Vetenskaps-Akad. Hand., xxxvi, No. 3. } Scorr, Studies in Fossil Botany, i, pp. 114-123. caw | CW SAY ieee! ain LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 665 in S. verticillatum. Poronti states* that on the whole the size of the leaves increases in passing from the older to the later horizons, but that the converse holds for the cutting of the blade. S. tenerrimwm from the lower horizons with | smaller leaves has quite narrow, linear, radiating segments, while S. Thoni from the “ Unterrothliegenden ” has large undivided leaves. This is an important point for comparison with the Ferns; but in this connection it must be remembered that in S. cuneifolium both linear and webbed leaves are found on the same specimen. This fact does not invalidate Porontk’s point, but it detracts somewhat from its value. FILICALES. In dealing with this, the largest group of the Pteridophyta, which shows so great a diversity in foliar structure, those which are held to be relatively primitive types will be taken first, and the descriptions will relate chiefly to their juvenile leaves. These commonly prefigure in some measure the characters of the adult leaf. But it will be seen from numerous examples that there is a greater similarity in the juvenile leaves of Ferns at large than is seen in their adult leaves. In the former, dichotomy, either with equal branching or sympodially developed, is in most cases a quite obvious feature. Osmundacee. It is convenient first to describe the case of Osmunda regalis, as its juvenile leaves show a condition which is unusually regular. But, however exceptional this degree of regularity may be in other Ferns, the study of it serves to bring into prominence that dichotomous structure which underlies the branching in the leaves of Ferns. The cotyledon has a single vascular strand traversing its stalk, which forks as it enters the lamina, and each resulting shank forks again twice (fig. 1, a). The two main branches of the first dichotomy occupy the two slightly unequal and not very clearly marked lateral lobes; but the left-hand branch in this particular leaf is slightly the stronger, and it shows signs of a second lobing. Older leaves taken from different plants, compared in succession of size, show at first a very striking equality of lobation and of dichotomous venation. Examples are seen in fig. 1, b-d, where this equality is strictly maintained, especially in the latter. But as we pass to older leaves, of larger dimensions, the equality is gradually lost, and in O. regalis the progressive steps may be more readily followed than in most Ferns. Thus in fig. 1, e, while the lobing and the dichotomy of the venation are not markedly different, the left lobe is larger than the right, and the acroscopic shank of the venation of the left lobe has correspondingly increased. Moreover, the basal parts of the alternate right and left shanks of its successive dichotomies have disposed themselves in a sympodial fashion. The left-hand lobe has also become clearly divided into two parts, of which the acroscopic projects the farthest, and with a * Lehrbuch, p. 176. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21). 98 Cb 666 PROFESSOR F. 0. BOWER ON straightening up and sympodial development of its venation it almost takes the position of a terminal lobe with a midrib. In (f) the pseudo-terminal position appears plainer, though its venation still appears as a very regular dichotomy. But in (g) the sympodial development of the veins, both in the pseudo-terminal lobe and in the apparently lateral lobes, has advanced so that each appears to be b ® S Fie, 1, a-g.—Successive juvenile leaves of Osmunda regalis, showing steps of progression from equal dichotomous venation to sympodial branching, and the establishment of a terminal lobe. (x 5.) provided with a midrib in its basal part. At the distal region of each, however, the venation reverts to the original type of almost equal dichotomy. In older leaves the terminal leaflet and lateral pinnee are themselves liable to similar lobing, all having the venation of the type already established in (g). And this architectural method may be continued without further essential modification till the fully formed adult leaf is reached. The general characters of these, with their high state of pinnation, are too well known to need redescription. But interest attaches to the origin of the pinne, their LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 667 relation to the venation, and to the developments seen at the leaf-base; for all of these points have their bearing, not only on this individual case, but also upon the behaviour of leaves at large. The origin of the lower pinne of the mature leaf is known to be monopodial.* They arise in acropetal sequence below the apex, and in the already massive organ from which they spring they cannot be traced as having any definite relation to its apical segmentation. But passing up from the base of the whole leaf to the distal end there is a gradual transition, similar to that already known in other Fern leaves ; this leads to a mode of origin of the last branchings being referable to a sympodial dichotomy, of the same type as that seen in the juvenile leaves. The base of the terminal lobe often gives indication of the relation which exists between these later a A b Fie. 2.—Portions of adult leaves of Osmunda regalis, showing the relation of pinnules to the venation and to the leaf outline. (x 4.) branchings of the leaf and its venation. The branch-system resulting from the dichotomy of each single lateral vein may in the distal part of the lobe be included within the margin of that lobe; but towards its base a convex crenation may some- times be seen, which includes the whole of such a branch-system (fig. 2, a). Lower down a more definitely projecting lobe may be found, and lower lobes still may successively be elongated outwards and their venation be sympodially developed. The separation of such lobes from the terminal lobe may be completed by involution of the margin down to the midrib. Such steps, which link together the branchings of the venation with the origin of the separate pinne, are seen at the base of the leaflet shown in fig. 2, b. Such a relation is found to hold very widely in Ferns. — It shows that even a pinnate leaf is constructed on the basis of sympodial dichotomy, while the pinne thus produced may be alternate, or apparently opposite, and owe their origin to monopodial branching. This seemingly alternate or paired disposal of the pinnee is prefigured by similar dispositions of the lateral veins. Uniformity in the mutual relations of the veins is not necessarily found throughout a single pinna. This is illustrated by fig. 3. In * Phil. Trans., part ii, 1884, p. 575. 668 PROFESSOR F. O. BOWER ON the right-hand pinna there shown the successive veins alternate throughout its length. In the left-hand pinna they alternate at the base, but in the upper part the lateral veins appear to be inserted opposite to one another. In reality, however, as Hormerster noted, they alternate still. But the distance between the successive dichotomies is so abbreviated that the appearance is given of paired veins, arising on opposite sides of a continuous midrib. This appearance, which is very common in Ferns, as it is also in the leaves of Angiosperms, is thus simply the result of a slight modification of the sympodial dichotomy. ‘The steps of that modification are well illustrated in the juvenile, leaves of Todea barbara (see below). : The leaf-base of the Osmundaceze shows developments which appear only in the older leaves, and have no relation to the distal branchings. They rank with similar Fic. 3.—Two pinnules of an adult leaf of Osmunda regalis, showing how their venation is essentially alternate, but may in places be apparently opposite. (x 4.) growths in a number of primitive types. In Osmunda these consist in wide lateral flaps which form a broad sheathing base to the leaf-stalk. This development, which - is biologically important for protection, is of the nature of an extension of those lateral wings which extend the whole length of the leaf in varying prominence, and merge finally into the marginal growth at the distal end.* Most Ferns are lable to occasional variations from the normal; these may ) illuminate their morphology. The “forma furcata” is such a variant, seen in Polypodium vulgare, Scolopendrium vulgare, Athyrium, Nephrolepis, ete., ete.t The forking is frequently a very perfect dichotomy, and may affect only the distal ends of the pinnz, or the apex of the leaf, or it may extend downwards, so that in extreme cases it may result in apparently two complete leaves seated on a common stalk. In a specimen of O. regalis from India the distal ends of the pinnz only were affected, and this gave an opportunity for examining the relation of the forking to the venation, and so to the regular architecture of the leaf. Three examples are * See Phil. Trans., 1884, p. 577. + See the writings of Moorn, Lowx, LurrsseEy, etc. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 669 shown in fig. 4, a, b,c. In (a) the forking is slightly unequal, the stronger being to the left. There is a regular alternation of the veins forming the sympodium, and the supply to the smaller right-hand lobe may be held to be one of these veins, which has developed more strongly and branched more frequently to supply the marginal outgrowth. On this view it would represent a partial reversion from the inequality of the forking in the sympodium towards the primitive equality, such as is seen in the juvenile leaves. A similar interpretation would apply to (6), but there the branching of the lateral lobe is more extensive. Such conditions lead easily to that in (c), where the two lobes are very equally developed, so that it is difficult to say whether one or the other is the apex of the sympodium. In other words, one branching of the sympodium has reverted very perfectly to the primitive equal dichotomy. " Fic, 4, a-c.—Various states of ‘‘furcation ” of the apex of the pinnule in Osmunda regalis. (x 4.) Seedlings of Todea barbara, kindly supplied by Messrs Hitt of Edmonton, have been compared, and they show an earlier departure from the equal dichotomy than in O. regalis. In fig. 5, a, b,c, a number of cotyledons, or at most second leaves, are shown, and they illustrate in the succession of their lettering gradual steps from equal dichotomy to sympodial branching. This applies not only to the venation, but also to the lobation, which is much more marked than it is in Osmunda. Fig. 5, a, isan almost diagrammatic dichotomy, and the length of the shanks between the first and second branchings is approximately equal. In (6) the dichotomy is slightly unequal, the right lobe being the larger, but still the distance between the first and second forkings is equal in the two shanks. In (c) that length is unequal, the longer being that of the stronger limb, which is unusual. The more frequent condition is that seen in (d), where between the first branching and the second the distance on the weaker limb is greater than on the stronger. This leads to the insertion of the vascular supply of the lower lobes, right and left, being nearly opposite to one another. It is still more nearly the case in (e). The transition is seldom so clearly seen as in these cases, and it accords with what has already been noted above in Osmunda. 670 PROFESSOR F. 0. BOWER ON Comparison may be made with the juvenile leaves of Todea superba, in which species the mature leaves are dissected into fine laciniz, each containing a single vascular strand. The terminal branchlets of the adult leaf or pinna show obvious dichotomy, and the relation of the laciniz indicates a sympodial origin. The first = V Fic. 5.—Juvenile leaves of Todea barbara. (x 4.) We Fie. 6, a-c.—Cotyledons of Todea superba, showing different complexity of construction. (x 38.) é leaves appear in strong antithesis to those of Osmunda, for they have from the first not only separate lacinize, but also sympodial development. This is shown in fig. 6, a, b, c, which are readily understood. They are all cotyledons, and it appears that with the narrow laciniz goes an early adoption of sympodial development. But none the less dichotomous branching underlies the structure of the cotyledon, as it also does that of the adult leaf. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 671 The developments at the base of the leaf in 7. superba go a step further than in Osmunda, for the broad lateral flaps extend across the face of the petiole, so as to form a continuous envelope, which ensheathes the next younger leaf. In this respect it is comparable with that seen in the Marattiaceze and Ophioglossacez.* A comparison of these three examples from the Osmundacee thus shows a near similarity in the venation-system, but a difference in the degree of laciniation. It lies near to hand to suggest that the entire cotyledon of Osmunda is a result of “webbing ” of a laciniate type, such as T. superba; the vascular skeleton remaining of the same fundamental type, and subject in either case to the same sympodial modifications. On the other hand, the sympodial development replaces the equal dichotomy relatively late in Osmunda, while in T. superba it is already advanced even in the cotyledons. The diversity thus seen in the Osmundacee is widespread in Ferns. The primary steps of equal dichotomy so well seen in O. regalis are apt to be abbreviated, or wholly omitted. The modifications of the dichotomy from the equal to the sympodial thus far seen have been described in plain words. They all fall under a well-known scheme, designated the scorpioid sympodial dichotomy.{ The differences which they show depend chiefly upon (i) the degree of sympodial development, (ii) the relative distance intervening between the successive forkings, and (iii) the number of these forkings. Schizxacex. Few personal observations have been made on the juvenile leaves of any Schizzeacez, excepting Anewma adiantifolia. But that is little matter for regret, since this family has been very adequately studied by Prantt.{ The genus Schizxa regularly maintains the dichotomous branching in the vegetative region of the adult leaf. But the matter is not so clear in the fertile region, where the branches are crowded together in a dense group and folded so as to face one another. If transparent preparations be made from the adult leaf, the branching may be traced through the vascular connections, as seen in S. dichotoma (fig. 6 bis). In (a) the leaf-tip has been halved, and the view from the inner surface shows seven pinne, of which the uppermost and lowest have themselves forked. The vascular strands all gather up to the main supply in a way that is made more clear by (b), which shows the adaxial aspect of another specimen. The pinne are in apparently opposite pairs, and the leaf ends in a bifurcation. The arrangement is to be interpreted thus: the pinnee are not in exact_pairs, but are really formed in a close sympodial series. The * Compare Phil. Trans., part ii, 1884, p. 577, plate xxxvii, fig. 8. t There have been misunderstandings in the use of terms applied to branch-systems. The use of the terms here adopted is that of Hormutster (Allgemeine Morphologie, p. 435) as applied to sympodial developments of monopodial branchings. The term “scorpioid” in this sense was introduced by Bravals, while Scummpmr designated this type of sympodium a Cicinnus. The terminology founded for monopodial systems may be adopted for designat- ing the similar developments in dichotomous systems, t Schizaeaceen, 1881. 672 PROFESSOR F. 0. BOWER ON actual relation of each apparent pair is like that of the lateral lobes already discussed in O. regalis, and the seemingly paired limbs are in each case forkings of different order. In fact, they constitute a -close scorpioid sympodium. The point is clearer in the vegetative leaf of Marsilia, a genus now acknowledged to have near affinity to the Schizeeaceee, which will be considered later. In other genera of the Schizeeaceze the equal dichotomous branching is widely departed from in the direction of sympodial forms. Pranru has analysed the steps towards this condition, both in the adult and in the juvenile leaves of Lygodium circnatum. He remarks* that notwithstanding the complexity of the adult leaf of Lygodiwm, the underlying architecture is simple. There is a “spindle” of ‘ unlimited apical growth, which bears alternating “ primary segments”: each of the latter has a short secondary spindle, and it bears regularly two secondary segments, a b Fic. 6 bis, a, b. Branching in the fertile region of Schizeea dichotoma. (x 4.) which vary in character according to the species. They show conditions grading between that of a dichotomy of the venation (shown for L. palmatum in PRANTL, pl. i, fig. 1, which is comparable with our fig. 1, d, of the young Osmunda leaf) and the sympodial development with a terminal lobe, having a midrib (in L. venustum, PranvTL, pl. i, fig. 9, which is comparable with our fig. 1, g, of an older leaf of Osmunda). In its secondary segments L. circinatum corresponds essentially with the former type. Having ascertained this transition by comparison of the secondary segments in different species, PRANTL examined the juvenile leaves of L. corcinatum, and found that, in their successive stages from the cotyledon onwards, they as a whole showed just the same transition from the dichotomous to the sympodial type, and he represents them by schematic drawings,t which with slight modifications are reproduced as fig. 7, a, b, c. Of these, fig. a represents the condition of the cotyledon, which shows a regular dichotomy. Fig. b shows the state of a later juvenile leaf, in which the shanks are unequally developed and a sympodial state is initiated. Fig. c represents the condition of the adult leaf, the arrow indicating the unlimited growth of the sympodial rachis, upon which the pinne now take a lateral position, characteristic of the scorpioid sympodium. * Schizueaceen, p. 8. + Pl. i, figs. 12-14. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 673 PRANTL applied a similar analysis to the leaves of Aneimia and Mohria, and found that they were susceptible of a similar interpretation ; but in them the apical growth is limited. Fic. 7.—Schematic drawings after PRANTL, showing the sympodial development in leaves of Lygodiwm circinatum. (See text.) As material was available from my Jamaican collections, it was thought worth while to examine and depict the successive stages of the juvenile leaves of Anewnua adiantifolia (fig. 7 bis, 1-11). The simplest leaf is seen in (1): it shows a single 1 2 3 6 “TI 10 9 8 Fic. 7 bis, 1-11.—Successive stages of development of the juvenile leaves of Aneimia adiantifolia. (x 4.) forked vein, and two equal distal teeth. Transitional stages (2, 3) lead to a double dichotomy, with four equal distal teeth. The two limbs of the dichotomy begin, in more elaborate forms (4—7), to develop unequally, while a deeper indentation marks off the two lobes. Each of these is supplied by the branches of one limb of the fundamental dichotomy, and each has distal teeth corresponding to the veins. The TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21). 99 674 PROFESSOR F. O. BOWER ON inequality in more advanced leaves (8, 9) becomes more pronounced, while the anadromic half of that which is the more strongly developed overtops the rest. In (10) it has assumed the character of a terminal lobe. In this lobe in the more advanced leaves a similar progression is repeated (11), and an apparently simple pinnate leaf is the result. Already in relatively small leaves the lateral pinnze may repeat the same method of advance. It is perceptible in the older pinne of (11). And thus a scheme of elaboration is initiated which has no theoretical limit. It corresponds to that of the other Schizeacez, as analysed by PRANTL, and it is essentially the same as seen in Osmunda. All are scorpioid sympodia. In such cases the chief differences he in the earlier or later; and the greater or less assertion of the sympodial development. There is, however, one additional complication. That is the formation of reticulate fusions of the veins, which appear in Lygodiwm, § Hydrodictyon, and in Aneimia, § Aneimidictyon. This may be held to have been a relatively recent acquisition. Marsiliaceer. The adult vegetative leaf of Pilularia is simple. That of Marsilia is normally four-lobed, but its juvenile leaf is simple.* The progression in M. Hrnesti from the 4 3 4 5 Ce Fic. 8.—Successive stages of the juvenile leaves of Marsilia. After BRAUN, simple to the adult was followed by A. Braun, and is discussed by Goxpst.t Braun’s drawings are reproduced as fig. 8, where 1-4 show successive expansions of the distal end, with dichotomous venation, the veins being linked at the margin * These facts compare with what is seen in Schizea rupestris, in which, as noted by GornsrL (Organography, Engl. ed., ii, p. 478), its sterile leaf is unbranched. The branching, such as is seen in the sterile leaves of S. digitata and elegans, appears here only in the fertile region. + Vergl. Entwickelungsges., p. 254. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 675 by arched commissures, as in the mature leaf. (5) shows an obvious dichotomy of the blade. GorseL* remarks the absence of an apex (Hndtheil) between them ; but this is in itself evidence of the dichotomy. He notes the occasional existence of trifid leaves. These would be explained on the same footing as the ternate leaf of Helminthostachys or Osmunda. Finally, (6) shows four lobes, the two distal being clearly the result of dichotomy, since again the apex is absent. The facts, viewed in the light of the juvenile leaves of other Ferns, seem to uphold the in- terpretation of Braun rather than that given by GorpeL. They indicate that the b Fic. 9.—«, base of insertion of the four pinne in Marsilia polycarpa, showing the vascular connections. (x 35.) b, scheme of their arrangement as result of three dichotomies. quatrefoil leaf is the result of three successive dichotomies developed in a scorpioid sympodium. The development of the adult leaf has been traced by Hanstern.{ The two lower pinne arise monopodially, one being inserted lower than the other.{ The leaf is then three-lobed. ‘‘ Next the apical cell as such ceases its activity, while laterally from it the marginal activity increases, and so the middle lobe gradually divides once more.” The interpretation of this, in accordance with experience of other related Ferns, is that the four-lobed leaf results phyletically from three successive dichotomies, though actually the two lower pinnee are produced mono- podially, as is the case with the lower pinne of so many other Ferns. But the original dichotomy is retained in the origin of the two later pinne. The vascular connections of the mature adult leaf lead to clearer views of these relations. They appear as in fig. 9, a. The insertion of the lower apparent pair is * Ec, p. 254, + Pringsh. Jahrbuch, iv, p. 249. t L.c., pl. xiv, fig. 2. 676 PROFESSOR F. O. BOWER ON not exactly opposite: in the drawing the right-hand pinna is lower than the left. The two distal pinnee are an obvious result of bifurcation. Thus, following the adult vascular connections, the whole is to be interpreted as a scorpioid sympodium with only three forkings, and it may be represented diagrammatically as in fig. 9, bD. Transverse sections of the mature region of insertion support this conclusion.* The fertile region in Schizea has been seen. to be constructed on the same plan, but carried further by repeated forkings. The fertile leaf of M. polycarpa bears a large number of sporocarps, and ac- cordingly offers opportunity for study of their morphological relations. GOEBEL has shown that in origin these are of the nature of pinne,t and Miss Auiison{ has demonstrated that the insertion of their vascular supply upon that of the leaf-stalk is marginal, as are all the pinna-traces in the Schizeeacez.§ The actual origin of the sporocarps as described by GoEBEL is monopodial. Phyletically the probable explanation of their peculiar arrangement is that in repeated forkings the shanks successively right and left of the forkings are constantly the weaker; in fact, it is a helicoid sympodium, examples of which will be described below for Matonia and Pteris semipinnata, etc.| The simple acicular juvenile leaves of Marsilia correspond on the one hand to the simple leaves of Pilularva, and on the other to the vegetative leaves of Schizea rupestris and other species. In all of these it may be held that dichotomy was probably present in the ancestry, as it is in the rest of the Schizeeacee. They may be held to have reverted to a condition where the simple juvenile leaf is maintained throughout life. The same may also be the true interpretation of the simple leaf- structure seen throughout life in Salvinia and Azolla. Marattiaceer. In Danexa, Marattia, and Angiopteris the ultimate venation of the adult leaf shows prevalent dichotomy. But in Kaulfussia the venation is reticulate. All of them have prominent midribs, from which, in the former cases, the dichotomising veins spring, while each strand extends in Angiopteris and Marattia into a short marginal tooth. The juvenile leaves have been figured in all the genera, but the transitions to the mature state have not been fully followed. Thus Camppeti has depicted the juvenile leaves of Danza, of Marattia,** of Angiopteris,t+t and of Kaul- * Aruison, New Phyt., vol. x, pl. iii. + Organography, Engl. ed., ii, p. 479. t Luc, p. 204. § Davin, Trans. Roy. Soc, Edin., vol. 1, pt. ii, No. 11, p. 354. || The use of this term is that adopted by Hormersrer, after Bravars (Allg. Morph., p. 435), as applied to monopodial branchings, and transferred to dichotomies also, An inversion of the use of the terms “helicoid” and “scorpioid” appeared both in the Textbook of Sacus, Engl. ed., 1875, p. 157, and in that of Van TIEGHEM, p. 38. But they are used in their original sense in the Textbook of STRASBURGER. | EHusp. Ferns, fig. 125, p. 148, OTe hig, Loe. Dad. tt L.c., fig. 122, p. 147; also Farmer, Ann. of Bot., vi, pl. xv. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 677 fussia.* From all of these drawings it would appear that dichotomy, as expressed in outline or in venation, is not an obvious character of the juvenile leaf. But a b c ad Fic. 10.—Juvenile leaves of Dancea nodosa. (x 4.) Fic. 11.—Young leaf of Angiopteris evecta, showing at its base, on the right, a pinna, and on the left another is indicated. (x 4.) comparison of such leaves themselves, and their interpretation in the light of experience from other Ferns, shows that dichotomy developed sympodially is the fundamental construction of them all. Observations were made on young plants of Danza nodosa, collected in Jamaica. The smallest cotyledon seen is shown in fig. 10, a, and, though the outline is entire, * Lc., fig. 128, p. 147 ; also Buit. Ann., 2nd series, vol. vii, pl. viil. 678 PROFESSOR F. 0. BOWER ON the vein dichotomises. In (b) there are three veins ; evidently the right shank has forked again. From this condition (c, d) illustrate how, by repeated right and left dichotomies, a sympodial venation may he built up, with a prominent midrib. This corresponds to what has been shown in CampBELL’s drawings for Angiopteris, Marattia, and Danea. It is an easy transition from fig.10, d, to fig. 11, which shows a young leaf of Angioptervs with incipient pinnation at the base. The venation of its upper region is a simple elaboration of that in fig. 10,d. It shows on the right hand a first basal pinna, and a second on the left is imperfectly indicated. The basal lateral vein, essentially dichotomous, has developed sympodially, with a marked midrib. The next vein on the left, though it supplies an evident lobe, or imperfectly formed pinna, has a venation that is clearly dichotomous, the sympodial development being hardly established. Upon these simple materials the scheme of construction of the leaf in the genera Danea, Marattia, and Angiopteris is wholly based. In all it is essentially dichotomous, but with a scorpioid sympodial character asserted early. In Kaulfussia the fundamental nature of the venation appears to be the same. But it is obscured even in the juvenile leaves by reticulate anastomosis, which may be held on ground of general comparison to be a character relatively recently acquired. This is borne out by the first juvenile leaves,* in which anastomosis is superposed upon a venation similar in its main branchings to that seen in Danea and Angiopteris. But the distal ends of the veins in Kaulfussia are connected by curved loops. Certain features of the adult leaf of Angiopteris were described many years ago.t It was found that the pinnz appear monopodially, in acropetal succession, and with regular alternation on the two sides of the leaf. Their number is small, and the “»hyllopodium” terminates abruptly in a blunt cone,{ which may sometimes still be recognised in the mature leaf. An appearance is thus produced as though the leaf were sympodial. Sometimes the last pimna takes an apparently terminal position, but examination of the development shows that this is not its real position. These features are important for comparison on the one hand with other Ferns, and on the other with the Higher Plants. It would appear that in Angiopteris, after the few monopodial pinne have originated, the apical growth usual in Ferns ceases, as it does in Cycads and Angiosperms. Nevertheless their arrangement accords with the scorpioid sympodium usual in the earlier-formed pinne of other Ferns. It was further concluded that the so-called “stipules” are basal growths origin- ating below the pinne, and not of pinna-nature. They are formed from those marginal wings which may be more or less clearly traced throughout the length of the leaves of Ferns. They appear to be of the same nature as the basal growths in the Osmundacez, but more massive. In Angiopteris there is a commissure * CAMPBELL, Dut. Ann., t. xiii, pl. viii, fig. 36. + Bower, Phil. Trans., part i1, 1884, p. 479. { L.c., figs. 16, 17. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 679 across the frontal face of the leaf, asin Todea. Thus there is nothing in the leaf of Angiopteris, as an example of the Marattiaceze, incompatible with what is seen elsewhere in Ferns; but certain features point a comparison with Seed-Plants. Ophioglossacee. The cotyledon of Helminthostachys has been depicted by Lane.* The ternate form of the leaves is shown, but no special attention was paid to their venation. Dr Lane has kindly supplied to me three juvenile leaves (either cotyledons or second leaves) from his Eastern collections, from one of which fig. 12 was drawn. The deeply ternate lamina has a venation which is evidently a scorpioid sympodial ~dichotomy. The junction of the pinna-supply at the base of the left-hand pinna is the lower, and the right-hand the upper; and it will be noted that the insertion of yy Dp Fic. 12.—A juvenile leaf of Helminthostachys, probably an actual cotyledon. From the collection of Dr Lane. (x 4.) the main shanks in the distal lobe alternate, the lowest being on the left, thus alternating with the right-hand pinna. In fact, the whole structure is disposed on the same plan as in fig. 1, g, of Osmunda. The chief differences are that the sinus between each pinna and the terminal lobe in Helminthostachys is deeper, and that the basal pinna in fig. 1, g, of Osmunda is to the right; this character is, however, inconsistent in Osmunda. We have seen that in Osmunda the stage of fig. 1, g, is only arrived at relatively late, being preceded by smaller leaves with regular dichotomy. Thus Helminthostachys steps direct into a position that is only secondary and derivative in Osmunda. The latter may then serve to interpret the real nature of the venation in the former, and the comparison justifies its recognition as a scorpioid sympodium, although the preliminary stage of equal dichotomy is not actually shown in its first leaves. The mature leaf of Helminthostachys has a venation with a strongly marked midrib in each segment, from which veins arise in pinnate fashion, forking once, or * Ann. of Bot., vol. xvi, pl. ii, fig. 61 ; and by CampBeLt, Husp Ferns, figs. 10, 45, 78. 680 PROFESSOR F. 0. BOWER ON sometimes twice, on their course outwards to the margin. No anastomoses are seen, Thus the venation of the segments of the adult leaf is merely an extension of the type shown in the juvenile leaf. But the adult blade as a whole has a rather unusual branching, which is not, however, constant in its details; but a central type is well shown in Frrcn’s drawing in Hooxer’s Second Century of Ferns, pl. xciv. It is, as a matter of fact, a relatively simple development from the type of the ternate coty- ledon. The place of the middle lobe is taken by a scorpioid sympodial series of lobes, equally developed on each side, so that they are arranged in apparent pairs, after the manner of the first pair. But each of the two lateral lobes of the cotyledon is here replaced by a sympodium of the helicoid type. It branches repeatedly, the ~ lower (katadromic) shank developing as a lobe, while the upper (anadromic) shank forks again, and so on. This development is already foreshadowed in the cotyledon. It is a type of branching known as an anadromic helicoid scheme. Fic. 18.—Cotyledon of Botrychium virginianum. (x 4.) A circumstance which may in some degree explain the advanced structure of the cotyledon is that the young plant springs from a bulky saprophytic prothallus. Botrychium also has a large prothallus, and the cotyledon in B. virginianum is a relatively large one. It has already been drawn by JEFFREY * and by CaMmpBELL.t The drawing of a cotyledon shown in fig. 13 is from a specimen collected in Jamaica. It is ternate like that of Helmunthostachys, but more deeply cut into secondary lobes, so that a single vein may supply each one of these. The plan of the venation is the same as in Helminthostachys, though simpler. It also is a dichotomy developed as a scorpioid sympodium. From the cotyledon of B. wirginianum to the adult leaf the development involves only continued progress along the lines thus initiated, the adult leaf being itself ternate. Fig. 14 shows its apex, but other species, such as B. ternatum, have a more robust habit, with thicker leaves, less finely cut. Fig. 15 shows a normal lobe of such a leaf, together with two imperfectly separated segments at its base. A comparison of this with fig. 11 of Angvopteris, and fig. 2 of Osmunda, shows a very striking similarity both as regards venation and the relation of the lobes. Notwithstanding the differences noted, the venation is the same as in * “Gametophyte of Botrychiwm virginianum,” Trans. Canadian Inst., 1896-1897, vol. v, pt. 2, pl. i, figs. 9, 10. + Husp. Ferns, fig. 7. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 681 B. virginianum, though more profusely branched, while the higher lobation of the latter is indicated in the segments of the former only by sinuosities of the margin. But B. ternatwm is not an extreme case. Fig. 16 shows the apical region of a leaf of B. Jenmani, Und, where the branchings in each lobe are more numerous, while the webbing is so complete that only a slight sinuosity of the margin is seen. Its condition approaches that well known in B. Lunaria, in which, though laciniate forms have been described,* the outline of the lobes in normal forms is virtually smooth. This is shown in fig. 17, in the case of one of the youngest plants observed. \ d | [\ Fie. 14,—Apex of adult leaf of Botrychiwm Fic. 15.—Apical lobe of an adult leaf of virgirianum. (x 2.) Botrychiwm ternatum. (x 4.) The venation is like that of B. Jenman, and it may be compared with that in Osmunda regalis, while the laciniate state of B. virginianum resembles Todea superba. lt has been suggested that the Osmunda type is the result of “ webbing” of a laciniate type such as 7. superba, and the same would apply to the forms of Botrychium. The co-existence of webbed and laciniate forms of the same genus or species, or even on the same plant, is a common feature in Ferns. It is seen very prominently in the species of Schizewa; S. dichotoma having narrow segments and S. elegans being webbed in varying degree, but with similar branching of the veins. Comparison may also be made of Aspleniwm dimorphum, quoted by GorBEL.t The cotyledons of B. ternatum and Jenmani have not been seen. But Mr Harotp Lyon sent me some years ago a photograph of young plants of B. (Scep- * Rab. krypt. Flora, iii, p. 558, fig. 176. + Organography, Engl. ed., ii, p. 478, fig. 316. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21). 100 682 PROFESSOR F. O. BOWER ON tridium) obliquum, in which the cotyledon is ternate, closely resembling that of Helnuanthostachys. The leaves of Ophioglossum show reticulated venation.* This feature may be Fic. 16.—Distal end of a leaf of Botrychiwm Jenmant. (x 4.) held as a strong indication of the advance of the genus along relatively modern lines. The juvenile leaves have not been examined in detail; but from CAMPBELL’S drawings of them in O. moluccanum and allied species f it would appear that the venation at the base of the blade is of the type seen in Helminthostachys, but 28) WZ Fic. 17.—Relatively juvenile leaf of Botrychium Lunaria. (x 4.) reticulately developed in the distal region. Thus the fusions of veins appear early in the ontogeny, as has been seen to be the case also in Kaulfussia, and others where the venation of the adult is reticulate. * See PRANTL, Bettrdge zur Systematek der Ophioglossaceen, 1884, Taf. 7, 8. 1 Eusp. Ferns, p. 12, fig. 2. - LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 683 Gleicheniaceer. In this relatively primitive family the leaves are profusely branched, and have the well-known apical growth, interrupted as a rule by periods of inactivity. The morphology of the leaf has been misunderstood in some quarters, though correctly interpreted by Gorsern and others.* But a necessary step to its complete elucidation would seem to be the examination of its juvenile leaves. This has been done by Fic. 18, a-e.—Juvenile leaves of Gleichenia (Dicranopteris) fulua. (a, c,d, x 5; b, x 4; e, x 10; e’, x 30.) CAMPBELL, but without fully taking up the poimts required.t Observations were therefore made on young plants of G. (Dicranopteris) fulva, collected in Jamaica. In some of the young plants the juvenile leaves develop as simple pinnate leaves (fig. 18, a). In the lowest pinnze the vein may show a simple dichotomy, but in those that are higher and larger the forking may be repeated with sympodial development. Passing up to the apex, as the pinna diminishes the vein may be unbranched, and the terminal lobe ends in a simple sympodial venation. The whole juvenile leaf is constructed on a dichotomous system, as an advanced scorpioid sympodium. The * Organography, Engl. ed., ii, p. 318. + Ann. Jard. Bot. Buit., 2 série, vol. viii, p. 93, pls. ix, x. 684 PROFESSOR F. O. BOWER ON same is the case for the older leaves, and it is illustrated in the successively larger pinne which they bear. Thus in fig. 18, b, the apical region of two older leaves is shown, with their distal scorpioid sympodium. The branches entering the upper- most pinne show a single forking; but passing downwards clear steps are seen leading to a scorpioid sympodium in each, with a marked midrib in the lowest, and even a secondary lobation. Thus is initiated a leaf-construction which may be pursued ad infinitum. Though relatively simple, it is to be interpreted on the same lines as in the previous cases. It may be noted that in the earliest pmne there is some inconstancy in the venation, which is sometimes anadromic (pinnee marked ( ) in fig. 18, b, c), but much more frequently it is katadromic, and the latter seems to be the constant condition to which it settles down in the mature plant: it is clearly indicated in the lower pinnee of fig. 18, b. These simple examples of instability, in a character which has often been used as distinctive, throw light on what is the real nature of it. Where the vein of a pinna has forked, either of the two shanks may take the further lead. Such simply forked pinnz are seen in all the plants here represented. If the acro- scopic shank takes the lead and forks again, the other appears as an appendage, and the venation would be termed “ katadromic.” If the basiscopic shank takes the lead and forks again, the acroscopic shank appears as an appendage, and the venation is “anadromic.” Such instability as Gleichenia shows may be considered intelligible in a Fern which in its early stages has not advanced far from the primitive equal dichotomy. The young plants of G. fulva afford an insight into the origin of the well-known “false dichotomy” of the leaves in the genus. Three young plants have been drawn (fig. 18, c, d, e), and in each of them the foot is still recognisable. The leaf opposite to it is the cotyledon (cot.), and in fig. 18, e, the insertion of its vascular strand proves that it is so. In (¢) the cotyledon is small and its apex fully developed, as is also that of the second leaf; but in the third, though it has developed four pinne, it has its apex still curled up, and, as no pinne of reasonable size are seen upon it, it may be held to be im a state of arrest. In fig. 18, d, a plant is shown in which the second leaf is in a corresponding state. In fig. 18, e, it is the cotyledon itself that is arrested, while the next leaf has its apex fully formed. This cotyledon is shown on a larger scale in fig. 18, e’, and it appears already to illustrate the fundamental features of the Gleicheniaceous leaf. Right and left are pinne fully formed, though on a quite rudimentary plan, while between them is the arrested tip, representing the whole upper region of the sympodium. Dichotomy is in actual fact present, and it is shown with equal shanks in each of the lobes. Comparison with other Ferns shows that dichotomy underlies the whole sympodium of this leaf. But the apparent or “false dichotomy” of the Glecchena leaf is not a simple dichotomy at all; it is a result of arrested development in a leaf constructed throughout as a scorpioid sympodium. ‘These observations accord with LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 685 those of CampBELL on the cotyledons of G. dichotoma.* But there the arrested apex is shown expanded, not still circinate as in the case of G. fulwa. The young plants thus provide all the morphological essentials for the formation of the most elaborate leaves of the genus, viz. a scorpioid sympodial branching, and a possible arrest of the apex, which may be periodical above any of the apparently opposite, paired pinne. Unlimited development along these lines, but with varying proportions, will account for the various types of leaf seen in the genus. Matonineer. The very striking leaf-architecture of Matonia has been repeatedly discussed, for instance by Dretst+ and by Tanstey.{ The latter collected young plants on the Malay ces Fic. 19, a-d.—Juvenile leaves of Matonia pectinata. a, though detached and broken, was the cotyledon of the plant, b. The drawings were made from specimens collected by Dr Lane on the Malay Peninsula, and kindly lent for the purpose. (x 4.) Peninsula, and depicted them. Though his observations help materially to a solution, they do not provide an exhaustive account. Dr Lane, who travelled with Mr Tawnstey, has kindly lent me his preparations of young plants of Matonia, and a more consecutive series illustrating the structure of the juvenile leaves can now be given. But it appears that the young plants and prothalli are only rarely met with in their native habitat. . The prothallus was still attached to the youngest specimens. The simplest form of juvenile leaf is that shown in fig. 19, a. It was originally attached to the plant (b), but broke away in the preparation, clearing, and staining, and it had already been broken across the middle. There is, however, no doubt that it was the coty- ledon, and that it was unbranched. The venation is a scorpioid sympodium, with * Tic., pl. x, figs. 55-57. + Natiirl. Pflanzenfam., i, 4, p. 343. t Ann. of Bot., 1905, p. 447. 686 PROFESSOR F. O. BOWER ON the lateral branches sometimes forked, but usually simple ; in construction it corre- sponds to that of Gleichenia (fig. 18, a), and the resemblance of the venation would be very close if the lobation of the Glevchenca leaves were undeveloped and the leaf entire. The next older leaf of the same plant showed equal dichotomy (b), each lobe having scorpioid sympodial venation. In fig. 19, c, the leaf appears to be the coty- ledon, but its branching is more advanced, the left-hand lobe having dichotomised again. The process has gone further in fig. 18, d, but the inequality in the two primary lobes is greater, though the number of lobes is the same in each. More- over, the branching is clearly that of an anadromic helicoid sympodium. Com- parison may, however, be made with the young leaf drawn by Tanstey.* It showed five lobes, one a “middle lobe,” while each of the laterals branched again. Assuming these branchings to be dichotomies, this frond “may be regarded as a sort of prototype of the adult frond.” Considering the facts that TansLEy shows, and the comparison that may be made with other Ferns, that assumption appears justified. . But if so, this is a case of a katadromic helicoid sympodiwm. We conclude that in the young leaves of Matonia pectinata the branching is dichotomous; but that it may be sympodially developed, and fluctuates between the anadromic and kata- dromic types. This is a particularly interesting conclusion in view of the archi- tecture of the adult leaves in Matonia and Dipteris. The latter will be shown below to be constructed on the anadromic scheme (Plate-fig. B); but the adult leaf of M. pectinata is a katadromic sympodium (Plate-fig. A). The transition from the state seen in TANSLEY’s fig. 2 to the adult leaf of Matonia only presumes a continua- tion of the katadromic sympodium thus initiated. The consequence of this method of — construction would be that the pinne-—that is, the weaker developed shanks of the successive dichotomies—would appear constantly on the acroscopic side (katadromic scorpioid sympodium) (Plate-fig. A). Comparing this with Helminthostachys, the difference is that there the weaker shanks are thrown off on the basiscopic side (anadromic scorpioid sympodium), while the branchings are fewer and less regular than in Matonia. ° The apparently divergent leaf of Matonia sarmentosa is built up of segments with a venation and form like those of the juvenile leaves of M. pectinata. Its branching was analysed by Diets.t He concluded that it is essentially on the same system as M. pectinata, but that the advance or suppression of the shanks of the dichotomies occur alternately on the inner (acroscopic) and the outer (basiscopie) limbs: “that is, they lead to a bostrychoid instead of a cicinnoid structure.” The matter has been reconsidered by Compron.{ He finds the frond far less stereotyped than in M. pectinata. “The rachis forks repeatedly, and one of two things may happen: (1) both branches may develop more or less equally, forming elongated axes, which fork again and again; or (2) one of the branches may develop fully, while the other, after bearing a very few (usually two) pinnz, ends in an aborted © Lc. fig...2; | Natirl. Pfhlanzenfam., i, 4, p. 343, fig. 180, B. t New Phyt., viii, p. 299. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 687 bud. The latter is the more usual process... . It seems probable that the appearance of pinnation in the frequent pairing of the pinne at the base of the rachis-segment is a secondary modification.” This analysis was based upon pre- served material, and supported by anatomical observations. It is probably correct. Moreover, it receives support from the facts now demonstrated for the juvenile leaves of M. pectinata. Itis in fact just the sort of instability of the sympodial development seen in them which, if continued in the larger adult leaf, would lead to something not unlike the leaf-construction of M. sarmentosa. In neither species of Matonia does the leaf-architecture coincide with that of Gleichenia, but both originate from a like sympodial source. But the modi- fications are different in them all. M. sarmentosa appears to be more like Gleichenia than M. pectinata. For in both of them arrest of branches is a feature. In M. sarmentosa short shanks of dichotomy bearing usually two limited pinnee and an aborted bud are produced, though not with regularity, on an approximately equal dichotomising system. In Gleichenia similar shanks, which may or may not lose their apical growth, and a bud, are borne on a sympodially developed scorpioid system. The juvenile leaves of Gleschenia provide the simplest type of this leaf- construction. It is thus possible to interpret all of them as variants of that sympodial dichotomy to which the juvenile leaves of the primitive Filicales so clearly point. Dipteridine. The adult leaves of most species of Dipteris are notable for their regular dichotomy, of which the shanks remain separate laterally in D. Lobbiana and quinquefurcata, but are webbed into two equal coherent lobes in D. conjugata (Plate-fig. B). Notwithstanding this simplicity of branching, there is an advanced reticulation of the veins of the type “ venatio anawetr.” But no one can doubt that the venation of the adult leaf is based on primitive dichotomy, notwithstanding the smali-meshed reticulation characteristic of relatively recent forms. In Matonia the ultimate venation is forked, and usually free, with few and irregular fusions, and it thus stands on a more primitive footing than Dzpteris. The reticulation appears even in the juvenile leaves of Dupteris, but it does not effectively obscure the dichotomy. An early leaf, perhaps the cotyledon, of D. Lobbiana is shown in fig. 20, a, which is simple in form and shows no completed reticulations. The single median vein and four lateral branches which it shows would be readily referable to a scorpioid sympodium. But there is no evidence, beyond comparison with Matonia and Gleichenia, that this is actually its character. Fig. 20, b, c, shows juvenile leaves which are bifureate both in form and venation. Accessory veins already forecast the mature reticulation. Later leaves fork again (fig. 20, d), and so the plant advances towards the mature state.* Similarly with D. conjugata, notwithstanding the extensive webbing and profuse reticulation of the adult leaf, the juvenile leaf shows * Compare SewakD, Phil. Trans., B, vol. 194, pl. xlviii. 688 PROFESSOR F. O. BOWER ON dichotomy of the main veins and fine reticulation from the first (fig. 21, a, bea The webbing is also pronounced, but variable in the older stages leading to the adult leaf.* Fic. 20, a-d.—Juvenile leaves of Dipteris Lobbiana. a, b, from specimens belonging to Dr LAnc ; ¢, d, after drawings by Miss bE BRuyNn. (a,b,c, x 2; d, x 2.) a b Fic. 21, a—-1.—Juvenile leaves of Dipteris conjugata. (x 6.) In the adult leaf of D. conjugata the dichotomy shows a distinctly sympodial tendency. The inner anadromic branches are favoured, as compared with the outer. This is apparent from Plate-fig. B. The same method, without the webbing, under- — lies the construction of the leaf of the probably related fossil, Dictyophyllum exile * Sewarp, l.c., pl. xlviii. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 689 (Plate-fig. C). The same again, with the added point ‘of a spiral curvature of the shanks, is seen in Camptopteris spiralis (Plate-fig. D). All of these are examples of sympodial dichotomies, but showing in increasing degree the consequences of the anadromic helicoid development. . Biologically these forms are interesting as showing how from a flattened organ the advantages of a radial symmetry may be acquired. The modern forms are rhizomatous, with long-stalked upright leaves. Probably the fossils were the same. The problem then would be how best to expose the assimilating surface. In Matonia the helicoid sympodium is expanded almost into the form of a widely open funnel, the effect of which is lost in pressed specimens.* This is still more the effect of the & b Fic, 22.—a, juvenile leaf of Chetroplewria ; b, a similar leaf of Platyceriwm Veitchit. (x 3.) webbed expanse of Dipteris comjugata.t But in Camptopteris, by the adoption of the spiral curvature of the sympodia, the pinne radiate outwards like bristles of a spiral wire bottle-brush. A good substitute for a radial vegetative shoot is the result. On the other hand, it is only a step in simplification from the condition of the leaf of D. conjugata to that in the allied Chetroplewria.t There the adult blade may be entire, or variously lobed. But its base shows the dichotomy very plainly, while it appears in the juvenile leaf in almost exactly the same way as in Dipteris (fig. 22, a). It may further be noted that Platyceriwm, in which the adult fertile leaves show often very equal dichotomy, has in its juvenile leaves a venation closely similar to that of Cheiropleuria (fig. 22, b). * Compare Sewarp, Fossil Plants, vol. ii, p. 293, fig. 228. + See Curis, Geographie der Farne, fig. 5. f Ann. of Bot., 1915, p. 495. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21), 101 690 PROFESSOR F. O. BOWER ON Such examples as those quoted serve to illustrate steps of transition between open venation and reticulation. The fossil record indicates that the former is relatively primitive. The structure of the juvenile leaves accords with this, the dichotomous venation being clearly the foundation, with the reticulation superposed upon it. Hymenophyllacee. This family of Ferns is now generally accepted as a relatively ancient one, but highly specialised in relation to a moist habit ; the filmy structure of its leaves is no longer regarded as an indication of affinity to the Bryophyta, as was formerly believed. Less importance may accordingly be attached now to the details of its leaf-architecture than at the time when Prantrui wrote his Monograph. Nevertheless, their importance should not be unduly discounted, as indicating primitive characters of general application, and anatomical comparison with the Zygopteridez strengthens this position. The branching and venation of their leaves has been studied very closely, and PRANTL in particular has analysed them in relation to dichotomous branching.* He referred all the diverse forms of leaf seen in the family to the development of a dichotomous system. In certain cases the leaf is simple in form, with entire margin, and no apparent branching of the blade. This is traversed by veins, and the general construction is like the “webbed” types in other Ferns. Where the leaf-blade is circular, the whole branching system of veins may show very perfect and equal dichotomy, as in Trichomanes reniforme, and both shanks of each forking develop with an almost diagrammatic regularity (fig. 23). But in certain other entire leaves that equality is not maintained, as in Trichomanes punctatum (fig. 24, a); and steps may be found leading by easy stages to a high inequality of the shanks, with the continual assertion of a sympodial false axis (fig. 24, b,c). But if such steps may be followed in the webbed or unbranched types, much more plainly is this to be seen in those which are profusely branched. The analysis of many forms from this point of view was carried out by Prantz.{ In view of his observations it will be unnecessary to enter into further detail here as to the relation of the branching to the cellular construction. He concluded{ that all complicated leaf-forms, such as he described for the Hymenophyllaceze, were referable to a simple type of dichotomous branching, and he suggested that a similar method might be applicable also to other families of Ferns. He applied it himself to the Schizzeacez, and we now see that its application extends to the juvenile leaves of other primitive types of Ferns. PRANTL was disposed to regard the fully webbed types, such as 7. renaforme, as relatively primitive. This is a position open to doubt, and especially so in view of the facts relating to many other Ferns. It would seem more probable, and it certainly accords equally with the facts, that the entire leaves are a result of “ webbing” in an * Hymenophyllaceen, Leipzig, 1875. ain Bete Hh allie Ie Dems) Ie LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 691 originally laciniate branch-system. Moreover, fig. 24, b, shows in the marginal sinus which separates certain of the dichotomising vein-systems what: may be held as a + Fic. 23.—Adult leaf of Trichomanes reniforme, showing details of venation, and especially the vascular connections before entering the petiole. + marks the limit between the right and left halves of the dichotomy. Drawn by Mr J. M. Toompson. (Natural size.) f Fic. 24, a-c.—Adult leaves of Zrichomanes punctatum. (x 10.) last relic of the merged leaf-segments. The evidence from the juvenile leaves would be valuable in this relation. Unfortunately, observations on juvenile leaves in this family are not extensive. But they show what would be theoretically expected, if the above conclusion were true. Thus the cotyledon of T. rigidum., Sw., as shown by GorBEL’s drawing quoted in Natiirl. Pflanzenfam., i, 4, p. 16, fig. 11, is single- 692 PROFESSOR F. O. BOWER ON nerved, and of simple form. Given dichotomy of the apex, the whole complexity of Hymenophyllaceous leaves can be derived from such a source, with “ webbing” thrown in secondarily, as an occasional but not a general factor. But the facts are not yet in hand upon which a demonstration of the point can be given for the Hymenophyllacez. NOTES ON SOME OTHER LEPTOSPORANGIATE FERNS. Representatives of all those families of living Ferns which are held to be relatively primitive have now been examined as to their juvenile leaves, and the relation which their structure shows to the adult state. It would be superfluous to pursue this inquiry exhaustively onwards through the enormous number of Lepto- sporangiate Ferns which are held to be later and derivative. Moreover, the juvenile leaves of these, as well as the adult, have been the subject of many observations. Thus Sumrinsxy shows for the cotyledons of Pteris serrulata a branching readily referable to dichotomy with a sympodial tendency.* This is also seen in the familiar figure of Adiantum capillus-veneris,t where the dichotomy seems quite perfect. Two other examples, selected from Ferns where the mature leaf has still an open venation, will be shown here in detail, the one (Cyathea) from the superficial, the other (Pteridium) from the marginal series. The first leaf in Cyathea insignis shows, as a rule, an obvious dichotomy (fig. 25, a), though the two shanks may not be exactly equal (b). It will be noticed on comparing these that while in (a) the distance along each shank from the first to the second dichotomy is approximately equal, in (b) it is longer in the weaker shank than in the stronger. This difference is still more marked in (c), where also the inequality of the branch-systems which spring from the two is greater. This leads to the condition seen in (d), where the distance between the first and second dichotomies is so short that the limb of the first appears almost exactly opposite to the left-hand limb of the second, giving the effect as of two almost equal pinne ; while the right limb of the second dichotomy appears as a terminal lobe. ‘This corresponds exactly to what is seen in Helmanthostachys and others. The same development is repeated in older leaves in the distal lobe itself (e), thus leading to the elaborate state of the pinnate adult leaf of Cyathea, which is thus referable to a dichotomous source. Comparison with the juvenile leaves of Pteridiwm aquilinum leads to similar results, though the two are widely apart in the system. The cotyledons are some- times very equally dichotomous in their branching (fig. 26, a, b); in others slight inequalities appear (c, d, e), with shortening of the distance between the first and second dichotomies on the stronger shank. This is more pronounced in cases where the inequality of the lobes is greater (f, g), giving the usual appearance of a ternate leaf. If development of the apparently distal lobe—really one half of the second * Entw. d. Farnkraiiter, Berlin, 1848. } Sacus’ Textbook, fig. 258. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 693 ra b Cc e ad Fic. 25, a-e.—Juvenile leaves of Cyathea insignis, illustrating their progressive development. (x 4.) Fic. 26, a-7.—A series of juvenile leaves, many of them actually cotyledons, of Péeridiwm aquilinum. They illustrate steps of progression from equal dichotomy to the scorpioid sympodium. The differences in size are largely due to greater or less advance of the individual development. (x 5.) 694 PROFESSOR F. 0. BOWER ON dichotomy—be repeated on the same plan (compare f, g), the result is reached as in (i) or (7). From these to the adult leaf of Pteridiwm the course is plain enough. It is interesting to note the close similarity of the juvenile leaves of Dennstedtia punctilobula, as shown by Conarp.* In this description of juvenile leaves of Ferns the types with reticulate venation have been purposely omitted, as they are with reason believed to be derivative forms in respect of that venation. But they must not be wholly neglected. Already the juvenile leaves of Dipteris have been described, and there the venation was found to be fundamentally and equally dichotomous (D. conjugata), or a sympodial dichotomy (D. Lobbiana), with reticulate loops superposed upon that branching. etal b) eae : a e Fic. 26 bis.—Juvenile leaves of Ceratopteris thalictroides, (x 4.) Another example has been examined in Ceratopteris, in which the seedling leaves had already been figured by Kny.{ The youngest leaves have a simple, unbranched, median strand (fig. 26 bis, a); slightly more advanced leaves may show dichotomy, with free branches (b), but later leaves may have the distal ends connected by a loop (c), while free-ending branchlets radiate outwards from it (c,d). From this state further advances may be made connecting these branchlets into further loops, and also by commissures between the main veins (e, f). But still the fundamental dichotomy may be recognised at the base of the blade, even in relatively advanced leaves, showing that the reticulation is a superstructure upon it. It seems unnecessary to repeat in additional examples progressions from the juvenile to the adult leaf which follow substantially similar lines. The above will suffice to illustrate, first for the Leptosporangiate Ferns with open venation, and also for those showing reticulation, how an apparently pinnate leaf may originate from a dichotomous source. A few notes are, however, appended here upon certain out- * Carnegie Inst., No. 94, pls. xxiv, xxv. + Parkeriaceen, pl. xxii, figs. 4-8, LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 699 standing types of leaf-formation. Some ferns retain a very exactly equal dichotomy of the primary venation even in the adult leaf. This usually goes along with a more or less orbicular blade, entire or more or less laciniate, borne distally on a relatively long petiole. Many genera with as a rule highly divided leaves include species with such a character. This raises the question of their phyletic relation to the more normal laciniate and pinnate representatives. Such characters are seen in Actiniopteris radiata (fig. 27), a Fern of rather uncertain affinity. It is also seen in the sterile leaf of Elaphoglossum (Rhipidopteris) peltatum, which is also an isolated type (fig. 28). In both cases the venation is that characteristic of RN) - Fie. 27+—Basal part of the lamina of an adult leaf of Actiniopteris radiata. (x 4.) juvenile rather than of adult leaves. They are deeply laciniate, with a single vein in each segment. They may be held to have retained and elaborated a relatively primitive construction. A somewhat similar condition is seen in those species with entire reniform leaves, which appear in several distinct genera, such as Adiantum, Gymnogramme, and Scolopendrium, their outline and venation being essentially the same as in Tricho- manes reniforme. A succession of the juvenile leaves of Adiantum reniforme, var. asarifolium, is seen in fig. 29. The first of these shows only a slight advance upon a simple dichotomy of the veins, with an entire blade (a). The further series (be) explains itself, while the similarity of the last to the drawing (fig. 24, a) of Tricho- manes punctatum is remarkable. But these Ferns are all widely apart in the system. They are probably results of independent development, possibly in response to similar requirements. They are perhaps best explained as extreme results of a 696 PROFESSOR F. O. BOWER ON webbed condensation of a profusely dichotomous system, while the dichotomy may -be held to be that of the juvenile leaf retained in the adult. They resemble the previous examples, but with the difference that all the veins are included in an entire webbed blade. Fic, 28.—Basal part of the lamina of an adult sterile leaf of Elaphoglossum (Rhipidopteris) peltatwin. (x 4.) a b e vy Ve Fic, 29, a-e.—A progressive series of juvenile leaves of Adiantum reniforme, var. asarifolium. (x 4.) A particularly good instance as bearing upon the probable secondary origin of the “ webbed” condition in these reniform leaves is seen in Scolopendrium Delavayt, Franch., a species collected by Henry in Yunnan (fig. 30). The petiole is long and the lamina reniform, with an entire margin. The venation is a sympodial dichotomy, leading to a ternate structure, as indicated by brackets in the drawing ; but there is no marginal involution. The distal group is a scorpioid sympodium; the lateral ones are katadromic helicoid, after the manner of Matonia. The double sori are LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PIERIDOPHYTA. 697 situated on pairs of the parallel veins, as typical of Scolopendriwm. The interest of this species is first in comparison with S. vulgare; the blade may here be held to be a shortened and condensed form of that there seen. But the structure in S. vulgare has recently been elucidated by comparison of Blechnum punctulatum.* This very variable leaf has been shown to be a derivative from the typical Blechnum sporophyll; the latter is a probable derivative from a Cyatheoid source, such as Matteuccia, and ultimately with probability from a Gleicheniaceous affinity. But Matteuccia has an ordinary pinnate leaf. Unless, then, all the comparisons given in detail in my memoir on Blechnum are at fault, the conclusion will be that the reniform blade of S. Delavayi is the result of a condensation of the leaf, with A Fic. 30,—Basal part of the lamina of an adult leaf of Scolopendriwm Delavayi, Fry. (x 4.) < simplifications of its outline, starting from a pinnate—that is, a scorpioid—type. This case is important in its bearing upon “ webbing,” and indicates that here an entire reniform leaf is probably derivative from a more elaborate and laciniate type of leaf. Another peculiarity which is striking, but by no means restricted to any given systematic group, is the unilateral branching of pinne, by helicoid development, as in Matona and others. It is seen in Pteris semipinnata in a very prominent form (Plate-fig. E). While Pteris quadriaurita and others show an equal lobation on either side of the pinna, or pinnule, in P. semipinnata the branching appears on the katadromic side only (fig. 31). This may be interpreted as the unequal development of successive dichotomies, so that the anadromic shank is constantly the stronger. A similar condition has been described for Stenochlena by Curist.{ On his plate iii he represents it for a species from Borneo, and on plate iv for one from ‘“ Deutsch- Neu-Guinea.” Thus it is not a local peculiarity. This anadromic helicoid branching is also represented in the fossil Odontopteris minor, Brongn., as shown by ZEILLER * Ann. of Bot., vol. xxviii, pp. 405-411. + Verhandl. d. Schw. Naturf. Ges., 89, p. 178. TRANS. ROY. SOC. EDIN., VOL. LI, PART II1 (NO. 21). 102 698 PROFESSOR F. O. BOWER ON (Plate-fig. F), while it may there be combined with an equal dichotomy of the whole blade. Lastly, this peculiarity of sporadic equal dichotomy must be considered. We see it in fossils. It is frequent in living Ferns. It appears sometimes at the distal ends of the pinnules ; sometimes it involves whole pinne, or the whole distal region of the blade, or even the lower part of the leaf, so that in extreme cases the appearance is as of two whole leaves borne on a single stalk. In one degree or another it is a common abnormality, and provides the “forma furcata” of Fern collectors. It has already been described above in Osmunda (fig. 4), and it is found in Polypodium, Scolopendrium, Nephrodium, and Athyrium, while it appears also frequently in Nephroleprs. An unexpected but very striking example is seen in Platyzoma, in a furcate form sent by Baron F. von Mituer to the Kew Herbarium. It has also been examined by Mr J. M. THompson,* in specimens supplied from the Sydney \ Fic, 31.—Part of a pinna of Pteris semipinnata, showing the relation of two of the pinnules to a pinna, and to its venation. (x 38.) NN ~ S th q { Herbarium. It may be there seen to affect the distal end of the leaf only, or to extend lower down to the petiole. It is shown also in the fossils, as in the Pteridosperm Odontopteris (Plate-fig. F). Definite furcate races have been estab- lished, where this character became permanent, as in Nephrolepis. Such a race of Nephrodium molle was cultivated some years ago in the Glasgow Botanic Garden. The petiole in such forked leaves showed a very perfect division of the vascular supply into equal parts. Such forkings .may be interpreted as the result of a reversion, in the ¢ase of individual branchings of a sympodium, to the primitive equal dichotomy. Their existence in so many genera and species may be held as an indication that, even where the sympodial development is strongly marked, the branching is still essentially dichotomous, and capable of quite equal development of its shanks. The interest in the various leaf-forms analysed in the above pages lies not so much in their mere existence, as in the fact that they are all capable of elucidation in terms of the underlying dichotomy. So that a general statement may be formulated with regard to the construction of the leaf in Ferns, including these peculiarities :—That the architecture of the Fern-leaf is fundamentally dichotomous ; * Trans. Roy. Soc. Edin., 1916. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 699 and that, as far as observation extends, the architecture of all leaves of Ferns may be interpreted in terms of dichotomy, with or without sympodial, and ultimately monopodial, developments. It has been seen above that in other phyla of Pteridophytes the leaf is either unbranched, or that it bifurcates. The only case of difficult interpretation is the three-lobed sporophyll of Cheirostrobus, and even there the distal end of each lobe forks equally. It is possible that it shows either trichotomy or a very condensed pair of dichotomies. Thus it may be stated generally for the Pteridophyta that in them the leaf is either unbranched or dichotomous ; or it shows branching and webbed venation that may be recognised as a sympodial derwative of dichotomy. CoMPARATIVE CONCLUSIONS. In order to help in obtaiming a scientific foundation for views of “leaf- architecture,” the leading facts have been summarised in the above pages as to the branching and venation of the leaf in those Vascular Plants which are held to be primitive. More especially attention has been given to their juvenile leaves, as elucidating the construction of the adult. But in many characters of form, of venation, of webbing, of basal growths, and even of apical arrest, the leaves of Pteridophyta prefigure the leaf-construction of Flowering Plants. We may now consider how far the conclusions stated above can be extended so as to include all leaves of Vascular Plants. And it will be found possible to base upon the data before us, together with other facts already well known, certain general conclusions relating to leaf-architecture at large, which will now be. briefly discussed and formulated. I. The conception of the shoot, and of its constituent parts, axis and leaf, as laid down by Sacus,* is adopted for the purpose of this discussion, but it is the leaf of the sporophyte only that is treated here. The leaf is normally a bifacial member, with the adaxial and abaxial relations indicated structurally and develop- mentally. Its form is asa rule flattened, with convex margins. But even where the form is cylindrical, as in Pilularia, the segmentation clearly indicates the bifacial character. Or, in the case of the leaf-stalks of certain Ferns that appear . cylindrical, the margins are still indicated by lines. Putting aside exceptional ceases of high specialisation, it may be stated that the leaf is a bifacial member with margins defined structurally, though these may be obscured. II. In the simplest cases the leaf has not a marked apical growth: it is un- branched, and may contain only a simple unbranched vein. This is the constant state in the Lycopodiales, fossil and modern. It is also generally seen in the Equisetales, though in Asterocalamites and Pseudobornia dichotomy existed. Dichotomy is, however, the rule in the Sphenophyllales and Psilotaceze, though in the Psilotaceze the vegetative leaves are simple. This is also seen occasionally in * Textbook, 2nd Engl. ed., p. 153 ; and Lectures on the Physiology of Plants, Lecture I. 700 PROFESSOR F. O. BOWER ON the- juvenile leaves of Filicales (Trichomanes, Marsilia) ; it is common in Gymno- sperms, and occurs in some Angiosperms. In all such cases the question arises whether the simple structure is really primitive, or the result of reduction. The fact that a consecutive series of ascending steps from these simple types to the most complex, and that such steps are commonly illustrated in more or less complete series in the individual life, as we pass from the juvenile leaf to the adult, seems to indicate quite clearly that there is no difference in kind, but only in degree, between the simple and the branched leaf of the sporophyte; it also accords with the view that primitive leaves were simple. But it is well to have in view a clear case of simplicity by reduction. It is seen in Platyzoma.* In this Fern THompson has shown that the leaves are dimorphic, the fertile being simply pinnate, the sterile beig apparently simple. But careful examination has shown various stages of abortion of the pinnee in the sterile leaf, leading to cases where the pinne are wholly absent. Thus by abortion a simple leaf may be derived from a pinnate leaf. The simple leaf of a Lycopod or an Hquisetum may actually be primitively simple, or the result of a simplification of a similar nature to that of Platyzoma. But whether the one or the other, the simple leaf may be held to be just as much a “phyllome” as are the more complex leaves of Ferns. We may conclude generally that a leaf may be simple or branched. Branching, whether seen in form or only in venation, 18 not of the essence of foliar nature, but is to be regarded as a means of elaboration of the member, which may be subject BE to simplrfication by suppression of the branches.t III. In the great majority of cases the leaf is buamelen and the relatively primitive types show that the branching is primarily distal. This may affect the outline of the leaf, and lead to its being more or less distinctly lobed; or it may be seen only in the venation, and the limb be entire. The relation of these states will be considered later (V). It has already been seen that the opinion expressed in the writings of HormersTer, Sacus, Pranti, GorBEL, and Poronr is that the fundamental type of branching in Fern-leaves is dichotomous.{ The first result of the observations here described is to establish the fact that from all the primitive types of Filicales the branching in the juvenile leaves is dichotomous. In this respect they accord with what is seen in the adult leaves of the Sphenophyllales and . of certain archaic Equisetales. Thus it follows that i all the prinutwe phyla of Vascular Plants where branching occurs, distal dichotomy is the fundamental mode of that branching. IV. Dichotomy may be apparent only in the venation, or it may characterise also the configuration of the leaf. Within near circles of affinity both states may be represented. A good example is seen on comparison of Todea superba (fig. 6) and * Trompson, Trans. Roy. Soc. Hdin., 1916. + Compare Lianrmr, “ Equisétales et Sphénophyllales,” Bull. Soc. Linn. de Normandie, 5 série, vol. vii, p. 93, 1903. | See paragraph on p. 659 as to the use of this term. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 701 the closely related Osmunda regalis (fig. 1), while Todea barbara (fig. 5) takes a middle position. Other examples may be found among the species of Hymeno- phyllum, of Botrychium, and of Dipteris. In fact, there is hardly any large genus of Ferns in which the gradations of “webbing” of the branched leaf between the laciniate and the webbed are not found. Sphenophyllwm illustrates the same pheno- menon; for the archaic SS. trichomatosum has deeply cut leaves, while C. cunetfoliwm has its leaves entire, with a slightly wavy distal margin, and Trizygia has even the margin entire, and the leaf expanded as a large assimilating organ. Such facts suggest that a progressive “ webbing” of leaves originally laciniate has taken place within the family. The facts for the genus Dipteris* are hardly susceptible of any other interpretation, and there is reason to believe that it occurred extensively among early Vascular Plants. Already Potronm 7} has indicated that in Sphenophyllum the species from the lower horizons are laciniate, while those of the more recent are entire, indicating a progressive webbing. To this view for the Sphenophyllaceze Dr Kinston assents. But, however cogent the evidence of this may be for certain cases, there is no doubt that in the course of descent the reverse has also happened. It is illustrated in the submerged species of Ranunculus and Potamogeton. The general conclusion will then be this: that 2t 2s probable that in leaves with branched venation a progressive webbing has occurred in the course of evolution. But where there 1s an actual state of laciniation seen in any given case, % may erther be really the primitive state preserved, or an adaptive reversion to it.} VY. The dichotomy in juvenile leaves, whether as seen in their laciniate outline or only in their venation, may be equal or unequal. In the latter case the system is commonly developed sympodially, and all possible gradations may be observed from the one to the other. These are well illustrated in the juvenile leaves of Pterrdium (fig. 26) and of Osmunda (fig. 1). It is important to note that the order of the ontogenetic progression is normally from equal dichotomy to sympodial development. Poronté also points to a similar progression in the juvenile leaves of Flowering Plants,§ in which the outline may be dichotomous, though not so in the adult leaves. He quotes Kny and WIncKLER to the same effect. But to take the mere outline into consideration without the venation may be misleading. Apparent dichotomy may be merely a result of deep emargination, and there are indications in the * Land Flora, p. 618, etc. + Lehrbuch, p. 176. { It may be a question whether all cases of complex venation in any leaf-expansion represent the result of webbing of laciniz originally distinct. It is possible that a branching of veins may be initiated de novo in a leaf- expansion. We know that some such new developments have produced the reticulate state. It would therefore be rash to deny that something similar may in some cases account for the origin of additional veins in an entire blade. We may conclude that in primitive types with open venation which are “webbed” the venation may be a near index of a primitive laciniation. But in derivative types, and especially where the venation is reticulate, as ia Angiosperms, this may have so obliterated the original scheme of construction that what is actually seen can no longer be trusted as indicating it further than in quite general features. § Ber. d. d. Bot, Ges., xiii, p. 245. 702 PROFESSOR F. 0. BOWER ON Cruciferee that it is so. The venation in such questions will have to be examined more carefully than hitherto. What emerges from examination of the illustrations in Luspock’s book On Seedlings is, that the prevailing venation in the cotyledons of dicotyledonous plants is of a type similar to that of an entire juvenile leaf of the Ophioglossaceze or Marattiaceze; but this subject will have to be gone over more eritically before comparisons can be applied in detail. The earliest juvenile leaves do not always show an equal dichotomy. ‘The series may be more or less abbreviated, and the construction typical of the primitive forms may be omitted. This is particularly obvious in Botrychiwm (fig. 12) and Helmin- thostachys (fig. 12). The Marattiaceze are interesting in this connection ; for while Danexa (fig. 10) may show an initially equal dichotomy, in Angiopteris* the cotyledon itself has distinctly sympodial development, the primitive equal dichotomy being omitted. This seems to be the prevailing type in the dicotyledons. Thus there is in the juvenile leaves of Vascular Plants an orderly progression of structure more or less fully indicated, but apt to be abbreviated in certain types. That pro- gression may be held to be probably phyletic in relatively primitive types. It proceeds from equal dichotomy to sympodial construction. VI. Where sympodial development of dichotomy is seen it may take various forms. Where the favoured shank is alternately right and left, which is the commonest state, the sympodium is scorproid, as in figs. 11, 14, 15. Where it is repeatedly on the same side the sympodium is helicoid, as in Plate-figs. A-F. But in leaves there are two possible types of this last, in relation to the leaf as a whole. The favoured shanks may be on the side towards the leaf-base, and the result may be designated a katadromic helicoid sympodium, as in Matonia, Plate-fig. A. Or it may be on the side towards the apex, when it is called.an anadromic scorpioid sympodium, as in Pteris sennpinnata, Plate-fig. EK. The various types of sympodial development are variously distributed wm the construction of leaves, and produce very characteristic effects wpon their outline and venation. VII. Where the size of the leaf and the complexity of its branching are consider- able, as in most adult leaves of Leptosporangiate Ferns, the primordium shows continued apical growth. The lower pinne are then asa rule initiated monopodially, budding out laterally below the apex. This adult condition is gradually led up to in the successive leaves of the individual plant. But there is also as a rule a pro- gression in the individual leaf from this adult mode of origin of the lower pinne to a sympodial origin, and finally to equal dichotomy.{ This is in fact a reversion in the later stages of the ontogeny of the leaf to that branching which is characteristic of the juvenile leaves. If the phyletic history be truly reflected in the steps of increasing complexity of construction in the successive leaves from the juvenile to the adult, then the steps in succession of origination would read thus :— (() Equal Dichotomy, modified to produce (vw) the Sympodial Dichotomy; and * CAMPBELL, Husp. Ferns, fig. 122. + See Land Flora, p. 627, and refs. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 703 (iii) where the development of the leaf is strong there may be transition to the Monopodial Branching. VIII. But in the individual Fern-leaf the order of events is developmentally the reverse of this. The first-formed branches (pinne) are monopodial, the later of sympodial origin, and the final branching dichotomous. Thus the distal end of the Fern-leaf may be held to retain its primitive characters. Anatomically it is found that ancestral features are also located in the leaf-trace at the base. It is thus possible to regard the middle region of the leaf as an evolutionary innovation, a view which accords equally with anatomical fact and with the comparative study of leaf-development. It may then be stated as a working hypothesis that the most archaic regions of the leaf (especially in Ferns) are the apex and base, and the middle region is derwative. IX. In some cases of highly organised leaves, especially in Ferns, the branching of the adult leaf may be throughout that of equal dichotomy. This is usually accompanied by a long petiole, so that the orbicular or laciniate lamina is expanded far from the leaf-base. Historically this may be either a retention of the primitive dichotomy or a reversion to it: an opinion can only be formed on the facts for each individual case (figs. 27, 28). Thus, in some highly organised leaves the sympodial or monopodial states may not be represented at all. X. In other cases the converse may happen, as in the adult leaves of Angzopteris,* where, though in the young leaf the branching is a scorpioid sympodium, the relatively few pinnee of the adult arise monopodially, and the phyllopodium is then arrested in its growth. It does not progress to the primitive dichvtomous stage, which is thus omitted from the adult leaf. A similar state is seen in the Cycadacee. t In them even the acropetal succession is not strictly maintained. This probably gives a correct clue to the architecture of the leaves in most Angiosperms, in which arrest of apical growth at an early stage is a marked feature of the leaf-development. Certain extreme examples of such arrest of apical growth involve the atrophy of the whole distal region. They are seenin simple form in the protective scales of the Osmundacez, where the whole branch system of the lamina is arrested in its early stages, while the sheathing base remains of normal size. The same is seen in the Cycadacee. In Angiosperms this is the case with bud-scales, in which in various degree the “lamina” is represented in atrophied state, while the intercalary petiole is not initiated.{| Such cases may be held as further advances in that arrest which is already seen in Angiopteris. Not only is the phyletically earlier, dichotomous region of the distal branching omitted, hut the whole of it; and the effective part of the scale originates from the “ hypopodium” (Blattgrund of Hicuier). Thus it may be stated that im some Ferns, in Cycads, and in almost all Angiosperms the apical growth of the leaf-primordium is arrested early, and the distal primitive stage of * Phil. Trans., pt. ii, 1884, figs. 15-17. + Ibid., pt. ii, 1884, figs. 24-44, {| GorBEL, Organography, Engl. ed., ii, p. 386. 704 PROFESSOR F. 0. BOWER ON equal dichotomy is omitted. In this case the pinne are all of monopodial origin, and it is the phylogenetically later region that constitutes the adult leaf. A step Further is seen in the bud-scales of the Osmundacex, Cycads, and Angiosperms, where the effective region originates basally. XI. The distal branching of the leaf-primordium results in a lobed, or webbed, blade. But other developments not referable to this appear in some cases at the leaf-base. They are often included under the term “stipules.” They may be absent from whole families, but constantly present in others; and this applies both to Pterido- phyta and to Flowering Plants. Among the former the Osmundacez, Marattiaceze, and Ophioglossaceee have them; others have not. They appear to be primarily the result of enlargement of the wings of the leaf-base, with or without a commissure. The latter is seen in the Marattiacez, in Todea and Stangeria, but not in Osmunda ~ or Cycas.* They may have been of distinct phyletic origin in the different families.+ Such developments, which are usually styled “ stipular,” are basal in origin, and distinct in nature from the distal branchings of the leaf. XII. The factor which becomes most prominent in determining the leaf-form in the Higher Vascular Plants is intercalary growth. In them what is lost by arrest of apical growth is largely made up by intercalary growth. This is effective in pro- ducing the petiole, but it is also active inthe blade. Here the marginal teeth (if present) are produced early, by monopodial branching upon the phyllopodium. They are carried outwards by transverse intercalary growth below their base, and appear at maturity as the marginal teeth. Thus i the leaves of the Higher Plants intercalary growth is prevalent, and its results disguise the similarity of the archa- tectural scheme to that in the Pteridophyta. XIII. But the intercalary growth may intervene at various points in the phyllo- podium, and thus in the mature state pinnze which were in close juxtaposition when formed may be separated. Especially it may intervene between the lowest pinne and the rest. They would then appear at maturity at the base of the petiole, and be lable to be ranked as “stipules,” though differing in origin and nature from the stipules described above. Something of this nature is foreshadowed in those basal pinnee seen in the Cyatheacez, and styled “aphlebiz.” Jt 1s on developmental and comparative grounds that the line will have to be drawn wn each special case between the two possible sources of the parts ranked as “ stupules,” viz. basal developments of wings, and pinnex left in a basal position by mtercalary growth above them. The attempt has been made in the above pages to study the architecture of the leaf in the Sporophyte on the broadest lines. All the three avenues of comparison named at the outset have been used: viz. a comparative study of adult leaves; a similar study of juvenile leaves; and reference to the fossil record. It will now * See Pll. Trans., pt. ii, 1884, pls. xxxviii—xl. + GoxBEL states pointedly that “the structures which were frequently considered to be stipules in the Ophio- glossaceez are not of that nature” (Organography, Engl. ed., ii, p. 365). He gives no reason for this statement, though he admits the use of the term for the similar growths of the Marattiacee. LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 705 appear that the study of the juvenile leaves in relatively primitive Vascular Plants is not the least valuable of them in eliciting the phyletic story of leaf-architecture. The view has been adopted throughout that the juvenile leaves of the Pterrdophyta are rudimentary, not arrested. For they are free-living leaves, that have never been controlled, like the cotyledons of Seed-Plants, by the exigencies of packing within a seed, or by adaptation to storage. It is held that the adult leaf 1s the result of promoted development from this relatively simple source. The theoretical position is entertained that the progression which successive juvenile leaves show with ereater or less perfection, from equal dichotomy to sympodial and ultimately to monopodial branching, reflects with some degree of accuracy a probable story of phyletic advance in leaf-architecture. The microphyllous Pteridophytes would then, speaking generally, be types which had progressed less far, though in some cases simple leaves appear to have become so by reduction.. This may have been so in Equisetum, and it is believed to be the case for the simple leaves of some Fermrs and Flowering Plants. But such an interpretation will not apply to the Lycopodiales : for in them the leaf is always simple, and the earliest fossils of the family bear this out. They appear to have retained a primitive simplicity of the leaf. Hacepting, then, such cases as are probably reduced, a general working hypothesis may be adopted, that the progressive steps seen with varying completeness in the successive Juvenile leaves of the individual, and reflected by comparison of the adult leaves in various early vascular types, indicate the steps of elaboration which prevailed wm the evolution of foliar structures in the Sporophyte. The complex scheme thus set up in the megaphyllous types was suitable enough for hygrophytic vegetation, such as the Filicales and Pteridosperms. But it was liable to modification under exposed subaerial conditions. The phyletic history of the Angiosperms may be uncertain, but the probability is that they sprang from a> megaphyllous rather than a microphyllous source. -And already among megaphyllous types we have seen evidences of certain modifications which would have the effect of producing a leaf like that of the Angiosperms. The chief of these are apical arrest, with omission of the phyletically oldest but biologically most vulnerable distal region, and the substitution of intercalary for apical growth. Thus, we may believe, there was produced a type of leaf with various added advantages; such as an early maturation of the apex, a more robust construction, a capacity for close packing in the bud, and a more adaptable development of form to meet the various biological needs of subaerial life. In fact, the biological probability that such changes should supervene as amendments of the Filical construction lends support to the conclusions arrived at on a basis of comparison. The application of the - results obtained from the Pteridophyta to the leaves of Flowering Plants can here be only suggested in general terms. - It is hoped on some future occasion to develop the points of comparison on the basis of fresh observations on them. But if the com- parisons here instituted are correct, they indicate as a general statement for Vascular TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 21). 103 706 PROFESSOR F. 0. BOWER ON Plants that their leaf-architecture is throughout referable to modifications of a branch-system originating phyletically in a simple leaf subject to equal dichotomy. PosTscRIPT. On Theories of the Ultimate Origin of the Leaf. The views put forward in this memoir depend upon the way in which the juvenile leaves are regarded. If they are, as is believed, relatively primitive in their char- acters, and the adult leaves are similar forms promoted to higher complexity, then the sequence of leaves in the individual will indicate the probable steps of phyletic advance. Similarly, a comparison of the microphyllous Pteridophytes with the megaphyllous gives a sequence of very similar steps. It seems not unnatural to suggest that along these parallel lines a true indication may be found of how the megaphyllous state originated in descent. Thus it would appear probable that the ultimate origin of the leaf was as a lateral appendage upon the axis, relatively small originally, but capable of enlargement by branching, together with such other steps as are exemplified in the comparisons given above. It may be noted that a lateral origin of the leaf accords with the facts of embryology of Vascular Plants. It has been shown by comparison of the embryos of the Pteridophyta that in all adequately examined cases the first segmentation of the zygote stamps the polarity of the embryo, and that the apex of the axis coincides, as nearly as is consistent with the detail of the segmentation, with the centre of the epibasal hemisphere.* Thus in point of fact the position of the axis is defined before the origin of the leaf is apparent. That is so also for the embryos of Gymnosperms and Dicotyledons. But till recently the case of the Monocotyledons, as exemplified by the commonly accepted type of Alisma, seemed to be an exception, for there the cotyledon appears to be terminal. But CouLTer and Lanp 7} have met this difficulty by showing that Alisma is not a general type of embryogeny for Monocotyledons. In Agapanthus and others there is a peripheral cotyledonary sheath, surrounding the apex of the axis, which is itself actually central, as in other types of embryos. In the case of Alisma the single cotyledon may then be held to have assumed the terminal position as a consequence of its early and strong development, by a sort of “phyletic slide.” If this be so, then the exception of the Monocotyledons as a whole falls away. The “terminal” cotyledon, where it does occur, is not phyletically terminal, but lateral. Thus the conclusion of primary origin of leaves laterally in the embryogeny of Vascular Plants is of general appli- cation, though lable to secondary displacement, as in some Monocotyledons. But this conclusion, and this way of regarding the leaf as a member originally microphyllous and progressively asserting itself in descent, is out of harmony with Opinions expressed by various recent writers. Thus CAMPBELL has advovated a * Land Flora, chap. xlii, p. 670. + Bot. Gaz, 1914, pp. 509-514, LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 707 megaphyllous origin, holding that the leaf preceded the axis in the evolution of the shoot. His view is, in fact, a phytonic theory akin to that of GauprcHauD, but stated in terms of embryonic development. He bases his argument largely upon facts relating to the Ophioglossacez, and particularly on the embryology of species of Ophioglossum itself. But it is necessary to point out, as invalidating his argument, first, that Ophioglossum is probably a highly specialised representative of its family. This is indicated by its reticulate venation, and by many facts in its anatomy and morphology. Secondly, in this family the dormant axillary buds observed by GwyNNE-VaucHan and Lane are themselves very minute, but none the less the axial nature is recognisable in them, bearing leaves in case of their further develop- ment. And generally, when the facts adduced by CampsBeLy for special cases such as Ophioglossum are placed against those from Pteriodophyte embryology as a whole, it does not appear that his thesis of the priority of leaf over axis is sustained by the actual facts. On the other hand, TansLey, in his Lectures on the Evolution of the Filrcinean Vascular System, has contemplated a megaphyllous origin of the leaf. He supported his view by comparisons between the vascular anatomy of the adult leaf and that of the axis in certain early fossils. Especially he traced his most convincing parallel in the leaves of those Zygopterids which, with an upright habit of the leaf, bore four vertical rows of pinne. He remarks :* “This tendency to radial organisation of the frond may perhaps be regarded as a relic of the time when, according to our basal hypothesis, the structure of the fronds of Ferns was but little differentiated from the structure of their stems.” On a previous page ft he had stated the theory thus: “I am inclined to believe that the Fern-leaf is in phylogenetic origin a branch, or rather a branch-system, of a primitive undifferentiated sporangium-bearing thallus, and not an appendicular organ, differing ab wmtio from the axis on which it is borne.” Miss Bancrort{ finds anatomical support for the theory in Rachiopteris cylindrica, and gives copious references to cognate literature. TaNSLEY refers the modern statement of this theory in the first instance to Poronié.§ But in 1884 [ had myself stated substantially the same hypothesis thus:|| “May we not with good reason think that, just as the phyllopodium gradually asserts itself as a supporting organ among structures originally of similar origin and structure to itself, so also the stem may have gradually acquired its characters by differentiation of itself as a supporting organ from other members originally similar to itself in origin and development? Thus the stem and leaf would have originated simultaneously by differentiation of a uniform branch-system into members of two categories.” Finding later that facts were,deficient for any approach to a proof of this hypothesis, I left it, so to speak, in the air; but now | return to the consideration of it after more than thirty years. * Tc. pp. 28-24. qy ne t Ann. of Bot., 1915, p. 562, § Deutsche Bot. Monatsschrift, xv, 1897. || Phal. Trans., ii, p. 605. 708 LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. The obvious rejoinder to this theory, supported as it now is by the study of relatively early fossils, is that it starts from forms of extraordinary complexity. [ do not anticipate that any settled conclusion can be arrived at for the present as to the ultimate origin of foliar structures in the Vascular Plants. Perhaps the most valid suggestions may come from newly discovered fossil-types. But im my view the methods pursued in this memoir, based as they are upon the general study of well-known living plants, are preferable meanwhile to comparisons of the mature anatomy of adult organs, in plants which are imperfectly known and appear to have been highly elaborated and specialised. A comparison based upon embryology; upon the successive juvenile leaves of the individual, leading up to the adult; and upon the form and venation of the adult leaves of all the leading types of Vascular Plants, and checked by reference to the fossil record, offers a footing, however precarious, upon which to approach some reasonably probable conclusion. That provisional prob- ability is, that leaves were initiated as lateral appendages of an axis; and that the more complex leaves resulted from elaboration of a simpler type. DESCRIPTION OF PLATE. Fig. A. Leaf of Matonia pectinata, R. Br.; reduced to 4 natural size. Fig. B. Leaf of Dipteris conjugata (Kaulf ), Reinw. ; reduced to 4 natural size. Fig. C. Leaf of Dictyophyllum exile; much reduced. After Natuorst, from SEwarp’s Fossil Plants. Fig. D. Leaf of Camptopteris spiralis; much reduced. After NatHorst, from Szwarp’s Fossil Plants. Fig. E. Leaf of Pteris semipmnata, reduced. : Fig. F. Leaf of Odontopteris minor, Brongn.; rather less than 4 natural size. After ZEILLER, from Sewarp’s Fossil Plants. Trans. Roy. Soc. Edin. Vou. LI. Professor F, O, Bower on “ Leaf-Architecture as illuminated by a Study of Pteridophyta.” Fig, (709 ) XXII.—Contributions to our Knowledge of British Paleozoic Plants. Part I. Fossil Plants from the Scottish Coal Measures. By Dr R. Kidston, F.R.S. (With Three Plates.) (Read July 3, 1916. MS. received July 10, 1916. Issued separately November 29, 1916.) INTRODUCTION. In this series of papers it is proposed to describe new or interesting Paleeozoic plants, many of which were collected several years ago. Some of these are in the possession of the Geological Surveys of England and Scotland, others belong to personal friends or are in the collections of public museums, while the remainder are in my own collection. They come from various localities and horizons, and a few from coalfields of which papers dealing with their fossil plants have already been published, and they are, in many cases, additions to these lists. I take this opportunity of tendering my thanks to all who have so kindly helped me in my investigation of the Carboniferous flora, to whose kind and sympathetic assistance I owe much of my knowledge of British Paleozoic botany. I am also indebted to the Executive Committee of the Carnegie Trust for a grant to defray the cost of the plates which illustrate this memoir. In this part of the series the following species are described :— Sphenopteris ncurva, n. sp. Sphenophyllum cuneifolium Sternb., sp., forma amplum, n.f. Sigilaria elegans Sternb., sp. Sigillaria incerta, n. sp. Sigillaria Strivelensis, n. sp. Stigmaria minuta Géppert. Lagenospermum parvulum, n. sp. Sphenopteris incurva Kidston, n. sp. (PI. III, figs. 1, la; text-fig. 1.) Description.—Frond . . .; penultimate pinne free, rachis smooth, winged ; ulti- mate pinne alternate, linear-lanceolate, rachis straight, winged; pinnules alternate, subcuneate or rhomboidal, and bearing 1-2 pairs of narrow upward-directed decurrent lobes with rounded apices, and a similar blunt, sometimes cuneate, terminal lobe, which is occasionally larger than the other lobes of the pinnule. The terminal pinnule frequently consists of two narrow, blunt, incurved segments with a sinus between them. A single vein enters each pinnule, which gives off a veinlet to each lobe. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 22). 104 710 DR R. KIDSTON. Remarks.—The specimen is shown natural size on PI. III, fig. 1, and a portion is enlarged two times at fig. la, to show the form of the pinnules. The remarkable feature of this fossil is the curious form of the pinnules which terminate the pinne. These consist of two narrow cusp-like lobes which bend towards each other, leaving a space between them and enclosing a triangular area. The two horns are some- times open above (fig. la, A A, and text-fig. 1, A and C, <7), but at other times they. meet (fig. la, B B, and text-fig. 1, B, x7). The structure is very suggestive of some organ, now removed, having been attached at the base of and between the two incurved horn-like segments. ad Text-Fie. 1. Sphenopteris incurva is allied to Sphenopteris herbacea Boulay,* but differs in its smaller and more rhomboidal pinnules, which usually only bear one or two pairs of alternate lateral lobes and a terminal one. The curious forked and incurved terminal pinnules are also not known to occur on Sphenopteris herbacea. Locality.—Barony Pit, Oldbyres Farm, about 15 mile west of Auchinleck, Ayrshire. Horizon.—Overlying Ell Coal. Westphalian Series. Collected by Mr D. Tarr. Sphenophyllum cuneifolium Sternb., sp., forma amplum, n.f. (Pl; TLL, fies.) Remarks.—The form of Sphenophyllum cuneifolium figured here approaches in its size to Sphenophyllum majus Bronn, sp.,t but the narrow wedge-shaped leaves, with their sharp teeth, show that the specimens are only an exceptionally large form of Sphenophyllum cunerfolium Sternb., sp., which, however, seems worthy of a distinctive name. ‘The leaves are 1°4 cm. long on the specimen seen at fig. 3, and 1°7 em. long * Bounay, Terr. houil. du Nord de la France, p. 27, pl. i, fig. 5, 1876. Zer~uER, Flore foss. bassin houil. d. Valen., p. 106, pl. vi, fig. 4; pl. vii, figs. 3-4. Sorotheca herbacea, Stur, Die Farne: Carb.-Flora d. Schatz, Schichten, p. 279, pl. xxxiv, figs. 4-8, + Rotularia major, Bronn in Biscuorr, Kryptogam. Gewiéichse, pp. 89, 181, pl. xiii, figs. 2a, 2b, 1828. Spheno- phyllum majus Kidston, Mém. Musée roy. @hist. nat. de Belgique, p. 221, pl. xiv, figs. 1-4, 4a; pl. xv, figs. 2, 3; text-fig. 35, 1911. i a CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALAOZOIC PLANTS. 711 on that at fig. 2. On the typical plant the leaves seldom exceed 0°80 mm. to 1 em. in length. The specimens were collected by Mr A. Macconocutn, and are preserved in the Collection of the Geological Survey, Edinburgh. Localities—Meadow Bank Pit, + mile south of Polmont, and No. 13, Callender Pit, ¢ mile S.S.E. of the Glen, near Falkirk, Stirlingshire. Horizon.—Shale resting on Ball Coal. Lanarkian Series. Sigillaria elegans Sternb., sp. (PI. II, figs. 1, 1a, 2, 3, 4; Pl. III, figs. 4 and 8, 8a.) 1826. Favularia elegans, Sternb., Hssai fore monde prim., vol. i, fasc. iv, pp. xiv and 48, pl. lii, fig, 4. 1836. Sigillaria elegans, Brongt., Hist. d. végét. foss., p. 438, pl. exlvi, fig. 1; pl. elv ; pl. clviii, fig. 1. 1886. Sigillaria elegans, Zeiller, Flore foss. bassin houil. d. Valen., p. 582, pl. Ixxxvii, figs. 1-4. ‘ 1887. Scgillaria elegans, Weiss, Siyillarien d. preuss. Steink.: I, Gruppe Favularien, p. 32, pl. (4) x, figs. 388-41, 43-50. (Includes varieties.) 1911. Sigillaria elegans, Kidston, “ Végét. houil. dans le Hainaut belge,” Mém. Musée roy. d’hist. nat., vol. iv, p. 185.* 1887. Stgillaria elegantula, Weiss, l.c., p. 44, pl. (7) xiii, figs. 74-78. (Includes varieties.) 1905. Sigillaria elegantula, Koehne in Poronm, Abbild. u. Beschreib. foss. Pflanzen-Reste, Lief. iii, No. 52, figs. 1-17 (718), 19-30. (Ineludes varieties.) 1887. Sigillaria loricata, Weiss, l.c., p. 18, pl. (1) vii, figs. 3, 4. (Includes varieties.) 1905. Sigillaria loricata, Koehne in Poront, l.c., No. 54, figs. 1-6. 1887. Sigillaria squamata, Weiss, l.c., p. 25, pl. (2) viii, figs. 17-22. (Includes varieties.) 1887. cf. Sigillaria microrhombea, Weiss, l.c., p. 17, pl. (1) vii, figs. 1, 2. (Includes varieties.) 1905. Stgillaria microrhombea, Koehne, l.c., No. 53, fig. 3 (? figs. 1, 2.) 1887. cf. Sigillaria fossorum, Weiss, l.c., p. 28, pl. (3) xi, figs. 25-32. (Includes varieties.) 1905. ef. Stgillaria fossorwm, Koehne in Poronts, /.c., No. 55, figs. 1-14 (? 15), 16-18 (? 19-22). 1887. Sigillaria Fannyana, Weiss, l.c., p. 35, pl. (5) xi, figs. 51, 52. Remarks.—Some paleobotanists hold the opinion that the plant figured under the name of Sigillaria elegans by BRoNGNIART is not the same species as that to which STERNBERG applied the name of Favularia elegans, and with which Bronentart identi- fied his specimens. This view was adopted by Wetss, who proposed to restrict the naine of Sigilaria elegans Brongt. (non Sternb.) for those specimens whose cushions were smooth or nearly smooth below the leaf scar, while for those whose cushions below the leaf scar are provided with two diverging lines or ridges he proposed the name of Sigilaria elegantula. Thesé differences are not constant and are found even | on the same specimen. Sigillaria elegans Brongt. is therefore now united, by those who believe that the S. elegans Brongt. is not the Favularia elegans Sternb., with Sigillaria elegantula Weiss. On the other hand, those who believe that BRONGNIART was correct in identifying his specimens with the Favularia (Sigillaria) elegans Sternb. unite Sigillaria elegantula Weiss with that species, and this is the view that is adopted here. From what has been already said, it will be seen that Sigillaria elegans is a variable species, and some interesting variations in the form of the leaf scars and in * Full references and synonymy are given here. 712 DR R. KIDSTON. the length of the cushion which separates them are shown on PI. II, figs. 1 4, and Pl. III, figs. 4 and 8. These specimens are preserved in the Collection of the Geological Survey of Scotland, Edinburgh. Near the top of the specimen given on PI. II, fig. 1, are seen some cone scars, in the neighbourbood of which the leaf scars are much compressed, being broader than long, and occupy the whole of the cushion except a small portion at both sides (fig. la). The upper and lower margins are straight or even slightly hollowed in some cases, while others have on the basal margin a prominent central little point. On the lower part of the specimen, the leaf scars assume the normal form of Sigillaria elegans (fig. 1), some of which are emarginate on their upper margin and rounded at the base, below which is only a very narrow band of smooth cushion. Fig. 2 shows portion of a specimen on whose upper part are some normal leaf scars, but immediately below them the cushions and leaf scars begin to elongate, the lower margin of the latter ending in somewhat sharp points. The cushion below the leaf scar is smooth. A piece of the counterpart is given at fig. 3. In both the cushions are clearly delimited by a straight furrow, and from some of the leaf scars a decurrent line descends from the lateral angles, which, bending slightly inward, runs parallel with the margin of the cushion. In fig. 4, Pl. II, the smooth leaf cushions are not quite so much developed as in the last example. The leaf scars, which in all cases are situated at the upper end of the cushions, have emarginate upper margins and generally end in a point at the base. The lateral angles are prominent. These characters can be clearly seen on the figures, which should be examined with a lens, Another small fragment is given on PI. III, fig. 4, enlarged two times. On the upper part of the fossil, in the neighbourhood of a branch whorl, the leaf scars are much transversely elongated and partially deformed through mutual pressure. At the bottom of the figure they assume the usual characters of Sigillaria elegans. The leaf scar is emarginate at the top and has sometimes a sharp point at the base, the lateral angles are well developed; the leaf scar occupies the whole width of the cushion, except at the margins, where there is a narrow border of cortex. This specimen is identical with the Sigillaria microrhombea? as figured by KorHne,* and very similar to Szgillaria nucrorhombea, var. acutissuma Weiss,t} only the leaf scars are not quite so sharp-pointed laterally as in Wess’ figure. With Sigillaria elegans Sternb., sp., must also be united the Srgiularia squamata Weiss.{ There 66 appear to be no fixed characters whereby these “‘ species” can be separated from Sigillaria elegans Sternb. and Brongt. or from Sigillaria elegantula Weiss. All the specimens described above occurred together. Another small specimen of Srgillaria elegans Sternb., sp., is given on Pl. III, fig. 8, natural size, and a part is enlarged two times at fig. 8a. Here the relative proportion of cushion to leaf scar varies considerably. At the upper end of the specimen, seen more clearly on the enlargement fig. 8a, the leaf scars are those of Sigillaria elegans, * L.c., No. 53, fig. 3. ft Lc., p. 17, pl. (1) vii, fig. 2. t Le. p. 25, pl. (2) viii, figs, 17-22. a CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALAMOZOIC PLANTS. 713 but, as the interfoliar area elongates, the leaf scars and cushions assume the form of Sigillaria Fannyana Weiss,* which can only be regarded as a form of Sigallarva elegans. Locality.—Stein’s Fireclay Works, # mile north-east of Castlecary Railway Station, Stirlingshire. (PI. II, figs. 1-4; Pl. III, fig. 4.) Horizon.—Upper part of Millstone Grit, Lanarkian Series. Collected by Mr D. Tarr. Locality—Shell Brook, branch of river Dane, near Mareknowles, fully } mile north of Barghleighford, North Staffordshire. (Pl. III, fig. 8.) Horizon.—Below Fifth Grit, near base of Millstone Grit. Lanarkian Series. Collected by Mr D. Tarr. Sieilaria incerta, n. sp. (PI. I, figs. 5, 5a, 6, 7; Pl. ITI, figs..5, 5a.) 1894. Sigillaria Brardii, Kidston (non Brongt.) (pars), Proc, Roy. Phys. Soc. Ed., vol, xii, p. 252. Description.—Sigillaria showing arrangement of leaf cushions intermediate between Favularia and Clathraria sections. Vertical rows of leaf cushions separated by a very strongly marked zigzag furrow. Cushions hexagonal, or rounded hexagonal, sometimes with feeble notch on the upper margin, resting on each other above and below for about one-third of their width, with prominent lateral angles, smooth. Leaf scar occupying almost the entire surface of the cushion or having a greater or less portion of the cushion prolonged beyond its lower margin, rounded above, lower margin rounded or ending in a point; lateral angles prominent. Cicatricules placed slightly above the centre, middle one (the vascular bundle scar) punctiform, the lateral (parichnos) slightly elongated. The cone scars form a verticil of one row deep. Remarks.—An impression of a specimen of this species is shown natural size on Pl. I], fig. 5, and a portion is enlarged two times at fig. 5a. The Favularian arrange- ment of the leaf scars is clearly seen on the upper portion of the fossil, though at the central part their distribution approaches to that of the Clathrarian section. To- wards the centre of the fossil is part of a verticil of cone scars, arranged in a single row. The leaf scars are subhexagonal with prominent lateral angles equidistant from top and bottom of the scar. The three cicatricules are placed slightly above the middle, the central punctiform, the two lateral elongate, converging to or diverging from each other. Immediately below the cone verticil (fig. 5a) the leaf scars occupy almost the entire cushion, which only forms a very narrow border—a border probably representing the elevated sides of the cushion before it was compressed. At the upper and lower ends of the fossil the cushion below the leaf scar is slightly elongated and the leaf scar occupies its summit, the lateral angles of which join the margin of the cushion some little distance above its lateral angles. These elongated cushions are hexagonal, smooth, and only rarely show indications of two diverging ridges from the base of the leaf scar. * WEISS, l.c., p. 35, pl. (5) xi, figs. 51 and 52. 714 DR R. KIDSTON. At figs. 6 and 7 two small portions of the outer surface of a compressed stem are given, enlarged two times. At fig. 6 the leaf scars are seen to be more pointed below and slightly flattened or emarginate on their upper margin. The smooth cushion here forms a distinct marginal area at the sides and especially below the leaf scar. At fig. 7, which holds the same level on the other side of the stem as fig. 6, the leaf cushion is still further developed below the leaf scar, and on the lowest cushion in the centre row there are indications of two decurrent ridges which descend from the base of the leaf scar. Some of the cushions and leaf scars on this example correspond very closely with some towards the lower end of figs. 5 and 5a. Another small specimen is given on PI. III, fig. 5, natural size, and enlarged two times at fig. 5a, where the leaf scars and cushions are more transversely elongated. The leaf scars are here placed very close to the upper end of the smooth cushions. It was on this example I recorded, in error, the occurrence of Srgillamia Brardu Brongt. from the Lanarkian Series. This species must therefore now be removed from the lists of fossil plants from that horizon. Sigillaria incerta is, however, easily distinguished from Sigillaria Brardii Brongt. - by the lateral angle of the leaf scar of Sirgillaria incerta joining the margin of its cushion some distance above the lateral angle of its cushion, except in those con- ditions where the leaf scar occupies almost the entire cushion. If the specimens of Sigillaria incerta be compared with those of Sigillaria elegans Sternb., sp., described in this communication, it will be seen that they possess certain points in common. On the specimen of Sigillavia elegans given on Pl. II, figs. 1 and la, it is seen that the lateral angles of the leaf scar and its cushion are on the same level, a condition we also find in the central portion of Sigidlarmia incerta, Pl. I, figs. 5, 5a; but where the leaf cushions become elongated below the leaf scar, the lateral angle of the leaf.scar is at a higher level than the lateral angle of the cushion. This is seen on some of the cushions of Sigillaria elegans figured on Pl. II, fig. 3. They also agree in having the cone scars arranged in a single verticillate row. They further agree in the lower margin of the leaf scar being occasionally pointed (cf. Pl. II, fig. 3 and figs. 6,7). These circumstances have raised a suspicion in my mind that Sigillaria incerta may be an extreme variation of Srgillaria elegans, especially as the fossils occur together. I am, however, unable to satisfy myself that this is really the case, so I have deemed it safer to keep separate the plant given at Pl. II, figs. 5, 5a, 6, and 7, and Pl. III, figs. 5, 5a, which is here distinguished as Sigilarva incerta. Locality.—Stein’s Fireclay Works, ~ mile north-east of Castlecary Railway Station, Stirlingshire. (Pl. I, figs. 5-7.) Horizon.—Upper part of Millstone Grit, Lanarkian Series. Collected by Mr D. Tart. Locality —Kast Newton, Wemyss, Fife. (Pl. III, fig. 5.) Horizon.—Coxtool Coal (=Six-Foot Coal), Lanarkian Series. CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALAOZOIC PLANTS. 715 Sigillaria Strivelensis Kidston, n. sp. (Pl. I, figs. 1, 2, 3, 3a, 4, 4a, 5, 5a.) Description.—Stem ribbed, ribs slightly convex, furrows strongly undulated. Leaf scars placed at the top of the inflations, with their lateral angles at the widest part and occupying almost the entire width of the rib, more or less distant, penta- gonal, notched on upper margin, basal margin pointed, rarely rounded, lateral angles slightly above the centre, prominent, short, and quickly reaching the furrow ; cica- tricules placed a little above the centre of the leaf scar, the central (vascular) transversely lunate or punctiform, the lateral (parichnos) on the same level, elongate, slightly diverging. Outer surface of ribs ornamented with three lines of diverging transverse bars (rarely uniting to form a single central band), which, descending from the base of the leaf scar, gradually broaden out till they reach the top of the leaf scar below. Immediately above the leaf scar, and arising from the notch, is a plumose arrangement of fine ridges. Ligule pit not observable. Subcortical condition finely striated longitudinally. Remarks.—As there is apparently here either an abnormal elongation of the cortex between the leaf scars in some of the specimens or a reduction in its extent in others, as shown by the varying distances which separate them, it is not possible to determine which is the normal form of the species. It has therefore been deemed best to accept provisionally a middle condition and describe it first. This is seen at figs. 1, 2, Pl. I, which are both enlarged two times. Fig. 1 shows the undulations of the furrows, with the leaf scars, which are about their own height apart, placed on the expanded portion of the rib. * They occupy almost its whole width, having only a narrow band of cortex between their lateral angles and the furrow, which the descend- ing ridges springing from the lateral angles speedily reach. The leaf scars are distinctly notched on their upper margin and some of them rounded at their base, though others are more pointed. The three diverging and broadening lines of ornamentation which descend from the base of the leaf scar abut against the top of the scar lying below. Immediately above the leaf scar indications of the small plume-like ornamentation can be seen to spring from the notch. The small example given at fig. 2 is very similar, only the leaf scars are slightly more distant and, with the exception of the lowest one seen in the figure, which is somewhat rounded below, the basal margin of all the others ends in a point. The plume-like marking above the leaf scar is well seen on this example. In all cases the leaf cushions: have formed prominent and probably considerably elevated bracket-like structures, their upper margins forming the base of an upward-rising surface to which the leaf was attached. On the specimen given at fig. 3, of which a portion is enlarged two times at fig. 3a, the leaf scars are less than their length apart and the inflations and con- tractions of the ribs become more marked. Owing to pressure, the interfoliar area seems to be pressed over the upper margin of the leaf scar, frequently obscuring the “notch” and hiding the plumose marking. On most of the jnterfoliar areas, the 716 DR R. KIDSTON. three descending lines of ornamentation are clearly seen. The basal margins of .the leaf scars are always, on this example, prominently pointed. The leaf scars are also longer than broad, At fig. 4 another small specimen is given where the inflation of the ribs is very marked, so much so that they form hexagonal areas whose bases being pressed against and over the upper margins of the leaf scars below, give a very Lepidophloid appearance to the specimen, somewhat similar to the Lepzdophloios tumidus Bunbury, sp.* The leaf scars on this example are also transversely extended and become broader than long; they are, however, of the same general structure as those on the other specimens, having prominent lateral angles and notched upper margin with pointed base. The ornamentation of the interfoliar area is also identical. The slight differences in the form of the interfoliar areas and the leaf scars seem to be the result of a lateral extension of the ribs. Another example of this species is illustrated at fig. 5, Pl. I, and a portion enlarged two times at fig. la. Here the leaf scars are separated by about three times their length. The furrows, though still distinctly undulated, are not so much so as in the specimens already described. The leaf scars possess the same characters, especially as those seen on figs. 2 and 8. They are notched on the upper margin and pointed at base, with prominent lateral angles, and on the cortex immediately above the leaf scar is the same plume-like marking, but the ornamentation on the interfoliar portion of the rib assumes more the form of a central band. Notwithstanding this difference, I have no hesitation in referring this example to the same species as figs. 1-4. The nearest allies to Srgillaria Strivelensis are Sigillaria contracta Brongt.t Sigillaria Feistmanteli Geinitz,t and Sigillaria polyploca Boulay.§ On the first two mentioned the leaf scars are pyriform, which at once distinguish them from Sigilaria Strivelensis, and further, the ornamentation of the interfoliar cortex in S. contracta is punctiform and covers the whole of the contracted portion of the rib, as seen on the type which is now preserved in the British Museum, and not in the form of a series of lines descending from a central ridge as given by BRoNGNIART. Such specimens, however, as those given on Pl. I, figs. 3 and 4, have a resemblance to Sigillaria polyploca Boulay, but they differ in the form of the ornamentation of the cortex, and the ridges which descend from the lateral angles are shorter and less prominent, and there is never any transverse furrow above the leaf scars in our species.|| The plumose-like ornamentation above the leaf scars is also absent in Sigilaria polyploca Boulay. * Lepidodendron ? tumidum, Banbury, Quart. Journ. Geol. Soc., vol. iii, p. 482, pl, xxiv, fig. 1, 1847. } Hist. d. végét. foss., p. 459, pl. exlvii, fig. 2. { Neues Jahrb., 1865, p, 392, pl. iii, figs. 4 and 4a; Frisrmanren, Vers, d. bohm. Kohl-Ablag., Abth. iii, p. 14, pl. ili, figs. 4, 4b,5, 5a, 1876. § Le terr. houil. du Nord de ia France, p. 47, pl. ii, fig. 8, Lille, 1876; Zuruner, Flore foss. bassin hourl. d. Valen., p. 540, pl. Ixxxii, figs. 7, 8, 1886. || Cf. figs. 1, 2, and 5. CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALAOZOIC PLANTS. 717 The specimens were collected by Mr D. Tarr, and are preserved in the Collection of the Geological Survey of Scotland, Edinburgh. Locality.—Stein’s Fireclay Works, 3 mile north-east of Castlecary Railway Station, Stirlingshire. Horizon.— Upper part of Millstone Grit, Lanarkian Series. Stigmaria minuta Géppert. (PI. III, fig. 6.) 1848, Goppert, Preisschrift iiber Steinkohlen, pl. xiv, fig. 24. 7 1852. Stigmaria ficoides, var. minuta, Gopp., Ubergangs, p. 246. 1866. Stigmaria Evenit,. Lesqx., Geol. Survey of Illin., vol. ii, Palzont., p. 488, pl. xxxix, fig. 9. 1879. Stigmaria Evenit, Lesqx., Atlas to the Coal Flora, p. 16, pl. Ixxv, fig. 1. 1886. Stigmaria Event, Zeiller, Flore foss. d. bassin houil. d. Valen., p. 618, pl. xci, fig. 7. 1890. Stigmaria Eveni, Grand’Eury, Bassin houil. du Gard, pl. xiii, fig. 13. 1880. Stigmarioides Evenit, Lesqx., Coal Flora, p. 333. 1890, Stiymariopsis Event, Grand’Eury, Bassin houil. du Gard, p. 248. 1894. Stigmaria ficoides, var. minima Nathorst (non Goldenberg), Foss. Flora d. Polarlander, Erster Theil, Erste Lief. : Zur palaeoz. Flora d. arktischen Zone, p. 44, pl. viii, fig. 9, and text-fig. p. 44. Description.—Rhizome bearing distant rounded or elliptical scars, from 1 mm. to 3 mm. in diameter, with a small central elevation surrounded by a depressed ring with a raised outer margin, somewhat irregularly placed and occasionally varying in ‘size on the same specimen. Cortex very finely granular, with or without irregular longitudinal furrows or foldings. Remarks.—This species of Stigmaria, originally described as a variety of Stigmaria ficoides by GoprprErtT, has the smallest rootlet scars of any described Stigmaria, and is easily distinguished from the common forms of Stegmaria ficoides by this character alone. The fine granular outer surface is generally distinctly seen, but on imperfectly preserved specimens may not be observable. Stigmaria minuta is also distinguished from Stigmaria minor Geinitz™ by its smal! rootlet scars and more distinctly granular cortex. The Stagmaria anabathra, var. minima Goldenberg,t must not be confused with the Stigmaria minima Nathorst, from which it is quite distinct. GOLDENBERG’S var. minima is closely related to, if not identical with, the Stagmaria minor of GEINITZ while Naruorst’s var. minima is indistinguishable from Stigmaria nunuta Gopp. Stigmaria minuta Gopp. is very rare in the Lanarkian Series, and hitherto I have only seen it from this series from the locality and horizon mentioned below. In the Westphalian Series it is more frequent, but by no means a common fossil. The specimen figured was collected by Mr A. Macconocutz and is ‘preserved 1 in the Collection of the Geological ee) of Scotland, Edinburgh. Locality.—Meadow Bank Pit, $ mile south of Polmont, Stirlingshire. Horizon.—Shale resting on Ball Coal, Lanarkian Series. * Stigmaria ficoides, var. minor Geinitz, Vers. d. Steinkf. in Sachsen, p. 49, pl. iv, fig. 6; pl. x, fig. 1, 1855. + Flora Sarepont foss., Heft iii, pp. 19 and 42, pl. xiii, figs. 3, 3a, 1862. TRANS. ROY, SOC. EDIN., VOL. LI, PART III (NO. 22). 105 718 DR R. KIDSTON. Lagenospermum Nathorst. 1914. Lagenospermum, Nathorst, Foss. Flora d. Polarldnder, Erster Teil, Vierte Lief. : Nachtrage zur palaeozoischen Flora Spitzbergens, p. 29. 1914. Radiospermum, Arber (pars), Annals of Botany, vol. xxviii, p. 101. a Description.—Seeds generally small, oval, elongate, or spindle-shaped, which are characterised in being entirely surrounded by a cupule, the upper margin of which is usually, if not always, divided into teeth. The cupule frequently shows ribbing, and this appears to correspond with the number of the teeth possessed by the cupule, which are frequently six. In some cases the cupules seem to be smooth, though a little furrow may pass down the cupule in a line corresponding with the margins of two neighbouring teeth. The seeds themselves appear in some cases to be ribbed (Lagenospermum Kidstoni Arber, sp.), while in other cases they are smooth.* The seeds may be borne on a compound branch system or terminate dichotomously divided branchlets. Remarks.—The genus is altogether provisional and was formed for the reception of seeds having a cupular structure, as in Lagenostoma Williamson, which was, however, founded on specimens showing their internal organisation. Several cupulate seeds, of which only the external form, and in some cases also the outline of the seed, is known, have been placed by their describers in WILLIAMsSon’s genus; but as no knowledge is possessed of their internal organisation, NaTHorst founded the genus Lagenospermum for their reception. Some of these may prove to be true members of the genus Lagenostoma Will., but as at present we are entirely ignorant of their structure beyond the fact that they possess a cupule, they have no claim to be included in a genus founded on clearly defined anatomical characters. The genus Lagenospermum has been intentionally made a very wide one, and its essential character is the presence of a cupule. There can be little doubt that, if we knew the organisation of the seeds included in it, it would most probably be found that they represent more than one type; but from our present imperfect knowledge of their structure it is much better to include such cupulated seeds, which are only known as incrustations or impressions, in a provisional, wide and comprehensive, “ venus.” In Lagenospermum Nathorst I would include the following British species :— 1905. Lagenostoma Kidstoni, Arber, Proc. Roy. Soc. London, vol. B lxxvi, p. 247, pl. i, fig. 1; pl. ii, figs. 1-6. : 1905. Lagenostoma Sinelairi, Kidston, MS. Arber, 2bid., p. 251, pl. ii, figs. 7-11. 1914. Lagenostoma oblongum Kidston, Trans. Roy. Soc. Edin., vol. 1, p. 160, pl. vii, figs. 1, la, 2, 2a; text-fig. 8. 1914. Lagenostoma’ urceolaris, Kidston, ibid., p. 161, pl. xvi, figs. 9, 9a, 10, 10a. Lagenospermum parvulum, Kidston, n. sp. * CaRPENTIER, “Sur quelques: fructifications et inflorescences du Westphalien,” Revue générale de botanique, vol, xxili, p. 1, pl. xii, figs. 1, 4, 1911. CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALHOZOIC PLANTS. 719 Lagenospermum parvulum Kidston, n. sp. (PI. II, figs. 7, 7a; text-fig. 2.) Deseription.—Small, seed entirely surrounded by the cupule, 4°5 to 5 mm. long, 2°15 to 2°65 mm. wide, oval, rounded at base and slightly contracted towards the apex, frequently longitudinally plicate, and terminating in six short teeth. Epidermal cells of cupule small, quadrate. Remarks.—Lagenospermum parvulum is closely related to LZ. Kidstona Arber, sp., but differs in its smaller size and in the cupule being slightly contracted upwards, which gives it a less obtuse form at the apex. The average size of L. Kidstoni is 6 mm. long with a width of 2°5 to 3 mm., whereas L. parvulum is only 5 mm. long and 2°15 to 2°65 mm. wide. The teeth also appear to be a little more clearly defined. Owing to the colour of the matrix it was difficult to get a good photograph of the specimen, but the text-figure (2, A and B, x7) gives a clear idea of the form of the J B TRxT-Fic. 2. cupule—the only part of the fossil exhibited, as the seed is entirely surrounded and obscured by it. Locality.—Pelbower Burn, fully + mile north of Kirkconnel Church, Dumfries- shire. Horizon.—In blaes, 3 feet above a 5-inch coal, a few fathoms above the “ Skipsey Band.” Westphalian Series. Collected by Mr D. Tarr. Specimen in the Collection of the Geological Survey of Scotland. EXPLANATION OF PLATES. Puate I, Fig. 1. Sigillaria Strivelensis Kidston, n. sp. x 2. (T. 2195 B.) Fig. 2. ie 6 5 F, eens 21208.) Fig. 3. = ee = ; Naturallsize. Fig. 3a, x 2. . (T. 2189 B.) Fig. 4 me - i 3 Fig. 4a, x 2. (T. 2198 B.) Fig. 5. - 4 3, cs Hs Fig. 5a, x 2. (T. 2394 B.) Locality.—Stein’s Fireclay Works, } mile north-east of Castlecary Railway Station, Stirling- shire. Horizon.—Millstone Grit. Lanarkian Series. 720 CONTRIBUTIONS TO OUR KNOWLEDGE OF BRITISH PALAOZOIC PLANTS: Puate II. Fig. 1. Sigillaria elegans Sternb., sp. Natural size. Fig. la, x 2. (T. 2400 B.) Locality.—Stein’s Fireclay Works, } mile north-east of Racnlereny pe ae: Staton, Sting shire. ot Horizon.—Millstone Grit. Lanarkian Series. Fig. 2. Sigillaria elegans Sternb., sp. Natural size. (T. 2393 B.) Same locality and horizon. Fig. 3. Sigillaria elegans Sternb., sp. Naturalsize. (T. 2396 B.) Same locality and horizon. Fig. 4. Sigillaria elegans Sternb., sp. Natural size. (T. 2398 B.) Same locality and horizon. Fig. 5. Sigillaria incerta Kidston, n. sp. Natural size. Fig. 5a, x 2. (T. 2197 B.) Same locality and horizon. Figs. 6, 7. Sigillaria incerta Kidston, n. sp. x 2. (T. 2395 B.) From different sides of the same specimen. Same locality and horizon. Puare III. Fig. 1. Sphenopteris incurva Kidston, n. sp. Natural size, Fig. la, x 2. (T. 1149 D.) Locality.—Parony Pit, Oldbyres Farm, 14 mile west of Auchinleck, Ayrshire. —_— Horizon.—Shale overlying the Ell Coal. Westphalian Series. ag Figs. 2, 3. apheneeiatie cunetfolium, Sternb., sp., forma amplum Kidston, n.f. Natural size, (Fig. 2, M. 1698 F., and fig. 3, M. 1754 F.) ee Locality. —Meadow Bank Pit, $ mile south of Polmont, Stirlingshire. Horizon.—Shale resting on Ball Coal. Lanarkian Series. g. 4. Sigillaria elegans Sternb., sp. x 2. (T. 2392 B.) om Locality.—Stein’s Fireclay Sea oe 8 mile north-east of Castlecary Railway Station, Stirl shire. me Horizon.—Millstone Grit. Lanarkian Series. : Fig. 5. Sigillaria incerta Kidston, n, sp. Natural size. Fig. 5a, x 2. (K. 2310.) Locality.—Kast Newton, Wemyss, Fife. . \ i ae Horizon.—Coxtool Coal (=Six-Foot Coal), Lanarkian Series. Collected by the late Mr J. W. Kirxsy. Fig. 6, Stigmaria minuta Goppert. Natural size. (M. 1717 F.) Locality.—Meadow Bank Pit, } mile south of Polmont, Stirlingshire. Horizon.—Shale resting on Ball Coal. Lanarkian Series. aa) Fig. 7. Lagenospermum parvulum Kidston, n. sp. Natural size. Fig. 7a, x 2. (T. 627 E.) “a Locality.—Pelbower Burn, fully }+ mile north of Kirkconnel Church, Sanquhar, Dumfries- shire. 5 Horizon.—In hlaes, 3 feet above a 5-inch coal seam, a few fathoms above the “Skipsey Band.” Westphalian Series. Fig. 8. Sigilliaria elegans Sternb., sp. Natural size. Fig. 8a, x 2. (T. 347 B.) Locality.—Shell Brook. branch of the river Dane, near Mareknowles, fully ~ nile nort| of Barghleighford, Staffordshire. Horizon.—About 100 feet below Fifth Grit, near base of the Millstone Grit, Tana ia Series, Fi ~ om All the figured specimens are preserved in the Collection of the Geological Survey of Scotland, Pl. II, fig. 8, which is in the Collection of the Geological Sanvey a England, ee _ Pls ‘Ul which is in the writer’s Collection. en: Vol. LI¢.—Puate I. Parr |. ozoIc PLANTS . v) British PA KIDSTON M‘Farlane & Erskine, Edin. ton : Photo. R. K Fias, 1-5. Sigillaria Strivelensis, Kidston n.sp. he 1% j PEP TRY ETN | } 4) Wes aA Ai) a, f Pig Pa rans. Roy. Soc. Edin" Vol. LI —Puate II. Kinston: British Pat#ozoic PLants—Parr | M‘Farlane & Erskine, Edin. q _ Fras. 1-4, Sigillaria elegans, Sternb. sp. Figs. 5-7. Sigillaria incerta, Kidston n.8p. ‘ans. Roy. Soc. Edin™ Vol. LI —Puate III. Kipston: British PALOZOIC PLANTS—Parrt I. . Kidsto} Photo. M‘Farlane & Erskine, Edin. 1G 1,$phenopteris incurva, Kidston n.sp. Fic. 5. Sigillaria incerta, Kidston n sp. am 2- Sphenophyllum cuneifolium, Sternb. sp., forma », 6. Stigmaria ficoides, Sternb. sp., var, minuta, Gépp. Ct | amplum, Kidston n.f. », 7. Lagenospermum parvulum, Kidston n.sp. Brn an (A2IRT XXIII.—A Revision of the British Idoteide, a Family of Marine Isopoda. By Walter E. Collinge, D.Sc., F.L.S., etc., Research Fellow of the University of St Andrews. Communicated by Professor M‘Inrosa. (With Eleven Plates.*) (MS. received December 11, 1915. Read January 10, 1916. Issued separately January 17, 1917.) CONTENTS. PAGE PAGH IL. Introduction . : ; é : . 721 | VIII. Systematic . : : . 734 Methods and Teaktliqte : : . 722 Family Tdoteidez, hee 1813 os eas 735 IL Hist ] 723 Key to the Genera of British Idoteidee . 736 coo ; , 1, The Genus Idotea, Fabricius, 1798 . 736 IIT. Classification and Affinities . 5 a . 724 Key to the British Species 3 . SH IV. Geographical Distribution . ; , PN (1) Idotea eae cee . . a “aie 2 pelagica, Leach : a, ue MMRRERNIGNGa =o) oo Oy Fe ee > 128 “ é se ewes od i VI. External Structure— (4) 4, emarginata (Fabr.). 74] (a) General Form. : : é . 730 (5) ,, granulosa, Rathke . _ 742 (b) The Cephalon. j P : 5 20 (6) sarsi, np. - 743 ee Woe Tiges peice io PP a @elekcridia(labber, . (>. 745 4 gel ; : 3 es tad (8) ,, metallica, Bose : » 46 ee a COS Taare (9) 4, linewris(Pennant). . 747 4. Oral Appendages.~ . ~. - . 7al 2. The Genus Zenobiana, Stebbing, 1895 749 Ree he Merokome: . - ; : 3 Spe (1) Zenobiana prismatica (Risso) . 749 ds Coxal Plates : 7 ; tee 3. The Genus Synisoma, nov. nom 750 2, Appendages . . F ; . 732 (1) Synisoma lancifer (Leach, MSS.) 751 ee mcnsone i i ; - 732 (2) 55 acuminata (Leach) . 752 1. Appendages . ‘ ; : - 733 Bibliography : : : : i : . Wee a Uropode Se et = s+ 1885) SoSplanation of Plates, 00. O°." <4)" . °°) 756 VII. Bionomics and Economics. 3 : . 733 Index . : ; . : - 7 : 5 be I. InrRODUCTION. Many of the earlier accounts of the group of Crustacea, known as Isopoda, suffer from want of more detailed diagnoses and figures. To some extent this has been remedied during recent years by the authors of different monographs, but there are still many families which remain somewhat obscure, or only partially understood, owing to the above-mentioned cause. Such a family is the Idoteidee. In the present communication | propose to revise the diagnoses of the British genera and species, to set forth in greater detail than has hitherto been done their structure, illustrating the same, and to discuss the classification and affinities of the family as a whole. This investigation has been carried out at the Gatty Marine Laboratory of St Andrews University, during my tenure of a Research Fellowship of the ‘University, and of a Walker Trust Research Scholarship. * The author desires to thank the Carnegie Trust for the Universities of Scotland for their grant to meet the cost of reproduction of the plates. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 23). 106 722 DR WALTER E. COLLINGE. My best thanks are here tendered to Professor W. C. M‘hyrosu for the facilities he has so kindly given me for carrying out the work, and for material from the University Natural History Museum and his private collection. [ also wish to record my thanks to the Rev. T. R. R. Sressrne for specimens of Zenobiana prismatica (Risso) and Stenosoma lanciferum (Leach) ; to Professor G. O. Sars for examples of the Norwegian species of Idotea; to Professor D'Arcy W. Tompson for the facilities he has at all times so kindly given me for examining the collections in the Museum of University College, Dundee ; to Dr Rarrar.e IsseEu for specimens of Zenobiana prismatica (Risso) and other species; to Dr K. STepHEn- SEN for specimens of Jdotea metallica, Bosc, and various species of Zenobiana and Stenosoma: to Professor L. A. L. Kine and Dr Tuomas Scorr for other species ; and to Dr W. T. Catman for his kind advice on certain matters of terminology. Finally, I have to thank Miss H. G. Krrx for the great pains she has taken with the figures, many made from my rough drawings, and others direct from the specimens themselves. METHODS AND TECHNIQUE. Whilst examples of both sexes of all the species here enumerated have been examined, many of them alive, dissections of all have been made and the different parts treated in various ways to bring out the minute structure. A brief account of the methods employed may prove useful. Killing.—Where living specimens have been obtainable these have been killed by placing them in small glass jars and slowly adding equal parts of fresh water and 90 per cent. alcohol. When dead the specimens were transferred to 75 per cent. alcohol, and finally to 85 per cent. Some species are apt to become very soft in the 75 per cent, alcohol, so that it is well to change them to the stronger alcohol after an hour or two. Staining.—Various fluids have been tried, but only those possessing great penetrative powers were found to be of any value. The best results were obtained with Mayer's alcoholic carmine. When the parts were not required as permanent mounts, very good results were obtained by leaving them for four or five hours in an aqueous solution of methyl-green. Mounting.—All permanent preparations were treated with alcohol, xylol, and mounted in Canada balsam (xylol). Temporary mounts were examined in xylol. For the removal of soft parts in the appendages, temporary preparations were held, after mounting in Canada balsam, over a spirit lamp for about the third of a minute. Where permanent preparations were desired, the different parts were boiled in a weak caustic soda solution, then washed in alcohol, stained, cleared in xylol, and mounted in Canada balsam. a A REVISION OF THE BRITISH IDOTEIDA. 723 Il. HistToriIcau. This family of Isopoda has received considerable attention from carcinologists | in all parts of the world. Musrs (44) in his well-known “Revision” of 1881 has reviewed the work of most of the earlier authors, and as I shall have occasion to mention some when discussing the classification and affinities, no useful purpose would be served in repeating these here. The genus Jdotea was constituted by J. C. Faprictus in 1798, the first description appearing at page 302 in his Supplementum Entomologia Systematica. Two years previously he had indicated that he intended describing the genus as new in the Supplement (cf. Index Alphabeticus, 1796, p. 86). In this latter publication he used the “nomen nudum” Idothea, but in'1798 at the head of his description he uses the name Jdotea, and I agree with Canon Norman (47) that the latter spelling of the name should be retained. Firstly, as he points out, because this is the spelling which is used, and intentionally used, with the description ; and secondly, because it is the spelling which has been almost universally employed for a hundred years. Miss RicHarpson (57) prefers to use the spelling Jdothea, pointing out that that of Idotea is preoccupied by WEBER, 1795. The earlier references to the family by Linnt, Patuas, Pennant, DE GeErr, Risso, Bosc, LatTreitteE, DesMAREST, BranptT, Kroyer, Mitne-Epwarps, Le&acu, RatHKE, WuiTs, and others need not be considered here, beyond the remark that frequently the definitions of these authors were insufficient, and ne they were not always clear as to the species they were describing. Dana (16) in 1853, in his great work on the Crustacea of the U.S. Exploring Expedition, described six new species of IJdotea, and diagnosed the new genera Epelys, Cleantis, and Erichsonia. Bate and Wxstwoop (3) gave a good account of the British species, as then known ; and Harcer (32) also gave a very full account of those genera and species found on the eastern coast of the Northern United States. Miers’ (44) work of 1881 was a great advance upon any previously published, and included a discussion of no less than forty-seven species, included in four genera. CHILTON (114) in 1885 described the interesting Idotea festiva from New Zealand. In 1894 (46) Norman drew attention to several distinct varieties of Idotea baltica, which have since been elevated by Sars to specific rank. Dotirus (20) in 1894-95 gave a brief account of the species found upon the coast of France. STEBBING (69) in 1895 proposed the name Zenobiana for the genus Zenobia, Risso ; and Sars (64) two years later showed, in connection with the Norwegian species, that what had generally been considered by recent authors as only varieties of one and the same species, were in reality referable to no less than six different species, all easily recognisable in both sexes. Brnepict (4) in 1897 published a useful revision of the genus Synidotea. CuHILTon (11) in 1899, in a paper which seems to have been 724 DR WALTER E, COLLINGE. overlooked by later authors, gave an account of the New Zealand species. STEBBING (70) in 1900 described the genus Pamdotea with the Oniscus ungulata of Pauas as the type; and in 1904 (68) the genus Glyptidotea for the I. lichtensteinm of Krauss. Norman (47) in the same year catalogued eleven species as occurring in the British Isles. Miss RicHarpson (55) in 1904 described a new genus, Symmius, for a Japanese species, in which the uropoda each consist of a single piece, the metasome has three segments, the flagellum of the antenne a single joint, and the maxillipede a three-jointed palp. In the same year (56) she described another Japanese genus, Pentias, which is characterised by the five-jointed palp of the maxillipedes and a uniarticulate metasome. In 1905 the same writer (58) redescribed the then known species of North America, amplifying the earlier diagnoses and adding the two new genera Mesidotea and Pentidotea. In 1910 Racovirza and Srvastos (52) gave an interesting cone of a new fossil species for which they erected the genus Prozdotea. In 1912 Miss RrcHarpson (60) pointed out that the Cleantis isopus of Mrmrs could not be retained in the genus Zenobiana, and suggested the new genus Cleantiella for its reception. IsseL (35) in 1913 redescribed and figured Zenobiana prismatica (Risso), pointing out various errors in BaTE and Westwoon’s figures and description, — further showed that Dana’s genus Cleantis was synonymous with this. Barnarp (Ann. Sth. Afr. Mus., vol. x) in 1914 described the new genus Engidotea, a new species of Synidotea (setifer), and three new species of Paridotea, Stebbing. In 1915 the writer published a diagnosis of IJdotea hectica (Pallas) with figures (14), and also gave a brief account of the species of the family occurring in St Andrews Bay (15). III. CuasstricaTION AND AFFINITIES. In the various attempts that have been made at a classification of the Isopoda, almost every part of the exoskeleton has at some time or other been employed. Many authors have put forth different systems, but few of them have proved successful or serviceable. Of the earlier systems very little need be said, for, as has been frequently pointed out, many of the errors and much of the synonymy arose owing to the imperfect descriptions and absence of figures, so that an author often confused one species with another; such conditions are, of course, common to all branches of zoology. | 3RANDT (7) was perhaps the first to realise that certain parts of the exoskeleton were more constant than others, amongst which he specially emphasised the import- ance of the antenne, the telson, and the uropoda. Mitne-Epwarps (45) very largely copied Branpt, but, accepting imperfect descriptions by earlier authors, his work, in this particular respect, has not proved of any great value, | See en SR A REVISION OF THE BRITISH IDOTEIDA. 725 Dana (16) in 1858 to a large extent followed the main lines of Mitne-Epwarpbs’ classification. Kinanan (37) in 1857 very thoroughly and critically examined the different systems up to that date, and at once recognised the importance of the antennz and the uro- poda, and, he adds, perhaps the form of the maxillipedes, but he states : “Two very important sets of characters have been omitted or overlooked, which must be taken into account in any classification which seeks to form natural groups; these are: the characters drawn from the form of the head and from the epimerals” (coxal plates). It is upon the whole of the above-mentioned characters that the classification in the present paper is based. The leading structural characters of systematic value are undoubtedly shown in the form of the cephalon and its appendages, the coalescence of the abdominal segments, and the operculiform inferior uropoda.. In all of these characters the Idoteidee exhibit a close relationship with the Arcturide. Both families are probably of ancient origin, and many of their genera afford evidence of more recent modifications and adaptations. Unfortunately many of the earlier species are in- completely described and still more incompletely figured, as already pointed out. OuLIn (48) very pertinently remarks, the classification of the different genera is as yet very uncertain, and is very much in want of a renewed critical revision. Such a revision the writer hopes to complete at no very distant date. LaTREILLE (39) in 1829 included in his fourth section of the family (the Idoteides of Leacn) the genera Idotea, Fabricius; Stenosoma, Leach; and Arcturus, Latreille. Mitne-Epwarps (45) in 1840 employed the same section for the genera Arcturus, Idotea, and Anthura, Leach, but this last-mentioned genus has little in common with either of the preceding ones, apart from the elongated form of the body. In 1853 Dana (16) removed the Arcturide to a new group, which he termed Anisopoda, and placed the Idoteide with the Cheetiliide, a proceeding which few subsequent authors have thought was warranted. Of the three genera he described, Epelys, Cleantis, and EHrichsoma, the first is now regarded as synonymous with Edotia, Guérin-Mén., the second with Zenobiana, Stebbing, and for the third name, which was preoccupied, BENEDICT substituted Hvichsonella. Bare and Wrstwoop (3) in 1867 associated together the families Arcturide and Idoteide. Craus (12) in 1871 placed both Chetilia, Dana, and Arcturus in his group Idoteides. Both Haremr (32) in 1880, and Mrmrs (44) in 1881, recognised the close affinity of the Idoteide with the Arcturide, as have practically all subsequent authors. Bonnizr (5) in 1887, following Ciavus, placed both Arcturus and Idotea in the family Idotheide. Hansen (30) in the same year gave an account of Glyptonotus entomon (Linn.), and figured and described some of the oral appendages. Dr Hansen informs me (vn litt. 7th Dec. 1915) that he now considers the specimens he examined to be referable to the G. sibiricus of Brruta. 726 DR WALTER E. COLLINGE. In 1901 Ontry (48) described from Patagonia a new genus and species (Pseud- idothea bonniert) as the type of a new family—Pseudidotheide,—which he regarded ‘as an intermediate link between Idotheide and Arcturide.” It is characterised by the transformation of the first pair of metasomatic appendages into a copulatory organ in the male, in addition to the usual style on the endopodite of the second pair, the coalescence of the metasomatic segments, the small antennule, the two- jointed flagellum of the antenna, and a five-jointed palp on the maxillipedes. The specimens very much resemble the Jdotea miersi of StupER (74), and whilst the characters given may serve to constitute a new genus, or even subfamily, I feel very doubtful whether they are sufficiently important to rank above this. THomMson (774) in 1904 described the new family Holognathide and genus Holognathus for the reception of the Idotea stewartr, Filhol, which differs from the true Idoteide in possessing a three-jointed mandibular palp. Racovrrza and Srvastos (52) in 1910 constituted a new sub-family, Mesidoteini, with the genus Mesidotea, Richardson, as the type, and including also Chiridotea, Harger, and Proidotea, Racovitza and Sevastos. In this interesting and thought- ful paper the authors point out that the family Idoteide is an ancient one, and especially the genera of the new sub-family, which are completely isolated from the remaining genera. They summarise as follows :— “(q@) La famille des Idothéides est formée par plusieurs lignées qui actuellement ne sont réunies par aucune forme intermédiaire. “(b) Au cours de leur évolution ces lignées ont subi un certain nombre de transformations orthogénétiques souvent identiques, ce qui a prog un grand apaibie adaptations convergentes ou paralléles. “(c) Les Mesidoteini littoraux septentrionaux sont d'une autre lignée que — Glyptonotus littoral austral avec lequel ils ont été réunis a tort. “(d) Cest également & tort que OHLIN rapproche son genre subantarctique littoral Macrochiridothea de Chiridotea littoral subarctique.” So far as I am aware, they are the only authors who have expressed an opinion upon the peculiarly modified uropoda, stating : “La transformation des uropodes en appareil protecteur des pléopodes, qui caractérise les Valvifera, doit étre fort ancienne puisque Proidotea est pourvue de ‘valves’ typiques. Il n’est pas possible de savoir actuellement si cette disposition a été acquise d’une fagon indépendante par les différents valviferes ou si elle est héritage d’un commun ancétre. Quoi qu il en soit elle a di s’effectuer de la facon suivante en partant d’une (ou plusieurs formes) & uropodes droits pourvus de protopodites plus petits que les uropodites : Habitude de replier les uropodes en dedans—augmentation progressive des proto- podites avec réduction consécutive des uropodites—allongement du pléotelson et de luropode—réduction et disparition de l’exopodite inutile précédent la réduction puis la disparition de l’endopodite qui seul joue d’abord un réle protecteur.” 7 IsseL (35) in 1913 showed that Dana’s genus Cleantis was synonymous with A REVISION OF THE BRITISH IDOTEIDA. Pont Zenobiana, Stebbing. Bare and Westwoop were evidently of this opinion, for writing in 1867 (3) they stated: “This species [Idotea parallela = Zenobiana prismatica (Risso)] appears to us to be very closely allied to, if not identical with, the genus Cleantis of Dana.” | Mirrs (44) in his “ Revision” states that Leacn’s diagnosis of Stenosoma “ does not permit of its being used for a sectional designation. . . . Leptosoma, on the other hand, will include all Idoteze with a uniarticulate postabdomen.” This author, however, overlooked the fact that Risso’s name Leptosoma was preoccupied. Leacu’s original diagnosis of Stenosoma (41, p. 365) is as follows :—* Antenne exteriores corporis longitudine, articulo tertio quarto longiore. Corpus lineare.” He included in this genus two such dissimilar species as his hectica (= Idotea linearis (Pennant)) and acuminata (=the Stenosoma acuminatum, Leach). The fact that Say (65), when describing his filiformis (= Erichsonella filiformis), included it in this genus, Goutp (26) his Idotea wrrorata, and Dana (17) his Idotea gracilima, sufficiently illustrates how imperfect the diagnosis is. The name has been used with so many varied conceptions that, with Miers, I agree that it cannot be employed for any section or division of the family. Seeing that neither of the above names can be employed, | am proposing that of Synisoma, and give an emended diagnosis of the genus. Symsoma, it may be remarked, differs from the genus Hrichsonella of BENEDICT in the following characters :— 1. The flagellum of the antennze does not consist of only a single clavate joint. 2. The coxal plates are distinctly separated and visible dorsally on all excepting the first segment. As pointed out above, Racovirza and Srvastos have already separated the genera Mesidotea, Richardson, Chiridotea, Harger, and Proidotea, Rac. and Sey., and placed them in the sub-family Mesidoteine. The remaining genera fall into at least three or four further subfamilies, of which the Idoteinze of Mrers would contain the genus [dotea, Fabr., and allied genera. Barnarp (Ann. Sth. Afr. Mus., 1914, p.-203) has recently described the new genus Hngidotea, with the I. lobata of Mrsrs as the type, which species is charac- terised by the inner lobes of the first maxille having only two setose spines, the second maxillz having only two lobes, and no setose styles on the uropoda. The palps of the maxillipedes are five-jointed. LV. GEOGRAPHICAL DISTRIBUTION. Whilst members of the Idoteide are found in all parts of the world, they would _ seem to be more numerous in the temperate and colder seas than elsewhere. Many species are found on the shore and in shallow water, whilst others frequent moderate depths, and a few have been taken at considerable depths. Harerr (30) records 728 DR WALTER E. COLLINGE. Edotia nodulosa, Kroyer, at a depth of 190 fathoms, off Halifax, U.S.A., and Mesidotea entomon (Linn.) has been recorded from 60 fathoms. Some species, like J. baltica (Pallas), are cosmopolitan, whilst others have a very restricted distribution. The whole question of the distribution of the different genera would well repay further investigation. Zenobiana, Stebbing, and Synisoma, Clige. (=Stenosoma, Auctt.), would seem to be rare in northern waters around the British Isles, although Barz and Wesrt- woop (3) mention having received examples of S. acwminata (Leach) from Cumbrae. The latter genus has probably a southern distribution. It is recorded from numerous localities on the shores of the Mediterranean, Adriatic, and Black Seas, and also from the southern coast of England. Of our British species of this family, the majority of the species of [dotea are common round the coasts of the British Islands. J. metallica, Bosc, and JI. viridis (Slabber) are perhaps rarer than the remainder. J. sarsv has, so far, only been received from St Andrews Bay. I have elsewhere (15) pointed out that of the eight previously recorded British species of Idotea, seven have been found in St Andrews Bay, viz. :— I. baltica (Pallas); I. pelagica, Leach; I. neglecta, G..O. Sars; I. emarginata, Fabr.; I. granulosa, Rathke; I. viridis (Slabber); I. linearis (Pennant). V. VARIATION. : Apart from the colour markings, which are endless, and slight variations in the mandibles and first and second maxille, very few variations have been met with in the large number of specimens I have examined. In I. baltica (Pallas) Bare and Westwoop (8, p. 381) give figures of variations in the form of the terminal segment of the body. Some of these are undoubtedly of young examples, or possibly were not referable to this species. In the form of the antenne, apart from the number of joints in the flagellum, and in the shape of the terminal segment, I have found very few variations indeed, and Sars (64, p. 81) states that he has found the form of the terminal segment of the metasome pretty constant, even in young specimens. According to CuiiTon (11), the segmentation of the metasome is liable to vary in a species. As an instance he cites the difference in Idotea peronw, M.-Edw., and I. stricta, Dana, and unites the two species. Miers (44) was of opinion that there were several distinct geographical sub- species or varieties of J. baltica, and I fully concur with this view. Thus in St Andrews Bay the ordinary large form occurs in great abundance, exhibiting innumerable colour variations ; but there is also a very much rarer and smaller form, which is found only, so far as my observations go, on the alge in the rock pools on the east shore. It is by no means common, for diligent search has so far only resulted in obtaining eight specimens. A REVISION OF THE BRITISH IDOTEID. 729 From the Mediterranean I have the same species, but in all the specimens the tridentate terminal segment of the metasome is less pronounced, and the mesosome in both sexes is comparatively wider. In examples of J. linearis (Pennant) received from Plymouth, eight specimens all show a depressed area on the ventral side of the fifth peduncular joint of the antenne, and in one example of the same species, received from the Bay of Nigg, Aberdeen, the left antenna has six peduncular joints. There appears to be a considerable amount of variation in the size of the different ‘species according to locality. Of the species treated of in this paper I give below a list, and the measurements given by different authors. The first line of figures refers to the length of the male, and the second to that of the female, where two are quoted. | Authority. Species. a | 6 iS a = ES Pe Q = Z Fe | oe F 3 aa] 2 (eq lal Z ah | 2 i a | f 5 I B A 3 eee eeciece a tee eg | 2 pai WS Pe lca clea ce vee | 4 o) / in. |mm.| mm, |mm.mm.} mm. |mm.) mm, | mm. | mm. 1. I, baltica (Pallas) . , ; . 1} | 30 | 30-38] 35 | 20 35 Sat 35 : | 90 | 20 2. 1. pelagica, Leach : : ool © Sha Pease REO bie + 13 13-14 | 9 | 9 9-10 3. I. neglecta, G. O. Sars . ; coil) hewn Th cee aes eemant [me AOBN eerevens |be=rero|| a8) 27°5 16 Oi 6 16-17 4. Iemarginata (Fabr.) . . lt AX | 8b Seales 30 30 18 ie its} 20 5. I. granulosa, Rathke . : wilh eis, ol uetsoes |e ema aA | 15 19-21 ode 11 11 16 6. I. viridis (Slabber) : . a Weeteam| | c ee 12 12 12 10 10 10 7. I. metallica, Bose ; ; Gl ieee meetelent ee | PS TS -S1 |... 23 Be 22 . 18-24 ; 18 8. I. linearis(Pennant) . . jee toes | S| aoe + | 28-38 25 9. Zenobiana prismatica (Risso) PW |e in Scrat | eco ae Ie 12 10-13] 13°5 10. Synisoma lancifer (Leach) 20 22°5 11, Synisoma acuminata (Leach). .| ... | 25 15 | VI. Exrernat Structure, The form and structure of the exoskeleton of this family have only been very imperfectly described and figured. The differences in a given species in the descrip- tions and figures of different authors are difficult to account for. Thus in such a cosmopolitan form as Idotea baltica (Pallas), Sars (64) states that there are sixteen TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 23). 107 730 DR WALTER E. COLLINGE. to twenty articulations in the flagellum of the antennz and figures eighteen. BaTE and Westwoop (3) state about twenty. Miss RicHarpson (58) gives fourteen as the number, whilst Harcer (30) states twelve to sixteen, and shows seventeen in his figure (pl. v, fig. 250). Both Miss Ricoarpson and Harcer in their figures of the maxillipedes show the basal segment, to which the four-jointed palp is attached, as one piece. Bare and Westwoop show it with a short lateral process on the inner side, but only figure three joints in the palp. Sars shows the joint and process, but the terminal portion is scarcely correct. I have elsewhere shown (13) that the oral appendages in the Isopoda are liable to considerable variation, but not to the extent of the fusion or total absence of the different parts. Numerous other instances might be quoted, for similar variations are to be found in the figures of the first and second maxille, the shape of the cephalon, the coxal plates, and the uropoda, but these are sufficient to indicate how imperfectly described and figured the various species are. “ No author, so far as I am aware, either describes or figures the groove on the third joint of the palp of the maxillipede. A certain amount of difference may be due to figuring and describing immature specimens (see p. 739), in other cases the differences are possibly due to differences in the sexes, many authors not stating % whether the examples were male or female; be this as it may, the need of more 1 accurate figures must have been felt by everyone who has studied the family. (a) General Form. The body is more or less broad and flattened, the various genera and species ranging from an oblong oval to an elongated linear shape. There is considerable range in the size of the different genera and species. The largest member of the family is Glyptonotus antarcticus, Hights, which attains a length of 90 mm., and the smallest is Idotea dana, Miers, which is only 2 mm. long. The males in practically all cases are larger and more robust than the females. (b) The Cephalon. Excepting in Chiridotea, Harger, the cephalon is quadrate in outline. In Idotea baltica (Pallas), I. neglecta, G. O. Sars, and I. pelagica, Leach, it is cleft on the posterior lateral margins, and in many species there is a more or less distinct sinuous line towards the posterior margin. 1. The Kyes.—Paired compound eyes are present in all the genera of the family. In most species they are large, and situated dorsally or dorso-laterally. In a few species they are quite small. In no genera or species of the family are they known to be absent. 2. Antennule. These are composed of a three-jointed peduncle with the flagellum A REVISION OF THE BRITISH IDOTEID . 731 consisting of a single clavate joint. On the dorsal surface of this joint are a varying number of groups of sensory setze, probably of an olfactory nature. 3. Antenne.—There are five peduncular joints in these appendages, and usually a multiarticulate flagellum. In EHvichsonella, Benedict, Husymmerus, Richardson, Zenobiana, Stebbing, and Symmius, Richardson, the joints are united to form a single piece. In Hdotia, Guérin-Ménéville, the antenne are rudimentary, and the flagellum consists of a single joint. Terminally there is a more or less conical style with a number of long fine setz at the apex. ( 4. Oral Appendages.—These include the upper and lower lips, the mandibles, the first and second maxille, and the maxillipedes. The lips have been regarded by most authors as inteeumentary modifications rather than true appendages, although Miss Arms “ maintains that the leaves of the lower lip ‘are independent outgrowths or buds from the integument, as much as any other pair of appendages; and the fact that the parts of the segment to which they must have belonged have disappeared, or cannot be readily found is, in her opinion, ‘an argument of doubtful weight’ ” (66). The upper lip is plate-like, and projects from the dorsal border of the mouth over the mandibles; a second and inner plate is present in some Isopoda. The lower lip is bilobed, and consists of a small inner pair of lobes and an outer larger pair. The mandibles are without palps; each consists of a stout robust body with a bidenticulated apical portion, below which is a broad truncated molar tubercle. There is considerable variation in the form of the teeth, according to age, ete. The first maxille each consist of a short robust outer lobe terminating in a variable number of stout curved spines, and an inner series of more slender ones, which may be simple or toothed. The inner lobe is a thin, flat, narrow chitinous band with a series of, usually setose, spines terminally. In Jdotea, Fabricius, Mes- idotea, Richardson, Glyptidotea, Stebbing, Zenobiana, Stebbing, Paridotea, Stebbing, Pentias, Richardson, and Synisoma, Collinge, there are three of these spines; in Synidotea, Harger, and Engidotea, Barnard, only two.* The second maxille each arise from the protopodite as two flat laminar lobes, the innermost of which is bilobed ; all of them are fringed with long sete. The maxillipedes each consist of a coxopodite which is divided into two, a basal segment or basiopodite composed of a single joint; articulating with this distally is the palp-or endopodite, on the outer side is an epipodite, whilst on the inner side is a short joint articulating with the basal seement, which joint is provided with one or more peculiar handle-shaped appendages known as the coupling processes. In Chiridotea, Harger, Synidotea, Harger, Edotia, Guérin-Mén., Chiriscus, _ Richardson, Symmius, Richardson, and Macrochiridothea, Ohlin, the palp is three- jointed; in Idotea, Fabricius, Colidotea, Richardson, Husymmerus, Richardson, and Hrichsonella, Benedict, four-jointed; in Mesidotea, Richardson, Pentidotea, * In an Australian form (genera and species as yet undetermined) there are four spines present. Partdotea ungu- lata also has four. 732 DR WALTER E. COLLINGE. Richardson, Paridotea, Stebbing, Zenobiana, Stebbing, Gilyptidotea, Stebbing, Cleantiella, Richardson, Engidotea, Barnard, and Pentias, Richardson, five-jointed ; and in Glyptonotus, Hights, the number varies from three to five joints. (c) The Mesosome. The mesosome or thorax consists of seven free segments articulating with each other. An exception to this is found in Crabyzos longicaudatus, 8. Bates (2), where the cephalon is partially coalescent with the first mesomatic segment. The first segment is usually narrower than the rest, and partly surrounds the cephalon. The pleural plates may be well developed, as in I. baltica, I. neglecta, and I. emarginata, or very small, as in J. linearis. 1. Coal Plates—In most members of this family the coxopodites of the appen- dages are developed to such an extent as to protrude beyond the pleural plates or overlap or replace them on the dorsal surface. Many authors have used the term “epimera” for these, but, as CALMAN (10) points out, that term has been used in so many different senses, that it seems better to abandon it altogether. In our British species the greatest development of these plates is seen in I. baltica, I. neglecta, and I. emarginata, and they are smallest in J. linearis and Zenobiana prismatica, and occupy an intermediate position in the members of the genus Synisoma. In I. hectica (Pallas), an Atlantic and Mediterranean species, they do not protrude beyond the ventral surface of the pleural plates. 2. Appendages.—Hach segment carries a pair of appendages, and each appendage is composed of seven joints. The coxopodite and basiopodite may be very small, whilst in the first and second appendages the distal joints are usually shorter than in the remaining ones. In many species the anterior appendages are smaller than the posterior ones; on the other hand, in Crabyzos longicaudata (S. Bate) the first pair are said to be the largest. The dactylopodite is chelate and unguiform. All the appendages are provided with sete#, sometimes of three or four kinds, and all are ambulatory in character. In the females the third and the three following appendages each have at their base a large membranous plate folded beneath the body to form an ovigerous sac or marsupial chamber. (d) The Metasome. The metasome in the Idoteide has undergone considerable modification by a consolidation of the terminal segments to form a strong dorsal shield. The number of segments varies in the different genera: thus in Glyptonotus, Kights, there are four or five segments; in Mesidotea, Richardson, Chiridotea, Harger, four; in Idotea, Fabricius, Chiriscus, Richardson, and Pentidotea, Richard- son, three are present; in Cleantiella, Richardson, there are two segments; in Zenobiana, Stebbing, three to five may persist; whilst in Synedotea, Harger, A REVISION OF THE BRITISH IDOTEID. 738 Colidotea, Richardson, Kdotia, Guérin-Mén., Crabyzos, 8. Bate, Husymmerus, Richardson, Hrichsonella, Benedict, Pentias, Richardson, Paridotea, Stebbing, and Synisoma, Collinge, the fusion is complete, the segments being represented by a single piece. In nearly all cases there are evidences of one or more suture lines indicating a further segment or segments. 1. Appendages.—There are five pairs of appendages, each of which consists of a basal segment carrying two elongate-oval lamelle fringed with long sete. The inner lamelle of the second pair, and of the first pair also, in Pseudidothea, Ohlin, in the male bears a fine stylet. 2. Uropoda.—This pair of appendages differs in this family, in common with other members of the suborder Valvifera, from the condition obtaining in any other Isopoda. Here they consist of a pair of somewhat flattened, opercular-like plates (the protopodites) closing within the metasomatic appendages, by meeting in the mid-ventral line or overlapping one another. Hach is attached to the metasome laterally. Posteriorly there is in most genera a hinge dividing off a posterior plate (the endopodite). In some genera, e.g. Mesidotea, Richardson, Chiridotea, Harger, Glyptonotus, Hights, and Macrochiridothea, Ohlin, there is an inner plate to each uropod (the exopodite). Articulating laterally on the outer side, or dorso-laterally, there is usually a plumose style. Sreppine (70) states that this is absent in Paridotea; and in Symmius, Richardson, the endopodite is not present, each uropod consisting of a single piece. VII. Bronomics anp Economics. Very little is known as to the life-history and habits of the family of Isopods here treated of. Roux (63) has given some interesting details of their habits, and Issex (34) has studied the tube-dwelling species of the genus Zenobiana. Whilst the family contains some of the largest known Isopods, some are quite small. Many species are exceedingly plentiful. Most of them seem to prefer decaying masses of algee in rock pools or still water, where they feed upon dead fishes, molluses, crustaceans, annelids, ete. Although the majority of species are marine, Loven has recorded Mesidotea entomon from deep Scandinavian lakes, and THomson (77) has described a species of Idotea (lacustris) from a fresh-water lagoon in New Zealand. Roux remarks that when the sea is rough they seek deep water, and are often thrown back by the action of the waves. This is so in St Andrews Bay, where, after a rough sea, large numbers of the commoner species are to be found clinging to the algz in the rock pools. Idotea baltica progresses by a series of short, jerky jumps, whereas J. linearis moves less rapidly, often apparently floating with the uropoda open and hanging 734 DR WALTER E. COLLINGE. perpendicular to the body. Most of the species are nocturnal, hiding under alge during the day. Jdotea linearis is fairly active during the daytime, gliding grace- fully through the water even when strong sunlight was shining on the vessel ‘in which it was contained.. In the rock pools also I have noticed this species active during the daytime. toux records that in the Mediterranean the fishermen collect these Isopods and use them as bait. The method of collecting is by tying together bundles of fucus, which are sunk below the water; on these the Idotes assemble, and from time to time the masses of alge are drawn up and the specimens collected. Fishes would seem to be the chief enemies of these crustacea. Mr W. Ramsay SmirH, in his papers on the food of fishes,* has given many records of their occurrence in the stomachs of various species, as has also Dr Tuomas Scorr in his “Observations on the Food of Fishes.” | I have records of the occurrence of all the British species of Idotea, excepting viridis and metallica, from the stomachs of the cod, haddock, and plaice. MarzporrF (43) has given a most interesting account of the colour variation in. I. baltica. The remarkable variability of this species has been frequently com- mented upon. He enumerates and figures a number of colour varieties, and shows that they are common to both sexes. The same individual changes colour by the expansion or contraction of dark brown ehromatophorous cells, adapting itself to its surroundings. Thus specimens found near the surface, on the top of alge, were light-coloured, while those on the bottom, in the same locality, were dark-coloured. After the extirpation of the eyes the animals cease to undergo these colour changes. VIII. Systematic. Many genera of doubtful value have been made by different authors, often upon very trivial differences. Dana (16), 1852, divided the family into the following five genera :—Idotea, Fabr. ; Hdotia, Guérin-Mén.; Hrichsona, Dana; Cleantis, Dana; and Hpelys, Dana. Barr and Westwoop (3), 1867, included all the British species in the genus Idotea, Fabr. Mrers (44), 1881, gives the following divisions :— Fam. IDOTEIDA, Dana. Sub-fam. 1. GLYPTONOTIN. Genus Glyptonotus, Kights. Sub-fam. i. [pore a. Genera Idotea, Fabr.; Hdotia, Guérin-Mén. ; and Cleantis, Dana. Harcer (30) included the five genera :—Chiridotea, Harger; Idotea, Fabr. ; Synidotea, Harger; Erichsonia, Dana; and Hpelys, Dana. StTEBBING (68) recognised * 7th, 8th, 9th, and 10th Reports of the Fishery Board of Scotland. + Tiid., 20th Report, 1902, pp. 486-538. A REVISION OF THE BRITISH IDOTEIDA. 735 the following genera:—Glyptonotus, Eights; Chetilia, Dana; Arcturides, Studer ; Idotea, Fabr.; Edotia, Guérin-Mén.; and Cleantis, Dana. Norman (47), 1904, mentions only the three British genera, whilst RicHarpson (58) in 1905 arranged the genera as follows :—Mesidotea, Richardson ; Chiridotea, Harger ; Idothea, Fabr. ; Pentidotea, Richardson ; Synidotea, Harger ; Colidotea, Richardson ; Hdotia, Guérin- Mén.; Husymmerus, Richardson ; Hrichsonella, Benedict ; and Cleantis, Dana. In the present communication I propose to divide the family as follows :— Fam. IDOTEIDA, Leach, 1813. * Sub-fam. i. Iporrrna, Miers, 1881. Genus 1. Idotea, Fabr., 1798. (1) (2) ,, pelagica, Leach. (3) ,, neglecta, G. O. Sars. (4) ,, . emarginata (Fabricius). (5) 4, granulosa, Rathke. (6) ,, Sarsi, n. sp. (7) ,, viridis (Slabber). (8) ,, metallica, Bose. (9) ,, linearis (Pennant). Sub-fam. ii. Genus 2. Zenobiana, Stebbing, 1895. (1) Zenobiana prismatica (Risso). Sub-fam. iii. Genus 3. Synisoma, nov. nom. (1) Synisoma lancifer (Leach, MSS8.). (2) % acuminata (Leach). Fam. Iporerp#, Leach, 1813. Body ovate, oblong or elongate, and more or less depressed. Cephalon and mesosomatic segments distinct; coxal plates separate from or confluent with the segments ; metasome with some or all of the segments consolidated into a terminal seutiform piece. Hyes usually dorso-lateral, occasionally dorsal, multifaceted. Antennules with three-jointed peduncle, flagellum uniarticulate. Antenne with five-jointed peduncle, flagellum usually multiarticulate, but may consist of a single joint or may be short or rudimentary. Mandibles without palps. First maxillee * Given on the authority of Agassiz, Nomenclator Zoologicus, 736 DR WALTER E. COLLINGE. with outer lobe terminating in a series of stout, curved spines, inner lobe terminating in two to four setose spines. Second maxillee each consist of three lobes fringed with setae, Maxillipedes operculiform, palp with three to five joints. Mesosomatic seg- ments, excepting the first, usually about equal in length, coxal plates distinct or coalesced with the segments. Thoracic appendages uniform in structure, ambulatory, varying in size from before backwards, usually stronger in male than in female, occa- sionally approaching a subcheliform character ; dactylus strong, unguiform, unequally bi-dentate. Metasome with some or all of the segments fused to form a large terminal scute-like segment. Metasomatic appendages laminate. Uropoda inferior, arching over the ventral side of the metasome and enclosing the appendages. Key To THE GENERA OF BritisH I[DOTEIDA. 1. Body ovate, oblong, or elongate, more or less depressed. Antennze with multiarticulate flagellum. Maxillipedes with four-jointed palp. Coxal plates of the mesosomatic segments, excepting the first, distinctly separated. | Metasome composed of three segments. Idotea, Fabricius. 2. Body somewhat slender, long and narrow, with parallel sides. Antenne short, with single-jointed flagellum, or with restricted number of joints. Maxillipedes with five-jointed palp. Coxal plates of the mesosomatic segments distinct, but small and narrow. Metasome composed of three to five segments. Zenobiana, Stebbing. 3. Body oblong, moderately convex, usually with indications of a keel. Antenne with multiarticulate flagellum. Maxillipedes with four-jointed palp. Coxal plates small, but distinct. Metasome composed of a single segment. Synisoma, Collinge. |. Genus Idotea, Fabricius. Idotea, Fabricius, J. C., Suppl. Entom, Syst., 1798, p. 302. Body ovate, oblong or elongate, more or less depressed. Cephalon subquadrate, not expanded laterally, sometimes with lateral clefts or posterior sinuous furrow. Coxal plates of mesosome usually well defined on the second to seventh segments. Kyes dorso-lateral. Antennule with short, clavate flagellum. Antenne with —~ A REVISION OF THE BRITISH IDOTEID. 737 elongated multiarticulate flagellum. Maxillipedes operculiform, palps (endopodites) with four joints, basipodite expanded. Metasome composed of three distinct segments and lateral suture lines in front of the long terminal segment. Uropoda laminar, with basal joint and setose spine. Male usually larger than female. Key vo tHe British SPECIES OF THE GENUS [DOTEA. I. Body oblong ovate. 1. Coxal plates wide. a. Metasome terminally produced in the middle and extending beyond the lateral angles. I. baltica (Pallas). b. Metasome terminally broad and rounded. I, pelagica, Leach. I. neglecta, G. O. Sars. -] c. Metasome terminally marginate. I. emarginata (Fabr.). 2. Coxal plates rather narrow. d. Metasome terminally drawn out and pointed. I. granulosa, Rathke. I. sarsi, Collinge. I. viridis (Slabber). e. Metasome terminally more or less truncate. I. metallica, Bose. II. Body elongated, filiform. 1. Coxal plates very small. a. Metasome terminally emarginate. I. linearis (Pennant). (1) Idotea baltica (Pallas). (Pl. I, figs. 1-14.) Oniscus baliicus, Pallas, Spite. Zool., 1772, vol. ix, p. 67, pl. iv, fig. 6. Idotea entomon, Pennant, Brit. Zool., 1777, vol. iv, p. 38, pl. xviii, fig. 5. Stenosoma trrorata, Say, Journ. Acad. Nat. Sci. Phila., 1818, vol. 1, p. 423. Idotea tridentata, Rathke, Fauna Norw., 1843, t. xx, p. 21. Idotea tricuspidata, Desmarest, Dict. des Sci. Nat., 1823, t. xxviii, p. 373, pl. xlvi; Roux, Crust. de la Medit., 1828, pl. xxix, figs. 11, 12; Milne-Edwards, Hist. Nat. Crust., 1840, vol. iii, p. 129; Sars, Forh. Vidensk. Selsk, Christ., 1859, p. 151; Norman, Rep. Brit. Assoc., 1866, p. 197; Bate and Westwood, Brit. Sessile-eyed Crust., 1868, vol. ii, p. 379. Idotea irrorata, Harger and Verrill, Rep. U.S. Fish Comm., 1873, pt. i, p. 569, pl. v, fig. 23; Verrill, Am. Journ, Sci. Arts, 1874, vol. vii, pp. 131, 134; Harger, Proc. U.S. Nat. Mus., 1879, vol. ii, p. 160. Idotea tricuspidata, Stebbing, Journ. Linn. Soc. Lond. (Zool.), 1876, vol. xii, p. 148; Dollfus, Feuille des jeunes Nat., 1895, p. 55. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 23). 108 738 DR WALTER E. COLLINGE. Idotea marina, Miers, Journ. Linn. Soc. Lond. (Zool.), 1881, vol. xvi, p. 25. Idothea baltica, G. O. Sars, Crust. Norw., 1897, vol. ii, p. 80, pl. xxxii; Richardson, Bull. 54, US. Nat. Mus., 1905, p. 364 ; Tattersall, Nord. Plank., 1911, p. 219, figs. 83-87. Idotea balihica, Norman, Ann. and Mag. Nat. Hist., 1904 (8. 7), vol. xiv, p. 441. Body oblong oval, moderately convex, dorsal surface perfectly smooth. Cephalon (fig. 1) wider than long, slightly emarginate anteriorly, posterior margin almost straight, posteriorly with lateral clefts continuing inwardly as dorsal ridges. Hyes large, round, situated dorso-laterally, slightly anterior to the median transverse line. Antennulee (fig. 2) extending to the end of the third peduncular joint of the antenne, first joint expanded, irregularly angular, second joint broad but rather smaller, third joint elongated ; flagellum single-jointed, clavate, with nine or ten groups of spatulate setee. Antenne (figs. 3 and 4) elongated, first peduncular joint very short, second longer and deeply incised ventrally, third and fourth increasing in length, but decreasing in diameter; flagellum normally with twenty joints, of which the first ten or eleven are somewhat cuboid, remainder elongated and slightly clavate, with terminal style and long sete (fig. 3). First maxille (fig. 5), outer lobe terminates in ten short curved spines and a single fine-pointed one, inner lobe narrow, terminating in three setose spines. Second maxille consisting of three laminar setose lobes. The segments of the mesosome, excepting the first, which is shorter, are about equal in length. Coxal plates large, occupying the whole of the lateral margins of second to seventh segments, gradually increasing in width (fig. 12). Maxillipedes (fig. 6) with four-jointed palp, the third joint grooved for the reception of the postero-lateral border of the first (fig. 7), basal segment large, with inner lobe distally, epipodite oblong oval, extending as far as the third joint of the palp. Thoracic appendages (figs. 8-11) more or less alike in structure, with few stout sete, first appendage short and stout, remainder gradually becoming larger from the second to the seventh. Metasome (fig. 13) has two short segments and indications of a coalesced third, terminal segment long, slightly convex, with the posterior margin tridentate, median tooth produced as an acute point, extending beyond the lateral ones. Uropoda (fig. 14) flattened, elongated, rounded anteriorly, sides nearly parallel, inner margins sloping obliquely towards the median line; articulating with the basal plate on its dorsal side is a short setose style; endopodite flat, posterior margin narrowed and slightly emarginate. Length of ¢ 35 mm., of ? 20 mm. Colour exceedingly variable. Geographical Distribution.—Common all round our coasts. Amongst many others I have the following records :—Shetland, Moray Firth, St Andrews, Firth of Forth, Yorkshire, Norfolk, Suffolk, and Essex coasts, Isle of Wight and South coast, Devon coast (west and south), Llandudno, Blacksod Bay (Co. Mayo), Irish Sea, St Kilda, Firth of Clyde, and Sutherland coast. It is also recorded as occurring on the Scandinavian and European coasts from Holland tothe Mediterranean and Black A REVISION OF THE BRITISH IDOTEIDA. 739 — Seas, on the Atlantic coast of North America (Harcer), West Indies (RicHarpson), Brazil, Java, and Red Sea (Mrers), and from New Zealand (CHILTON). Remarks.—This species is easily distinguished from all other members of the genus by the characters exhibited in the antennules, antenne, cephalon, the meso- somatic segments, in the tridentate character of the posterior margin of the last metasomatic segment, and the form of the uropoda. Minor characters are to be found in the form of the first maxille, the maxilli- pedes, and the thoracic appendages. Young examples, measuring 2°5 mm. in length, found on the metasomatic appendages beneath the uropoda, strongly resemble the parents, excepting in the form of the antennules, antenn, and the terminal segment. The maxillipedes of this species are larger and better developed than in any other British member of the genus. On the anterior border of the third joint of the palp there is a deep groove (Pl. I, fig. 7) in which the thickened posterior border of the fourth joint moves. In examining the antennze of this and the other species here enumerated, I have found the shape of the terminal style exceedingly constant, and have therefore given figures of it in all the species. The actual number of joints in the flagellum varies according to age. I have endeavoured to figure those in what I consider to be fully adult specimens. In a young specimen of J. baltica, measuring 2°5 mm. in length, the flagellum was composed of only three joints and a style. | (2) Idotea pelagica, Leach. (Pl. II, figs. 15-25.) Idotea pelagica, Leach, Trans. Linn. ‘Soc. Lond., 1815, vol. xi, p. 365; Bate and Westwood, Brot. Sessile - eyed Crust., 1868, vol. ii, p. 384, figs.; Dollfus, Mewille des jeunes Nat., 1895, p. 8, fig. 23. Tdothea pelagica, G. O. Sars, Crust. Norw., 1897, vol. i, p. 81, pl. xxxiil. Idotea pelagica, Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), vol. xiv, p. 442; Tattersall, Mord. Plank., 1911, vol. vi, p. 220, figs. 88-93. Body comparatively short and stout. Cephalon (fig. 15) wider than long, anterior margin slightly concave, posterior margin straight, with sinuous line in front. Hyes comparatively large, rounded, situated anterior to the median transverse line. An- tennule (fig. 16) small, first joint widely expanded, second less so, and third ex- panded distally. Antenne (figs. 17 and 18) short and stout, extending just beyond the first segment of the mesosome, first three joints almost subequal, as also fourth and fifth ; flagellum short, composed of eight joints, terminal style short, with three or four stout sete. First maxille (fig. 19), outer lobe terminating in seven stout curved spines, a single fine-pointed one, and two toothed ones, inner lobe with three setose spines. The segments of the mesosome (fig. 23) are about equal in length, anterior angles of the pleural plates of the first segment bluntly pointed. Coxal 740 DR WALTER E. COLLINGE. plates occupy the whole of the lateral margins of the second to seventh segments, widening from before backwards. Maxillipedes (fig. 20) with elongated basal plate and epipodite, distal inner lobe fairly large, no groove on the third joint of the palp. Thoracic appendages (figs. 21 and 22) short and strongly built.* Metasome (fig. 24) with two short segments and well-marked lateral sutures indicating a further coalesced one, terminal segment with slight median ridge, lateral margins converg- ing slightly to an obtuse point. Uropoda (fig. 25) rather broad, posterior margin obliquely transverse, outer margin slightly curved inwards, endopodite terminally obliquely truncate, outer margin converging a little towards the distal end, setose style short. Length of ¢ 13-14 mm., of P 9-10 mm. Colour dark brownish. Geographical Distribution.—Not a plentiful species in St Andrews Bay. I have examples from Lowestoft, Boscombe, and Plymouth. Recorded from the Firth of Forth; Bell Rock (Bare and Westwoop); Aberdeen (Scorr); Firth of Clyde; Port Erin Bay (Hewirr) and South-West Ireland (Normay) : also coasts of Norway (G. O. Sars) and France (BoNNIER). Remarks.—Although for long confused with I. baltica (Pallas), this is a very distinct species, and there is no difticulty in at once separating both sexes from the former species. (3) Idotea neglecta, G. O. Sars. (PI. III, figs. 26-36.) Idotea marina, Dollfus, Feuille des jeunes Nat., 1895, p. 7, fig. 22. Idothea neglecta, G. O. Sars, Crust. Norw., 1897, vol. ii, p. 84, pl. xxxv, fig. 1. Idotea neglecta, Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), vol. xiv, p. 442; ‘Tattersall, Mord. Plank., 1911, p. 225, figs. 106-110. Body oblong oval, wider anteriorly than posteriorly. Cephalon (fig. 26) wider than long, shehtly emarginate anteriorly, posterior margin almost straight, pos- teriorly with slght lateral clefts and dorsal ridges. Hyes rather large, situated dorso-laterally, anterior to the median transverse line. Antennule (fig. 27) extend- ing to the end of the third peduncular joint of the antenne, first and second joints expanded, third elongated ; flagellum single, rather long, clavate joint. Antenna (figs. 28 and 29) of moderate length, peduncular joints gradually increasing in length from the first to the fifth; flagellum composed of eighteen to twenty joints and short setose style. First maxillee (fig. 30), outer lobe terminating in eight stout curved spines and three toothed ones; inner lobe narrow, terminating in three setose spines and a short spine on the outer side distally. The segments of the mesosome (fig. 34) almost subequal, anterior angle of the first pleural plates directed forwards, obtusely pointed. Coxal plates of male occupy the whole of the lateral * In the male there is usually a dense fringe of fine setze bordering the inside of the outer joints ; in’some cases, however, these were not present, and no difference was observable from the condition obtaining in the female. A REVISION OF THE BRITISH IDOTEIDA. 741 margins of the second to seventh segments, gradually increasing in width, smaller in female. Maxillipedes (fig. 31) with four-jointed palp, groove on the third joint very slight, basal plate elongated, with inner lobe fairly broad, epipodite extending to the middle of the third joint of the palp. Thoracic appendages (figs. 32 and 33) moderately strong. Metasome (fig. 35) with two short segments and sutural indications of a third coalesced segment, terminal segment long, raised in the mid- dorsal line, slightly narrowing distally, terminating in an obtuse point, lateral angles rounded. Uropoda (fig. 36) broad, bluntly rounded anteriorly, with raised inner margin, posterior margin obliquely excavate, endopodite terminally transversely truncated. Length of ¢ 27°5 mm., of ? 16-17 mm. Colour dark brown or greenish-black, sometimes variegated with irregular lighter-coloured patches. Geographical Distribution.—St Andrews Bay, Firth of Forth, Lowestoft, and Plymouth ; Clyde Area (Patience); Moray Firth (Scorr); Shetland and Falmouth (Norman); west coast of Ireland (TatrersaL) : also Norwegian coast (NorMaAN and Sars). (4) Idotea emarginata (Fabr.). (Pl. VI, figs. 37-47.) Cymothoa emurginata, Fabricius, Entom. Syst., 1793, vol. 11, p. 508. Idotea emarginata, Fabricius, Entom. Syst. Suppl., 1798, p. 303; Latreille, Hist. Nat. Crust. et Ins., 1803, vol. vi, p. 370; Milne-Edwards, Hist. Nat. Crust., 1840, vol. ii, p. 130; Bate and Westwood, Brit. Sessile-eyed Crust., 1868, vol. ii, p. 386, figs.; Miers, Journ, Linn. Soc. Lond., 1881, vol. xvi, p. 43. Idotea cestrum, Leach, 7'rans. Linn. Soc. Lond., 1815, vol. xi, p. 8365; Desmarest, Consid. Crust., 1825, p. 289. Idothea emarginata, G. O. Sars, Crust. Norw., 1897, vol. ui, p. 85, pl. xxxv, fig. 2. Idotea emarginata, Vollfus, Feuille des jeunes Nat., 1895, p. 6, figs. 17,18; Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), vol. xiv, p. 443; Tattersall, Nord. Plank., 1911, p. 226, figs. 111-115. Body oblong oval, slightly convex, surface smooth. Cephalon (fig. 37) wider than long, anterior margin slightly emarginate, posteriorly with sinuous line. Eyes large and-round, situated dorso-laterally. Antennuleze (fig. 38) extending to the end of the third peduncular joint of the antenne; first joint much expanded and produced on the inner side, second smaller, third elongated. Antenne (figs. 39 and 40) long, first three joints excavated on their outer side, fourth and fifth elongated ; flagellum composed of from fourteen to sixteen joints, with elongate setose style. First maxille (fig. 41) with outer lobe terminating in eight stout curved spines and two toothed ones, inner lobe with three elongated setose spines, rather wide at their bases, and with small setule distally at the outer border. The segments of the mesosome (fig. 45) almost subequal, anterior angle of the pleural plates of first segment sharply pointed. Coxal plates wide and occupying the whole of the lateral margins of the second to seventh segments, posterior angle of the seventh produced backwardly as sharp spine. Maxillipedes (fig. 42) elongated 742 DR WALTER E. COLLINGE. and somewhat narrow, basal plate and epipodite long, groove on the third joint of the palp only small. Thoracic appendages (figs. 43 and 44) stout. Metasome (fig. 46) with two short segments and lateral indications of a further coalesced one, terminal segment broad, slightly convex, lateral margins curved, posterior margin slightly emarginate, with truncate median portion, lateral angles projecting. Uropoda (fig. 47) with almost parallel sides, posterior margin slightly emarginate, endopodite with straight internal lateral margin, externally sloping, posterior margin emarginate. ; Length of ¢ 30 mm., of ? 20 mm. Colour variable. Geographical Distribution.—Common all round our coasts. I have records from numerous localities, amongst which may be mentioned the following :—Wick, Caith- ness, Moray Firth, Aberdeen, St Andrews Bay, Firth of Forth, Northumberland and whole of the eastern coast, Sussex, Dorset, and Devon coasts, Welsh coast, Firth of Clyde, Firth of Lorn, Skye and Sutherland coast. Also recorded from Kattegat (MrtNeRT) and coasts of France and Spain. I have what I believe to be young examples of this species from the Mediterranean. Remarks.—This species stands out as distinct from any other member of the genus, its large size and robust form contrasting strongly with any other species, excepting I. baltica, from which it differs, however, in quite a number of struc- tural details. Some small, dark-coloured male forms which have been examined, at first sight bear a superficial resemblance to I. metallica, but in nearly all cases they have a slight median stripe running down the middle of the mesosomatic segments, which at once serves to distinguish them, as also the more convex mesosome, and in being wider anteriorly than posteriorly. Like I. baltica (Pallas) this species is subject to great variation in its colour markings. (5) Idotea granulosa, Rathke. (Pl. V, figs. 48-58.) Idotea granulosa, Rathke, Beitrage zur Fauna Norwegens, 1843, p. 23. Idotea phosphorea, Harger, Rep. U.S. Comms. Fish and Fisheries, 1874; part i, p. 569; Proc. U.S. Nat. Mus., 1879, vol. ii, p. 160; Rep. U.S. Comms. Fish and Fisheries, 1880, part vi, p. 347, pl. v, figs. 27-29. Idotea marina, var. phosphorea, Miers, Journ. Linn. Soc. Lond., 1881, vol. xvi, p. 31. Idothea granulosa, G. O. Sars, Crust. Norw., 1897, vol. ii. p. 82, pl. xxxiv, fig. 1. Idotea granulosa, Norman, Ann. and Mag. Nat. Hist,, 1904 (s. 7), vol. xiv, p. 441; Tattersall, Mord. Plank., 1911, p. 223, figs. 94-99. Idothea phosphorea, Richardson, Bull, No. 54, U.S. Nat. Mus., 1905, p. 367, 3 figs. Body oblong oval, moderately convex, dorsal surface finely granular or almost smooth. Cephalon (fig. 48) broader than long, with the anterior margin almost straight, posteriorly with transverse sinuous line. Hyes moderately large, situated just in front of the median transverse line on the lateral margin. Antennule A REVISION OF THE BRITISH IDOTEID A. 7438 (fig. 49) with first joint expanded, second less so, third elongated and somewhat expanded distally ; flagellum clavate. Antenne (figs. 50 and 51) stout, first three joints almost subequal, fourth longer, fifth longest; flagellum with sixteen joints. First maxille (fig. 52), outer lobe terminating in eleven curved spines, inner lobe with three setose spines. The segments of the mesosome (fig. 56) are almost sub- equal, except the first, the coxal plates occupy the anterior two-thirds of the lateral margin of the second segment, the third rather more, and the remainder the whole of the lateral margins, increasing in breadth from the fourth to the seventh segments. Maxillipedes (fig. 53) with four-jointed palp, groove on the third joint small, basal plate elongated, with narrow epipodite extending to the third joint of the palp; inner lobe also natrow. Thoracic appendages (figs. 54 and 55), first short and stout, gradually becoming larger from the second to the seventh. Metasome (fig. 57) composed of two narrow segments and lateral sutures indicating a coalesced third, the terminal segment tapering to a pointed extremity with the lateral angles rounded. Uropoda (fig. 58) flattened, elongated, rounded anteriorly, sides nearly parallel, inner margin posteriorly extended, exopodite bluntly pointed posteriorly, anterior margin convex, short setose style. Length of d 19 to 21 mm., of ? 16 mm. Colour (in alcohol) light reddish, or yellowish-brown with minute black dots. Geographical Distribution.—St Andrews Bay, Firth of Forth, Isle of Wight; Bay of Nigg (Scorr); Port Erin Bay (Hewirr) ; Northumberland, and Berehaven, Ireland (Normay) ; South and West Ireland (TarrErsa.t) : also Norwegian coast (G. O. Sars). Remarks.—The narrow lanceolate form of the terminal segment at once serves to distinguish this species.from any other member of the genus. Sars (64) states that the coxal plates are “comparatively small,” but in all the British specimens | have examined they are as figured (PI. V, fig. 56). Hxamples received from Professor Sars are rather smaller than ours, the largest being 18 mm. in length, but the coxal plates are, comparatively, only very slightly smaller on segments 2 and 3. That there is considerable variation in size in different localities is evident. Sars gives the length of the male and female as 15 and 11 mm. respec- tively, whereas those I have examined from this country average 19 and 16 mm. HarGer (32) states that the length of his 1. marina, var. phosphorea, was 25 mm. Although the dorsal surface is usually finely granulated, specimens with a perfectly smooth surface are by no means rare. (6) Idotea sarsi, n. sp. (Pl. VI, figs. 59-69.) Body oblong oval, moderately convex, dorsal surface rough. Cephalon (fig. 59) wider than long, anterior margin emarginate, posterior margin narrower than the _ anterior one. KHyes fairly large, situated dorso-laterally. Antennule (fig. 60) with first joint expanded, second joint small, third joint elongated ; flagellum single clavate joint, slender. Antenne (figs. 61 and 62) short, first joint small, fifth joint with 744 DR WALTER E. COLLINGE. tooth-like spine on the anterior inner border ; flagellum with thirteen to fourteen joints. First maxillee (fig. 63) with outer lobe terminating in eleven curved and a single straight spines, inner lobe with three setose spines, and a small curved spine on the outer side distally. The segments of the mesosome (fig. 67), excepting the first, are almost subequal; the pleural plates of the first are laterally truncate, with the anterior angle pointed. Coxal plates large, occupying the whole of the lateral margins of the-second to seventh segments, gradually increasing in width ; posterior angle of the seventh produced backwardly. Maxillipedes (fig. 64) with four-jointed palp, no groove on the third joint, basal plate, epipodite, and inner lobe elongated. Thoracic appendages of normal type (figs. 65 and 66). Metasome (fig. 68) has two short segments and sutures indicating a coalesced third, terminal segment long, slightly convex and raised in the mid-dorsal line, sides straight and gradually tapering, lateral angles obtuse, posterior margin medially produced in a blunt rounded process extending beyond the lateral angles. Uropoda (fig. 69) somewhat narrow, endopodite with posterior margin bluntly rounded, anterior margin slightly produced. Length of d 19 mm., of 2 15 mm. Colour (in alcohol) one brown. Geographical Distribution.—At present known only from St Andrews ne Remarks.—Externally, small examples of this species are not unlike J. wiridis (Slabber) ; indeed, I had referred such to that species until meeting with the larger specimens. Later, on examining the minute structure numerous important differ- ences were found. Fortunately, whilst engaged upon this species, Professor G. O. Sars very kindly sent me examples of J. viridis from Norway, and it was at once evident that I had something very different. Large adult males might easily be mistaken at first sight for I. granulosa, Rathke. From I. granulosa this species differs in the shape of the cephalon, which is more deeply excavated on the anterior margin, and the posterior lateral portions slope inwards. The antennule are more slender, as also the antenne, the style of the latter being longer. There are minor differences in the form of the first maxillee and maxillipedes. The coxal plates form the complete lateral margin of the mesosomatic segments 2-7, whereas in I. granulosa those on the second segment occupy the anterior two-thirds and not the whole of the third segment. The segments of the metasome differ in form, especially the terminal one, which has the sides gradually sloping, with the posterior margin truncate laterally and a prominent median tooth. The endopodites of the uropoda are more triangular in shape, being cut away laterally. From I. viridis (Slabber) the differences, apart from its greater size, are more striking ; thus it differs in the shape of the cephalon, the antennule, and the antenne. The style of the antennze is much shorter than in I. viridis. In the form of the first maxillze and in the maxillipedes the differences are more pronounced, especially the A REVISION OF THE BRITISH IDOTEID. 7495 shortened basal segment and the elongated outer lobe of the latter appendages. In I. viridis these two parts are about equal in length, and the coxal plates do not form the complete lateral margin of the second to fourth mesosomatic segments, as in this species. The posterior margin of the terminal segment of the metasome in J. viridis is not laterally truncate, whilst the endopodites of the uropoda are broader and more truncate in that species than in this. It gives me great pleasure to associate with this species the name of Professor G. O. Sars, to whom I am indebted for many interesting species of this family. (7) Idotea viridis (Slabber). (Pl. VII, figs. 70-80.) Oniseus viridis, Slabber, Naturk. verlustigingen, 1778, p. 104, pl. xu, figs. 4, 5. Idotea phosphorea, Hoek (nee Harger), Crust. Neerlandica, 1889, ii, p. 7, pl. vii, fig. 2, 27. Idotea salinarum, Dollfus, Fewille des jeunes Nat., 1895, p. 7, fig. 21. Idothea viridis, G. O. Sars, Crust. Norw., 1897, ii, p. 83, pl. xxxiv, fig. 2. Idothea angusta, G. O. Sars, zbid., p. 84. Idotea viridis, Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), xiv, p. 442. Body oblong linear, somewhat depressed. Cephalon (fig. 70) wider than long, anterior and posterior margins slightly emarginate. Hyes moderate in size, situated dorso-laterally. Antennule (fig. 71), first joint only very slightly expanded, second and third elongated ; flagellum with single clavate joint. Antenne (figs. 72 and 73) slender, fourth and fifth joints subequal; flagellum with twelve joints, mostly elongated, style conical with long sete. First maxille (fig. 74), outer lobe narrowed distally, with six stout curved spines and three toothed ones. The segments of the mesosome (fig. 78) are somewhat unequal in length, pleural plates of the first with rounded anterior and posterior angles. Coxal plates of the second to fourth segments narrow and not occupying the whole of the lateral margin, those of the fifth and seventh segments much wider and occupying the whole of the lateral margins. Maxillipedes (fig. 75) elongated, palp four-jointed, basal plate narrowed on the inner side, epipodite oblong oval, distal inner lobe fairly large. Thoracic appendages (figs. 76 and 77) much more slender than in most other species, and elongated. Metasome (fig. 79) with two short segments and well-marked lateral sutures indicating a further coalesced one, terminal segment slightly convex, lateral margins almost parallel, narrowing a little towards the posterior end, posterior margin terminating as a drawn-out point, with obtuse lateral corners. Uropoda (fig. 80) narrow, endopodite obtusely truncated terminally, outer margin cut away, setose style normal. Length of ¢ 12 mm., of ? 10 mm. Colour greenish or greenish-yellow. Geographical Distribution—S8t Andrews Bay and Isle of Wight; St Kilda (Hewitt); Clyde Area (Pattence); Suffolk coast, Weymouth, and Guernsey (NorMay) ; South and West Ireland (Tarrersatt) : also Norwegian coast (G. O. Sars) ; coasts of - Holland and France (Horx). ° Remarks.—This species must be considered rare in St Andrews Bay. Thanks to TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 23). 109 746 DR WALTER E. COLLINGE. the kindness of Professor G. O. Sars, [ have been able to examine examples from Norway. Apart from being slightly smaller, they do not differ at all from the specimens taken in St Andrews Bay, and they are practically identical with specimens from Freshwater Bay, Isle of Wight, kindly sent to me by Mr W. OmEr-Cooprr. (8) Idotea metallica, Bose. (Pl. VIII, figs. 81-91.) Idotea metallica, Bose, Hist. Nat. des Crust., 1802, t. ii, p. 179, pl. xv, fig. 6; Latreille, Hast. Mat. Crust., 1803, t. vi, p. 373. Idotea peloponesiaca, Roux, Crust. de la Médit., 1828, pl. xxx, figs. 10-12. Idotea atrata, Costa, Fauna del R. Napolt (Crust.), 1838, pl. xi, fig. 3. Idotea rugosa, Milne-Edwards, Hist. Nat. Crust., 1840, vol. iii, p. 131. Idothea robusta, Kroyer, Naturhist. Tidssk., 1846 (s. 2), vol. ii, p. 108, pl. xxvi, fig. 3. Idothea compacta, White, List Crust. Brit. Mus., 1847, p. 95. Idothea algtvica, Lucas, Anim. artic. Hapl. Sct. Algérie, 1849, vol. i, Crust., p. 61, pl. vi, fig. 2. Idotea robusta, Harger, Rep. U.S. Comms. Fish and Fisheries, 1880, p. 349, pl. vi, figs. 30-32. Idotea metallica, Miers, Journ, Linn. Soc. Lond., 1881, vol. xvi, p. 35; Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), vol. xiv, p. 443; Tattersall, Fisheries, Ireland, Set. Invest., 1904, No. 2 [1905], p. 50; Stebbing, Zrans. Linn. Soc. Lond. (Zool.), 1910, vol. xiv, p. 108; Tattersall, Nord. Plank., 1911, p. 227, fig. 116; mec Stephensen, Rep. Danish Oceanog. Exp., 1908-1910, No. 3,.1915, p. 12, fig. 4. Idothea metallica, Richardson, Bull, No. 54, U.S. Nat. Mus., 1905, p. 362, 3 figs. Body oblong ovate, moderately convex, surface more or less rugose. Cephalon (fig. 81) wider than long, anterior margin slightly emarginate, posteriorly with deeply impressed furrow. Eyes large and round, situated dorso-laterally. Anten- nulee (fig. 82) extending to the end of the second peduncular joint of the antenne. Antenne (figs. 83 and 84) short and robust, joints 1 to 4 of the peduncle short, fifth joint longer ; flagellum short, with eight to ten joints. First maxille (fig. 85) with outer lobe terminating in nine plain spines and three toothed ones, inner lobe with three elongated setose spines. The segments of the mesosome (fig. 89) almost subequal, anterior angle of pleural plates of first segment bluntly rounded. Coxal plates occupy the whole of the lateral margins of the second to seventh segments, posterior angles of those of the fifth to seventh produced backwardly as sharp spines. Maxillipedes (fig. 86) broad, joints of the palp and basal plate short and broad, as also the epipodite ; inner distal lobe large. Thoracic appendages (figs. 87 and 88) robust. Metasome (fig. 90) with two short segments and strongly marked lateral sutures, indicating a further coalesced one, terminal segment with a strong median ridge, lateral margins converging to a truncate extremity. Uropoda (fig. 91) oblong, with almost parallel sides, endopodite with slightly sloping lateral margins, posterior margin truncate, setose style short. Length of 6 23 mm., of ? 18 mm. Jolour (in alcohol) bluish-green. ; Geographical Distribution.—As previously pointed out, this species has not as yet been found in St Andrews Bay. Dr Scorr informs mé that he took it off the N.H. coast of Scotland in 1909. I have examples from South-West Ireland (whence - A REVISION OF THE BRITISH IDOTEIDA. 747 it is recorded by NormaN and TaTTERsALL), also from the Adriatic and Mediterranean Seas, Japan, and the Atlantic coast of North America. Recorded also from Greenland, Iceland, and Patagonia (Ricnarpson) ; New South Wales and Borneo (Mrers). Remarks.—This species is widely separated from any other British species. It approaches J. emarginata (Fabr.) in the form of the cephalon and uropoda, and less so in the form of the coxal plates of the mesosome. In the form of the antennules and antenne, the first maxille, the maxillipedes, and in the shape of the metasome, there are marked differences. The maxillipedes are short and broad, whilst in I. emarginata they are narrow and elongated. Miss Ricnarpson (58, p. 363, fig. 393) wrongly figures the coxopodite as a single piece, and SrTEPHENSEN’s figure (73, p. 13, fig. 4) of LZ. metallica is really of a new and allied species which I am describing elsewhere. The lateral margins of the metasome in J. metallica are straight, whereas in J. emarginata they are slightly curved. On the posterior region of the cephalon there is in this species a deeply impressed transverse furrow. Referring to this, TATTERSALL (75) remarks that this species “may be distinguished from I. emarginata very readily by the presence of a small supplementary segment between the cephalon and the first segment of the thorax.” The term “supplementary segment” is somewhat unfortunate, as examination shows that this furrow is in no sense a segment or even a suture indicating one, but simply the sinuous line, common to a large series of genera and species of this family, near the posterior margin of the cephalon. In J. baltica and other species it is laterally continued as a cleft ; in J. metallica this is not so, only it is more deeply impressed. In the closely allied species (the /. metallica of STEPHENSEN) this is not so pronounced. Somewhat similar lines or furrows are present on the segments of the mesosome in many species of Isopoda, and as a sinuous line it is present on the cephalon of I. baltiea, I. neglecta, I. pelagica, I. emarginata, 1. linearis, and many other species. There is perhaps less difference in the form of the body in the two sexes in this species than in any other British species of the genus. (9) Idotea linearis (Pennant). (Pl. IX, figs. 92-102.) Oniscus linearis, Pennant, Brit. Zool., 1777, vol. iv, pl. xviii, fig. 2. Tdotea tridentata, Latreille, Gen. Crust. et Ins., 1806, vol. i, p. 611. Idotea hectica, Leach, Edinb. Encycl., 1833, vol. vii, p. 404 (nec Pallas). Stenosoma lineare, Leach, Trans. Linn. Soc. Lond., 1815, vol. xi, p. 366. Idotea linearis, Milne-Edwards, Hist. Nat. Crust., 1840, vol. iii, p. 132. Idothea sexlineata, Kroyer, Naturhist. Tidssk., 1846 (s. 2), vol. ii, p. 88; Voyage en Scand., etc., pl. xxvi, fig. 1. : Idotea linearis, Bate and Westwood, Brit. Sessile-eyed Crust., 1868, vol. ii, figures on p. 388; Miers, Journ. Linn, Soc. Lond., 1881, vol. xvi, p. 47; Norman, Ann. and May. Nat. Hist., 1904 Guct); vol. xiv, p. 443; Tattersall, Nord. Plank., 1911, p. 229, tig. 117. Body narrow, elongated, somewhat depressed dorsally, surface uneven. Cephalon (fig. 92) wider than long, with anterior margin deeply emarginate, lateral lobe 748 DR WALTER E. COLLINGE. broadly rounded, posterior margin almost straight, with slight impressed line anterior to the margin. Eyes large, situated dorso-laterally immediately behind the lateral lobes. Antennule (fig. 93) extending beyond the third peduncular joint of the antenna, first joint comparatively small, second joint very small. Antenne (figs. 94 and 95) large, with last two joints of peduncle nearly subequal, and each nearly twice as long as the two preceding ones; flagellum normally with twenty joints, terminal style long, distally with dense brush of sete. ~ First maxille (fig. 96) with outer lobe terminating obliquely truncate, with nine curved spines, inner lobe terminating in three fine setose spines. The segments of the mesosome (fig. 100), excepting the first, almost subequal, pleural plates small, coxal plates small, occupy- ing the anterior portion of the lateral margin of segments 2 to 4, the middle portion of the fifth segment, and the posterior portion of the sixth and seventh. Maxillipedes (fig. 97) with four-jointed palp, groove on the third joint scarcely visible, basal segment short and broad, inner distal lobe large, epipodite roughly oval. Thoracic appendages (figs. 98 and 99) only moderately stout, protopodite of the seventh deeply grooved. Metasome (fig. 101) has two short segments and indica- tions of a coalesced third, terminal segment long, lateral margins slightly incurved anteriorly, widening a little beyond the middle and contracting towards the posterior margin, which is excavate with median obtuse point, and longer and more acutely produced laterally. Uropoda (fig. 102) flattened, elongated, rounded and narrower anteriorly, widening towards the joint, endopodite narrower than the terminal margin of the basal plate, longer than the breadth, and slightly cut away on the outer postero-lateral border. Length of 6 28-380 mm., of 2 25 mm. Colour (in alcohol) pale brownish-green, with darker-coloured longitudinal stripes. Geographical Distribution.—Fairly common all round our coasts. I have records from a large number of localities, amongst which the following may be mentioned:— Moray Firth, Aberdeen, St Andrews, Northumberland coast, Norfolk and Suffolk coasts (very small examples), Weymouth, Plymouth, Bristol Channel, Blacksod Bay (Co. Mayo), Irish Sea, and Firth of Clyde. I have examined very fine specimens in the Cambridge University Museum of Zoology, measuring 38 mm. in length, from Lowestoft. Also recorded from Netherlands, Denmark, Mediterranean, and Java (Miers); Channel Isles and coasts of France and Spain (Norman). Remarks.—I. linearis stands out distinct from any other species of the family occurring in the British Isles. Its long filiform body, graceful antenne, and small coxal plates at once serve to identify it. , A REVISION OF THE BRITISH IDOTEIDA. 749 2. Genus Zenobiana, Stebbing. Zenobia, Risso, Hist. Nat. de V Europe Mérid., 1826, v, p. 110; Miers, Journ. Linn. Soc. Lond., 1881, xvi, p. 19; Dollfus, Fenille d. jeunes Nat., 1895, p. 9. Cleantis, Tana, Amer. Journ. Sci. and Arts, 1849 (s, 2), viii, p. 427; ibid., 1852, p. 300; U.S. Haplor. Exped., 1853, pp. 697, 707. Zenobiana, Stebbing, Ann. and Mag. Nat. Hist., 1895 (s. 6), xv, p. 24; Norman, tbid., 1904 (s. 7), xiv, p. 443; Tattersall, Nord. Plank., 1911, p. 231; Issel, Ann. Mus. Zool. Napoli, 1913, iv, p. 2. Body somewhat slender, long and narrow, with parallel sides. Cephalon wider than long. Eyes small, situated dorso-laterally. Antenne short; flagellum short, with single joint, or restricted number of joints. Maxillipede with five-jointed palp. Coxal plates distinct, usually small and narrow. Metasome with three to five stements, lateral margins of the terminal segment partly folded over the dorsal side towards the depressed area at the posterior extremity. As already pointed out, Issen (35) regards the genus Cleantis, Dana, as synony- mous with Zenobiana. TatTrERSALL (76) in 1911 had previously pointed this out, but was in favour of retaining Dana’s name. He states, p. 231: “ Es scheint mir dass die Genus Cleantis, Dana, sich als Synonym mit Zenobiana erweisen wied, und dass daher, da der Name Zenobia sich nach STEBBING als schon vergeben erweisen hat und daher fiir das in Rede stehende Genus nicht wahlbar ist, das letztere den Namen Cleantis tragen muss der urspriinglich von Dana fiir dasselbe vorgesehen ist.” As it stands at present the genus is an unsatisfactory one, including as it does species widely separated from one another. Miss RicHarpson (60) has very rightly placed the Cleantis isopus of Miers in a new genus, Cleantiella, “owing to the differences in the shape of the body, which is broader and more flattened, and in the character of the legs, and to the fact that the abdomen is composed of but two segments.” (1) Zenobiana prismatica (Risso). (Pl. X, figs. 103-114.) Zenobia prismatica, Risso, Hist. Nat. del’ Kurope Mérid., 1826, v, p. 110, pl. v, fig. 24. Zenobia mediterranea, Risso, tbid., p. 111. Idotea chelipes, O. G. Costa, Fauna del regno di Napoli, 1838, p. 2, pl. xi, fig. 2a, 0, c. Idotea prismatica, Heller, Verh. zool.-bot. Ges, Wien, 1866, xvi, p. 729. Idotea parallela, Bate and Westwood, Brit. Sessile-eyed Crust., 1867, ii, p. 391, fig. Idotea (subg. Zenolia) prismatica, Miers, Journ, Linn, Soc. Lond., 1881, xvi, p. 21. Zenobiana prismatica, Stebbing, Ann. and Mag. Nat. Hist., 1895 (s. 6), xv, p. 24; Norman, zbdd., 1904 (s. 7), xiv, p. 444; Tattersall, Nord. Plank., 1911, p. 231, fig. 120; Issel, Ann. Mus. Zool. Napoli, 1913, iv, p. 1, figs. 1-9. Body elongated, narrow, with the sides of the metasome parallel. Cephalon (fig. 103) wider than long, and wider anteriorly than posteriorly, anterior marein curving very slightly inwards towards the median line ; dorsal surface convex, sloping towards the anterior margin. Eyes small, dorso-lateral. Antennulz with first and second joints slightly expanded, third joint small; flagellum club-shaped, with terminal sete. Antenne (figs. 104-108) short, first joint small, second widely ex- . 750 DR WALTER E, COLLINGE. panded, third short and wide, third and fourth subequal and deeply indented distally on the ventral side ; flagellum with three or four joints in the male, female usually unjointed. First maxillee (fig. 109) with outer lobe terminating in four stout curved spines and three toothed and one small one, inner lobe with three long setose spines. The segments of the mesosome (fig. 112) are slightly depressed, the pleural plates of the first extending forwards and partly flanking the cephalon. Coxal plates of the second to fourth segments very narrow and not extending over the whole of the lateral margin, fifth and seventh much wider, forming the entire lateral margin with the posterior angles produced into an elongated point. Maxillipedes (fig. 110) elongated, palp five-joited, with well-marked groove on the third joint, epipodite wide and long, distal inner lobe pointed distally. Thoracic appendages (fig. 111) fairly robust. Metasome (fig. 113) with three short segments and lateral indications of a further coalesced one, terminal segment with lateral margins straight and _ pos- teriorly elevated, folding over on to the dorsal side, posterior margin rounded. Uropoda (fig. 114) wider anteriorly than posteriorly, setose on the inner and outer lateral margins, endopodite narrower than the basal plate. Length 13°5 mm. Colour (in alcohol) variable. : Geographical Distribution.—F irth of Clyde, south and south-west coast of Eneland (STEBBING) ; Jersey (SINEL) ; west coast of France (DotiFus); Adriatic Sea (CLAUS) ; Mediterranean (ISSEL). Remarks.—This species has been very fully described by Issmn (35), who has given outline figures of the animal and various appendages. In addition to correct- ing Bate and Westwoon’s description and figures, he has pointed out that there is considerable variation in the flagellum of the antenne of the female. Three of his figures are here reproduced illustrating this (figs. 106-108). 3. Genus Synisoma, nov. nom.=Stenosoma, Auctt. Body oblong, widening slightly towards the middle of the length of the meso- some, usually keeled, surface smooth, tuberculated or sculptured. Cephalon with well-developed lateral lobes. Antennulee with first joint expanded, flagellum com- posed of a single joint. Antennze with large second joint, flagellum multiarticulate, style short. Palp of maxillipede four-jointed, distal inner lobe usually wide. Segments of the mesosome keeled or raised, pleural plates of the first segment produced laterally or forwards. Coxal plates small, but visible dorsally on all segments excepting the first. Appendages somewhat slender. Metasome composed of a single segment, with lateral sutures indicating coalesced segments ; terminally the segment is usually pointed. Uropoda flattened, endopodite more or less tri- angular in shape. Dotirus (20) wrongly confounded the genus Synidotea, Harger, with this genus, A REVISION OF THE BRITISH IDOTEID. 751 (1) Synisoma lancifer (Leach, MSS.).* (Pl. XI, figs. 115-128.) Leptosomu lancifer, Leach, MSS, in Brit. Mus. Idotea appendiculata, Bate and Westwood (nec Risso), Brit. Sessile-eyed Crust., 1867, ii, p. 396, figs. Stenosoma lancifer, Dollfus, Fewille des jeunes Nat., 1895, p. 5, fig. 13. Stenosoma lanciferum, Norman, Ann. and Mag. Nat. Hist., 1904 (s. 7), xiv, p. 444; Tattersall, Nord. Plank., 1911, p. 230, fig. 118. Body oblong, widest at the third mesosomatic segment, with well-marked dorsal keel. Cephalon (fig. 115) wider than long, and wider anteriorly than posteriorly, anterior margin produced in the median line as a depressed sharp point, lateral lobes well developed, sinuous in front of the posterior margin. LHyes moderate in size, situated dorso-laterally. Antennule (fig. 116) with first joint expanded. Antenne (figs. 117 and 118) fairly stout, first joint short, second the largest, third and fourth subequal, fourth about the length of the first; flagellum with sixteen joints, 1 to 6 cubical, remainder elongated, terminal style short and conical. First maxille (fig. 119), outer lobe terminating in five stout curved spines and six inner ones, of which two are toothed ; inner lobe terminally rounded, with three long setose spines and a small spine on the outer anterior border of the lobe. The segments of the mesosome (fig. 126) slope from a strong dorsal keel, third segment the widest and longest, gradually becoming smaller anteriorly and posteriorly ; pleural plates small, that of the first segment truncate anteriorly. Coxal plates occupying the anterior half of segments 2 and 3, slightly more of 4, 5, and 6, and nearly the whole of the seventh. Maxillipedes (fig. 120) broad and stout, basal plate medium size, epipodite elongated, narrowing anteriorly, distal imner lobe very wide, with marginal stout spines and setose spines distally (fig. 121). Thoracic appendages (figs. 122-125) somewhat slender, sparsely setose. Metasome (fig. 127) composed of a single segment, with lateral sutures anteriorly indicating three coalesced segments, ° lateral margins almost straight anteriorly, expanding towards the posterior end and rounded, terminating in a drawn-out rounded process. Uropoda (fig. 128) flattened, endopodite triangular in shape, short setose. style. Length of ¢ 22°5 mm. Colour (in alcohol) yellowish-brown. Geographical Distribution.—South and south-west coast of England (NorMAN and Srepsinc); Channel Isles (KopHLER); also west coast of France (DoxuFus) ; Mediterranean (STEPHENSEN). I have to thank Dr EH. J. Auien, F.R.S., for kindly sending me for examination very fine examples from Wembury Bay and Looe Island. Remarks.—This handsome species is easily separated from any of its allies by the prominent keel in adult specimens, and the form of the terminal segment. STEPHENSEN’S (73) S. appendiculatum (Risso), 1826, is the S. acuminata of * As the International Rules of Zoological Nomenclature (1905) prohibit the use of an “ unpublished” name, it might be advisable to adopt DouuFus as the authority. 752 DR WALTER E. COLLINGE. Lracu, 1815, whilst his S. acwminatum (Leach)—represented by two examples from different localities—is in one case referable to S. capito (Rathke), the other approaching S. lancifer (Leach), but I am inclined to regard it as-a distinct species. Judging from the material I have examined, I am of opinion that there is a further British species, allied to S. lanczfer, in which, however, the metasome has the lateral margins of the terminal segment almost straight, and the posterior margin gradually converging to a blunt point. STEPHENSEN’s figure of his S. acununata (73, p. 15, fig. 5) approaches this very closely. 2. Synisoma acuminata (Leach). Stenosoma acuminatum, Leach, Trans. Linn. Soc. Lond., 1815, xi, p. 366; Hdinb. Eney., 1833, vii, p. 433. © Idotea acuminata, White, List Crust. Brit. Mus., 1847, p. 95; Bate and Westwood, Brit. Sessile-eyed Crust., 1867, ii, figures on p. 394 ; Miers, Journ. Linn. Soc, Lond., 1881, xvi, p. 59. Stenosoma acuminatum, Dollfus, Feuille des jeunes Nat., 1895, p. 5, fig. 14; Norman, Ann. and May. Nat. Hist., 1904 (s. 7), xiv, p. 444; Tattersall, Nord. Plank., 1911, p. 231, fig. 119. Stenosoma sepiicuiiure Steptenseh, Rep. Danish Oceanog. Exp., 1915, p. 17, fig. 7. I regret that I have not been able to obtain any British examples of this species. The following description is taken from Migrs (44), with some slight alterations in the terminology :— : “Body narrow, elongated, moderately convex, with indications of a longitudinal median dorsal carina. Cephalon with its anterior margin somewhat excavated, and its antero-lateral angles rather prominent. First four segments of the mesosome each widest in the middle, and with their lateral margins (in a dorsal view) more or less angulated ; the posterior segments of the mesosome are widest at or near their postero-lateral angles. Metasome ovate-lanceolate, with the lateral margins . at first straight and then curving regularly to the distal extremity, which is subacute or acute, or even acuminated, and with more or less distinct traces of lateral sutures near its base, indicative of two coalescent segments. Hyes small, placed in the middle of the lateral margins. Antennules do not reach the extremity of the ante- penultimate joint of the peduncle of the antennz, with their basal joints moderately dilated. Antennze reaching sometimes to the posterior margin of the sixth meso- somatic segment; the last two joints of the peduncle slender, subequal, and each longer than the preceding; flagellum about nineteen-jointed, and longer than the peduncle. Legs very slender, subequal. The coxal plates (in a dorsal view) are very small, and in the second to fourth segments occupy the middle of the lateral margins ; in the fifth and sixth segments they are placed near to, and in the seventh segment quite at, the postero-lateral angle of the segment; in the last two segments they are of nearly triangular form. The terminal plates of the uropoda are considerably longer than broad, and rounded at their distal ends. “Length of the largest Beer in the British Museum collection about 1 inch (25 mm.), breadth rather less than $ inch (5 mm.).” A REVISION OF THE BRITISH IDOTEIDA, 753 Miers further states : “This is a very variable species, and I have been obliged to unite under one name several types that have usually been considered distinct. “It occurs on the shores of the Mediterranean and Adriatic, in the Black Sea, and on the South British coasts, and northward apparently as far as the island of Cumbrae on the Clyde. ' “Dr Leacn’s designation of I. acuminata may apply to what may be considered the typical form of this species, in which the body is less distinctly carinated, the epimera [coxal plates] less distinctly angulated, and the terminal segment lanceolate, with the sides rounding off to the distal extremity, which is acute or subacute, but not produced and acuminated. Besides Leacn’s typical specimen (which is in very bad etaon there is but a single specimen, from Tripoli, in the British Museum collection presenting these characters.” Geographical Distribution.—* On the shores of the Mediterranean and Adriatic, in the Black Sea, on the South British coasts, and northward apparently as far as the island of Cumbrae on the Clyde” (Miers). ‘Channel Isles (KoEHLER). BIBLIOGRAPHY, (1) Barz, C. Spznce, “On the British Edriophthalma,” Rep. Brit. Assoc. for 1855, London, 1856, pp. 18-62, pls. xii-xxii, (2) —— “On some new Australian Species of Crustacea,” Proc, Zool, Soc. Lond., 1863, pp. 498-505, pls. xl, xli. (3) Bartz, C. Spence, and J, O, Westwoop, A History of the British Sessile-eyed Crustacea, London, 1868: Isopoda, vol. ii, pp. 99-495. _ (4) Benzovicr, J. E., “A Revision of the Genus Synidotea,” Proc. Acad. Nat. Sci. Phila., 1897, pp. 389-404. (5) Bonnirr, J., Catalogo des Crustacés Malacostracés recueillis dans la baie de: Concarneau,” Bull. Sct. du Dépt. du Nord, 1887 (s. 2), t. x, pp. 3-190. (6) Bosc, L. A. G., Histotre naturelle des Crustacés, Paris, 1802, pp. 1-296, 18 pls. (7) Branprt, J. F., “‘Conspectus Monographie Crustaceorum Oniscodorum Latreilli,” Bull. Soc. Imp. Nat. ’ Moscou, 1833, vol. vi, pp. 171-193. (8) Branort, J. F., and J. T. C. Rarzesurc, Medizinische Zoologie, Berlin, 1830-34, vol. ii. (9) Branpr, E., “Du systéme nerveux de l’Idothea entomon (Crustacé isopode),” Compt. Rendus, 1880, pp. 713-715. (10) Cauman, W. T., “Crustacea,” part vii, fasc. 3, in Lankester’s Treatise on Zoology, London, 1909, (11) Cuturoy, CHarues, “Revision of the New Zealand Idoteide,” Trans. New Zealand Inst., 1890, vol, xxii, pp. 189-204. **On a new Species of Idotea,” Ann. and Mag. Nat. Hist., 1885 (s. 5), vol. xv, pp. 123, 124, pl. Va., figs. 1-3, (12) Ciaus, Grundziige der Zoologie, French trans., Paris, 1871. (13) Cotniner, Watrer E., ‘On the Range of Variation of the Oral Appendages in some Terrestrial Isopods,” Journ. Linn. Soc, Lond. (Zool.), 1914, vol. xxxii, pp. 287-293, pls. 20, 21. (14) “Some Observations on the Isopod Idotea hectica (Pallas),” Ann, and Mag. Nat. Hist., 1915 rei (s. 8), vol. xvi, pp. 162-164, pl. ix. (15) —— “On the Isopoda of the Family Idoteide occurring in St Andrews Bay,” Scottish Nat., 1915, pp. 331, 332. TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 23). 110 (11a) 754 DR WALTER E. COLLINGE. (16) Dana, J. D., “Crustacea,” U.S. Huplor. Exped., 1852, vol. xiii, pp. 691-713, pls. xlvi-liii. (17) -—— “Catalogue and Descriptions of Crustacea collected by Dr John L. Le Conte,” Proc. Acad. Nat. Sci. Phila., 1854-55, vol. vii, pp. 175-177. (18) Desmarzst, A. G., “ Malacostracés,” Dict. d. Sci. Nat., Paris, 1823, t. xxviii, pp. 138-425, 56 pls. (19) Ds Gurr, C., Mémoires pour servir ad Vhistotre des Insectes, Stockholme, 1778, vol. vii. (20) Doutrus, A., “Les Idoteide des cotes de France,” Feuille des jeunes Nat., 1894-95, No. 279, pp. 1-5; No. 290, pp. 17-18; No. 291, pp. 38-40; No. 292, pp. 41-56. (21) Fasrictus, J. C., Enioipoton Systematica, ute Hafaes, 1792-94, 4 vols. (22) Index Alphabeticus, 1796, pp. 1-175. (23) —— Supplementum Entomologia Systematica, Hafnie, 1798, pp. 1-572. (24) Index Alphabeticus, 1799, pp. 1-53. (25) Gapzixizwicz, W., “Die Gréssenvariation von Idothea tricuspidata,” Biol. Centralbl., 1907, Bd. xxvii, pp. 505-508. (26) —— “Sur la biologie de I’Idothea tricuspidata,” Bull. Acad, Sci. St Pétersb., 1906 [1907] (s. 5), t. 24, pp. 263-272. (27) Gounp, A. A., Invertebrata of Massachusetts, Cambridge, U.S.A., 1841, pp. 336-338. (28) Guinysss, A., ‘Structure du systéeme digestif de I'Idotea hectica, Latr.,” Bull. Soc. philom. Paris, 1906 (s. 9), t. vili, pp. 203-209. (29) Guitrin-Minivinte, F. E., Iconographie du Réegne animal de G. Cuvier, etc.: Crustacea, 1829-1844, t. ili, pp. 1-46, pls. i-xxxv. (30) Hansen, H. J., “Oversigt over de paa Dijmphna-Togtet indsamlede Krebsdyr,” Dijmphna-Togtets zool.-bot. Udbytté, 1886, pp. 185-286, pls. xx—xxiv. (31) Harcer, Oscar, ‘‘ Descriptions of new Genera and Species of Isopoda from New England and adjacent Regions,” Amer, Journ. Sct. and Arts, 1878 (s. 3), vol. xv, pp. 373-878. “Report on the Marine Isopoda of New England and adjacent Waters,” Rep. U.S. Comms. Fish and Fisheries for 1878, part vi, 1880, pp. 297-462, pls. i—xi. (33) Heuter, C., ‘Carcinologische Beitrige . . . Adriatischen Meeres,” Verh. zool.-bot. Gesell. Wien, 1866. (34) Issen, Rarragve, “ Richerche di etologia sull’ Isopodo tubicola Zenobiana prismatica (Risso),” Arch, Zool. exp. et gén., 1912, t. li, pp. 450-479, figs. 1-6. (35) —— “Nota sulla Zenobiana prismatica (Risso) (Idotea chelipes, Costa) e sulla identita del gen. Zeno- biana Risso col gen. Cleantis Dana,” Napoli Ann. Museo. Zool., 1913, pp. 1-8, figs. 1-9. (36) Kowatxvsky, A., “Anatomie der Meerassel Idothea entomon, und kurzer Uberblick tiber die in den Gewassern des Peterburger Gouvernements verkommenden Krebse,” Hstest, izsleadov. S. Peterburg. Gubernti, 1864, t. i, pp. 241-265, pls. iv-ix (in Russian). (37) Kinanan, J. R., “ Analysis of certain Allied Genera of Terrestrial Isopoda ; with descriptions of a new Genus, and a detailed List of the British Species of Lagia, Philougria, Philoscia, Porcellio, Oniscus, and Armadillidium,” Nat. Hist, Rev., 1857, vol. iv, pp. 258-282, pls. xix—xxii. (38) Lucas, H., “Histoire naturelle des animaux articulés Crustacés,” Explor. Scient. de V Algérie, 1840-42, Paris, 1849, vol. i, pp. 1-88, 8 pls. (39) Larreruin, P. A., Histoire naturelle générale et particuliére des Crustacés et des Insectes, Paris, 1802- 1805, vols. vi, vil. (40) Genera Crustaceorum et Insectorum, Paris, 1806, vol. i. (41) Leacn, W. E., ‘A Tabular View of the External Characters of four Classes of intel which Linné arranged under Insecta, ete.,” Trans. Linn. Soc. Lond., 1815, vol. xi, pp. 8306-400. (32) (42) ——- “Crustaceology,” in Edinb. Ency., 1833, vol. vii, pp. 383-437. (43) Marzporrr, Cart, ‘“ Ueber der Farbung von Idotea tricuspidata, Desm.,” Jena. Zettschr., 1883, Bd. xvi, pp. 1-58. (44) Miers, Epwarp J., ‘‘ Revision of the .Idoteide, a Family of Sessile- eyed Crustacea,” Journ. Linn, Soc. Lond., 1881, on xvi, pp. 1-88, pls. i-iii, (45) Mitng- nova H., Histoire naturelle des Crustacés, Paris, 1840, vol. iii, pp. 115-284, pls. xxxi—xxxiii. _ (46) Norman, A. M., at Month on the Trondhjem Fiord,” Ann. and Mag. Nat. Hist., 1894 (s. 6), vol. xiii, pp. 112- 133, pls. vi, vil. A REVISION OF THE BRITISH IDOTEIDA. 755 (47) Norman, A. M., “ British Isopoda of the Families Aigide, Cirolanide, Idoteide, and Arcturide,” dbed., 1904 (s. 7), vol. xiv, pp. 430-448, pls. xii, xii. (48) Ouuin, Axet, “Isopoda from Tierra del Fuego and Patagonia,” I. Valvifera, Svenska Kuxped. till Magellanslanderna, 1901,.Bd, ii, pp. 261-306, pls. xx—xxv. (49) Oxivisr, M., “ Histoire naturelle des Insectes,” Hneycl. meéthod., 1789, vol. iv, pp. 246-256. (50). Patuas, P. S.,.Spicilegia Zoologica, Berolini, 1772. (50a) Patiznce, A., “On the occurrence of Idothea neglecta, G. O. Sars, and Idothea viridis (Slabber), within the Clyde Sea Area, and some notes on other Clyde species of Idothea,” Trans. Glasgow Nat. Hist. Soc. (1905-06), 1908, pp. 42-46. (51) Pennant, T., British Zoology, London, 1777. (52) Racovirza, E. G., and R. Sgvastos, “ Proidotea haugi, n.g., u.sp., Isqpode oligocene de Roumanie et les Mesidoteini, nouvelle sous-famille des Idotheide,” Arch. Zool. exp. et gén., 1910 (s. 5), t. vir pp. 175-200, pls. ix, x. (53) Raraxs, Hernricn, “ Anatomie der Idothea entomon, oder des Schachwurmers,” Neuste Schriften der Naturf. Gesell. Danzig, 1820-25, vol. i, pp. 109-136. (54) Rionarpson, Harriet, ‘ Description of a new Species of Idotea from Hakodate Bay, Japan,” Proc. U.S. Nat. Mus., vol. xxii, pp. 131-134, figs. 1-6. (55) —-— “Isopoda collected in Japan in the Year 1900 by the U.S. Fish Commission Steamer Albatross, and in the Year 1881 by the U.S. Steamer Palos,” zbid., 1904, vol. xxvii, pp. 32-45, 22 figs. (56) “‘Tsopoda collected in Japan by Jordan and Snyder,” ibid., pp. 46-51, 6 figs. (57) —— “Further Changes in Crustacean Nomenclature, ” Proc. Biol, Soc. Washington, 1905, vol. xviii, ppy 9, LO: (58) —— “A Monograph on the Isopods of North America,” Bull. No. 54, U.S. Nat. Mus., 1905, pp. liii+ 727, 740 figs. (59) —— “ Description of a new Genus and Species of gue G@rastcent of the Family Idotheide from the mouth of the Rio de la Plata, Argentina, South America,” Proc. U.S. Nat. Mus., 1911, vol. xl, - pp. 169-171, figs. 1-5. (60) “Description of a new Species of Isopod of the Genus Cleantis from Japan,” ibid., 1912, vol. xlii, pp. 27-29, 1 fig. (61) Risso, A., Les Crustacés des Environs de Nice, 1816. (62) —— Histoire naturelle de l Europe méridionale, 1826. (63) Roux, J. L. F. P., Crustacés de la Méditerranée, etc., Marseilles, 1829-1830. (64) Sars, G. O., Account of the Crustacea of Norway: Isopoda, vol. ii, 1897. (65) Say, Tuomas, “An Account of the Crustacea of the United States,” Jowrn. Acad. Nat. Sci. Phila., 1818, pp. 393-401 and 423-433. (66) Scorr, THomas, “Notes on some Rare and Interesting Marine Crustacea,” 22nd Ann. Rep. Fish. Bd. Scotland, 1903, pp. 242-261, pls. xiii-xv. (67) Sressine, T. R. R., “A new heatalien Spheromid, Cyclura venosa, and notes on Dynamene rubra and trridis,” Journ. Linn. Soc. Lond. (Zool.), 1876, vol. xii, pp. 146-151, pls. 6, 7. (68) A History of Crustacea, London, 1893. (69) —— “Notes on Crustacea,” Ann. and Mag. Nat. Hist., 1895 (s. 6), vol. xv, pp. 18-25, plage (70) —— “South African Crustacea,” part i, 1902, pp. 14-66, pls. i-iv (published July 20, 1900); part ii, 1904, pp. 1-92, pls. v-xvi (published October 7, 1902), Marine Investigations in South Africa. (71) —— “General Catalogue of South African Crustacea,” Ann. Sth. African Mus., 1910, vol. vi, pp. 281-593, pls. xv—xxii. (72) “Tsopoda from the Indian Ocean and British East Africa,” T’rans. Linn. Soc, Lond. (Zool.), 1910, vol. xiv, Bs 83-122, pls. 5-11. (73) SvepuEnsen, K., ‘‘Isopoda, Tanaidacea, Cumacea, Amphipoda (excl. Hyperiide),” Rep. on the Dutch Oceanog ae Expeds. 1908-1910 to the Mediterranean and Adjacent Seas, Copenhagen, 1915, vol. ii, No. 3, pp. 1-53, figs. 1-33, (74) Stuper, Tu., ‘“‘Isopoden, gesammelt wiihrend der Reise S.M.S. Gazelle um die Erde 1874— Gre Abh. Akad. Wiss, Berlin, 1883 [1884], pp. 1-28, pls, i, ii. 756 DR WALTER E. COLLINGE., (75) Tarrersaty, W. M., “The Marine Fauna of the Coast of Ireland,” part v, Isopoda, Fisheries, Ireland, Sci. Invest., 1904 [1905], pp. 1-90, pls, i—xi. (76) “Pie nordischen Isopoden,” Nordisches Plankton, Kiel, 1911, pp. 181-313, figs. 1-340, (77) Tuomson, Geo, M., “ Descriptions of New Crustaceans,” Trans, Proc. New Zealand Inst., 1884, vol. xvi, pp. 234—240, nik xil, xiii, figs, 1-5. “A new Family of Crustacea Isopoda,” Ann. and Mag. Nat. Hist., 1904 (s, 7), vol. xiv, pp. 66- 69, pl... (78) Weser, M., ‘‘Die Isopoden gesammelt wihrend der Fahrten des Willem Barents in das nordliche Eismeer,” Bijdr, Dierk., 1884, pp. 1-39, 3 pls. (79) Wurrer, Apam, List of the Specimens of Crustacea in the Goltection of the British Musewm, London, (774) 1847. EXPLANATION OF PLATES. (All the figures are drawn from male specimens, unless otherwise stated.) Idotea baltica (Pallas). Fig. 1. Dorsal view of the cephalon. x 7. ult Fig. 2. Dorsal side of the left antennule. x 20. Fig. 3. Terminal style of the antenna. x 110, Fig. 4. Dorsal side of the left antenna. x 11. Fig. 5. Ventral side of the terminal portions of the inner and outer lobes of the right first maxilla. x 130. Fig. 6. Ventral side of the right maxillipede. x 21. Fig. 7. Portion of the third and fourth joints of the right maxillipede, showing the thickened edge of the latter and the groove on the former, Fig. 8. Ventral view of the second thoracic appendage. x 14. Fig. 9. Claw of same enlarged. Fig. 10, Ventral view of the eighth thoracic appendage. x 10. Fig. 11. Claw of same enlarged. Fig. 12. Dorsal view of the lateral portions of the mesosomotic segments, showing the coxal plates. ~ 35. Fig. 13. Dorsal view of the metasome. x 5, Fig. 14. Left uropod. x 7, Idotea pelagica, Leach. Fig. 15. Dorsal view of the cephalon. x 7. Fig. 16. Dorsal side of the left antennule. x 30. Fig. 17. Dorsal side of the left antenna. x 16. Fig. 18. Terminal style of the antenna. x 110. Fig. 19, Ventral side of the terminal portions of the inner and outer lobes of the right first maxilla, x 130, Fig. 20. Ventral side of the right maxillipede. x 2]. Fig. 21. Ventral view of the second thoracic appendage of P. x 9. Fig. 22, Ventral view of the eighth thoracic appendage of P. x 6:5. Fig. 23. Dorsal view of the lateral portions of the mesosomatic segments, showing the coxal plates. x 3:5. : Fig. 24. Dorsal view of the metasome. » 5. Fig. 25. Left uropod. x 7. Idotea neglecta, G. O. Sars. Fig. 26. Dorsal view of the cephalon. x 7, Fig. 27, Dorsal side of the left antennule. » 30, eal. . Dorsal side of the left antennule. x 30, . Dorsal side of the left antenna. x 11. . Terminal style of the antenna. x 110, . Ventral side of the terminal portions of the inner and outer lobes of the right first maxilla. . 64, 2 GD, 66; . Dorsal side of the left antenna. x 16. aos . Ventral side of the terminal portions of the inner aud outer lobes of the right first maxilla. A REVISION OF THE BRITISH IDOTEID. ar ‘ Terminal style of the antenna. x 110. x 1380. . Ventral side of the right maxillipede. x 21. 2 32, . Ventral view of the eighth thoracic appendage. x 7. . Dorsal view of the lateral portions of the mesosomatic segments, showing the coxal plates. Ventral view of the second thoracic appendage. x 14. rot. . Dorsal view of the metasome. x 5. . Left uropod. x 7. Idotea emarginata (Fabr.). © Dorsal view of the cephalon. x 7, SO: . Ventral side of the right maxillipede. x 21. . Ventral view of the second thoracic appendage. x 10. . Ventral view of the eighth thoracic appendage. x 7. - Dorsal view of the lateral portions of the mesosomatic segments, showing the coxal plates. Maso . Dorsal view of the metasome. x 3°5. . Left uropod. x 7. Idotea granulosa, Rathke. . Dorsal view of the cephalon. x 7. . Dorsal side of the left antennule. x 30, . Dorsal side of the left antenna. x 11. . Terminal style of the antenna. x 110. . Ventral side of the terminal portions of the inner and outer lobes of the right first maxilla. x 130. . Ventral side of the right maxillipede. x 21. . Ventral view of the second thoracic appendage. x 14, . Ventral view of the eighth thoracic appendage. x 14. - Dorsal view of the lateral portions of the mesosomatic segments, showing the coxal plates. ae a is . Dorsal view of the metasome. x 5D. . Left uropod. x 7. Idotea sarsi, 1. sp. . Dorsal view of the cephalon. x 7, - Dorsal side of the left antennule. 30. . Dorsal side of the left antenna. » 11. 3 . 62, . Ventral side of the terminal portions of the inner and outer lobes of the right first maxilla. Terminal style of the antenna. « 110. x 130. }Ventral side of.the right maxillipede. x $7. : Ventral view of the second thoracic appendage. x 14. Ventral view of the eighth thoracic appendage. x 10. 758 lp 5) ies} (Sd fer ee gg 9@ OQ a9 = ge Gd ‘ig, 103. Dorsal view of the cephalon. x 8. ‘ig. 104. Dorsal side of the left antenna. x DR WALTER E, COLLINGE. . Dorsal view of the lateral portions of the mesosomatic segments, showing the coxal plates. » ” 20 Silicious sandstone with sandy layer and lenticular masses of silicified peat formed ‘i of Rhyniu. + N ( Silicious sandstone with irregular beds of chert full of Rhynta. Sandstone formed OF bs) 29 of very thin layers separated by black lines, irregularly bedded, and full of | carbonised plant-remains. ( 18a | Silicified peat mixed with sandy matter. Vegetable remains much broken up. | Layer of silicious sandstone at top and at base, the latter about 1 inch thick. M |/o 9/ 188 Intermediate portion formed of silicified peat composed of Rhynia. Vegetable remains in lower part of bed much broken up. Upper part contains stems little or not compressed. ; 17 Bedded silicified peat composed of Rhyn/a, in part much decomposed, with L occasionally some sandy material. 1 0| 16 Bed of silicious sandstone with intercalated lenticular patches of peat formed of | Rhyma. / K | 0 8] 15 | Silicified peat with much broken-up vegetable remains. Thin sandy layer at top. I O 3th iad. Bed of silicified peat with thin-bedded silicious sandstone at top about 4 inch thick. \ Rhynia slightly compressed. H | 0 11] 13 | Much decomposed bed of silicified peat composed of Rhynia. {a 0 31 19 Bed of silicified peat formed of Rhyniu, a good deal decomposed and broken up, : z, with thin layer of very fine-bedded silicious sandstone at top and base. 11 Irregular thin beds of silicious sandstone. Silicified peat formed of broken-up and F a decomposed Rhynia. O See Bedded silicified peat formed of Rhynia with layer of fine-bedded silicious sand- stone, about $ inch thick at top, with another of about the same thickness at base. Vestas ‘ | Silicious bed of sandstone in very thin lamine at top, less than 1 inch thick, with | 10 black lines on bedding surfaces. Remainder of bed composed of more or less | distinctly bedded silicified peat formed of Rhynia. The stems in this, as in most of the other beds, lie horizontal with the bedding and in many cases are E almost uncompressed, while at other places in this bed the vegetable matter is | much decomposed and broken up. At one place a sandy layer occurs which has followed the natural irregularities of the surface of the peat at the time of | 9 deposition (see Pl. I, fig. 1), a feature characteristic of the sand beds which \ occur in the silicified peat. * The exact horizon of the Rhynie Old Red Sandstone has not yet been definitely determined, but it cannot be younger than the Middle Old Red Sandstone. + The site of this trench is shown on the map (fig. 1) in the Report of the Committee on “The Plant-bearing Cherts at Rhynie, Aberdeenshire,” Brit. Assoc. Rep., 1916, Newcastle Meeting. SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 763 — Continued. ft. in. 5 ‘ , : D 0 38 8 | Silicious band with thin sandstone layer at top'and base with plant remains, | O° 3 7 | Fine light grey silicious sandstone. 6 Band of bedded silicified peat formed of Rhynia, with stems usually much decayed. B { At top occurs a thin bed of silicious sandstone. oS : 5 | Cherty sandstone with carbonaceous matter. A { 0 6 4 ”» 9 ” ” sd 3 | Silicious sandstone with carbonaceous matter. \ 2 | Highly silicious sandstone with Asteroxylon. = Is 1* | Thick bed of impure peat with Rhynia and Asteroxylon. Stems generally much | decomposed. | aa | Grey clay. 2 0 White plastic clay, greenish tint and rusty spots, bedding obscure but more distinct near bottom. 2 0 Clay or clayey shale, not so light in colour as that above. Rhynia extends throughout the whole section. Asteroxylon has only been observed up to the present in bed A”. t is not our sphere to treat of the detailed geology of the Rhynie area; that will be done by other and more competent writers. As intimately connected with the petrifactions which have been placed in our hands for description, it is, however, necessary to make some remarks on the nature of the chert band in which they occur. : Our material has consisted of a complete set of specimens taken from the chert band, lettered and numbered as noted in the first and third columns of the section given above. In addition, some of the loose specimens have yielded most valuable information. The chert is much jointed and fissured, and when fresh is of a blackish or dark grey colour in which, when visible, the plants are only indicated by darker spots or streaks. Specimens naturally weathered sometimes assume a light buff colour, and the plant-remains are clearly exhibited (PI. II, fig. 2). When blocks are treated with a weak solution of hydrofluoric acid, which etches the surface and brings out more or less distinctly the structure of the chert, the plant-remains are also rendered more visible (Pl. I, fig. 1; Pl. II, fig. 5). The blocks themselves and the numerous microscopical sections studied prove that the whole chert zone was originally formed of a series of peat beds which, during their formation, had been subject to periodic inundation, when thin layers of sand were spread over the surface. Hach bed of peat appears to have a thin layer or bed of sand at its top. Several of the isolated specimens have shown the plants growing vertically from * This column gives the original numbers of the specimens as collected. ’ 764 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS the old land surface represented by the peat (Pl. II, figs. 2 and 5). One specimen (fig. 5) seems to contain a tuft of Rhynia which must have grown on the surface at the time the peat bed was brought to a final close by infiltration with silica. This example shows the upright stems for a length of 6 inches above the bed from which they grew. The uppermost. parts of the stems are embedded in almost pure silica without any admixture of peaty substance, and with only a little fine, dust-like matter disseminated through the silica (Pl. V, fig. 20). The whole history of the formation of the Rhynie Chert Zone, at least of that portion from which our specimens were taken, can be clearly read. One can in imagination see a land surface, subject at intervals to inundation, covered with a dense growth of Rhynia Gwynne-Vaughanm. By the decay of the underground parts of Rhynia and the falling down of withered stems (for this plant had no leaves) a bed of peat was gradually formed varying from an inch to a foot in thickness. The peat was then flooded and a layer of sand deposited on its surface. Again the Rhynia covered the surface, and this process of the formation of beds of peat, with the deposition of thin layers of sand, went on till a total thickness of 8 feet had — accumulated. After the formation of 8 feet of alternating peat and sand local physical con- ditions must have altered, for water with silica in solution, possibly discharged from fumaroles and geysers, poured over the peat bed and sealed it up. Thus the whole was converted into a band of chert, the structure of the plants being preserved in many cases in great perfection. It may be mentioned that the presence of geysers or hot springs has been suggested by Dr Mackt® as an explana- tion of the occurrence of so many cherty dew lonmenee in the rock series of Rhynie (i.c., pp. 283—236). Rhynia Gwynne-Vaughani, Kidston and Lang. MoprE oF OCCURRENCE. As mentioned in the introduction, the silicified peat is almost entirely formed of the prostrate stems and the rhizomes of Rhynia, while in one or two fortunate cases the closely crowded aerial stems were seen standing vertically (Pl. II, figs. 2 and 5). In these cases the ancient land surface with its vegetation is recorded in a manner very rarely met with. The horizontal bedding of the silicified peat is distinctly seen in Pl. L fig. 1, which represents a block from bed E 10 of the vertical section on page 762. In the lower parts of the individual peat beds the plant-remains can be seen to become more compressed than above. This is also seen in the microscopical section shown in PI. III, fig. 9, where the stems in the upper part are round and separated by a considerable amount of amorphous peaty matrix, while in the lower part they are crowded and compressed. In fig. 10, from the lower part of another bed the i SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 765 stratification of the silicified peat layers in different degrees of decay is still more clearly visible; two thin bands a and 6b occur in which the stems are much more decayed and crushed together. These two photographs give an idea of the abundance of the stems of Rhynia in the peat. From the fact that they are practically all ¢ut transversely in these sections, it can be inferred that for some reason or other the stems lay horizontally and more or less parallel. A horizontal surface of a block, on which the closely crowded and flattened stems of the plant show that the whole growth had been laid-down and compressed, is represented in PI. II, fig. 3. The small portion of this, enlarged two diameters, in fig. 4 shows at a a median line on the flattened stem. This will be seen below to indicate the position of a central vascular strand.* In contrast to this mode of preservation, the unaltered cylindrical stems of the free aerial portion of the plant can be exposed by fracturing the chert. Hxamples of such stems, which have preserved their shape perfectly in the almost pure silicious matrix, are shown enlarged fourteen times in PI. III, figs. 6-8. EXTERNAL MORPHOLOGY. Owing to the favourable nature of the material, it has been possible to arrive at a clear conception of the morphology of the plant as a whole. It will assist the reader if we deal briefly with the general organisation of Rhynia before entering into the details of its structure. The plant formed a practically pure growth, and its erect cylindrical stems stood closely crowded. These stems probably attained a height of 8 inches or more,t and range from 6 mm. to under 1 mm. in diameter. The plant was rootless and had no leaves, being composed entirely of a system of cylindrical axes or stems. Its lower portion consisted of branched underground rhizomes attached to the peaty soil by numerous rhizoids. Branches of the rhizome turned gradually or abruptly up and assumed the characters of the aerial stems. The latter were occasionally branched dichotomously and tapered gradually upwards. They bore small hemispherical projections which were more or less closely placed without apparent regularity. On some of these bulges tufts of rhizoid-like hairs were borne, while in other cases the projections developed into adventitious branches, usually attached by a narrow base. Some of these branches appear to have been readily detached, and their occurrence free in the peat suggests that they served to propagate the plant vegetatively. Even in the most complete specimens, preserved as they grew from the peat, the terminal parts of the aerial stems are wanting. The reproductive organs have not been observed attached to the complete plants, and it is impossible to say * Such impressions, with or without an evident midrib, according to their preservation, might easily be described as linear leaves or even alge. } The stems in PI. II, fig. 5, measured 6 inches, but were incomplete. 766 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS whether the sporangia terminated the main axes or some of the lateral branches. It is, however, certain from detached specimens found in the silicified peat that some aerial axes ended in large, elongate-pointed sporangia. That the rhizomes, aerial stems, and sporangia were portions of the same plant might have been inferred from their association in a bed composed of one type of plant only. Their continuity has, however, been directly traced, and is established by anatomical evidence. Quite apart from any question of affinity, the build of Rhynia, with its rootless rhizomes of delicate structure sending up xerophytic aerial stems, finds its closest parallel among existing plants in the general morphology of Psilotum. ANATOMY OF THE VEGETATIVE ORGANS. The plant consists throughout of cylindrical axes with an epidermis, a relatively wide cortex, and a simple central cylinder. The latter has a solid strand of tracheides surrounded by a zone of phloem. It will be convenient to describe the rhizomes in the first place and then to deal with the aerial stems. Rhizome. The rhizomes seem to have been formed of more delicate tissue than the stems arising from them, so that the peat often consists entirely of the latter, the rhizomes having decayed. In other portions of the peat, however, well-preserved rhizomes have been found; thus Pl. IV, fig. 13, shows a small portion of a section through the peat in which rhizomes are seen in their natural position. ~The section passes through three pieces of the rhizome. The uppermost is cut transversely, and shows the broad cortex and the single stele. The lowest section is also transverse, but goes through a rhizome where dichotomous branching is about to take place, the stele having already divided. The middle section passes longi- tudinally through a rhizome, just missing the xylem of the stele except on the extreme right. The lower side of this rhizome shows two large hemispherical bulges of the outer cortical tissues. Portions of these bulges, more highly magnified, are represented in figs. 11 and 12 on Pl. HI. The greater development of the cortex on the lower side of the two transverse sections in fig. 13 is due to the plane of section passing through similar bulges. From the downwardly directed surfaces of all three rhizomes, and especially from the hemispherical] bulges of the middle piece, numerous rhizoids extend into the peat.™ Sections through other rhizomes are shown on Pl. IV. In fig. 17 two rhizomes: of different sizes are seen in transverse section. The stele of the larger one has divided preparatory to dichotomous branching. From the epidermal cells of its ) * The peculiar oval bodies in the cortex of the two lowest rhizomes in fig. 13 are the reproductive organs of saprophytic fungi. The detailed description of the numerous fungi which occur in the deposit is reserved for a future comnfunication, but the reader must allow for their presence in many of the illustrations to this paper. SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED,- ABERDEENSHIRE. 1767 lower side rhizoids are given off without the cortex forming a very definite projection. On the other side of this rhizome a distinction is evident between the narrow zone of outer cortex and the broad inner cortex. The smaller section in this figure has a single stele, and its outer tissues continue on the lower side into a marked bulge which would doubtless later have borne rhizoids. The position and general appearance of the rhizoids in all the examples examined show that the rhizomes were growing naturally in the peat when preserved. In fig. 18 a portion of the rhizoid-bearing surface of another rhizome is more highly magnified, and shows very clearly the relation of the rhizoids to the peaty soil. The rhizoids are seen to be non-septate, and do not appear to have been divided off from the epidermal cells bearing them. In fig. 14 the rather large rhizome seen in transverse section is attached by numerous rhizoids below, while from the opposite side an aerial branch ascends vertically. The small rhizome shown rather more highly magnified in fig. 15 is preparing to branch. The slender, transversely extended strand of tracheides is cut some- what obliquely. Another rhizome in transverse section is shown still more highly magnified in fig. 19. Around the central strand of xylem («.) comes the thin- walled phloem (ph.), passing without a sharp limit into the inner cortex (7.c.). The two or three outermost layers of the cortex (0.c.) contrast with the inner cortex in their appearance. The epidermis (ep.) of the rhizome has its outer cell walls thin as compared with the aerial stems to be described below, and the cuticle is less developed. Stomata have not been observed in the epidermis of the rhizome, the cells composing which are four- to six-sided in surface view and almost isodiametric (fig. 16). On the vertical branch in fig. 14, which was presumably the base of an aerial stem, the epidermal cells assume a more elongated form and a stoma occurred near to the base of the branch. The histological details of the various regions will be described more fully in connection with the aerial stems. It is sufficient to recognise here, as shown in the sections of rhizomes on Pl. IV :—1. The epidermis, the cells of which can grow out as rhizoids; 2. The narrow zone of outer cortex; 8. The broad inner cortex ; 4. The phloem; 5. The slender strand of tracheides forming the xylem. These tissues are often less sharply defined than in the aerial stems, but, except for the differences in the epidermal layer, there is no fundamental distinction between the anatomy of the rhizome and that of the aerial stem. The epidermis, the outer cell walls of which are thin as compared with those of the aerial stem, is seen to be fairly well defined in fig. 19 and on the upper side of the rhizomes in figs. 15 and 17. On the lower side, where numerous rhizoids are borne, no cuticle is recognisable, and the epidermis itself is less sharply dis- _ tinguished from the cortex on account of periclinal divisions having taken place in the superficial cells. In other cases, as is best seen in fig. 13, this cell division in the epidermis and outer cortical layers has led to the formation of large hemi- 768 DR R KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS. spherical protuberances. As figs. 11 and 12 on Pl. IE show, this has involved a large number of the superficial cells, but only affects the epidermis and outer cortex. The result of this is that the large bulges, like the smaller bulges on the aerial stems to be described below, are growths from the superficial tissues only, and have no vascular supply from the stele of the rhizome. The cell contents have disappeared or are unrecognisable, but no evidence of the presence of a fungus forming a mycorhiza has been found, although the saprophytic fungi in the peat are well preserved. Aerial Stem. The aerial stems, as already shown (PI. II, fig. 5), tapered gradually upwards, and slight modifications of the structure will have to be taken into account. Moreover, the lower portions of the stems must have been surrounded by the peat, and would constitute an ill-defined transition region. The range in diameter of stems can be seen at a glance in figs. 21-25 on Pl. V. These transverse sections are all of the same magnification (x 20), but the largest (fig. 21) is of a stem of only medium size, a little under 3 mm. in diameter, while stems of 6 mm. diameter have been observed. The smallest (figs. 24-25) are under 1 mm. in diameter. Fig. 20, Pl. V, shows a number of stems from the upper region of the plant embedded in an almost pure silicious matrix. As the portions isolated in the round show (PI. III, figs. 6-8), the stem had a well-marked epidermis, was destitute of leaves, but bore small hemispherical projections which were irregularly distributed. — The sections of stems in figs. 21-25 do not pass through any of these small pro- jections, but the transverse section in fig. 27, and the longitudinal section in fig. 26, show them. The bulges of similar size shown in figs. 28-29 differ by bearing rhizoid-like hairs. It is a reasonable supposition, though we have no direct evidence for it, that these rhizoids were mainly developed in the transition region. In one case, however, a rhizoid was observed on.a bulge in the upper region of a stem. Fig. 30 shows an aerial stem preparing for dichotomous branching. The stele is already divided. A characteristic aerial stem of good size is shown in transverse section in fig. 21. It exhibits the well-marked epidermis (ep.) with a thick outer wall and cuticle. The two or three succeeding layers of clear cells form the narrow outer cortex or hypoderma (0.c.). The broad inner cortex (2.c.) is composed of smaller rounded cells with intercellular spaces. The phloem (ph.) contrasts with the cells of the sur- rounding cortex by the smaller diameter of its thin-walled elements. In the centre is the solid strand of xylem (x.). The corresponding arrangement of the tissues is seen in longitudinal section in fig. 26, and even more clearly on Pl. VIII, fig. 59. The several tissues may now be considered in order from without inwards. Epidermas and Stomata.—The appearance of the epidermis in transverse section is well shown in Pl. VI, fig. 35, which includes portions of the outer surface of two 2 * SHOWING STRUCTURE, FROM THE RHYNIE CHER’ BED, ABERDEENSHIRE, 769 adjacent stems. Its cells are smaller than the underlying cortical cells. Their outer wall is thick, and in favourable cases exhibits a distinction into several layers, the outermost layer being the strongly developed cuticle. More usually, as in this figure, it is so preserved as to appear thick and uniform. The lateral and inner walls are thin. The epidermal cells are longer than broad, and, viewed from the outside, either on the surface of exposed stems (PI. III, figs. 6-8) or in tangential sections (Pl. VI, figs. 31-32) are seen to be broadly fusiform. They are often characterised by a peculiar dark median line (fig. 31). Suitable transverse sections show that this line is the expression of a sharp cuticular ridge springing from the middle of the slightly convex surface of each epidermal cell. This ridge is shown in fig. 36, where, however, a slight obliquity exaggerates the thickness of the outer cell wall. In other cases this ridge is wanting, sometimes at least as the result of imperfect preservation, but in some regions of the stem the epidermal cells were wider and normally lacked the central ridge (PI. III, fig. 6; Pl. VI, fig. 32). This has been verified from the study of outer surfaces of stems by reflected light as well as from sections. Stomata occurred in the epidermis, but in no region do they seem to have been at all numerous. They have not been observed on the rhizome, but in the case of the stem springing from a rhizome in PI. IV, fig. 14, a stoma was present on the base of the branch at sé. They were thus present in the epidermis of the transition region, though doubtless more abundant on the upper portions of the aerial stems. _ A stoma is shown in surface view in Pl. VI, fig. 32, and figs. 33 and 34 are of stomata in transverse section. Fig. 37 passes longitudinally through a stoma at st. As these figures show, there is nothing peculiar about the shape of the guard cells, and they are not depressed below the general surface, though the sparseness of the stomata and the thickness of the cuticle indicate the xerophytic construction of the plant. Cortex.—The outer cortex (Pl. VI, fig. 35, 0.c.) consists of one to four layers of cells which are large relatively to the cells of the epidermis and of the inner cortex. It is best marked in the lower and thicker portions of the aerial stems. The cells of the outer cortex generally appear clear and empty-looking, and are slightly longer than broad. The cells of the inner cortex are round in transverse section and were separated by fairly large intercellular spaces (Pl. VI, figs. 88 and 40). They are longer than broad, this being more marked on passing from above. downwards in the stems (of. Pl. VII, fig. 42, with fiz. 43). As preserved, the inner cortex often has a darker brown colour which is specially marked in the cells immediately below the outer cortex (Pl. V, figs. 21, 22, and 24). Ina few cases the cells of the inner cortex were filled with closely crowded bodies suggestive of chloroplasts or possibly starch (Pl. VI, fig. 39). The inner cortex was especially liable to decay, and stems are often found in which it had wholly disappeared while the stele and outer cortex remained. That the inner: cortex in the aerial stems probably constituted the assimilating TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 24), 112 770 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS tissue while the outer cortex is to be regarded as a hypodermal layer, is shown not merely by the characters of their respective cells, but by their relations to each other in the neighbourhood of a stoma. This is seen in the transverse section on Pl. VI, fig. 34. The hypoderma, which in this case was clearly marked and con- sisted of two layers of large cells (fig. 34, 0.¢.), is not developed beneath the stoma (st.), but the inner cortex (7.c.) here extends to the surface. Thus the well- developed system of intercellular spaces in the inner cortex was placed in communi- cation with the external atmosphere. In the longitudinal section shown in fig. 37 the inner cortex had disappeared, but the interruption in the hypoderma beneath the stoma is beautifully shown. This, as fig. 34 showed, was originally filled by cells of the inner cortex. While there is no doubt as to the hypodermal nature of the outer cortex in the aerial stems, it must be borne in mind that a distinction of this narrow zone of cells from the inner cortex is traceable throughout the plant. This holds even for regions which there is no reason to think had a specialised hypoderma or assimi- lating inner tissue. Thus it has been seen in the rhizomes (Pl. IV, figs. 15 and 19), and even in small axes without a stele (Pl. X, fig. 73). The stem shown on Pl. X, fig. 74, is peculiar in having one sector of the transverse section composed of cells of larger size. This applies both to the hypoderma and the inner cortex. Stele-—The stele throughout the plant is composed of a central strand of tracheides surrounded by a zone of phloem. It is not delimited from the inner cortex by any layers which can be interpreted as endodermis or pericyele. The stele exhibits a considerable range in size, partly in relation to the region of the plant, and partly in relation to the stoutness of the individual stems (PI. V, figs. 21— 25; Pl. VII, figs. 44-46). Thus the stele represented in Pl. VII, fig. 45, isa fairly large one, and has a strand of xylem composed of numerous tracheides; while that in fig. 46, which belongs to the slender stem represented in fig. 25, had only two tracheides. Slender axes are also met with in which no vascular tissues have been differentiated (Pl. VIII, fig. 60; Pl. X, fig. 73). The phloem in transverse section (Pl. VII, figs. 41 and 44) is composed of thin- walled elements, four- to six-sided, and fitted closely together. No distinction can be made between the various elements in this zone, which is about four or five cells in depth. In longitudinal section (Pl. VI, fig. 42) the elements composing the phloem are much longer than the cells of the inner cortex. They further differ from the cortical cells in their end walls being oblique instead of transverse. As comparison of figs. 42 and 43 will show, the elements forming the phloem become shorter and less characteristic in the stele of the more slender upper region of the stem. Although sieve plates have not been found, the position of this zone, its clear appearance, and the form of its component elements seem ‘to justify its — recognition as true phloem. — 7a: - ‘ : SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 771 The xylem strand, whether composed of few or many tracheides, is always solid, no parenchyma being mixed with the tracheides. No distinction between proto- xylem and metaxylem can be drawn, all the tracheides being alike. The thickening of the tracheides was annular (PI. VII, figs. 47 and 48). It had the form of rather broad rings, which give the tracheides the appearance of being transversely barred. Occasionally two of the bars converge and, uniting at one end, take the shape of a Y or V, but neither definite spiral thickening nor the passage to a scalariform type of thickening has been seen. Hemispherical Projections.—In Pl. II, fig. 7, a number of the definite little bulges which occur on the stems are seen in the round. Several of these projections are seen in the transverse section of a stem on PI. V, fig. 27. Their structure is illustrated in greater detail in figs. 49-51 on Pl. VII. In fig. -49 a small bulge is seen from the outside on an obliquely viewed epidermal surface. In _this view the. surface cells of the projection are isodiametric. They fit closely together without intercellular spaces. They contrast in form with the surrounding epidermal cells and have a much thinner outer wall than these, but we have satisfied ourselves that there is unbroken continuity between the two. Two bulges- in longitudinal vertical section are represented in figs. 50 and 51. As is especially clearly shown in fig. 51, the bulge is due to periclinal division in the epidermis and outer cortical cells. The thick outer wall and the cuticle cease to be marked over the projection; this often exhibits a brown discoloration of its outer cells, which are sometimes broken down. Though much smaller, these projections agree in their relation to the tissues with the large bulges on the rhizome (Pl. III, fig. 11). In connection with this it is interesting to find that the superficial cells of some of the small projections grow out as rhizoids similar to those on the rhizome. Examples of these rhizoid- bearing projections will be found on PI. V, figs. 25 and 29, and more highly magnified on Pl. VIII, figs. 52-54. Lateral Branches.—Many of the small projections were the seat of a further development of considerable interest, adventitious lateral branches being produced from them. This is illustrated in figs. 55-61 on Pl. VIII. The tissues at the base of the branch are continuous with the epidermis and outer cortex of the parent axis. The branch in fig. 57 has its own vascular strand, but this does not exhibit any connection with the stele of the parent axis. Such transverse sections through the main axis and branch did not, however, exclude the possibility that the stele of the branch might have continued obliquely downwards as a trace through the cortex of the main axis to join with the stele of the latter. Where, however, as in fig. 61, we see the bases of two fairly large branches in a longitudinal section that includes the stele of the parent axis throughout, and there is no indication of traces passing out, it seems safe to conclude that these lateral branches had no vascular connection with the main axis. 772 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS In fig. 55 several small bulges are present around the stem, only one of which has given rise to a branch. The branch in fig. 56 is itself branching. These adventitious branches appear to have arisen from all regions of the stem. The particular example shown in fig. 58 is from a stem about 4 inches above the ground, in the block shown on PI. II, fig. 5. Some of the branches had a fairly wide base of attachment (Pl. VIII, figs. 56— 59, 61), while others widen out from a very narrow attachment (fig. 60). All intermediate forms are found. The adventitious branches appear to have been readily detached. In fig. 55 this separation of the branch has almost taken place. Specimens with a very narrow stalk-like base of attachment have been met with free in the peat, a good example being shown on PI. X, fig. 72. This branch widens rapidly but has not developed a stele. The distribution in the peat of such detached branches suggests that new plants started from them, and that this was an important method of vegetatively propagating the plant. Doubtless, as the branch grew on, a stele was differentiated in it, but small cylindrical stems of various diameters without any indication of a stele have been frequently met with (Pl. X, fig. 73). Another example is shown in fig. 60, lying’ beside the attached branch, which itself has no indication of a stele. The sear left by the separation of a branch often became the seat of degenerative changes. Its position is then marked by a small black patch of dead tissue. The adjoining cells had frequently elongated at right angles to this, the wound-reaction showing that the separation of the branch had occurred during the life of the plant. The foregoing description of the vegetative organs of Rhynia Gwynne-Vaughan has been based upon the best-preserved specimens. We may add a remark on some results of less perfect preservation in producing a different structural appearance, especially in the region of the stele. The fairly large stem represented in PI. X, fig. 75, shows the commencement of the decay of the epidermis and of the inner cortex. When this is more advanced it leads to the complete disappearance of the inner cortex, only the stele and the thin cylinder of the outer cortex being preserved. A further change affecting the stele has been observed in numerous examples of — partially decayed stems of various sizes. It is illustrated in the large steles shown in figs. 76-78, Pl. X. The markings on the xylem disappear, while dark material is deposited throughout the stele (fig. 76). This leads to a dark core representing the xylem surrounded by a dark zone in the position of the phloem. A somewhat — similar condition is represented in fig. 77, where, however, the inner portion of the — phloem has broken down, leaving a clear space between a dark central mass and a — surrounding dark ring. In fig. 78, though decay is less advanced, the xylem (#.) actually appears lighter in tint than the phloem (ph.), and has lost all its thickening, while the zone of phloem appears to consist of elongated and very dark pointed elements. : This condition of preservation will be referred to further in the concluding remarks, . SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 773 SPORANGIUM. The aerial stems of Rhynia Gwynne-Vaughani have been traced for at least 6 inches from the surface of the soil, but the block shown in PI. I, fig. 5, unfortu- nately did not contain the terminal portions of the stems. Both dichotomous and lateral branching were occasionally observed in the upper regions. No reproductive organs have, however, been seen attached to these stems, though presumably they would occur on the higher portions. It is impossible therefore to say whether the main axis terminated in a sporangium or whether sporangia were borne on special lateral branches, though, judging from the size of the sporangium, the latter sup- position appears less probable. A number of specimens of sporangia have, however, been met with in the sub- stance of the silicified peat (Pl. [X), while free spores were disseminated throughout the matrix. It might have been inferred with reasonable certainty from the purity of the vegetation that these sporangia were the reproductive organs of Rhynia. The small specimen shown in figs. 63 and 63a fortunately places the matter beyond doubt, for the sporangium is here borne terminally on a slender axis with the characteristic vascular strand and tracheides described above. Another and larger sporangium was cut in a series of transverse sections (slides Nos. 2417-2422), and following the series down, the basal region of the sporangium was found to be continuous with a badly preserved but characteristic axis of Rhynia (cf. Pl. IX, fig. 69). The sporangium attained a length of at least 12 mm. and a breadth of 2°5 mm., and the axis it terminated was about 1°5 mm. in diameter. More precise measure- ments cannot be given, as the sporangia varied considerably in size. It was cylindrical in form, though as enclosed in the peat it is generally slightly flattened (figs. 64 and 65). The form of the base of the sporangium and its junction with the stalk is best shown in fig. 62. The apical region of this specimen appears somewhat more rounded than was actually the case, owing to the section being slightly tangential. As fig. 64 shows, the sporangium was more pointed. In several cases a pair of sporangia have been met with lying side by side in the silicified peat (figs. 65, 69). This is suggestive of a possible junction of their stalks, but proof of this is wanting. The wall of the sporangium is + mm. thick, and is differentiated into several layers (figs. 66-67). In fig. 66, which represents more highly magnified a portion of the wall of the sporangium shown in fig. 65, the wall is seen in true transverse section. The cells of the epidermal layer are narrow, and greatly extended at right angles to the surface. They have thick walls, and are covered by a well-marked cuticle. In this specimen the middle layers of the wall had wholly decayed, their position being represented by the clear space which separates the epidermal layer 774 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS (ep.) from what we interpret as a persistent tapetal layer (tap.). The cells of the latter appear rounded, have thin walls, and vary in size. For the most part the tapetum consists of a single layer with a rather irregular surface towards the cavity of the sporangium, but at places it is two cells thick. The tissue intervening between the epidermis and the tapetum was evidently more delicate and is not well preserved in any of our specimens. It is best shown in fig. 68, Pl. IX, and consisted of a considerable number of layers of small, thin- walled cells. The corresponding layers are recognisable in the sporangial wall when cut longitudinally (fig. 67). The tapetal cells were seen to be longer than broad. In this view the epidermal cells appear almost square. This difference from their appearance in transverse sections (fig. 66) is explained by the actual form of these cells as seen from the outside, being narrowly fusiform, like the epidermis of the stem, though on a smaller scale. The thick-walled epidermal layer appears to have been uniformly continuous over the surface of the sporangium. No satisfactory indication of any lines of dehiscence has been detected in any of the sections, though specially looked for. The sporangium contained an enormous number of spores. In some cases these were still united in tetrads (Pl. X, fig. 70), each spore showing a convex outer wall and a three-sided inner face where it adjoined its sister cells. .In other sporangia the spores were separate. They exhibit some variation in form and size but on an average measure about 65 m in diameter. The spores found scattered through the peat (Pl. -X, fig. 71) resemble those in the sporangia, but had increased slightly im size. Only the cuticularised wall is preserved. One of the spores in fig. 71 shows the triradiate marking on its inner face. : No stages in the germination of the spore have been seen, nor has the gameto- phyte been found. . SUMMARY. 1. The plants grew in a gregarious fashion in a peaty soil practically composed of the decaying remains of the same species. This land surface was probably in the neighbourhood of water, and liable to periodic mundations. 2. The plant had no roots and no leaves. It was entirely composed of branched cylindrical stems. 3. The branched underground rhizomes were attached to the peat by numerous rhizoids, most abundant on large, downwardly directed bulges of the outer cortex. 4. Some of the branches grew upwards as tapering aerial stems. 5. The aerial stems bore small lateral projections irregularly scattered over the surface. 6. Some of the projections, possibly in the lower region, developed rhizoids. —_ i SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 775 7. Some of the projections at various levels on the stem gave rise to adven- titious lateral branches. 8. Some of the lateral branches, attached by a narrow base, were readily detachable and probably served for vegetative propagation. 9. Dichotomous branching of the stem occurred sparingly. 10. In the rhizomes and stems, epidermis, outer cortex, inner cortex, and stele can be distinguished. 11. The epidermis in the aerial stems had a thick outer wall and stomata were sparingly present. 12. The cortex consisted of a narrow outer zone, which in the aerial stems had the character of a hypoderma, and a broader inner cortex. The more delicate tissue of the inner cortex had intercellular spaces and was in relation with the stomata. It possibly represented the assimilating tissue. 13. The vascular system consisted throughout of a simple cylindrical stele composed of a slender solid strand of tracheides with broad annular thickenings and no distinction of protoxylem and metaxylem. Surrounding the xylem was a zone of phloem consisting of elongated thin-walled elements. 14. No vascular strands were given off to the small projections on the stem. 15. No vascular connection existed between the stele of a lateral branch and the stele of the parent axis. 16. In the dichotomous branching of the stem the stele divided to supply the two branches. _ 17. The plant bore large cylindrical sporangia. The sporangium had a thick wall, and terminated a stout stalk which corresponded to a small stem. 18. The sporangium contained numerous spores which were all of one kind. ConcLUDING REMARKS. ‘The plant which has been described (whatever the precise age of the Old Red Sandstone Beds in which it is found may prove to be) is the most ancient land plant of which the structure is at all fully known. By a fortunate circumstance of its preservation in large quantity as it grew, its external form, structure, and sporangia are known almost as well as if we were dealing with an existing species. It will be evident that the simplicity of the general organisation and of the anatomy of Rhynia has important bearings on the origin of the sporophyte and its differentiation into stem, root, and leaf in the Pteridophyta. While fully alive to the interest of this, we do not propose in the present paper to consider the bearing of the new facts here brought forward on these speculative questions. We hope to _ consider them later. In attempting to indicate the position of Rhynia Gwynne-Vaughani in the vegetable kingdom, it will be sufficient to briefly compare it on the one hand with 776 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE: PLANTS the existing group of the Psilotales, and on the other hand with one less perfectly preserved Devonian plant, Psilophyton princeps, Dawson. Even these comparisons will not be developed fully until Asteroxylon has been described in the next paper of this series. The Psilotales, with the two existing genera Psilotwm and Tmesipteris, have always presented difficult morphological problems. ‘They are rootless, the under- ground parts consisting of rhizomes bearing rhizoids. The leaves are small and without vascular system in Psdlotwm, larger and with a vascular supply in T’mesi- pteris.. Their reproductive organs consist of bi- or tri-locular synangia, subtended by a pair of leaf-like lobes. This fertile structure has been variously interpreted as a bifid sporophyll subtending an adaxial synangium or sporangiophore, or as a lateral branch bearing a pair of leaves and terminating in a synangium. The Psilotaceze agree with Rhynia and differ from all other Pteridophyta in the absence of roots and (if the second interpretation be accepted) in the position of the synangia terminating a short branch. ‘There is, further, a striking general agreement between Psilotwm and Rhynia in the plant consisting of more or less delicate subterranean rhizomes, bearing a system of xerophytic stems, and in the occurrence of both dichotomous and lateral branching. We also find in the rhizome or in the ultimate branches of Pszlotwm a parallel to the simplicity of the stele of Rhyma. The Psilotaceze exhibit, however, some important points of difference from Rhynia, such as the presence of leaves, the specialisation of the fertile branches or sporophylls, the more complicated anatomy, the synangia, and the shape of the — spores. The comparison will not be carried further at present, but it may be noted in passing that it would lead us to regard the Psilotaceze as having preserved many — primitive characters and not as reduced. On this view the Psilotaceee would be the little modified survivors in the existing flora of a type of plant that existed in early geological times, the most fully known example of which is now Rhynia Gwynne- Vaugham. It does not, however, follow that a direct line of descent is to be drawn between Rhynia and the Psilotaceze as we know them.“ | With regard to the comparison of Rhynia with extinct plants, it is only necessary at present to consider Psilophyton princeps, Dawson, including under this the variety ornatum, which probably merely represents the lower portions of the stems. We take the description of this species as given by Dawson in the “ Fossil Plants of the Devonian and Upper Silurian Formations of Canada,” + though the plant has been described by him in several of his other works. The main characters * The primitive nature of the general organisation of the Psilotacec is clearly held on other grounds by C. Ee. BERTRAND in his Recherches sur les T'mesipteridées, Lille, 1883, pp. 313-316; and by LianieEr in his “ Equisétales et Sphénophyllales leur origine filicinéenne commune,” p. 95, Bull. Soc. Linn, de Normandie, 5° sér., vol. vii, 1903, p- 93, Caen. + Geological Survey Canada, Montreal, 1871. SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 777 of Psilophyton princeps, as we understand it, are shown on pl. ix, figs. 97-110; pl. x, figs. 112-114, 118-120; pl. xi, figs. 127, 128, 133a, b, c, and 134, 134a, b,c of the memoir cited above.* Psilophyton princeps as thus limited} consisted of upright stems which frequently dichotomised, the weaker branches often appearing as if borne laterally. The stouter stems bore numerous spiny outgrowths but no definite leaves. The growing regions of the stem were circinately coiled. The stems were marked by longitudinal ridges. The finer branches appear to have been destitute of spines. They dichotomised repeatedly, and their ultimate branches terminated in oval sporangia which were borne singly or in pairs. The material of Psilophyton princeps is mainly in the form of impressions. Dawson has, however, figured a few specimens in which the structure was imperfectly preserved,{ and has based a reconstruction § on such remains. The most interesting of these is the axis given natural size in his fig. 133c, and magnified in fig. 134. This had a wide cortex and a single central cylinder. Little of the structure of the cortex was preserved, but the central cylinder is figured and described || as consisting of “an axis of scalariform vessels surrounded by a cylinder of parenchymatous cells and by an outer cylinder of elongated woody cells.” Dawson’s figures are of such interest that they are reproduced in the accompanying text-figs. 1 and 2, together with his description of-the figures. The description of the central bundle of Psilophyton princeps as composed of ~ scalariform vessels” surrounded by ‘‘ woody fibres” is at first sight difficult to understand, but seems to become intelligible in the light of the imperfectly preserved examples of Rhyna described above on page 772. While we have not seen Dawson’s original specimens, we venture to interpret his published figures as exhibiting a stem with a stele consisting of a solid strand of tracheides surrounded by a zone of phloem which had become partially decayed and discoloured much as in the specimens of Rhynia shown in our Pl. X, figs. 76-78. In particular, Dawson’s fig. 134 (see text-fig. 2) may be compared with our fig. 77, and his reconstructed fig. 127 (see text-fig. 1) with our fig. 78. Dawson’s fig. 134@ suggests that the tracheides in his specimens were annular rather than scalariform. What Dawson speaks of as “the outer fibrous cylinder” or “ bark fibres” would correspond to the persistent outer cortex of Rhynia. The “cellular cylinder” in text-fig. 1 would correspond to the inner cortex; this has almost completely dis- * It will be observed that we exclude from Psilophyton princeps Dawson’s figures of “rhizomes,” pl. x, figs. 111, 115,116, and 117. We also exclude at present from Psilophyton all the other species which have been referred to it by Dawson and other writers, as they do not seem to show the characters necessary for their definite reference to this genus, + The course we adopt is in general agreement with the views of Soum-LauBacu, Fossil Botany, pp. 189-192, and more recently of P. Bertranp in “Note preliminaire sur les Psilophytons des gres de Matringhen,” Ann. Soc. géol. du Nord, yol. xlii, p. 157, 1913. { Lec., pl. xi, figs. 133 and 134, § L.c., pl. xi, fig. 127. || L.c., p. 37. TRANS. ROY. SOC, EDIN., VOL. LI, PART III (NO. 24). 113 778 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS CREE G62 0C: fe. X-. ph. Fo. Oe, Trxt-FIc. 1.—Psilophyton princeps, Dawson. ———- +) =) Pe ‘Restored section of magnified stem, showing (qa) scalariform uxis, (b) woody cylinder, — (c) cellular cylinder, (d) outer fibrous cylinder.” (Dawson, J.c., p. 90, pl. xi, fig. 127 ) The letters placed below the figure indicate our int~~»retation of DAwson’s reconstruction. 0.¢., outer cortex ; @.c., inner cortex ; ph., phloem ; #., xyle j PT a A | se es ag SS a Trxt-FIG, 2.—Psilophyton princeps, Dawson, ‘Fic, 188.—Rhizoma, transverse sections showing axis. Natural size. ‘Wig, 134.—The same magnified. 184a, scalariform tissue ; 1346, woody fibres; 134c, bark fibres, 100 diams.” ——— SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 779 appeared in text-fig. 2. The “woody cylinder” or “woody fibres” (6 in text-figs. 1 and 2) would correspond to the phloem of Rhynia, while the “scalariform axis” (a, text-figs. 1 and 2) is the equivalent of the central strand of xylem. To make this interpretation clear, we have added letters below text-fig. 1 corresponding to the lettering used throughout our plates. This text-figure may further be compared with the longitudinal sections shown in figs. 26, 42, and 59 of this paper. If this interpretation be correct, there would be a substantial agreement in structure between Psilophyton princeps and Rhynia Guwynne-Vaugham. The two plants further agree in bearing large oval sporangia on the ends of ultimate branches of the stem. Psilophyton princeps differs from Rhynia, however, in the presence of spines, in. the more profuse dichotomous branching, in the subordination of some of the branches to a sympodial main axis, and in the absence, so far as we know, of lateral adventitious branches. On these grounds we regard Psilophyton as exhibiting characters which ally it with the more fully known Rhynia, and separate these two genera from all other vascular plants. The differences b:tween Rhynia and Psilophyton, however, warrant their being treated as distinct genera. Tt will be clear that Rhynia and Psilophyton belong to the Vascular Cryptogams or Pteridophyta, but they ca not be placed in any of the Classes of this great group as at present known and defined. These Classes are the Filicales, Hquisetales, Sphenophyllales, Psilotales, and Lycopodiales. It is therefore necessary to recognise another group of Pteridophyta, of equivalent value to those mentioned, to include Rhynia Gwynne-Vaughani and certain of the specimens described under the name of Psilophyton princeps. This Class 1s characterised by the sporangia being borne at the ends of certain branches of the stem without any relation to leaves or leaf-like organs. For this Class we propose the name Psilophytales. This name is derived from that of the earlier described though less perfectly known genus Psilophyton, and further suggests the resemblance between the plants of this class and the existing Psilotales. Whether the other characters in which Rhynia and Psilophyton seem to agree will prove to be common characters of the whole group, must be left open for the present, but they are not essential for its definition. 780 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS CLASSIFICATION AND DIAGNOSIS. PsitopHytTaLes: A Class of Pteridophyta characterised by the sporangia being borne at the ends of certain branches of the stem without any relation to leaves or leaf-like organs. Rhyna Gwynne-Vaugham, Kidston and Lang, n.g. and n. sp. Psilophyton princeps, Dawson (pars). “Fossil Plants of the Devonian and Upper Silurian Formation of Canada,” Geol. Survey Canada, 1871, p. 37, pl. ix, figs. 97-110; pl. x, figs. 112-114, 118-120; pl. xi, figs. 127, 128, 133, 134. Raynta, Kidston and Lang, n.g. Diagnosis.—Aerial stems without leaves or spines but bearing small protuber- ances ; sporanginm large, cylindrical, terminal on ultimate branches. Rhyna Gwynne-Vaughan, Kidston and Lang, n. sp. Diagnosis.—Plant gregarious, rootless and leafless. Underground rhizomes with rhizoids, generally situated on large, downwardly directed protuberances of the cortex. Aerial stems cylindrical, tapering upwards, about 8 inches in height, bearing small hemispherical protuberances. Stems sparingly dichotomous and also bearing lateral adventitious branches. Sporangium large, cylindrical, and terminating an aerial stem. Sporangial wall thick, of many layers of cells. Homosporous, spores developed in tetrads about 65 p in diameter. Stele throughout the plant small, cylindrical, consisting of a solid strand of annular tracheides, surrounded by a zone of thin-walled phloem. Cortex consisting of an inner and outer zone. LEpidermis of aerial stems with cuticularised outer wall and stomata. Locality.—Muir of Rhynie, Aberdeenshire. Horizon.—Old Red Sandstone (not younger than the Middle Division of the Old Red Sandstone of Scotland). In conclusion, we wish to express our thanks to Dr W. Mackie, Elgin, for placing his slides in our hands for description as well as supplying us with material ; to Dr J. 8. Fier, F.R.S., Director of the Geological Survey of Scotland, for kind assistance ; and to Mr D. Tarr for much help given when examining the chert band at Rhynie. We also gratefully acknowledge our indebtedness to the Executive Committee of the Carnegie Trust for a grant towards defraying the expense of the plates illustrating this memoir, SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE 781 EXPLANATION OF PLATES. (All the figures are from untouched photographs. ) Rhynia Grwynne-Vaughani, Kidston and Lang. Prats I. Fig. 1. Vertical surface of a block from the chert band (E 10 of section on page 762), etched with hydrofluoric acid to render the plant-remains more distinct. The dark layers s,s and s’, s’ are silicious sandstone with carbonaceous matter. From s, s to the base of the block and between s, s and s’, s’ are two layers of the silicified peat (P 1, P 2) composed of the remains of Rhynia Gwynne-Vaughant. The structure of some of the larger stems is clearly shown, especially in the lower layer of peat. Natural size. (No. 5282.) Puate II. Fig. 2. Portion of the vertical surface of a block of the Rhynie chert which had been weathered naturally. The rounded stems of Rhynia stand in low relief from the surface. Up to the level a-a, which marks an original land surface, the stems lie horizontally in the silicified peat, but above this level they are seen to become vertical. Natural size. Fig. 3. Portion of the horizontal surface of another block of the Rhynie chert showing the crowded stems of Rhynia flattened down and compressed. Natural size. (No. 5283.) Fig. 4. A small portion of fig. 3 enlarged. At a and at other places the narrow ribbon-shaped com- pressed stems show a raised median line marking the position of the stele. 2. (No. 5283.) Fig. 5. A loose block of the Rhynie chert, etched with hydrofluoric acid and viewed from the side. The lower portion up to the level of the old land surface a, a shows remains of Rhynia lying in all directions in the silicified peat. From this level to the summit of the block the tapering aerial stems can be followed bending over to the right. Natural size. Puate III. Fig. 6. Small stem exposed on a fractured surface of the chert showing the rounded form of the stem and the epidermal cells which in this case have no median line. x14. (No. 5286.) _ Fig. 7. Similar specimen showing a number of the small projections or bulges. The epidermal cells in this and the following figure have the median ridge. x14. (No. 5284.) Fig. 8. Hollow impression from which the stem has been removed showing the epidermis and three of the small projections. x14. (No. 5285.) Fig. 9. Portion of a ‘vertical section through the peat. The stems in the lower part are more compressed and decayed, while those above are uncompressed and separated more widely by interstitial matter. x54. (Slide No. 2410.) Fig. 10. Portion of a similar vertical section through the silicified peat showing still more marked decay and crushing of the stems ataand b. x54. (Slide No. 2388.) Figs. 11-12. Enlargements of portions of fig. 13, Pl. IV. For description see explanation to fig. 13. Puate IV. Fig. 13. Portion of section through peat composed of much decayed stems in which three rhizomes are shown in their natural position. On the downwardly directed sides of all the rhizomes the epidermis and outer cortex show increased development, which in the case of the longitudinal section is seen to have resulted in the formation of two definite rhizoid-bearing bulges, a and 6. A portion of the bulge a is more highly magnified in fig. 11, Pl. III, and of the bulge 0 in fig. 12. These figures, and especially fig. 11, show the origin of the bulge by increased growth and division of cells of the epidermis and outer cortex. 0.¢., outer cortex ; t.c., inner cortex. Fig. 13, x14. Figs. 11-12, x 60. (Slide No. 2412.) Fig. 14. Transverse section of a rhizome attached to the peat by rhizoids (rh.), and sending up an aerial stem. Close to the base of the latter at st. a stoma was recognised. x20. (Slide No. 2396.) Fig. 15. Small rhizome preparing to branch, the xylem of the stele having divided, x20, (Slide No, 2396.) 782 DR R. KIDSTON AND PROF. W. H. LANG ON OLD RED SANDSTONE PLANTS Fig. 16. Tangential section showing the isodiametric epidermal cells of a rhizome in surface view. rh., rhizoids. x60. (Slide No. 2396.) Fig. 17. Two rhizomes in the peat in transverse section. Further description in text. «14. (Slide No. 2396.) Fig. 18. Section through a rhizoid-bearing bulge of another rhizome showing the long non-septate rhizoids extending downward into the peat. rh., rhizoids, x28. (Slide No. 2411.) Fig. 19. ‘Transverse section of rhizome. 60. ep., epidermis; 0.¢., outer cortex ; z.¢., inner cortex ; ph., phloem ; 2, xylem. (Slide No. 2396.) Puate V. Fig. 20. Section through chert showing cylindrical aerial stems in an almost pure silicious matrix free from foreign vegetable matter. x9. (Slide No. 2408.) Figs. 21-25. Transverse sections of aerial stems of different diameters. x20. The various regions of the stem distinguished by lettering in figs. 21 and 23 can also be recognised in the other figures. ep., epidermis ; 0.c., outer cortex ; 7.¢c., inner cortex ; ph., phloem; w., xylem. Fig. 21 (Slide No. 2398). Fig. 22 (Slide No. 2398). Fig. 23 (Slide No. 2397). Fig. 24 (Slide No. 2399), Fig. 25 (Slide No. 2416). Fig. 26. Portion of a longitudinal section of an aerial stem. ep., epidermis; 0.¢., outer cortex ; zc, Inner cortex; ph., phloem; «., xylem; pr., small projection. x20. (Slide No. 2390.) Fig. 27. Transverse section of au aerial stem showing three projections. pr., projections. x 20. (Slide No. 2406.) : Figs. 28-29. Transverse sections af two stems showing small’ projections bearing EES pr.) pro- jections ; 7h., rhizoids. x20. (Slide No. 2389.) Fig. 30. Transverse section of an aerial stem showing the stele dividing preparatory to dichotomous branching. x20. (Slide No, 2392.) Puate VI. Fig. 31. Epidermis in surface view showing the form of the cells and the dark median line. x60. | (Slide No, 2390.) ; Tig. 32. Stoma with surrounding epidermal cells in surface view; median line absent. Cf. fig. 31. | x 160. (Slide No. 2394.) Fig. 33, Stoma in transverse section showing the two guard cells and the pore. ~x160. (Slide No. 2408.) Fig. 34. Transverse section of the epidermis and the underlying tissues in the neighbourhood of a | stoma. st., stoma; ep., epidermis; 0.c,, outer cortex interrupted beneath stoma ; 7.c., inner cortex extending outwards to beneath stoma. x 160. (Slide No, 2405.) y Fig. 35. Epidermis and underlying tissues in two adjacent stems, . ep., epidermis; 0.c., outer cortex ; >| z.c., inner cortex. x60, (Slide No, 2405.) = Fig. 36. Transverse section showing the cuticular ridges of the epidermal cells. Cf. fig, 31. x60. | (Slide No. 2389.) zl Fig. 37, Longitudinal section through the epidermis and outer cortex in the region of a stoma (st.). The inner cortex which filled the interruption in the outer cortex has decayed. x60. (Slide No, 2395.) : Fig. 38. Transverse section showing the rounded cells and intercellular spaces of the inner cortex. x60. (Slide No. 2398.) > | Fig. 39. Transverse section of inner cortex showing the cell contents partially preserved. 160. | (Slide No. 2395.) " Fig. 40. Transverse section of an aerial stem. ep., epidermis ; 0.c., outer cortex, not well marked ; iby inner cortex with large intercellular spaces ; ph., cee ;,%., xylem, x ‘60. (Slide No, 2397.) = . Puate VII. , Fig. 41. Transverse section of stem with well-marked stele. ph., phloem; a, xylem. x60. (Slide No. 2392.) : Fig, 42. Portion of longitudinal section of stem shown in fig. 26 more highly magnified. ep., epidermis ;_ 0.¢., outer cortex ; 7.¢., inner cortex ; ph., phloem; z., xylem. x60. (Slide No. 2390.) — i SHOWING STRUCTURE, FROM THE RHYNIE CHERT BED, ABERDEENSHIRE. 783 Fig. 43. Longitudinal section of stele of the upper region of an aerial stem. 7.c., inner cortex; ph., phloem ; x, xylem. x60. (Slide No. 2415.) Fig. 44, Transverse section of small stele to show the phloem. ph., phloem; a, xylem. x 160. (Slide No. 2391.) Fig. 45. Transverse section of xylem of stele represented in fig. 41. ph., phloem; #., xylem. x 160. (Slide No. 2391.) Fig. 46. Transverse section of the stele of the small stem shown in fig. 25. 7c, inner cortex; ph, phloem ; x., xylem consisting of two tracheides. 160. (Slide No. 2416.) Figs. 47 and 48. Longitudinal sections of the xylem showing the broad annular thickening of the walls of the tracheides. 250. Fig. 47 (Slide No. 2403). Fig. 48 (Slide No. 2400). Fig. 49, Part of an oblique section of an aerial stem showing the epidermis (ep.) in surface view, and also a surface view of one of the small projections (p7.). x60, (Slide No. 2390.) Figs. 50 and 51. Two longitudinal sections passing through small projections. ep., epidermis ; 0.c., outer cortex; i.c., inner cortex. x60. Fig. 50 (Slide No. 2390). Fig. 51 (Slide No. 2415). Puate VIII. Figs. 52, 53, and 54. Examples of small projections (pr.) bearing rhizoids (7h.) shown in transverse sections of stems. Figs. 52-53, x60. Fig. 54, x 160. (Slide No, 2389.) Fig. 55, Transverse section of a stem showing two projections (p7.) aud a partially separated branch (ir.) which is leaving a scar (sc.), x60. (Slide No. 2413.) Fig. 56. Obliquely transverse section of a stem bearing a lateral branch (b7.) attached by a fairly broad base. The adventitious branch appears to be itself branching. «., xylem of parent stem; a’., xylem of branch, x20. (Slide No. 2409.) Fig. 57, Transverse section of stem with an adventitious branch (br.) attached by a fairly narrow base. %., xylem of parent stem; «’., xylem of branch. x20. (Slide No. 2402.) Fig. 58. Longitudinal section from the higher region of the stem showing on the right a branch (br.) with a fairly wide base, while on the left the base of another branch is cut obliquely. x20. (Slide No, 2414.) . Fig. 59. Longitudinal section, becoming oblique on the left, of a stem with well-marked tissues. The branch (dr.) is attached by a broad base. ep., epidermis; 0.¢., outer cortex ; 7.¢., inner cortex; ph., phloem ; x., xylem of parent stem; 2’., xylem of branch. x20. (Slide No. 2401.) ig. 60. Transverse section of a stem with a small ‘projection (pr.) and a branch (07.) attached by a very narrow base. The branch has no stele, and a transverse section of another small axis without a stele lies beside it. x60. (Slide No. 2413.) Fig. 61. Median longitudinal section of a stem showing at br’. and br?. the bases of two lateral branches. No vascular connection exists between these and the parent axis. w., xylem of parent axis; #., xylem of one of the branches, x20. (Slide No. 2404.) Puate IX. Fig. 62. Slightly tangential longitudinal section of a sporangium terminating a fairly stout stem. The sporangium, which is filled with an enormous number of spores, is lying in the peat. x5}. (Slide No. 2393.) Fig. 63, Small ESEy sporangium (sp.) borne on a slender stem (aa.) and lying in the peat. x 5}. (Slide No. 2392.) Fig. 63a. The sporangium shown in fig. 63 more highly magnified. ep., thick-walled epidermis ; tap., tapetal layer ; a., xylem of axis; a, accidentally contracted junction of axis and base of sporangium. x 20. (Slide No. 2392.) Fig. 64. Longitudinal section of another sporangium showing the pointed apex. 7. (Slide No. 2396.) Fig. 65. Two sporangia (s. and s’.) cut transversely as ey lay side by side in the peat. x 14, (Slide “No. 2411.) i Fig, 66. Portion of the sporangium s. in fig. 65 more highly magnified. ep., thick-walled epidermis ; m.l., decayed middle layers of wall; tap., tapetum; sp., spores. x60. (Slide No. 2411.) 784 OLD RED SANDSTONE PLANTS FROM RHYNIE CHERT BED, ABERDEENSHIRE. Fig. 67. Portion of the wall of the sporangium in fig. 62 more highly magnified. ep., thick-walled epidermis ; m./., decayed middle layers of wall; tap., tapetum ; sp., spores. x60. (Slide No. 2393.) Fig. 68. Transverse section of another sporangium. ep., thick-walled epidermis ; m./., middle pipe of wall; ¢ap., tapetum; sp., spores. 14. (Slide No. 2395.) Vig. 69. Section roth the silicified peat passing transversely a sporangium (s.) and an axis which bore another sporangium, The axis is shrivelled and altered, but in a manner characteristic of Khynia, c., cortex of axis; v.s., vascular strand. x11. (Slide No. 2421.) PuatTe X. Fig. 70. Spores, some still united in tetrads, from the sporangium shown in fig. 68. x 160. (Slide — No, 2395.) ¥ Fig. 71. Spores free in the silicified peat, one showing the triradiate marking. x 160. (Slide — No. 2395.) Fig. 72. Adventitious branch widening out from very narrow base and occurring free in the silicinedl peat. x33. (Slide No. 2406.) Fig. 73. Transverse section of small stem without any stele. 60. (Slide No. 2391.) Fig. 74. Transverse section of stem, the cortical cells of which are of larger size in one sector than elsewhere. x20. (Slide No. 2392.) Fig. 75. Transverse section of large stem showing commencing decay of the i inner and gee - the: outer cortex. x20, (Slide No. 2387.) +4 Fig. 76. Transverse section of a partially decayed stem. o.c., remains of outer cortex ; ph., phloem 5 z., xylem. ‘The space marked 7.c. was originally occupied by inner cortex. x60, (Slide No, 2393. yo 3 her Fig. 77. Transverse section of a similar stem to that in fig. 76, but with the inner portion of the phloem ‘ broken down. Lettering as in fig. 76. 60. (Slide No. 2393.) | Fig. 78. Longitudinal section of the stele of a partially decayed stem; the xylem (z.) vee lest its | hiekenmes while the phloem (ph.) has assumed a dark colour. x 60. (Site No. 2395.) a (All he figured specimens are in the collection of Dr R. Krpstoy. ) { Vol. LI. \Trans. Roy. Soc. Edin® PLATE I. KIDSTON anp LANG. Bemrose, Collo., Derby. RHYNIA GWYNNE-VAUGHANT, K&L. Trans. Roy. Soc. Edin® VoL Lt KIDSTON anp LANG. PLATE II. Bemrose, Collo., Derby. RHYNIA GWYNNE-VAUGHANI, K&L. Vol. LI. Trans. Roy. Soc. Edin® PLaTE III. KIDSTON anv LANG. Photo iston, Bemrose, Collo,, Derby, HANI, K&L. = I RHYNIA GWYNNE-Vauc ? Z\ae Mu 5 Up hoe * > (yeas & ) ae Sura. WS : } . is. Roy. Soc. Edin* Vol. LL. STON anD LANG. PLATE IV. Bemrose, Collo., Derby RHYNIA GWYNNE-VAUGHANI, K.& L. s. Roy. Soc. Edin’ : | Vol. LI UpSTON anp LANG. . oe Ghis is De eet Bemrose, Collo.,Derby RHYNIA GWYNNE-VAUGHANI, Kia I. = > 4 = — [ea 2 et & 5 > = : Wee: =| & < a os) 5 ‘a é Wes a re Z. = ae 5 eal ; 3 i) © 2.4 0 B : eee 4 4 eS § e is 3 wa ON a ae aS on ae, Vol. LI. hns. Roy. Soc. Edin= PLATE VII. RHYNIA. GwWYNNE-VAUGHANI, K & L. ins. Roy. Soc. Edin* 4}STON anv LANG. wR PSS > veo r ; aS, oF TT Bemrose, Collo., Derby. RHYNIA GWYNNE-VAUGHANI, K&L. Rs ie ans. Roy. Soc. Edin* . Vol. LL \STON anv LANG. PLaTE X. an Bemrese Colla Denny RHYNIA GWYNNE-VAUGHANI, K & L. ( 785 ) XXV.—tThe Prothallus of Tmesipteris Tannensis. By Professor A. Anstruther Lawson, D.Sc., F.L.S. (With Three Plates.) (MS. received May 9, 1916. Read June 4, 1916. Issued separately April 21, 1917.) The Psilotacez constitute one of the most interesting of existing Pteridophyte groups. This interest is mainly due to their phylogenetic isolation. They are perhaps the most isolated of existing types—showing no close affinity to other Pteridophytes. They are very highly specialised in their anatomy, habit, and habitat, and limited to a comparatively narrow geographical distribution. They include two probably monotypic genera—T'mesipterrs and Psilotwm—which are closely related. Although both genera are limited to the tropics and sub-tropics, Psilotum has a much wider distribution than TJmesipteris. The latter is confined to the South Sea Islands, Australia, New Zealand, and parts of Polynesia. Up till recent years they have been classed with the Lycopodiales, but this has been more a matter of convenience than an indication of close relationship, and has only served as a temporary classification until more knowledge has been obtained regarding them. The peculiar nature of these plants, as shown in their structure and in their habitat, suggests that they are highly specialised remnants of a much larger group which has practically become extinct. Recent inquiry into the structure of the sporophyte of both plants has made this much more than a suggestion. As a result of the careful investigations of Scorr (1908), Boopiz (1904), Bower (1894-1908), Forp (1904), and others, there has been a marked tendency to remove the Psilotacex from their temporary position among the Lycopodiales and associate them with the extinct Sphenophyllales. Scorr (p. 631) states that it seems best to regard the Psilotacex as forming a class of their own, the Psilotales, under the main division Sphenopsida—their closest affinity under this class being the Sphenophyllales. This probable affinity of the Psilotacez to such an ancient and long extinct race as the Sphenophylls, and the great uncertainty of their relationship to other existing types, have awakened a great interest in these two specialised and isolated genera. When, in addition to this, we consider that the Psilotacee is the only group of Pteridophytes in which the gameto- phyte generation and embryo are not yet definitely known, our interest in these plants becomes twofold, and any new information throwing additional light on their life-histories will be welcomed. Our knowledge of the gametophyte generation of the Pszlotacee consists entirely of a description of a structure supposed to be the prothallus of Psilotum. This supposed prothallus was described by Lane in 1904, and his account of it is based upon the study of a single specimen. It was found embedded among the adventitious ‘roots of a tree-fern and associated with a Psilotwm sporophyte which had already produced aerial shoots and rhizomes. It was more or less cylindrical in form, about TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 25). ; 114 786 PROFESSOR A. ANSTRUTHER LAWSON ON a quarter of an inch in length, and about three-sixteenths of an inch at its broader end. It tapered off to a point at the opposite end, and had the general appearance and symmetry of a prothallus of a certain species of Lycopodium. Lane identified — two regions in the prothallus, as in Lycopodiwm—a lower vegetative region and a wider upper region to which the reproductive organs were confined. The lower vegetative region was of a brown colour and bore numerous rhizoids. The structure is subterranean in habitat, wholly saprophytic, and seems to correspond to the type of prothallus of Lycopodium clavatum or complanatum. It bore several antheridia, — but no evidence of archegonia or embryo was revealed. Since only one specimen was found, the knowledge derived from its study is very fragmentary, and there was no real evidence to prove that the prothallus described is really that of Psilotwm. Doubt certainly exists, and the uncertainty is freely acknowledged by the discoverer. — In regard to the other genus of the Psrlotacew, nothing whatever concerning the prothallus has so far been recorded. We have no knowledge of the gametophyte — generation or embryo of Tmesipteris. With the exception of the doubtful and very uncertain prothallus described by Lane, our knowledge of the gametophytes and embryo of the Pszlotacee may be regarded as a complete blank. It seemed therefore to me, upon my arrival in Australia in February 1913 to occupy the Chair of Botany in the University of Sydney, that an opportunity had at last offered itself for filling in this interesting gap in our knowledge of the Pteridophytes. A day or two after my arrival I learned that both 7Tmesipteris and Psilotum were to be found growing in great abundance in the immediate vicinity of Sydney. It was hoped that a careful search would confirm Lane’s account of Psilotum and reveal the gametophyte of Tmesipteris. After close attention to this matter; I am now happy to be able to report that my efforts have been partly successful. I have discovered several specimens of a structure which I believe un- doubtedly to be the prothallus of Tmesipteris, and a single specimen of a structure that I believe to be the prothallus of Psilotum. The latter structure, however, bears no resemblance to the supposed prothallus described by Lane. In the search for the prothalli of these plants two methods were followed. Search was made in the field for the prothalli as they occur in nature; and spores were sown in various media with the hope of their germinating and developing prothalli. In the latter method many attempts were made to cause the germination of the spores in the soil in which T'mesipteris and Psilotum sporophytes were growing. A soil was made by taking the surface from the trunks of Dicksonia, Alsophila and Todea, on which Tmesipterts grows very freely. It is also a fact to note that Tmesipte ‘is grows quite freely in soil in nature, as well as on tree trunks; and such soil was also taken as a medium in which to germinate the spores. Cuts were made in the tree trunks with a sharp axe, and in the cleft thus made spores were abundantly sown Similar experiments were made with Psilotum. It sometimes happens that Psilotum and Tmesipteris grow together under very similar conditions, but this is rather the THE PROTHALLUS OF TMESIPTERIS TANNENSIS. 787 exception than the rule. The general rule is that the two plants require very different habitats. Psilotwm flourishes abundantly in exposed rocky places—especially in the crevices of sandstone cliffs; while Z’mesipteris, on the other hand, requires shady, moist conditions, especially in gullies near waterfalls. Spores of both plants were sown in definitely marked places in the field, and have been examined from time to time. Many of these experiments are still under careful observation. When the results of these germinations are more complete, I hope to describe a detailed series of stages. But in the meantime I| propose to give a preliminary account of the mature prothallus of 7’mesipteris, as I have found several specimens associated with young sporophytes in nature. The first of these specimens was found in the substance of the outer part of the trunk of Todea barbara in the month of April 1914, at Bulli Pass, N.S.W. In May of the same year a second oh ee was found. Of the first specimen found I had some doubts, for it showed neither antheridia nor archegonia, but I now believe it to be a portion of the prothallus which was broken off from the rhizome of a young sporophyte. This was to some extent confirmed by the discovery of the ‘second specimen, which showed undoubted antheridia and archegonia. From these two specimens | felt certain in my own mind that I had at last been successful in finding the prothallus, but as only one of them showed the reproductive organs, I thought it better to delay publication until I had confirmed these preliminary observations by the finding of other specimens. It was not until March and April of 1915 that I was able to find additional prothalli, and these I found at Mount Wilson, on the trunks of Dicksonia. In this locality, which is quite one hundred miles north of Bulli Pass, where the first specimens were found, two additional prothalli were found. These were both in close association with very young sporophytes, and showed exactly the same sort of structure as those found at Bulli. In one case the prothallus was in actual contact with the rhizome of a young 7’mesipteris plant, but it became detached while removing the particles of soil and sand with a camel’s- hairbrush. Later on in the same year an additional specimen was found at Sommerby Falls, near Gosford, New South Wales, and I have since found others at Mount Wilson. I have therefore no hesitation in definitely describing these specimens as the gametophyte generation of Tmesipteris Tannensis. GENERAL FEATURES OF THE PROTHALLUS. Compared with the Lycopodiales and other Pteridophytes the prothallus of Tmesipteris is small. The largest specimen examined measured just about one-eighth of an inch in length and varied in thickness, being three or four times longer than broad. It is practically cylindrical in form, but slightly wider at one end than at the other, as shown in figs. 1, 2, and 3. The form, although in the main cylindrical, varies considerably ; no two specimens showed the same shape, some of them even showed branching. The method of its branching does not seem at all 788 PROFESSOR A. ANSTRUTHER LAWSON ON > oe as SS ee. regular. It may bifurcate in a dichotomous fashion or a branch may grow out at right angles to the main axis, as shown in figs. 2 and 3. One of the most striking features of the prothallus is the presence of numerous long rhizoids. These rhizoids are borne on all sides. They are of great length, frequently attaining a length half that of the prothallus itself, and in growing out | from the surface take on very curious twists and spiral shapes, as indicated in figs. — 1,2, and 3. All of the specimens were found at least half an inch below the surface. From this as well as for other reasons I conclude that the prothallus is strictly subterranean. No evidence whatever was found of their appearing on the surface of the substratum in which they were growing. It was with the greatest difficulty that they were isolated from the particles of soil and sand which clung very closely — to the rhizoids. The tissue of the prothallus itself is extremely soft and fragile, and more than one specimen was destroyed in my attempts to remove the particles of — sand and soil by means of a soft camel’s-hair brush. It was only with the greatest Sealy See eee ee of care that such particles could be removed. Such removal was quite necessary in — order to observe accurately the surface cells of the prothallus and the reproductive ; organs. The least particle of grit left on the surface made it impossible for micro- — tome sections. : The colour of the prothalli is a feature to be noted. They are of a light brown, very similar to the colour of the hairs and soil clinging to the surface of the trunks of the tree-ferns, or in the soil in which they.are growing. This close harmony in the colour of the prothallial tissue and the substratum in which they grow has no~ doubt been one of the reasons why these structures have not before been discovered. As a matter of fact, the specimens that I have found were first recognised while examining the substratum with a binocular microscope. With the naked eye it is difficult to distinguish a prothallus from a fragment of a young rhizome. The first specimen found was really thought to be a portion of a rhizome, but later study and the discovery of additional specimens by the same methods has proved this first specimen to be an undoubted prothallus. : As stated above, the colour is characteristic and an. brown, and in none of the specimens found was there any trace whatever of the pigment chlorophyll. In this connection it is of importance to note that the prothalli studied consisted of living tissue. This was proved to my satisfaction by the appearance of the living | cytoplasm and nuclei in the cells. Had the prothalli been dead when found, the absence of chlorophyll might be accounted for on basis of decay and disintegration of the protoplasm. The cells, however, were living, and I am inclined to believe that chlorophyll is normally absent, and that these prothalli do not depend upon phot synthesis for their initial food-supply. This absence of chlorophyll is quite 1 harmony with their subterranean habitat. When we consider the saprophytic nature of Lycopodium, Botrychiwm, and othe subterranean Pteridophyte prothalli, it is not surprising to find this same habit in THE PROTHALLUS OF TMESIPTERIS TANNENSIS. 789 Tmesipteris. When such structures become strictly saprophytic they not only lose all trace of chlorophyll in their cells, but they nearly always become symbiotically associated with a mycorrhizal fungus. To this rule 7mesipteris is no exception. It is strictly saprophytic. It bears no chlorophyll in its tissues, but becomes infected with an endophytic fungus. In many saprophytic Pteridophyte prothalli the endo- phytic fungus becomes localised in certain definite regions of the prothallus. This seems not to be the case in T’mesipterts. The fungus here seems to infect any of the cells in the vegetative tissue of the prothallus. This is shown in figs. 4, 5, and 6. It is a very definite and conspicuous thread-like mycelium which enters into the interior of the cells, forming long coil-like structures in the protoplasm of the host cells. The endophytic fungus is certainly more conspicuous in the surface cells and those near the surface, but, as shown in sections and as illustrated in fio. 6, the infection may extend into the very interior of the prothallus. I cannot say whether the fungus ever infects the archegonial cells, because none of my specimens showed the early unfertilised stages of the archegonium. Although many older archegonia were found, none of their cells showed the endophytic fungus. This was also observed to be the case in regard to the antheridia. Here both young and old stages were found, but in none of the antheridial cells was the fungus observed. In this connection it should also be noted that in the case where the young stages of the embryo were observed, as shown in figs. 6 and 16, none of the embryo cells appeared to be infected. This absence of the fungus from the embryo cells—which of course may not prove to be general—made it possible for a sharp contrast to be _ drawn between the cells of the young sporophyte and the cells of the surrounding -gametophyte which contain the fungus (fig. 6). The young sporophyte, however, does later ‘become infected by what certainly appears to be the same fungus that infects the gametophyte. In the gametophyte tissue the endophytic fungus evidently provides all the necessary food for its host. And if one may judge from the appearance of sections, it would seem that the fungus may remain within the protoplasm of the host cells without causing immediate destruction. But the appearance of older infected cells makes it quite clear that the fungus eventually causes the death and disintegration of the cell contents it infects. As indicated in figs. 4 and 5, the nucleus seems to be quite normal, but as the fungus increases within the cell the nucleus appears to undergo a gradual change, and finally disintegrates along with the other protoplasmic structures of the host cells. As far as 1 have been able to judge, it would seem that the prothallus becomes infected at a very early period, and I am inclined to believe that the endophytic fungus is necessary for the germination of the spore. I shall, however, go into this matter more in detail in a future paper. There is evidently no differentiation of the prothallus into reproductive and vegetative regions. As we would expect to find in prothalli of subterranean habitat, the antheridia and archegonia are produced on the same individual. These reproduc- 790 PROFESSOR A. ANSTRUTHER LAWSON ON tive organs were confined to no particular region. The antheridia were compara- tively large structures, and were found variously distributed over the surface. The archegonia, on the other hand, were quite small and very numerous. This may be — observed in figs. 3 and 6. In this respect Tmesipterrs does not show a feature which is characteristic of Lycopodium, where the reproductive sexual organs are limited in their distribution. So, taking these main features of the prothallus of Tmeszpteris into account, there appears to be very little resemblance to other known Pteridophytes. Its saprophytic habit and subterranean habitat do recall Lycopodiwm, but structually there is no real resemblance. On the other hand—with the probable exception of the general — form of the archegonia and antheridia—there is not the -remotest suggestion of Equisetum, which, of course, is a free-living, chlorophyll-bearing prothallus growing upon the surface of the soil. If one may draw conclusions from these few specimens, which form the basis of this study, it would seem that we have in the Psvlotales a type of prothallus that is not closely related to any other existing Pteridophytes; and _ this lends support to the more recent ideas that their relationships to the Lycopodiales — on the one hand, and the Hqwisetales on the other, are very remote. The only other — alternative would be a close relationship to the Sphenophyllales, which has already been suggested by both Scorr and Bower. THE ANTHERIDIA. All of the specimens found—with one exception—bore antheridia. While all of ~ the stages in the development of these structures were not found, sufficient material — was obtained to make a considerable study of them. In the first place, the antheridia seemed not to be confined to any particular region of the prothallus. They were found at both ends, in the middle and on both sides. Compared with the archegonia, — they were very large. They grow out from the surface as large, almost spherical, — structures. I do not know of any Pteridophyte that has larger antheridia. The distributing of the antheridia over the surface of the prothallus is shown in figs. — 2 and 3. These figures also show the relative size of the two kinds of reproductive F organs. In none of the cases examined were there archegonia found on the prothallus — without being accompanied with antheridia. The antheridia and archegonia were nearly always found in close proximity to one another. The antheridia are almost spherical in form, and extend out quite conspicuously from ‘the surface of the prothallus as rounded protuberances. They are quite evidently superficial structures, and in this respect stand in contrast to Lycopodium and Eguisetwm and also to the description given by Lane (1904) of the supposed prothallus of Psilotum, types in which the mature antheridium comes to lie below the surface. As shown in figs. 8, 7, 8, 9, 10, and 11, the antheridium here in Tmespiteris has a distinct character of its own, both in size and surface position. THE PROTHALLUS OF TMESIPTERIS TANNENSIS. 791 The very early stages in its formation were not found, but the series shown in figs. 8, 9, 10, and 11 gives a fair idea of the development. Fig. 8 represents a section through the antheridium, and here we observe the nature of the wall, which even in this early stage shows seven cells in section surrounding the spermatogenous tissue. A little later stage is shown in fig. 9: here eleven cells are to be seen in a single section of the wall, and this completely surrounds a large mass of sperm mother cells. In fig. 7 we have represented two antheridia in section lying close together. In fig. 10 the antheridium is shown not in section but in surface view, and it contains a very large number of spermatocytes. This figure also shows the relative size of the mature antheridia in relation to the mature archegonia, which are situated close beside it. At the very apex of the spherical antheridium the main wall cell is larger than the others, and, judging from figs. 10 and 11, it is this cell which becomes free, forming an exit for the spermatozoids. The mature spermatozoids were not found, but if one may judge from the appearance of the nuclei in the sperm mother cells shown in figs. 9 and 10, they must be of considerable size and quite numerous. . Old antheridia from which the spermatozoids had been discharged were frequently met with, and some of these are shown in figs. 2 and 3. A more highly magnified example is shown in fig. 11. From the large number of spermatozoids that are evidently formed and the close proximity of the antheridia to the numerous arche- gonia, fertilisation is easily accomplished. From the evidence to hand one cannot say definitely whether these spermatozoids are multiciliate or biciliate. One would expect to find the former condition. As I now know how to obtain the prothalli, I can safely promise a more thorough investigation of spermatogenesis and fertilisation in the near future. As far as this brief description has gone, however, it is sufficient to convince one that there is not a very strong resemblance of the antheridium of Tmesipteris to either Lycopodium or Equisetum. | THE ARCHEGONIA. The archegonia of Tmesipteris are quite small in size and very simple in structure. As already stated above, there is clearly no differentiation of a vegetative and repro- ductive region of the prothallus. Almost any part of the prothallus apparently bears archegonia. They were found at both ends and on all sides, as shown in figs. 3 and 6. Fig. 3 is drawn from an entire prothallus, and one surface view shows fourteen archegonia associated with three antheridia. On the opposite side of this same prothallus, which was studied by turning the slide over, no less than sixteen archegonia were observed associated with antheridia. Fig. 6, which represents a fairly median section through another prothallus, shows clearly that the archegonia may be distributed on at least two opposite sides. In this section we may see eight archegonia in the upper side of the figure and four on the lower side. In fig. 2 we have represented an entire prothallus with only two archegonia showing, so that 792 PROFESSOR A. ANSTRUTHER LAWSON ON from these figures it seems that the distribution of the archegonia has little if any — regularity. When they are produced in the older prothalli they are quite numerous and set closely together. ‘ Fig. 12 is a representation of a more highly magnified view showing four archegonia as they appear in a surface aspect. This figure is instructive, as it shows — the relative size of the archegonia to the ordinary vegetative cells of the prothallus — and to the rhizoids. It will be observed from this aspect that the opening of the canal is surrounded by four neck cells, and that these neck cells are very much smaller than the ordinary surrounding vegetative cells. At the point where the four — neck cells meet there is a distinct opening or canal, and this region in the mature prothalli always took on a deep brown colour. These brown regions were quite helpful, for they indicated the presence of the archegonia from a mere superficial — study with the hand lens. Before these structures were actually found the bases of old detached rhizoids — were at first mistaken for archegonia, but a close study of them with the high- — power magnification soon revealed their true nature. The archegonium, as stated above, is a very simple structure. Although unfortunately the very early stages in its development were not found, scores of older ones were, and a study of these makes — it clear that in their form and simplicity the archegonia are unique. No evidence of | canal cells or of ventral canal cells presented itself; and while no doubt a more careful future study will reveal these structures, they cannot form a very conspicuous — feature. The youngest archegonium found is represented in fig. 138, which was drawn from a median longitudinal section. Here two of the neck cells are shown, and also the egg cell which occupies the central position of the venter. The latter structure, as in all Pteridophytes, is buried in prothallial tissue. The nucleus of the egg cell - is clearly shown, occupying a central position in a dense mass of protoplasm. My interpretation of this figure is that it represents a mature archegonium ready for fertilisation, and if canal cells and. ventral canal cell were formed they have disin- tegrated and disappeared. Figs. 14 and 15 show two other mature archegonia which have not been fertilised. A great many sections of archegonia of this nature were found, and they all show a distinct peculiarity in the form of the neck cells. In every specimen examined it was found that these neck cells extend very little indeed beyond the surface of the prothallus, and in section they appear as terminating in curiously pointed margins, and these all four together form a flat disc concave in the middle but otherwise parallel with the surface of the prothallus. The concavity of the disc formed by the four neck cells varied somewhat with the maturity of the archegonium, as shown in the figs. 3 and 12. The shaded portion of the archegonia ph in fig. 12 is intended to represent the concavity of the neck cells just above the ’ sie canal. This curious form of the archegonium would suggest an adaptation for the THE PROTHALLUS OF TMESIPTERIS TANNENSIS. 793 not curved back and overlapping the surface of the prothallus, but growing straight forward in the form of a straight tube. It will be seen from this brief description that the structure of the archegonium in Tmesipteris bears no very striking resemblance to either Hqwisetwm or Lycopodium or to any other Pteridophyte. In its general form and simplicity one is inclined to regard it as reduced. In certain respects it recalls Selaginella, but the resemblance here is a remote one. It should be remembered that we are dealing here with a plant whose sporophyte and gametophyte are both reduced and highly specialised and adapted to definite habitats. It is, therefore, not surprising to find a reduction also in at least one of the sexual organs. THE EMBRYO. The discovery of the gametophyte structure of Tmesipteris will no doubt prove of very great interest, not only because it fills in an important gap in our know- ledge of the Pteridophytes as a whole, but also because it helps to establish the phylogenetic position of the Psilotales. But I fully realise that in respect to the latter the structure of the embryo sporophyte will prove of more importance than the prothallus itself. It should be remembered that in both Psilotwm and Tmesipteris no roots have ever been found as a structure of the sporophyte. It will be of very great interest to know whether the very young sporophyte ever possessed a root, and if so at what time such a root disappears. It will also be of interest to know if a suspensor is present in the very young embryo. Added to this, the facts revealed from a study of the vascular anatomy of the embryo will prove of great value, especially in view of their supposed relationship to the fossil sphenophylls. Realising the importance of these matters, I believe it will be best to leave the matter of the embryo for the present until more material is obtained. I found only one specimen of the very young embryo, but this unfortunately was not sufficiently advanced to show any vascular tissue, and was too old to show the initial divisions of the egg cell. The embryo found is shown in figs. 6-and 16, and is represented in median section. The region of the prothallus occupied by the embryo bulges forward asa distinct protuberance. Carried forward on the prothallial tissue one may see two archegonia, one of which has been fertilised and had developed the embryo. The general position of the embryo in relation to the prothallus is seen in fig. 6. The details of the embryo at this stage and as seen in median section are shown in fig. 16. There is clearly here an upper and lower portion of the embryo. The upper part forms the main axis of the embryo. But it was impossible to distinguish the cotyledon region from the initial stem. The lower half of the embryo shows three lobes in section. One of these, marked R, I interpret to be the root. The two other lobes constitute the foot. But the curious division of this latter region is significant, and the smaller portion of it may prove to be a suspensor. But this is merely suggestive, and must be confirmed by the study of additional material. TRANS. ROY, SOC, EDIN., VOL. LI, PART III (NO. 25). 115 794 As I now know how and when to obtain material, I have hopes of procuring a fresh supply in the near future, and these details of the embryo will be carefully studied. In conclusion, | beg to acknowledge my indebtedness to Dr 8. J. Jonnston, of the Zoology Department, University of Sydney, who showed me where Tmesipteris was growing at Bulli Pass, N.S.W. Iam also greatly indebted to my friend Mr DaRneLt- Smirn, of the Bureau of Biology, Department of Agriculture, N.S.W., who was kind — enough to show me where Z7'mesipteris grows abundantly at Bonner Falls. At both of these localities | found prothalli, two at Bulli Pass and one at Sommerby Falls. Sydney, I am also indebted for some young sporophyte plants which he obtained at Mount Wilson. 1894. 1908. 1904. 1904, 1904. 1904. 1908. Fig. 1. A portion of a prothallus of Z’mestpteris as seen from the surface. Fig. 2. Another specimen seen from the surface and showing its branching. Fig. 3. Another specimen of prothallus seen from above and showing the ohaveckone rhizoids, — antheridia, and archegonia. Fig. 4. A section showing a few prothallial cells containing the endophytic fungus. Fig. 5. Another of the same showing the fungus more in detail. Fig. 6. A median section through the prothallus showing the distribution ef the archegonia on the surface, the position of the embryo, and the distribution of the infection by the endophytic fungus. Fig. 7. Sections of two young antheridia, Fig. 8. A young antheridium seen in section more highly magnified, Fig. 9. A later stage in development of the antheridium as seen in section. Fig. 10. An antheridium seen from the surface. Fig. 11. An old antheridium from which the contents have been discharged. Fig. 12. A surface view showing four archegonia. Fig. 13, A mature archegonium ready for fertilisation aud seen in section. Figs. 14 and 15, Older stages of the same. Fig. 16. A section through a portion of a Rabin iilse containing an embryo, THE PROTHALLUS OF TMESIPTERIS TANNENSIS. To Mr Jonn M‘Lucxin, B.Sc., of the Department of Botany, University of LITERATURE CITED. Bowsr, F. O., ‘Studies in the Morphology of Spore-producing Members,” Phil, Trans. Roy. Soc., 1894. Bower, F. O., The Origin of a Land Flora, Macmillan & Co., London, 1908. Boopiz, L. A., ‘On the Occurrence of Secondary Tracheides in Psilotum,” Ann. Bot., vol. 1xxi, p. 505, 1904. Booptz, L. A., New Phytologist, 1904. Forp, S. O., “The Anatomy of Psilotum triquetrum,” Ann. Bot., \xxii, p. 589, 1904. Lane, W. H., “On a Prothallus provisionally referred to Psilotum,” Ann. Bot., xxii, p. 571, 1904. Scort, D. H., Studies in Fossil Botany, Black, London, 1908. DESCRIPTION OF FIGURES. as G JUN I9i/ Wo EEE C.Hodges & Son, imp. Lon. Vol. LI. TMESIPTERIS TANNENSIS.—PLarTE II. C Hodges & Sonamp. lon Vol. LI. PROTHALLUS OF TMESIPTERIS TANNENSIS.— PLATE LIL. STRUTHER LAWSON: C. Hodges & Son, imp, Lon. Vi. VII. VIII. IX. X. XI. SUT. XIII. XIV. ae XX. XX. XXVIL. XXVIIL XXIX. XXX. a Pte ” XXXI. XXXIL January 1917.—Volumes or parts of volumes not mentioned in the above list are not for the p to the public. the Society reprints from time to time parts of its publications which have become scarce, the absolute cor! mWONE DE whore KN wh SWOOP RODOHOH OH HEHE OOH COHHOOODOROOCOH Bewne | Fellows or others who may specially desire to obtain them must apply direct to the S Price to the Public, 16 se aSoOMorkNT — SSOOSC SOOO Cco OSS: SHSOBSOOHOKR OHH COOH COHHOOOCOCCOO Fellows. £ s. 9 15 14 17 16 12 12 15 I 6 14 — ial SISO WOH IDWS following reduced Prices :-— Vol. |. XXXIV: ”» ” 3) ” by) 9 ” | Pe | XXXIX. Pt. Pt. of this list cannot be guaranteed beyond this date. The papers published in this Part of the Transactions may be had on application to the Publish . XL aN. . XV., yr: SEV IL VDL. : eae Price to Public. | XXXITI. Pt. Bits Pt. | XXXV.* Pt. Pt. Pt. ie XXXVI. Pt. Pt. . ye XXXVII. Pt. Pt. Pt. | ‘Be Pt. IXXXVIIL. Pt. Pt. Pt. WNW PWN RwWNHE RW NER WDE Whe Re Goh NEE RH HE HON OCH OH HEP WOOKP HEH HEP HS DE NWNONHE Price to the Public. x > — — ROOK rPNwWrNwWNWhlWr* — — — — HOH OCAwntnowoorntdseNnontae — — 4s. Price yp bO | Fellows. Vol. =HSBOSDOHM OOH OHOH OH OOOHOHOOHHEHH OHO | at the following prices :— 3s. ls. Is. Price to Fellows. | oe ae aK. OKT, OX Kir SIV. ROWS Ss FOSPOARODSOROBNOSCONOOROSOACSCORBSCSOSOO Ben Nevis Volumes. \ XLVIL. Xs XLUL XLIV.Pt.1 rae Price to Public. Tf 3s. 4s. 6d. 1s. 9d. 6s. 4s, 6d. ls. 6d, Ll eel el eel eel el eel oe ll oe OO an nol ear) Price to the “ete Public. | Fello oe — — — NMHnIw IAN OW NAMB OWA SCHHOATHBRWISHEOAWW — — = . — SSSCSSSESSOSADOSONISCSCHWDNORDOSOCOARCO! — SHODOR HH OHH Hh <— ie _— ed eh pg rom Owoowmornreucj§9 —" os a TRANSACTIONS -RovAL SOCIBTY OF BDINBURGH. VOLUME 1) a og. 8 374 IV, —SESSIONS 1915- 16-17. CONTENTS. ‘ ’ , Page ician and Silurian Brachiopoda of the Gurvan District. By F. R. C. Ruexp, Aue e.D., EGS. Communicated by Dr Horns, F.R.S. (With Twenty-Four Plates), 795 ~ (Issued February 2, 1917.) ; bir Wyre and the Titterstone Clee Hill Coal. Fields: (With Five Plates and Six (NE beet eh rn a rr a , 3 3 i a 99 SC., F G.S.—The Fossil Plants of the Titterstone Clee Hill Coal Field. By : R. Kinston. Wi th an Appendix on he Fossil Plants collected ay om the Cont of the Clavérley Trial Boring. By Dr R. Kinston. : - Sagat March 9, 1917.) EDINBURGH: +" _ “oa MDCCCCXVII. Price Thirty-one Shillings and Sixpence. Se See ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 813 of the types and other material. There is no perforation of the beak, as he mentions, , but there is a distinct pedicle-groove which is represented as a ridge in his figure (op. cit., pl. xvi, fig. 32) of the interior of a pedicle-valve. This groove is well exhibited in the specimen which he figured as Lingula Ramsayi. The weak median depression in the brachial valve which widens but dies out anteriorly is best seen in the Minuntion specimens. This character is present in Schizambon (Siphonotreta *) mannesotensis, Hall and Clarke,* from the Trenton Formation, and the characters of the pedicle-valve also closely agree with those of S. fissus, var. canadensis, Ami.t S. Dodgei, Winchell and Schuchert,{ and S. duplicimuratus, Hudson,§ are likewise allied species. But Sch. fissus, var. canadensis, was regarded by Hatu and CLARKE as probably not congeneric with the type species of the genus, Sch. typicalis, Walcott,|| on account of the structure of the pedicle-passage, and it may be necessary to put this whole group of species in a separate genus or subgenus. Schizambon scoticus (Davidson) var. (Plate III, figs. 15-17.) There is an imperfectly known shell closely resembling Sch. scotecus which occurs in the Starfish Bed, and probably it will have to be regarded as a distinct species when we have better-preserved specimens for study. But at present it may best be placed as a variety of Sch. scoticus. It differs from the Stinchar Limestone form in having (1) a shorter foraminal groove ; (2) stronger and more prominent {mbricating ridges on the surface ; and (3) fewer and coarser spines. Dimensions.—Length, 7-15 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Genus ACROTHELE, Linnarsson. Acrothele medioradiata, sp. nov. (Plate III, figs. 18-22.) Shell corneous, circular or subcircular. Pedicle-valve low conical, with long anterior concave slope and shorter, steeper posterior slope without a definite false area ; apex sharp, prominent, pointed, situated at about one-third the length of the valve, with minute circular foramen at apex and internal narrow groove ending at about three-fourths the distance between apex and posterior margin. Brachial valve * Hawt and CrarKkE, Palzont. New York, vol. viii, 1892, Brach., i, p. 177, pl. iv, figs. 37, 38. + Haut and Crarxg, zbid., p. 115, pl. iv, figs. 32-36. t WincHett and Scuucuert, Palzont. Minnes., iii, p. 361, pl. xxx, figs. 5, 6. § Crarkn, Rep. State Palzont. 1903, New York State Mus., 1905, Bull. 80, p. 284, pl. 5, figs. 6, 7. || Waxcorr, Mon. U.S. Geol. Surv., vol. viii, 1884, “ Paleeont. Eureka Distr.,” p. 69, pl. i, fig. 3; ¢d., Camb. Brach. (Mon. U.S. Geol. Surv., vol. li 1912), p. 624, pl. Ixxxiv, figs. 1, la-1d. 814 DR F. R. C. REED ON THE gently convex, beak marginal or submarginal, not much raised. Surface of both valves marked with fine concentric striz and several (6-8) strong low radiating wrinkles running from the apex to the middle of the anterior margin. Interior of shell finely punctate. : Dimensions.—Diameter, 3-7 mm. ; height, 1-2°5 mm. Horizon.—Balclatchie Group. Localities.— Balelatchie, Dow Hill. 4 Remarks.—It is probable that this shell is referable to the genus Acrothele, and in its external ornamentation and shape it resembles A. subsidua (White) * and its varieties. A. Matthewi (Hartt) + and its varieties also seem to be allied. Acrothele (?) sp. a. (Plate III, figs. 23, 242? 25 ?) 1883. Beniitcia Nicholsont, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., pp. 213, 224 . Shell corneous, subcircular to elliptical, wider than long. Brachial valve flattened, slightly raised towards beak, with posterior edge nearly straight and a shallow bro median depression on surface towards anterior margin; beak small, submargi slightly raised, with traces of a pair of minute cardinal muscle-scars ; interior w short median septum. Surface of shell marked with concentric lines very minutely granulated. _Dimensions.—Diameter about 5-6 mm. Horizon.—Balelatchie Group. Localities.—(1) Balelatchie, Ardmillan, Dow Hill. Remarks.—Some of the minute shells, labelled by Davipson Acrotreta Nicholsom may be referable to Acrothele, but others (figs. 24, 25) may belong to Acrotreta. pedicle-valve which can be associated with them has so far been discovered. Acrothele (Redlichella *%) sp. b. (Plate III, figs. 26, 27.) 1883. Siphonotreta micula, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 224. Non 1851. Siphonotreta micula, M‘Coy, Ann. Mag. Nat. Hist., ser. 2, vol. viii ee Dp. 389 ; Brit. Pal. Foss, Woodw. Mus., p. 188, pl. ix, fig. 3. There is a small horny brachiopod occurring at Balclatchie which much resemb the small shells from the Llandeilo of Co. Down, referred by Davipson { to Acrot granulata, Linnarsson,§ of the Middle Cambrian of Sweden, but it is probable tha the Llandeilo and Cambrian forms are nevertheless distinct, though Watcorr || puts * Watcort, Camb. brach. (Mon. U.S. Geol. Surv., vol. li, 1912), p. 656, pl. 1x, figs. 1, 8, 9. + Ibid., p. 650, pl. 1xi, figs. 1, 2, 3, 5, t Davipson, op. ctt., v, p. 214, a xvi, figs. 29, 30. § Linwansson, Bih. t. k. svensk. vet, Akad. Handl., Bd. iii, No. 12, 1876, p. 24, pl. iv, figs. blow : || Watcorr, Camb. Brach. (Mon. U.S. Geol. Surv., “ih li, 1912), p. 663, pl. lvi, figs. 2, 2a—n. ‘ . ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 815 them together in his list of references to the species. Some of these Girvan specimens were labelled by Davipson Siphonotreta micula, M‘Coy, but they do not possess the shape or characters of that species, being subcircular in outline, and having irregular sublamellose finely fimbriated concentric strize ; the beak also is intra-marginal, and seems to form the apex of a low cone. In size they average 3 mm. in diameter or less. A. granulata; Linnarss., is taken by Watcotr as the type of his subgenus Redlichella (op. cit., p. 662). - Horizon.—Balclatchie Group. Locality.—Balclatchie. Genus AcrotrETA, Kutorga. “ Acrotreta costata,’ Davidson. 1883. Acrotreta ? costata, Davidson, Mon. Brit, Foss. Brach., vol. v, Silur. Suppl., p. 213, pl. xvi, fig. 24. The one small specimen on which this species is based is very imperfect, and Davinson’s figure is a highly imaginative restoration, unless the specimen has sub- sequently been much injured, which does not seem to have been the case. The shell may belong either to the genus Conotreta or Acrotreta, but the “radial ribs” which are now visible seem to be only the imperfect cast of the posterior face of the pedicle- valve showing the pedicle-groove and edges of the false-area. But the specimen is too poorly preserved and too fragmentary for a satisfactory determination. WatcoTT* says: “The description and figures suggest a genus allied to Cyrtia, Hall and Clarke, of the Silurian” ; but, as stated above, Davinson’s figures are untrustworthy. Horizon.—Balclatchie Group. Locality.— Balclatchie. Acrotreta Nicholsonr, Davidson ? 21868. Acrotreta % Nicholsont, Davidson, Geol. May., vol. v, p. 313, pl. xvi, figs. 14-16. ‘ 1871. Acrotreta 1 Nicholsont, Davidson (pars), Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 343, pl. xlix, figs. 36, 37, 40 (non figs. 38, 39). Non 1883, Acrotreta Nicholsoni, Davidson, op. cit., vol. v, Silur. Suppl., p. 213, pl. xvi, figs. 21-23. 1912. Acrotreta Nicholsoni, Walcott (pars), Camb. Brach. (Mon. U.S. Geol. Surv., vol. li, 1912), p. 696, pl. Ixxiii, figs. 1, le, 1d, 1h, 12, 1k (non figs. la, 18, le, Lf, 19). In describing the new species Micronutra Davidson: it has been mentioned above that the type of A. Nicholsoni came from Dobbs’ Linn, Moffat. Larwortu f only records the latter from the zone of Dicrano. Clingani in that region. The Girvan specimens referred by Davipson to this Moffat species cannot be retained in it or in the same genus, though Watcort has adopted Davipson’s conclusions and figures. An examination of the figured specimens shows that the drawings are incorrect and mis- leading. The median groove from the apical foramen of the pedicle-valve to the * Watcort, Camb. Brach. (Mon. U.S. Geol. Surv., vol. li, 1912), p. 674. + LapwortH, Quart. Journ. Geol. Soc., vol. xxxiv, 1878, pp. 313, 331. 816 DR F. R. C. REED ON THE - posterior margin is wanting, and this seems to be an essential character of Acrotreta Nicholsoni. In the brachial valve of Acrotreta, Watcort (op. cit.) mentions a long, _ well-defined median ridge and a pair of strong cardinal tubercles near the margin of the area, but these cannot be clearly detected in any of the Girvan specimens. Some of Davrpson’s figured specimens of A. Nicholsont from Girvan have been removed to Micromitra Davidsoni (p. 797), and others must be placed in a new species of Conotreta (C. conoidea, q.v.); while others labelled A. NMicholsoni by Davipson seem to belong to the genus Acrothele rather than Acrotreta. Consequently itis highly doubtful if the Moffat form A. Nicholsoni really occurs in the Girvan area at all. Acrotreta (%) sp. a. (Plate IIT, fig. 28.) There are some small caleareo-corneous brachial valves occurring in the Stinchar Limestone and Balclatchie Groups which are probably referable to some species of this genus, and somewhat resemble Acrotreta curvata, Walcott,* from the passage beds between the Cambrian and Ordovician of Eureka, Nevada. The valve is sub- circular in shape; the beak is small, acute, and marginal, and from it a shallow depression may run forward across the flattened surface of the valve to the front edge. The ornamentation consists of fine concentric lines. The cardinal scars can be rarely detected. | q Dimensions.—Diameter about 2°0 mm. Horizons.—(1) Stinchar Limestone Group ; (2) Balelatchie Group. Localities.—(1) Craighead ; (2) Balclatchie. Acrotreta (2) sp. b. (Plate ITI, figs. 29-31.) Shell transversely subcircular. Brachial valve gently convex, with small pointed marginal beak ; interior with long straight septum extending about three-fourths the length of valve; pair of small triangular cardinal scars contiguous to septum in mediat line close to beak, and a pair of similarly placed small subcircular “‘ central” (2) scars at about one-third length of valve. External surface of shell with concentric erowull lines, and a few fine radial strize in interior. Dimensions.—Length, c. 4mm. ; width, ¢. 5 mm. Horizon.—Stinchar Limestone Group. Localities.—Bennane Burn, R. Stinchar ; Craighead. | Remarks.—One specimen of the interior of a brachial valve [2809, Geol. Sur Mus., Edinburgh] shows the muscle-scars and internal septum which suggest it ‘its reference to the genus Acrotreta. The others merely show the internal septum. * WaLcort, op. cit., p. 682, pl. Ixviii, figs. 1f—12. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT 817 Genus Conotreta, Walcott. Conotreta conordea, sp. nov. (Plate IV, figs. 1-5.) 1883. Acrotreta Nicholsoni, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur, Suppl., p. 213, pl. xvi, figs. 21, 23 (non fig. 22). 1912. Acrotreta Nicholsoni, Walcott (pars), Camb. Brach. (Mon. U.S. Geol. Surv., vol. li, 1912), p. 696, pl. Ixxiii, figs. le, 1f, 1g (non cet.), Shell very inequivalve. Pedicle-valve high, acutely conical, rounded; height usually greater than width or length; width slightly greater than length ; apical angle, 20°-30° ; apex sharply pointed, excentric, nearer posterior edge, perforated by minute circular foramen ; posterior face below apex vertical, somewhat flattened, not defined, with two very weak obscure furrows running down its centre from apex to margin and faintly enclosing very narrow slightly depressed median triangular area (often indistinct). Interior of pedicle-valve with several longitudinal ridges running about one-half of the distance down anterior face, of which two or three bifurcate. Surface of valve with rather strong concentric striz. [? Brachial valve nearly flat, transversely subcircular, with small pointed inconspicuous beak and thin median internal ridge ; surface ornamented with rather strong concentric striz. | Dimensions.— Pedicle-valve. Brachial valve. (1) (2) Height . Paya OD Oe mime Length . . 43 mm. Width . Veo too 4 mm Width . ~~ 456 mime Horizons.—({1) Stinchar Limestone Group; (2) Balclatchie Group. Localities—(1) Craighead ; (2) Ardmillan, Balclatchie, Dow Hill. Remarks.—The type of this shell is from Craighead, where the best-preserved specimens occur. Davipson’s unfortunate figures of Acrotreta Nicholson have been the cause of much confusion. Acrotreta Nicholsom has a much lower pedicle-valve, and the posterior face and interior have different characters. Our Craighead shells and some from Balelatchie and Ardmillan agree much more closely with Conotreta Rusti, Walcott,* of the Trenton Formation, not only in their external shape and features but in the internal impressions in the pedicle-valve. Davinson’s figure of the Ardmillan specimen is misleading, for there are no such grooves on the posterior face as he has indicated The brachial valve of the genus Conotreta is not described by Hatt and Ciarkg, and the small flat valves above described as probably belonging to C. conoidea may possibly belong to Acrothele. * Waxcort, Proc. U.S. Nat. Mus., xii, 1890, p. 365, figs. 1-4; Haun and Cuarkg, op. cit., Brach., i, p. 104, pl. ivk, figs. 16-21. 818 DR F. R. C. REED ON THE Genus Trematis, Sharpe. Trematis craigensis, sp. nov. (Plate IV, figs. 6-9.) Shell obovate to subcircular, widening slightly anteriorly. Brachial valve gently convex ; beak blunt, intramarginal, not projecting ; interior with median muscle-sear formed by pair of elongated subtriangular “central” muscles situated behind centre of valve, separated by very low weak median septum ; marginal scars indistinct. In- ternal surface of shell marked by very fine radial striee; external surface covered with small closely placed hexagonal cells separated by narrow walls; cells becoming smaller near margins. Pedicle-valve unknown. Dimensions.—Lenegth, 13 mm. Horizon.—Stinchar Limestone Group. Locality. Craighead. Remarks.—This species is undoubtedly allied to 7. punctata (Sow.)* and T. millepunctata, Hall, but the shape of the shell is rather different and the external ornament is finer, the cells being smaller than in the former species, and the beak of the brachial valve in the latter is more prominent and the central muscle-scars more anteriorly placed, being almost in the middle of the shell. Trematis melliflua, sp. nov. (Plate IV, fig. 10.) Shell transversely subcircular ; brachial valve gently convex [? with weak median depression from beak to margin]; beak marginal, blunt, slightly incurved ; surfe of shell marked with a few concentric growth-ridges and covered with very min closely placed hexagonal pits, arranged in quincunx manner; no radiating | present. Dimensions.—Length, 5°1 mm.; width, 6°3 mm. Horizon.—Balclatchie Group. Locality.—Balclatchie. Remarks.—The absence of radiating lines and the densely placed very small pits give a minutely cancellated or honeycomb appearance to the shell-surface, as it Trematis millepunctata, Hall,t from the Hudson River [Lorraine] Group, with which species ours may be compared. I am only acquainted with one specimen of interesting little brachiopod, and it is in Mrs Gray’s collection. The me depression may be due to crushing and thus not be an original or specific “ The shape of the shell and the much finer ornament distinguish this species f that of the Stinchar Limestone termed 7. craigensis. * Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 69, pl. vi, fig. 9. + Haut, 24th Ann. Rep. N.Y. State Cab. Nat. Hist., 1872, p. 221, pl. 7, figs. 22-25. { Haut and Cxiarke, Palxont. New York, vol. viii, Brach., i, p. 139, pl. ive, figs. 4-10 ; Scavensnr, Bull Ul, 8 U.S. Geol. Surv., p. 452 (for references), ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 819 Trematis silurvana, Davidson. (Plate IV, fig. 11.) 1866. Discina (Trematis) stlurtana, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 71, pl. vi, figs. 8, 8a, 80. : This species, of which the type comes from the Bala beds of Horderley, was not recorded by Davipson from the Girvan district. Only three imperfect specimens are known to me, and these occur in Mrs Gray’s collection. They show the characteristic radiating lines of transversely subquadrate to subcircular pits lying between the radial ridges, and one specimen consists of the pedicle-valve which has not been previously described in this species. This valve forms a low depressed cone with a narrow triangular fissure running inwards from the margin to the subcentral apex, and it resembles in these respects the pedicle-valve of 77. millepunctata, Hall, as figured by Haut and CiarKeE.* Dimensions.— Diameter about 20 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Genus ScH1zocraNIiA, Hall and Whitfield. Schizocrama asterordea, sp. nov. (Plate IV, figs. 12-14.) 1851. Orbicula? crassa, Hall, Salter, Proc. Geol. Soc. London, vol. vii, p. 175, pl. x, figs. 3, 4; p. 151, woodcut 7. 11866. Discina crassa (Hall?), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 69, pl. vi, fig. 7. 41883. Discina crassa, Davidson, op. cit., vol. v, Silur. Suppl., p. 224. Non 1847. Orbicula’ {Obolelia} crassa, Hall, Paleont. New York, i, p. 290, pl. 79, fig. 8. Shell oval; brachial valve gently convex, most so just behind centre; apex sub- marginal, low, close to posterior margin, blunt, slightly incurved. Surface covered with fine rounded equal and equidistant thread-like lines, increasing in number by interpolation, and curving back slightly on each side. Pedicle-valve unknown. Interior of brachial valve with a pair of elongate closely placed posterior muscle- scars converging and narrowing to apex of shell, and with a weak median ridge running forwards from apex. Dimensions.—Length, 10°3 mm.; width, 8°0 mm. ; height, 2°2 mm. Horizon.—(1) Drummuck Group (Starfish Bed); (2) Whitehouse Group ? Locality.—(1) Thraive Glen ; (2) Whitehouse Bay. Remarks.—This shell resembles Schizocrania filosa (Hall) + in its shape and ornamentation, and in the muscle-scars of the brachial valve, so far as it is known. The specimen figured by Davipson from Penwhapple Glen as Discina crassa, Hall (*) (and of which the occurrence at Thraive was subsequently recorded), is probably the * Haut and Cuarxe, Palzont. New York, vol. viii, Brach., i, p. 138, pl. ive, fig. 8. + Haut and Cxarkg, tbid., p. 142, pl. ive, figs. 22-29 (230). TRANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 26). 119 820 DR FE. RB. -C. REED ON THE same species. But Haut’s Orbicula ? [Discina] crassa is referred by ScrucHeErt * to the genus Obolella, and is a Lower Cambrian species; and Watcorr likewise places it in this genus. | Genus OrpicuLorpEa, D’Orbigny. Orbiculoidea Armstrongi, sp. nov. (Plate IV, figs. 15-17.) Shell circular or slightly subcireular. Pedicle-valve gently convex, low conical; apex subcentral; interior with pedicle-tube extending rather more than half the distance from centre to posterior margin. Brachial valve low conical, with apex ex: central, situated at about one-third the diameter or less from posterior margin. Surface marked with strong regular rounded concentric ridges of rather unequal size. Dimensions.—Average diameter 8 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Remarks.—This species is undoubtedly very close to Discina perrugata, M‘ Coy, the type of which came from the Llandovery beds of Kilbride, Ireland. But I am not convinced that it is identical, especially as it comes from a different horizon. M‘Coy says that the apex is situated at one-third the diameter from the margin; this would apply only to the brachial valve. Davipson does not record Orbic. [D.j perrugata from the Drummuck beds, but he states§ that it occurs at Balclatchie. I have not, however, seen any specimens attributable to it from this locality. It m be named Armstrongi, after one of the authors of the Catalogue of Western Scottish Fossils, 1876. ; Orbiculoidea shallochensis, sp. nov. (Plate IV, figs. 18-20.) 1883. Discina perrugata, Davidson (pars), Mon, Brit. Foss. Brach., vol. v, Silur. Suppl., pp. 210, 224 — (non pl. xvii, fig. 47). Shell circular to subcircular. Pedicle-valve low conical, with sharp central ape x interior with pedicle-tube extending nearly three-fourths the distance from centre t posterior margin; edge of valve somewhat flattened or thickened. Brachial valve very low conical, more or less flattened ; apex very excentral, situated at about one fourth the diameter from posterior margin. Surface with numerous fine concentri¢ close rounded lines. — Dimensions.—Length, 9°0 mm. ; width, 8°5 mm. Horizon.—Whitehouse Group. Locality.—W hitehouse Bay, Shalloch Mill. Remarks.—The finer surface-ornamentation and the longer pedicle-tube and mor re excentric apex of the brachial valve distinguish this species from O. Armstrongi. — * Scaucumrt, Bull, 87, U.S. Geol. Surv. (1897), p. 373. + Watcorr, Camb. Brach., 1912, p. 592. — t M‘Coy, Syn. Stlur. Foss. Irel., 1846, p. 24, pl. iii, fig. 2. § Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 65, pl. v, fig. 23, ORDOVICIAN AND SILURIAN’ BRACHIOPODA OF THE GIRVAN DISTRICT. 821 Orbiculoridea stincharensis, sp. nov. (Plate IV, figs. 21-24.) 1883. Discina perrugata (M‘Coy), Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., pp. 210, 224, pl. xvii, fig. 47. Shell slightly oval or elliptical, longer than wide. Pedicle-valve low conical, more or less flattened ; apex subcentral, rather nearer posterior than anterior end ; interior with pedicle-tube extending about half the distance from centre to margin. Brachial valve with apex more excentric, situated about two-fifths to one-third of the diameter from posterior margin. Surface covered with coarse rounded low concentric ridges. Dimensions.—Length, 7°5 mm. ; width, 6°5 mm. Horizon.—Stinchar Limestone Group. Locality.— Craighead. Remarks.—The oval shape of the shell, the shorter pedicle-tube, and the coarser and fewer concentric ridges distinguish this form from the other Girvan species. lt appears to resemble O. lamellosa, Hall,* from the Trenton Formation, but has more rounded concentric ridges. Subgenus scuizoTreTA, Kutorga, Orbiculoidea (Schizotreta ?) Forbesi, Davidson. (Plate IV, fig. 25.) 1848. Orbicula Forbesi, Davidson, Bull. Soc. Géol. France, sér. 2, vol. v, p. 334, pl. iii, fig. 45. 1866. Orbivuloudea Forbesi, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 73, pl. vii, figs. 14-18. 1883. Orbiculoidea Forbesi, Davidson, Mon, Brit. Foss. Brach., vol. v, Silur. Suppl., p. 224. The specimens attributable to O. Forbes: from Woodland Point are broken or poorly preserved, and the internal and apical characters are. not clearly recognisable. One pedicle-valve is circular or subcircular in shape, and shows minute transverse striz in the rather broad grooves between the narrow concentric ridges on the surface. The brachial valves have a more excentric apex, and are oval or rather narrower posteriorly, as Davinson mentions. The Niagaran species, Orbic. (Schizotreta) tenuilamellata (Hall) + is probably closely allied. Horizons.—(1) Saugh Hill Group ; (2) Mulloch Hill Group ? Localities—(1) Woodland Point; (2) Mulloch Hill. “ Orbicula oblongata.” The specimens attributed by Davipson{ to “Orbicula” [Discina] oblongata, Portlock,§ and described as Discina oblongata from the Balclatchie Beds, do not appear to be brachiopods at all. Only the so-called pedicle-valve has been found, and Davipson himself remarked (op. cit., p. 66) that no “lower” [brachial] valve was * Hawt, Palzont. New York, vol. i, 1847, p. 99, pl. 30, fig. 10. + Hawt, Palzont. New York, vol. ii, 1852, p. 250, pl. 53, fig. 3. { Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 66, pl. vii, figs. 4, 52; ibéd., Suppl., v, p. 209, pl. xvii, figs. 43-46, § Portiock, Rep. Geol. Londond. (1843), p. 445, pl. xxxii, fig. 13. 822 DR F. R. GC. REED ON THE known. ‘The “ upper” valve shows no foramen and no internal characters which would lead us to place it with the Discinidz : there is no trace of a pedicle-tube or groove The shape of the shell, the incurved pointed apex, and its general aspect suggest that it should be referred to the gasteropod genus Archinacella or to some allied patelliform genus. Probably Porttock’s specimens from Desertcreat should also have the same reference. We may remark that it bears a considerable resemblance to Archinacella orbiculata (Hall),* A. deleta (Sardeson),+ and allied species from the Ordovician of America. 4 Genus PHILHEDRA, Koken. Philhedra drummuckensis, sp. nov. (Plate IV, figs. 26-29.) Shell circular. Upper valve conical, capuliform, with pointed apex situated excentrally at about one-fourth to one-fifth the diameter of valve from poste: end, with steeper posterior slope. Surface of valve covered with closely set fine radiating lines of minute radially elongated pustules, and a few irregular concentric orowth-ridges. Dimensions.—Diameter, 5-5'5 mm. ; height, 2-2°5 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Remarks.—The general affinities of this shell seem to be with Ph. rivulosa (Kutorga),{ but it is much more closely allied to the species Ph. mulloche described below, though it has a more pointed apex. The internal characters are not satisfactorily preserved. Philhedra mullochensis, sp. nov. ? (Plate IV, figs. 30-83.) 1883. Crania siluriana, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur, Suppl., p. 215, pl. xvii figs. 51-53 (non figs. 49, 50). Non 1866. Cranta siluriana, Davidson, op. cit., vol. iii, pt. vii, p. 82, pl. viii, figs. 19, 20. “oa Shell subcircular. Upper valve convex, capuliform, rounded, subconical, with blunt apex situated near posterior margin at about one-fourth to one-third the diameter of the shell. Surface of valve covered with minute pustules arranged in very numerous closely placed regular radial rows; one or two coarse irregular low rounded concentric ruge usually present. Interior of upper valve with pair of faintly impressed large anterior occlusors, and slightly thickened internal margin. — Dimensions.—Diameter, 9-11 mm. ; height, 3-4 mm. Horizons.—(1) Mulloch Hill Group ; (2) Saugh Hill Group. Localities.—(1) Mulloch Hill, Craigens ; (2) Newlands. * RuEpEMANN, Bull. 525, New York Educ. Dep. (Mus. Bull. 162), 1912, p. 108, pl. 7, figs. 1-6. 4 + Unrice and Scorre.p, Palxont. Minnesota, vol. iii, pt. ii, 1897, p. 831, pl. 1xi, figs. 16-20. | Von Hurns, Verhandl, Russ. Miner. Gesellsch. St Petersburg, ser, 2, vol, xxxvi (1899), p . 200, t. iv (xii), figs. 8 8- If ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 823 Remarks.—This species differs from the typical Philhedra silurvana (Davidson) * of the Wenlock Beds by its more elevated rounded capuliform shape and less central apex and by its pustulate ornamentation. Several of the species of Philhedra described by Von Huene j from the Baltic provinces have a general resemblance to it and some- what similar ornamentation, but none seem very closely allied. It differs chiefly from Crama setifera, Hall,t of the Niagara Group, by its more excentric and blunter apex. Philhedra * penkillensis, sp. nov. (Plate IV, fig. 34.) Shell subcircular, conical, less than half as high as wide; apex excentric, situated at about one-third the diameter of shell from posterior margin. Interior of upper valve with two? pairs of subcircular muscle-scars, the anterior pair subcentral and large, posterior pair indistinct. Surface of shell ornamented with very fine closely placed equal raised thread-like radial lines and a few concentric striz and growth- ridges. Lower valve nearly flat. Dimensions.—Diameter about 5°5 mm.; height about 2 mm. — Horizon.—Penkill Group. Locality.—Penkill. Remarks.—The single specimen of the upper valve with the shell preserved, on which this species is founded, was labelled by Davipson as Discina sp. and marked as drawn, but I cannot find out where it was figured. The position of this shell may be either in the genus Philhedra or Pseudocrania, and its affinities may be with Ph. solaris (Kichwald).§ The muscle-scars are somewhat indistinct and uncertain, but the anterior pair is fairly well seen in an internal cast. Philhedra Playfoiri, sp. nov. (Plate IV, figs. 35, 36.) Shell subquadrate, widening anteriorly, slightly emarginate on anterior and posterior borders. Upper valve convex, rounded ; apex obtuse, excentric, situated nearer posterior than anterior margin ; posterior face short, steep. Interior of upper valve with weak transversely oval posterior occlusors close to margin, and deeply sunk subcentral circular anterior occlusors situated closer together, with narrow longitudinal median ridge in depression between them ; strong lateral vascular sinus on each side, extending round in front of anterior occlusors and nearly meeting in middle and giving off to margins short radiating lateral branches. Surface of shell ornamented with thin sublamellose concentric growth-ridges somewhat irregular, and bearing small sub-equidistant pustules. * Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 82, pl. viii, figs. 19, 20. + Von HuEns, op. cit., pp. 297-315. t Hatt, 11th Ann. Rep. State Geol. Indiana, 1882, p. 283, pl. 21, figs. 8-10, § Von HvENE, op. cit., p. 308, t. iv, figs. 23, 24. 824 ‘DR F. R. C. REED ON THE Dimensions.—Length, 12'0 mm.; width (maximum), 15°0 mm. Hoiizon.—Balclatchie Group. ; Locality.—Dow Hill. Remarks.—Only one internal cast of the upper valve and a portion of the impression of the shell of the same specimen are available. The internal characters suggest a relationship to Ph. rivulosa (Kutorga),* but the ornamentation of the surface is more like Pseudocrania petropolitana (Pander).t The interior is much like Crania Crofti, Dav.,{ from the Bala beds of Wales. This species is dedicated to the famous Scotch professor, JoHN PLAYFarrR. Philhedra siluriana (Davidson). (Plate IV, fig. 37.) 1866. Crania siluriana, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vil, p. 82, pl. viii, figs. 19, 20. — 1883. Crania siluriana, Davidson (pars), op. cit., vol. v, Silur. Suppl., p. 215, pl. xvii, figs. 49, 50 (non figs. 51-53), This species was recorded by Davipson from Woodland Point and Mulloch Hill, but nearly all those from the latter locality seem to belong to the species here described as Ph. mullochensis. The typical Ph. siluriana is much flatter, more depressed, and the ornamentation consists only of concentric growth-ridges. Hatt’s§ Crama siluriana from the Niagara Group is probably identical in this restricted sense. In a few examples from Woodland Point some of the muscle-scars can be distinguished, and it is seen that the anterior occlusors are circular and larger than the posterior ones (which are very faint) and situated their own diameter apart. Von HuEneE|| refers this species to the genus Phalhedra. Horizons.—(1) Saugh Hill Group; (2) ? Mulloch Hill Group. Localities —(1) Woodland Point ; (2) ? Mulloch Hill. Philhedra sp. (Plate IV, fig. 38.) One internal cast of the upper valve of a species of Philhedra from the Starfish Bed occurs in the British Museum, but it cannot be referred to the above-described species Ph. drummuckensis from the same horizon. "4 The shell is conical, and semi-elliptical in ae being wider than long * Von HuENE, op. cit., p. 300, pl. iv, figs. 8-16. + Von Huens, op. cit., p. 231, pl. i, figs. 4-14. { Davipson, Mon. Brit. Foss. Brach., vol. v, Silwr. Suppl., v, p. 215, pl. xvii, figs, 54-56. § Hawt, Trans. Albany Instit., vol. iv, p. 208 (1863) ; id., 11th Ann. Report State Geol. Indiana, 1882, P.! 82, pl. 21, figs. 3-7 ; Bescner and Ornate: Mem. New York State Mae , i, 1889, p. 13, pl. 1, figs. 1, 2. || Von Huznsg, op. cit., p. 217. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 825 longitudinal lines. The surface of the shell appears to have been covered with rather strong numerous rounded radii which project over the edge as minute marginal teeth, but no pustules can be detected. Indeed, the whole character of the shell suggests its relationship with Davipson’s Crana Crofti,* which Von HuENE j places in the genus Plilhedra; but the marginally projecting radii especially resemble Psewdo- erama pectinata, Gagel,{ from the Jevesche Bed of Estland. Dimensions.—Lenegth, 7°5 mm. ; width, 9°0 mm. ; height, c. 3°0 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Genus PuHouipors, Hall. Pholidops antiqua (Schlotheim) ? (Plate V, figs. 1-3.) 1822. Patellites antiquus, Schlotheim, Petrefaktenkunde, Suppl., i, p. 62, t. xii, fig. 2. 1899. Pholidops antiqua, Von Huene, Verh. Russ. Miner. Gesell. St ee: ser. 2, Bd. xxxvi, p. 270, text-fig. 11, t. 1, figs, 19-21 (and references). A few badly preserved specimens of a species of Pholidops in Mrs Gray’s collection from Shalloch Mill appear to have the muscle-scars, scutellum, and pallial line developed in the same manner as Ph. antiqua, Schloth., which they resemble rather than the Silurian species Ph. implicata (Sow.).§ But the material is too imperfect for a precise determination. The same remarks apply to a specimen from the Starfish Bed. Horizons.—(1) Whitehouse Group; (2) Drummuck Group (Starfish Bed). Localities.—(1) Shalloch Mill; (2) Thraive Glen. Pholidops implicata (Sowerby). (Plate V, fig. 4.) 1839. Patella ? implicata Sowerby, in Murchison’s Silwr. Syst., p. 625, pl. xii, fig. 14a. 1866. Crania implicata (Sowerby), Davidson, Mon. Brit. Foss, Brach., vol. iii, pt. vii, p. 80, pl. viii, figs. 13-18. 1883. Pholidops implicata (Sowerby), Davidson, op. cit., vol. v, Silur. Suppl., p. 216, pl. xvii, fig. 48. This interesting little species is represented on several horizons in the Llandovery of Girvan, and Davipson recorded it from the Penkill and Woodland Point Beds. The curious central scar, well described and illustrated by. Von HueEns, || is clearly seen in several specimens. The relations of this species with Ph. antiqua (Schl.) are discussed by Von HvENE (op. cit., p. 281), and again by Mosere and GRONWALL.4 * Davipson, op. cit., vol. v, Silur, Suppl., p. 215, pl. xvii, figs. 54-56. + Von HUENE, op. cit., p. 217. { Gaegn, Brach. camb. silur. Gesch., 1890, p. 25, t. i, fig. 11. g Davinson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 86, pl. viii, figs. 13-18. || Von Huxng, op. cit., p. 276, t. ii, fo 22-25 ; text-fig. 12. ‘S| Mosrre and Gronwatt, “ Om Fyledalens Gothandauien Lunds Univ. Arsskrift, N.F., Af. 2, Bd. v, No. 1, 1909, p. 30, t. il, figs. 8, 9. 826 DRE R. Co REEDON THE Horizons.—(1) Saugh Hill Group ; (2) Penkill Group; (3)Camregan Group. ‘ Localities.—(1) Woodland Point, Newlands; (2) Penwhapple Glen; (3) Bargany Pond Burn. 3 Genus PsEupocranta, M‘Coy. Pseudocrama divaricata, M‘Coy. 1851. Pseudocrania divaricata, M‘Coy, Ann. Mag. Nat. Hist., ser. 2, vol. viii, p. 388. 1852. Pseudocrania divaricata, M‘Coy, Syn. Brit. Pal. Foss, Woodw. Mus., p. 187, pl. in, figs. 1, 2. 1853. Crania divaricata, M‘Coy, Davidson, Mon. Brit. Foss. Brach., vol. i, Introd., p. 122, figs. 246, 247, 1866. Crania divaricata, M‘Coy, Davidson, op. cit., vol. iti, pt. vii, p. 78, pl. vill, figs. 7-12. ’ This species was not recorded by Davipson* from the Girvan area, but a poor specimen occurs in Mrs Gray’s collection from the Starfish Bed, measuring about 25 mm. in length, and showing the typical characters, Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Pseudocrama multifilosa, sp. nov. (Plate V, fig. 5.) Shell broadly oval to elliptical, convex, capuliform, forming low cone with excentric subacute apex directed forwards and situated from posterior margin about one-third the length of the shell. Surface of shell marked with a few concent i growth-lines and wrinkles and with closely placed very numerous delicate radiating fine lines, curving slightly back at sides. Dimensions.—Length, 8°0 mm.; width, 6°5 mm.; height, 3°2 mm. Horizon.—Drummuck Group iStareen Bed). Localitves.—Drummuck, Thraive Glen. Remarks.—This species does not seem closely allied to any previously described British species. The interior is unknown. Pseudocrama aft. depressa (Kichwald). (Plate V, fig. 6.) aff. 1842. Orbicula depressa, Kichwald, Urwelt Russland, ii, p. 76, t, i, fig. 11. Shell subquadrate, widening somewhat anteriorly, truncate posteriorly. Brachial valve convex, rounded, capuliform, most elevated towards posterior end; post face steep, short, abrupt; interior of valve showing (1) broad flattened marg limb ; (2) weak median longitudinal ridge leading to shallow emargination of front margin ; (3) pair of large oval anterior occlusor muscle-scars well defined except i front and closely placed to median ridge which separates them, situated behind mic dle of shell; (4) low broad rounded concentric ridge at posterior end of valve, halfway * Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, pp. 79, 80 ; Von HuENE, op. cit., p, 224, ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 827 between muscle-scars and edge of valve; whole interior of valve densely covered with minute pits arranged in concentric lines. Exterior of shell unknown. Dimensions.—Length, 9°5 mm ; width (maximum), 10°5 mm. ; height (maximum), 4°0 mm. Horizon.—Balclatchie Group. ‘ Locality.—Ardmillan. Remarks.—Only one internal cast of the shell above described has come under my notice. It appears to be related to Ps. depressa (Hichwald), as interpreted by Von Hurne.* The minute pits show of course as tiny granules in the internal cast. Order PROTREMATA. Genus Ortuis, Dalman. The classification of the numerous species of this genus has been attempted by several paleontologists with more or less success, but the most comprehensive and detailed system is that put forward by Messrs Hatt and CiarKks.t The probability of further additions to the number of groups or subgenera equivalent in value to those which they proposed was, however, recognised by these authors, and it must be confessed that their scheme is not entirely comprehensive or sufficient when applied to British species. Some new divisions of subgeneric rank have accordingly been here introduced. It may be mentioned that ScuucHErT{ has elevated several of Hatt and Crarke’s groups to the rank of genera, and has also used several new subgeneric terms. But in the present state of our knowledge it seems wiser to group all the species in one genus Orthis, while recognising the fact that it contains many more or less distinct subgeneric divisions. Subgenus ORTHIS, sens. rest. Orthis calligramma, Dalman, var. nov. craigensis. (Plate V, figs. 7-9.) 11868. Orthis calligramma, Dalman, var., Davidson, Mon. Brit. Foss. Brach., vol. iii, pl. xxxv, fig. 6 (non cet.). A form of O. calligramma occurs at Craighead in considerable abundance, but it does not agree strictly with the usual Welsh and English shells referred to this species. It closely resembles Wiman’s type§ from the Orthoceras Limestone, except in the more conical shape of the pedicle-valve and smaller incurvature of the beak. In this Craighead variety the shell is subquadrate to subcircular and of rather large size, averaging 20-25 mm. in length; the pedicle-valve is subconical and * Von HUENE, op. cit., p. 249, t. i, figs. 24-27 ; t. ii, figs. 1, 2. + Haut and Crarke, Palzont. N.Y., vol. viii, Brach., i, 1892, pp. 186-229. { ScuucHERrt, in Zittel’s Text-book of Palzxontology, vol. i (translation, 2nd edit., 1913), pp. 381, 382. _ § Wa1an, Bull. Geol. Instit. Upsala, vol. viii, 1907, p. 108, pl. viii, figs. 28-30. TRANS. ROY. SOC, EDIN., VOL. LI, PART IV (NO. 26). 120 828 DR F. R. C. REED ON THE slightly convex, with the beak forming the apex and rising high above the hinge line, but not much incurved ; the hinge-area is large and steeply inclined, and has a large triangular delthyrium. The muscle-scar in this valve is very short, being Ret almost entirely to the beak, and it is deeply sunk and well circumseribed, .The brachial valve is flattened, but has occasional traces of a weak median depression, There are 24-30 simple equal.regular straight rounded ribs, separated by cuba or slightly narrower rounded interspaces, and both the ribs and interspaces are ornamented by fine longitudinal lines, while the interspaces usually show strong concentric striz. Horizon.—Stinchar Limestone Group. | Localities.—Craighead, Bogang. Orthis caligramma, Dalman, var. nov. subplicata. (Plate V, figs. 10-15.) 1883. Orthis calligramma, Dalman, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, ‘Stee Suppl., . 225, a 1883. Ree rustica, Sowerby, Davidson (pars), dbid., p. 226. ae Shell plano-convex, subquadrate, rounded to semielliptical ; hinge- -line rather | less than width of shell or equal to it ; cardinal angles obtuse or subrectangular. Pedi valve strongly convex, swollen; beak rather high, incurved; hinge-area lar triangular, lying in plane of valve or steeply inclined. Brachial valve flattened, y very weak median depression widening anteriorly ; beak small, inconspicuous, not raised ; hinge-area narrow, nearly at right angles to valve. a Interior of pedicle-valve with broad subtruncate short triangular muscle-sear rather deeply impressed, less clearly defined in front, composed of two narr diductors enclosing adductors of about one-third the width of the scar; a pair of © short broad rapidly disappearing subparallel vascular sinuses run forward : diductors; teeth short, small. | Interior of brachial valve with thickened cardinal plate, long narrow linear cardinal process, and very short stout triangular crural plates ; muscle-scars obscure. | Surface of valves ornamented with 20-26 rounded ribs of equal strength and equidistant on margin; about half the number are primaries, the others being i calated in the interspaces or branching off one side of the primaries at about one fourth to one-half their length. Interspaces wider than ribs, rounded, with strong concentric striation. Ribs curve back slightly on both sides towards cardinal : ang! les Dimensions.—Length, 20 mm. ; width, 24 mm. Horizon.—Mulloch Hill Group. Locality.—Mulloch Hill Wood. Remarks.—These shells were mostly labelled by Davipson O. rustica, but are quite distinct in their convexity and ribbing. The interior of the brachial valy ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 829 is much like O. plicata, Sow.,* and the interior of the pedicle-valve and its shape and contour somewhat resemble O. virgata, Sow.,t but from the former it differs by its shorter hinge-line and from the latter by its different ribbing. O. sowerbyana, Day., is much more transverse and has more numerous and closer ribs. The interior of our shell and general characters suggest that it is closely allied to O. flabellites, Foerste,{ of the Clinton Beds. Orthis calugramma, Dalman, var. ? (Plate V, figs. 16-19.) 41883. Orthis calligramma, Dalman, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., pp. 181, 225. It is possible that some small shells, measuring 5-10 mm. in width, from the Starfish Bed are referable to this species; but as no larger examples have been found, they may be only the young of some other species. In shape, ribbing, number of ribs, and internal structure, so far as known, they possess the usual characters of this species in its wide sense, and indeed more resemble the type-form than the Stinchar Limestone and Mulloch Hill varieties above described. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Orthis Playfair, sp. nov. (Plate V, figs. 20-32.) 1883. Orthis calligramma, Dalman, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., pl. xiii, fig. 24 (non cet.). 11883. Orthis unguis, Sowerby, Davidson, zbid., p. 117, pl. xiii, fig. 28. Shell subquadrate to subcircular, unequally biconvex to plano-convex ; hinge-line rather less than maximum width of shell; cardinal angles obtuse or slightly rounded. Pedicle-valve regularly convex, rounded ; beak more or less raised and swollen, somewhat incurved, rising well above hinge-line ; hinge-area rather large, triangular, lying in plane of valve or steeply inclined; delthyrium wide, triangular. Brachial valve flattened or very weakly convex, much less deep than opposite valve, with weak median depression ; beak low, small, inconspicuous; hinge-area narrower than in opposite valve. Interior of pedicle-valve with large and stout teeth, and with elongated sub- triangular muscle-scar occupying beak, bounded in front by short strongly convergent ridges merging into low broad thickening of shell forming semioval boss; diductors long, subtriangular, separated by narrow adductor impressions ; ovarian areas marked with weak radial grooves. Interior of brachial valve with very thick transverse cardinal plate bearing straight * Davipson, Mon. Brit. Foss. Brach., vol. iii, pl. xxxvii, fig. 1. + Ibid., pl. xxxvii, fig. 2. { Gragav, Bull. N.Y. State Mus., No. 45, vol. ix, 1901, p. 186, fig. 94. 830 . DR F. R. C. REED ON THE linear cardinal process and very stout short crural plates; adductor scars rather deeply sunk, separated by broad rounded ridges about one-third the length of valve. Ribs 12-24 in number, angular or subangular, simple, coarse, regular, straight, decreasing in size towards the cardinal angles, separated by angular or subangular interspaces equal in size to ribs, both ornamented with a few fine longitudinal lines and closely set concentric strive. Shell substance pierced with a few large punctee or tubules. Ribs show marginally inside valves. * ; Dimensions.—(1) Craighead form, average length, 10-12 mm.; (2) Balclatchie form, average length, 12-15 mm. ; (3) Shalloch Mill form, average length, 10-14 mm, Horizons.—(1) Stinchar Limestone Group; (2) Balclatchie Group; (3) White- house Group. B a Localities.—(1) Craighead, Minuntion, Tramitchell, Doularg, Bogang; (2) Bal- clatchie, Ardmillan ; (3) Shalloch Mill. = Remarks.—This shell undoubtedly belongs to the group of O. calligramma as understood by Wysocorsk1,* but cannot be regarded as a typical example of this species, though included in it by Davirpson and most authors. It-differs in the angularity, greater coarseness, and smaller number of the ribs, and in these respects resembles Wysocorskrs O. Schmidti and O. callactis, Dalm. The variety of O. callactis figured by Linpstrom + differs in having the ribs less angular, but otherwise is much like our Girvan form. The Craighead specimens have rather fewer ribs and a lower hinge-area on the pedicle-valve and are usually smaller than those from the Balclatchie Beds, but otherwise seem specifically inseparable. Wysocorskr’s O. lyckholmiensis (op. cit.), as interpreted by Wriman,{ may be com- pared as regards ribbing, but it has a lower beak to the pedicle-valve and the shell is biconvex, besides occurring at a higher horizon. Gace.’s§ O. calligramma may be identical with our form. It appears advisable to separate the latter from the true O. calligramma, and it may be distinguished as O. Playfairi. It is probable that Davipson’s 0. calligramma of his pl. xxxiv, figs. 1 and) 3), trem Moelydd, is identical, but it is doubtful if the specimen from Craighead, fig. 4, belongs to the same species, as the ribs seem rounded and more widely spaced. Dr Marry || has recently discussed the synonymy of O. proava, Salter, which is closely allied to our Girvan shell; but I am not prepared to consider them identical, because the Welsh shell has a more convex shape, a more deeply lobed brachial valve, and a less” distinct angularity of the ribs. The coarseness and especially the angularity of the ribs recall certain examples of O. actonix ; but the internal characters, the shape of the shell and the less swollen and less incurved beak of the pedicle-valve distinguish the present species. The internal characters are chiefly known from the specimens * WysoGorskI, Zertsch. dewt. geol. Gesell., lii, 1900, p. 230, t. viii. + Linpstrom, Fragm. Silur., p. 26, pl. xiv, figs. 1-3. f {| Wrman, Arkiv f. Zool., Bd. iii, No. 24, 1907, p. 8, t. ii, figs. 9-12. a § Gacen, Brach. camb. silur. Gesch., 1890, p. 33, t. ii, fig. 18. : || Martmy, Appendix III, Swmmary of Progress of Geol. Surv. for 1911, 1912, p. 78. fy ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 831 from the Balclatchie Beds, which mostly occur as internal casts with external impressions ;' and it is possible that these should be separated as a variety. The Shalloch Mill examples more closely resemble the shells from Craighead than those from Balclatchie, and are especially comparable to the above-mentioned O. lyckholmiensis, Wysog.; the longitudinal striations of the Craighead and Balelatchie forms seem, however, missing. This species is dedicated to the distinguished Scotch professor, JoHNn Puayratr. Subgenus PLECTORTHIS, Hall and Clarke. Orthis (Plectorthis) ardnullanensis, sp. nov. (Plate V, figs. 33-39.) Shell semicircular to semielliptical ; hinge-line equal to width of shell; cardinal angles rectangular or slightly pointed. Pedicle-valve subconical, slightly convex ; beak pointed, not incurved ; hinge-area large, at right angles to plane of valve ‘or steeply inclined, with triangular delthyrium. Brachial valve flattened or slightly concave except near beak, with a broad very shallow median depression widening to margin ; beak small, inconspicuous ; hinge-area steeply inclined, smaller than that on pedicle-valve. Interior of pedicle-valve with small broad subcircular or transverse well-defined muscle-scar about one-fourth (or less) the length of valve; teeth small, supported on dental plates. Brachial valve with thickened hinge-plate, thin laminar cardinal process, and large stout crura. Posterior adductors well marked, oval, widely separated ; anterior adductors smaller, weaker. Surface of valves ornamented with 28-35 simple rounded or quadrate equal and equidistant ribs, separated by rounded interspaces of equal size. Ribs and interspaces crossed by delicate concentric striz, and usually marked with a few fine longitudinal strie, of which the one in the middle of each interspace is the most prominent. Ribs show near margins of interior of shell. Dimensions.—Lenegth (average), 10 mm.; width (average), 12 mm. Horizon.—Balclatchie Group. Localities.—Ardmillan, Balclatchie (conglomerate) ? Remarks.—In its essential characters this new species much resembles O. tiice- nara, Conrad,* and Davrpson + himself described a shell from Piedmont Glen, apparently belonging to our species, as O. tricenaria with a query. In its ribbing it much resembles O. calligramma, but in its general shape and in the contour of the valves it is different. The ribbing is not quite the same as in O. tricenaria, and is more like that in the Girvan species described below as O. subplicatella, but the nature of the muscle-scar in the pedicle-valve and the conical shape of that valve distinguish it. O.duftonensis, Reed, { differs by the greater convexity of the brachial * WINCcHELL and ScnucHert, Palzont. Minnesota, vol. iti, p. 418, pl. xxxii, figs. 18-20. t+ Davipson, op. cit., iii, p. 276, pl. xxxviii, fig. 28. i REEp, Geol. Mag., dec. v, vol. vii, 1910, p. 295, pl. xxiv, figs. 5-11. 832 DR F. R. C. REED ON THE valve and the less transverse outline of the shell and by the shape and characters of the muscle-scars. The best preserved and most typical examples of O. ardmillanensis come from Ardmillan. There is a variety (figs. 37-39) occurring at this locality and in the Balelatchie conglomerate which seems to differ only in the greater number (40-45) of its ribs, and probably it can only be separated as a variety of the type. Orthis (Plectorthis) duftonensis, Reed, var. (Plate V, figs. 40,41; Plate VI, figs. 1, 2.) 11883. Orthis calligramma, var. plicata, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 118, pl. xiii, fig. 26 (excl. cet.). A few shells from Ardmillan showing the external impression and internal casts of both valves seem to be specifically inseparable from O. duftonensis, Reed,* described from the neighbourhood of Melmerby. The shape, contour, and type of ribbing are identical ; and the internal characters, including the remarkable vasculai sinuses in the pedicle-valve, agree in all essential particulars ; the number of ribs is also relatively the same, varying from 40-45. In the typical O. duftonensis from Melmerby the ribs are rather narrower than the interspaces, but in the smaller Ardmillan specimens the ribs are equal in width to the interspaces or rather wi than them. The Girvan examples vary from 10-25 mm. in width. Davip labelled them O. calligramma, var. plicata, and the general resemblance of O. tonensis to this form was pointed out by me in the original description of the spec Vascular sinuses of much the same character occur in O. wnsculpta, Hall,t and in O. Rankin, Dav. (sens. restr.), as described below. Horizon.—Balclatchie Group. Locality.—Axrduillan. Orthis (Plectorthis) rustica, Sowerby. 4 (Plate VI, figs. 3-8.) . 1839. Orthis rustica, Sowerby, in Murchison’s Silurian System, p. 624, pl. xii, fig. 9. . a 1868. Orthis rustica, Sowerby, Davidson, Mon. Brit. Foss. Brach., vol. iii, p. 238, pl..xxxiv, figs. 13-1 1883. Orthis rustica, Sowerby, Davidson (pars), op. ctt., vol. v, Cane Suppl., p. 226. 1883. Orthis plicata, Sowerby, Davidson (pars), tbid., p. 226. The typical form of O. rustica occurs at Woodland Point, but is much less abundant than its variety rigida. Externally the much closer and more numerous ribs distinguish it from the latter as well as the more marked convexity of the brachial valve which in var. rigida is nearly flat. But in some specimens the are fewer and mostly simple. Internally the muscle-scar in the pedicle-valve is ‘“Davipson described, saucer-shaped or rather subcircular to subpentagonal, but nearly as broad as long and not elongated, and is only about one-fourth the length * Rugp, Geol. Mag., dec. v, vol. vii, 1910, p. 295, pl. xxiv, figs. 5-11. + Hatt and Crarxn, op. cit., Brach., i, pl. va, fig. 13. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 833 of the valve, while the diductors are not separated by a ridge, and there is a pair of strong, outwardly curved divergent vascular sinuses starting from its front end and enclosing the ovarian areas. In the brachial valve the two pairs of adductors are strongly marked, but rather small; and there are distinct vascular sinuses diverging oh each side from them. Davipson* showed these features fairly well in his figs. 16, 17. Horivzon.—Saugh Hill Group. Locality.—W oodland Point. Orthis (Plectorthis) rustica, var. nov. paucicostata. (Plate VI, figs. 9-12.) 1883. Orthis rustica, Sowerby, Davidson (pars.), op. cit., Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., . 226. 1883. Orthis calligramma, Dalman, var. plicata, Sowerby, Davidson, ibid., p. 226. Some of the larger shells from Woodland Point and Cuddystone Glen, which have the general characters of O. rustica, differ by the smaller number, greater simplicity, and regularity of the ribs on the surface, which also show strongly on the interior, as in the variety mgida. Thus there are only from 25-30 primary ribs on a shell with a width of about 20-30 mm. ; these ribs are all rounded and simple and straight and of equal or subequal size, while at about half their length there may be intercalated a shorter, smaller rib, and there is an absence of the concentric rugosities generally noticeable. ‘The internal characters appear to be identical with O. rustica, but the cardinal angles are usually more pointed and the ribbing is more like that of O. plicata, Sowerby. This variety is difficult to separate completely from O. rustica and its variety O. rigida, as it occurs associated with them, and transitional forms are observable. Davipson labelled those from Cuddystone Glen O. rustica, and those from Woodland Point O. calligramma, var. plicata. It seems best to regard them as a separate variety of UO. rustica, and they may be designated paucicostata. Horizons.—(1) Saugh Hill Group; (2) Camregan Group. Localities. (1) Woodland Point ; (2) Cuddystone Glen. Orthis (Plectorthis) rustica, Sowerby, var. rigida, Davidson. (Plate VI, figs. 13-19.) 1847. Orthis rigida, Davidson, Geol. Journal, p. 64, pl. xiii, figs. 1-4. 1868. Orthis rustica, var. rigida, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 238, pl. xxxiv, figs. 18,19. * 1883. Orthis rustica, var. rigida, Davidson, op, cit., vol. v, Silur. Suppl., pp. 178, 226. This form is very abundant at Woodland Point, and ranges in size from 10-20 mm. in length. The pedicle-valve is conical to moderately convex and is always swollen about the beak, which is somewhat incurved; the brachial valve is nearly flat or only weakly convex, with a faint median depression in the less flattened * DAvipson, op. cit., iii, p. 238, pl. xxxiv, figs. 13-17. 834 DR F. BR. Cy REED ON TEE examples. The ribs are subangular, and there are about 20-30 primaries with shorter and smaller ones intercalated rather irregularly as regards number and length, but usually at about one-third and again at about three-fourths ; rarely do the primaries divide. The interspaces between them are somewhat unequal and irregular in size, The lateral ribs curve back slightly. On the margin the ribs are unequal in strength. All are crossed by close concentric lamellations. In no case are they as numerous asin the typical O. rustica from the Wenlock Limestone. The internal characters of this variety were not depicted by Davinson, but show some noticeable features, In the pedicle-valve the muscle-scar forms an elongated pentagon, nearly twice as long as wide, and is sunken and well defined; it is rather less than one-third the length of the shell and is truncated in front ; the diductors are distinet, and separated by a narrow median ridge. In the brachial valve the cardinal process is narrow, short, and linear, and the crural plates are short and widely divergent ; the two pairs of adductors are usually feebly impressed. Davipson* was incorrect in saying the muscle-scar in the pedicle-valve was saucer-shaped if he meant this description to apply to var. rigida. It may be necessary to regard this variety as a distinet, species on account of this difference, which sharply distinguishes it from the typical O. rustica. ; Horizon.—Saugh Hill Group. Localities.—Woodland Point, ?Shalloch Forge. Orthis (Plectorthis) rustica, var. walsalliensis, Davidson. (Plate VI, fig. 20.) : 1848, Orthis Walsulltt, Davidson. Bull. Soc. Geol. France, ser. 2, vol. v, 1848, p, 322, pl. iv, fig. 7. 1868. Orthis rustica, var. wallsalliensis, Davidson, Mon. Brit. Foss. Bra+h., vol. iii, pt. vii, p, 238, pl. xxxiv, figs. 20-22. } 11883. Orthis rustica, var. walsalliensis, Davidson, tbid., vol. v, Silur. Suppl., p. 226. Davipson described this variety originally from the Wenlock Limestone of Walsall, but subsequently figured examples from Rushall Canal and Buildwas, and mentioned it as occurring in the Girvan area at Penkill. Some poor specimens from Penwhapple Glen appear to be referable to it, judging from their shape an ad numerous ribs. Horizon.—Penkill Group. Locality.—Penwhapple Glen. Orthis (Plectorthis) subfissicosta, sp. nov. (Plate VI, fig. 21.) 1883. Orthis intercostata, Portlock, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 226. — Non 1843. Orthis intercostata, Portlock, Geol. Rep. Londond., p. 454, pl. xxxvii, fig. 3. Shell transversely semielliptical; hinge-line equal to width of shell; card angles rectangular. Pedicle-valve gently convex, most so near beak; beak small * Davinson, op. cit., vol. iii, p. 238, pl. xxxiv, figs. 18, 19, i ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 835 pointed, very slightly incurved, rising a little above hinge-line; hinge-area small, triangular, steeply inclined, with triangular delthyrium. Brachial valve (not well known) flatter than pedicle-valve, with smaller inconspicuous not elevated beak and narrower hinge-area. Surface of valve with 12—14 primaries dividing into two or three unequal ribs close to beak, the middle one usually the strongest; a shorter rib of intermediate size is usually intercalated between the primary fascicles at one-third to one-fourth their length, and near the hinge-line all the ribs are smaller, subequal, and more closely placed; all the ribs are subangular and straight. Dimensions.—Length, 13 mm. ; width, 19 mm. Horizon.—Stinchar Limestone Group. Locality.—Aldons. Remarks.—The two specimens on which the above description is based were labelled O. intercostata, Portl.,* by Davripson, and one} of his figured specimens of this species is stated in the explanation of the plate to be from Girvan, but he does not give this locality in the description on p. 236, though in his Supplement it is entered in the list on p. 226 as occurring at Craighead, Aldons, and in the Balelatchie conglomerate. However, these specimens from Aldons do not agree in their ribbing with the type of O. intercostata, Portlock, from Tyrone (Davinson, op. cit., iii, pl. xxxviii, figs. 1, 3), and they more resemble O. fissicosta, Hall,} particularly in respect of the ribs, though the latter divide closer to the beak in our species than in the American one. Orthis (Plectorthis) subplicatella, sp. nov. (Plate VI, figs. 22-26.) Shell transversely subelliptical ; unequally biconvex ; hinge-line rather less than maximum width of shell; cardinal angles obtuse. Pedicle-valve strongly convex, somewhat flattened anteriorly in middle; beak rounded, incurved; hinge-area moderate, triangular, nearly in plane of valve. Brachial valve much less convex than opposite valve, somewhat flattened ; beak small, inconspicuous, not incurved ; hinge-area smaller than that of opposite valve, steeply inclined or at right angles to plane of valve. Surface of valves covered with about 30 straight simple regular equal quadrate or rounded ribs with vertical sides, equidistant, rather wider than the interspaces, which are deep, regular, rounded; both covered with close regular con- centric striations. Interior of pedicle-valve with small teeth, and with oblong to subpentagonal muscle-scar, truncate anteriorly, about one-third (or less) the length of valve, deeply sunk, with sharp lateral marginal edges and long subparallel sides: slightly indented at about half length; diductors duplicate on each side, enclosing narrow adductor scar. * PortLocK, Geol. Rep. Londond., p. 454, pl. xxxvii, fig. 3. 7 Davipson, op. cit., iii, p. 286, pl. xxxviii, fig. 4. { Hatt, Palwont. N.Y., vol. i, 1847, p. 121, pl. xxxii, fig. 7; Fomrsra, Bull, Denison Univ., vol. xiv (1909), p. 221, pl. iv, figs. 54, B; ibid., vol. xvi (1910), p. 48, pl. vi, be 4, TRANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 26). 121 - 836 DR F. R. C. REED ON THE Interior of brachial valve with sublanceolate cardinal process, stout triangular crural plates; adductor scars widely separated, posterior pair subcircular, more distinct, and larger than anterior pair. Ribs show internally round margins of shell. : Dimensions.—Length, 10-11 mm. ; width, 15— 16 mm. Hovrizon.—Balelatchie Group. Localities.—Balclatchie, Ardmillan. ; Remarks.—This shell agrees in shape and general characters of ribbing and in- ternal characters with O. plicatella, Hall,* of the Hudson River and Trenton Groups. The shape of the muscle-scar in the pedicle-valve recalls O. Whitfieldi, Winchell.} Our species is allied to O. calligramma and O. ardmillanensis in the regularity, shape, and simplicity of its ribs, and likewise to O. duftonensis, but it cannot well be associated with any of these species in any strict conception of their “pea fic limitations on account of the nature ot the muscle- -scars. in w en orem oe O. Whitfieldt, and of which O. scoehliy Miller, is the gene a type. Orthis (Plectorthis) thraivensis, sp. nov. (Plate VI, figs. 27-32; Plate VII, fig. 1.) Shell transversely semielliptical to subquadrate, plano-convex, or slightly concavo- convex ; cardinal angles rectangular ; hinge-line equal to maximum width of shell. — Pedicle-valve forming a low cone, flattened or slightly concave, less deep tha 0 opposite valve except at beak. Beak forming apex of cone, pointed, not incurved, higher than that of brachial valve. Hinge-area large, triangular, steeply inclined or nearly at right angles to plane of valve; delthyrium large, triangular. Brachial valve gently convex; beak small, not prominent, slightly ineury hinge-area smaller than that of pedicle-valve, vertical or nearly so. Interior of pedicle-valve with subcircular to subpentagonal well-circumseribed obscurely separated by narrow linear adductors, with sharp curved marginal ridges defining it at sides but weaker in front. z Interior of brachial valve with narrow thin compressed linear cardinal process, stout cardinal plate produced into low broad median ridge separating obscurely marked adductor scars; dental sockets with thick anterior walls. Surface of walls covered with narrow rounded regular straight ribs of equal si * Hau, Palxont. New York, vol. i, 1847, p. 122, pl. xxxii, fig. 9; MumK, Paleont. Ohio, vol. i, 1873, a pl. viii, fig. 7: Hann and CiarKe, op. cit., pl. v, figs. 18-20. ; + Wincaett and ScaucuErt, Paleont. Minnesota, vol. iii, pl. xxxiii, fig. 8, p. 437. { Forrsts, Buli. Denison Univ., vol. xiv, 1909, p. 225. re § Forrsts, ibid., vol. xvii, 1912, p, 131. || Foxrrsre, 2bid., p. 130, pl. viii, figs. 8A, B,C, — ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 837 nearly all primaries (especially in young individuals), and increasing in number by regular intercalation close to beak (and very rarely again), separated by equal rounded or subquadrate interspaces ; all ribs equal in size and equidistant on margin and numbering 50-60 in the larger shells. Ribs and interspaces crossed by strong close concentric striations. Ribs show strongly on interior of shell. Dimensions.— (rie. (2) (3) Length . ; ia EF 135 14°5 mm. Width = . é vie we ONE 18°0 20°0 mm. Horizon.—(1) Drummuck Group (Starfish Bed) ; (2) Whitehouse Group ? Locality.—(1) Thraive Glen; (2) Shalloch Mill. Remarks.—This shell may be placed in the subgenus Plectorthis. In the shape, regularity, and simplicity of the ribbing it recalls many shells referred to the species O. calligramma by different authors, and may also be compared to O. duftonensis.* The interior of the brachial valve specially resembles that of O. plicata, Sow.t Fragmentary specimens are difficult to distinguish from O. porcata, M‘Coy. Some poorly preserved shells from Shalloch Mill, of which the internal characters are unknown, were labelled by Davipson O. sowerbyana and O. calligramma, but agree in shape and ribbing with the Drummuck examples of O. thraivensis. Subgenus prnorruHis, Hall and Clarke. Orthis (Dinorthis) flabellulum, Sowerby, var. nov. carrickensis. (Plate VII, figs. 2-6.) 1883. Orthis flabellulum, Sowerby, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., p- 179, pl. xiii, figs. 3-5 (non fig. 6). Shell plano- to concavo-convex, semicircular to semielliptical; length to width about 4:5; hinge-line less than width of shell; cardinal angles rounded or obtuse. Pedicle-valve subconical, flattened or slightly concave, shallow, deepest at beak, re- supinate ; beak small, not incurved nor much elevated, pointed ; hinge-area at right angles to plane of valve, moderately large, triangular. Brachial valve gently con- _yex; beak small, inconspicuous; hinge-area smaller than that of opposite valve and lying im plane of valve. Surface of valves covered with straight radiating rounded ribs, of equal size, increasing by fairly regular bifurcation at about one-third (or less) their length, numbering 50-60 on margin; interspaces equal to or narrower than ribs, rounded; ribs and interspaces crossed by fine concentric striz. Interior of pedicle-valve with short subquadrate well-defined muscle-scar, about one-fourth (or less) the length of valve, weakly emarginate behind. Remarks.—The large resupinate shells from Craighead which Davipson referred * Regd, Geol. Mag., dec. v, vol. vii, 1910, p. 295, pl. xxiv, figs. 5-11. + Davipson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 245, pl. xxxv, figs. 25, 26; pl. xxxvii, fig. 1. 838 DR F. R. C. REED ON THE to this species differ to some extent from the typical form of O. flabellulum.* The ribs are more numerous and divide more regularly and more often, splitting up into two or three smaller ones at one-third to two-thirds their length, and as there are 24-30 primaries the total number of ribs on the margin is much greater than in the type. Indeed, the shells approach O. porcata in their ribbing and shape, but the ribs are coarser and stronger, the pedicle-valve is shallower, and the beak less prominent. Davipson’s figures of this Girvan variety do not correctly represent the ribbing. There is no good interior preserved, but a fragmentary cast of a pedicle-valve shows the characteristic muscle-scar. The specimens from Craighead are all of large size and measure 30-35 mm. in length. Smaller examples of the same variety occur at Ardmillan Brae. We may note a considerable resemblance to Dinorthis Ulrichi, Foerste,t from the Ordovician of Kentucky. . oe Horizons.—(1) Stinchar Limestone ; (2) Balclatchie Group. Localities —(1) Craighead ; (2) Ardmillan Brae. Orthis (Dinorthis) porcata, M‘Coy. : (Plate VI, figs. 7-12.) 1846. Orthis porcata, M‘Coy, Syn. Silur. Foss. Ireland, p. 32, pl. iui, fig. 14. 1848. Orthis inflata, Salter, var. retrorsa, Salter, Mem. Geol. Surv., vol. ii, p. 373, pl. xxvii, figs. 3, 4. 1852. Orthis porcata, M‘Coy, Syn. Brit. Pal. Foss. Woodw. Mus., p. 223, pl. in, figs. 41, 42. = 11860. Orthis Carleyt, Hall, 13th Rep. New York State Cab. Nat. Hist., p. 120, text-fig. 1870. Orthis porcata, M‘Coy, Davidson, Mon. Brit. Foss, Brach., vol. iii, pt. vii, pl. xxxi, figs. 12 16 (217-20). ze 1883. Orthis Sowerbyana, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., pp. 78, 226 1883, Orthis flabellulum, Sowerby, Davidson (pars), ilid., p. 179, pl. xiii, fig. 6 (non figs. 3-5, 20, 2: 11892. Orthis (Plesiomys) retrorsa, Salter, Hall and Clarke, Palwont. New York, vol. viii, Brach., iy pp. 197, 222, pl. va, figs. 14-16. by 11912. Dinorthis Carleyi, Hall, Foerste, Ohio Naturalist, vol. xii, No. 3, pp. 449, 453, pl. xxii figs. Sa—p. I The resupinate shells from Girvan which I refer to O. porcata were identified b Davison partly as O. flabellulum, Sow., and partly as O. sowerbyana. All of: come from the Upper Bala or Starfish Bed of Thraive Glen. Davrpson apparen relied on the internal characters of the pedicle-valve for his identification of some a O. flabellulum (Suppl., p. 179, pl. xiii, fig. 6), for there is a note in his writing on one marked specimen saying: “ This is a very characteristic interior of the ventral valve of O. flabellulum, making the identification certain.” But the characters of exterior of the shell are quite sufficient to separate it from this species, wh internally there are certain differences which show that it cannot be associated with the typical O. flabellulum, Sow., from Boduan. ‘ Some of the Girvan shells are more convex than others, and such were labellec O. sowerbyana by Davipson and quoted by him from Thraive Glen (op. cit., p. * Davinson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 248, pl. xxxiv, figs. 1-12. + Forrste, Bull. Denison Univ., vol. xiv, art. 17 (1909), p. 320, pl. vii, figs. 74—-c. | | ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 839 This corresponds to the shell called inflata, var. retrorsa, by SaurEr.* The internal characters of the brachial valve are precisely those figured by Davinson in O. porcata (op. cit., ili, pl. xxxi, fig. 15); it also agrees with the specimen from Coniston figured by M‘Coy + under this name. A comparison with the original types proves this fact. With regard to external characters, the ribbing is identical with the Coniston type, but M‘Coy was incorrect in stating that the ribs only increase by division, for they also increase by intercalation, and only occasionally bifurcate at one-third to one-half their length in the pedicle-valve, while in the brachial valve they appear to inerease mainly by bifurcation at one-fourth to one-third their length and often again at two-thirds, and sometimes close to the margin. The interspaces between the ribs are subequal, rounded, and usually narrower than the ribs themselves. The "primary ribs number about 20-25 and continue right through to the margin without increasing in size, and are regular and equidistant, but as a result of the intercalation or division of other ribs there are usually about 75-80 on the margin. All the ribs of whatever length are uniform and equal in size on the margin, and are straight, except towards the cardinal angles of the brachial valve, where they curve back slightly. There is considerable variation in the amount of convexity of the brachial valve, but the shells are always strongly resupinate, and between the large swollen and the gently convex condition all intermediate stages can be found. The pedicle- valve likewise varies in the degree of concavity or flatness. In the brachial valve it should be remarked that the sides of the oval or lanceolate cardinal process are obliquely channelled by 3-4 distinct grooves, and this feature is also well seen in the type-specimen of Portiock’s O. grandis, { which Davinson (op. cit., pl. xxxi, fig. 19) refers to O. porcata. With regard to the muscle- sears, the anterior adductor scars are roughened and smaller than the posterior ones, as in M‘Coy’s Coniston types. In the interior of the pedicle-valve the muscle-scar is subpentagonal, well circum- scribed, but not deeply sunken: it is weakly emarginate in front, and each half seems composed of two unequal elements side by side, this feature being indicated by the bilobation of the lateral margins. -In this respect the scar is distinguished from that of O. flabellulum. The original specimen of O. porcata on which the species was founded by M‘Coy § came from Portraine, but the description given by him of the ribbing does not apply well to the shells from Coniston subsequently figured by him under the same name, for he states that “ the ridges [ = ribs] are angular, of unequal length, and frequently branch into three or four at once, but alternating with these are some quite simple from beak to margin and others dividing into two.” The Tyrone shell described by * Sarmr, Journ. Geol. Soc., vol. i, 1845, name in table facing p. 21; id., Mem. Geol. Surv., vol. ii, 1848, p. 373, pl. xxvii, figs. 3, 4. + M‘Coy, Syn. Brit. Pal. Foss. Woodw. Mus., 223, pl. iH, figs. 41, 42. } Porriock, Geol. Rep. Londond., p. 452, pl. xxxii, fig. 25. § MCoy, Syn. Silur. Foss, Ireland, p. 32, pl. iii, fig. 14. 840 DR F. R. C. REED ON THE PortLocK as O. grandis, and based on the internal cast of a brachial valve, was referred, as above stated, by Davipson to O. porcata; but, while it agrees in internal characters with M‘Coy’s form from Coniston and with our Girvan specimens, the ribs appear to be much finer and more numerous, and the cardinal angles are more rect- angular and the whole shell more subquadrate in shape, so that its specific separation may be necessary. The interior of the pedicle-valve of the Tyrone shell was not figured by Porrrock, but Davrpson figured (op. cit., pl. xxxi, fig. 20) one of his specimens (now in the Jermyn Street Museum), and it agrees in the muscle-scar, ete., with the Coniston O. porcata, but the valve is less concave and more subquadrate and is rather abruptly deflected at the margin. A series of other examples from Tyrone indicates that these features are constant. Our Girvan shell, on the other hand, closely resembles O. Cates Hall,* from the Hudson River Group of America, and has frequently been identified with O. retrorsa, Salter, which, as above stated, is placed by Davipson in O. porcata, M‘Coy. Hatt and CLARKE put both in the subgenus Plesiomys, but subsequently ScHUCHERT { in- cluded this subgenus in Dinorthis, and the separation of the two seems inadvisable. ForRSTE } is inclined to separate O. Carleyi from O. retrorsa, but puts both in Dinorthis. In, the Girvan area O. porcata appears to be confined to the Drummuck Group, whatever its vertical range may be elsewhere. Horizon.—Drummuck Group (especially Starfish Bed). Locality.—Thraive Glen. Subgenus HEBERTELLA, Hall and Clarke. Orthis (Hebertella) balclatchiensis, Davidson. (Plate VII, figs. 13-20.) 4 1888. Orthis Bouchardi, Davidson (pars), Mon, Brit. Foss. Brach., vol. ili, pt. vil, pl. xxvi, _ 23 (non figs. 16-22). 4 1883. Orthis balcletchiensis, Davidson, op. cit., vol. v, Stlur. Suppl., p. 176, pl. xiii, figs. 12-14, 1883. Orthds crispa, Davidson (pars), ibid., p. 176, pl. xiii, fig. 8. Davipson’s description of this species is correct so far as it goes, but the interic of the shell is only figured without any remarks on its characters. Internal casts 0 both valves from Balclatchie and Ardmillan in Mrs Gray’s collection show that the pedicle-valve had a strongly marked triangular muscle-scar, bluntly truncated im front, and raised on a distinct thickening of the shell; the diductor scars are narrow and are separated by a shorter adductor scar of equal width; and the genital a are marked by radiating broken sinuous ridges and vascular sinuses, a specially strong pair usually running forwards divergently from the front end of the diducto scar and dying out at about two-thirds the length of the shell. % In the brachial valve the crura are short, stout, and triangular; there is a narro\ * Hawi and Ciarke, Palxont. New York, viii, pt. 1 (1882), pp. 197, 222, pl. va, figs. 14-16. + Scnucuert, Bull. 87, U.S. Geol. Surv., 1897, pp. 134, 215. t Forrstn, Ohio Naturalist, vol. xii, No. 3 (1912), pp. 449, 453, pl. xxii, figs. 84—D. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 841 thin low laminar cardinal process and a broad subquadrate muscle-scar composed of two pairs of deeply impressed adductors of which the anterior pair is the larger and expands anteriorly ; a strong median ridge is present, and the genital areas are pitted and ridged. As regards the external shape it may be mentioned that the pedicle-valve is usually deeper and more conical, and has a larger hinge-area than the other valve, but the median sinus is less marked or may be absent. The number of ribs varies, and occasionally the median rib divides into 2 or 3 subequal ones a short distance from the beak. This species is closely allied to O. crispa, M‘Coy, which occurs on a_ higher horizon in the Girvan area, but the former differs in the shape of the muscle-scar in the pedicle-valve, in the beaks of the valves being more equal in size, in the brachial valve being more convex, in the teeth being larger and stouter, and in the ribs being fewer in number. The shell is also usually smaller. The American species O. (Heb.) bellarugosa, Conrad,* from the Galena shales of Minnesota, bears a considerable resemblance to it. Horizons.—(1) Stinchar Limestone Group ; (2) Balclatchie Group. Localities.—(1) Craighead ; (2) Balclatchie (shales and conglomerate), Ardmillan. Orthis (Hebertella) bellatrix, sp. nov. (Plate VII, figs. 21-28; Plate VIII, figs. 1, 2, ? 31.) 1883, Orthis sowerbyana, Davidson (pars), Mon. Brit. Foss, Brach., vol. v, Silur. Suppl., pp. 178, 226. Shell unequal, biconvex, transversely subelliptical ; hinge-line straight, rather less than width of shell; cardinal angles obtuse or slightly rounded. Pedicle-valve less convex than brachial valve, with very shallow broad median sinus or flattening of: surface causing wide weak sinuation of front margin; beak moderately high, slightly ineurved, prominent, with high triangular hinge-area steeply inclined to plane of valve; delthyrium large. Interior with elongate subpentagonal to oval muscle-scar, about one-fourth the length of the valve, deeply sunk and circumscribed, composed of a pair of small elongate diductors with narrow linear adductor scar between them ; teeth strong, short. Brachial valve swollen, more convex and deeper than opposite valve; beak low, small, inconspicuous, not projecting; hinge-area narrow, nearly in plane of valve. Interior with long narrow linear cardinal process, swollen at base; strong short erural plates; adductor scars separated by low broad rounded ridge; posterior adductors large, circular, deep, well marked ; anterior adductors smaller, indistinct. Surface of valves ornamented by numerous subequal subangular ribs, slightly curved back on sides, increasing rather irregularly by division into 2 or 3 at about one-fourth their length, and frequently again at half or two-thirds their length, and rarely by intercalation ; ribs finer and closer together towards cardinal angles, * WINCHELL and ScuucHErt, Pal. Minnes., vol. ili, p. 434, pl xxxiii, figs. 1-4, 842 DR F. R. C. REED ON THE but elsewhere on margin nearly equal in size and numbering in all 100-130. Coarse concentric close striation over ribs and interspaces. Ribs apparent near margins it interior of shell. j a Dimensions.—Average length, 24 mm. ; width, 30 mm. ; thickness, 14 mm. Hovizons.—(1) Balclatchie Group (including conglomerate) ; (2) Stinchar Lime- stone Group ? hae Localities.—(1) Balclatchie, Dow Hill; (2) Doularg. Remarks.—The best-preserved specimens come from Balclatchie; those from the conglomerate, which Davipson labelled O. sowerbyana (op. cit., Swppl., p. 178), have rather finer ribs, but there is in each case some variation in their strength. This shell must undoubtedly be put in the subgenus Hebertella, and is allied to O. alveata-richmondensis, Foerste,* O. subjugata, Hall,t O. occidentalis, Hall,} and O. sinuata, Hall§ One specimen from Doularg (fig. 31) labelled by Davipsoy as a “‘most characteristic specimen of O. confinis” seems to be referable to our new species. a Orthis (Hebertella) crispa, M‘Coy. (Plate VIII, figs. 3-7.) 1846. Orthis crispa, M‘Coy, Syn. Silur. Foss. Ireland, p. 29, pl. iii, fig. 10. } —. 1870. Orthis crispa, M‘Coy, Davidson, Mon. Brit, Foss. Brach., vol. iii, pt. vii, p. 256, pl. xxxviii, figs. 5-10, 1883. Orthis crispa, M‘Coy, ibid., vol. v, Silwr. Suppl., p. 176, pl. xiii, fig. 7 (non 8). er Shell unequally biconvex, subquadrate to semicircular or transversely semi elliptical, from 1} to 1% times as wide as long; hinge-line usually rather less width of shell or just equal to it; cardinal angles obtuse or slightly roun Pedicle-valve moderately convex, somewhat flattened in middle near anterior mar ‘or weakly depressed; beak rather high, pointed; not much incurved or swollen; hinge-area moderate, steeply inclined. . Brachial valve strongly convex, more swollen than pedicle-valve, and usué without any median depression ; beak much lower and smaller (but more ineurved) and hinge-area much narrower than in opposite valve. Shell subquadrate, nearly as long as wide, unequally biconvex, widest at hi line; cardinal angles rectangular or slightly less than right angles, rarely al Hinge-line furnished with 6-10 small oblique transverse crenulations for a s. distance on each side of delthyrium. Pedicle-valve gently convex, most so alor median line, often subcarimate, deepest just behind middle, more or less flattens towards cardinal angles; beak small, acute, rising well above hinge-line; hinge-ar triangular, moderately high, steeply inclined, with wide triangular delthyri frequently narrowed by lateral ingrowths of margins. Interior of pedicle-valve wit pair of stout teeth inclined at 60°-70° to hinge-line; muscle-scar deeply impr behind, but weak and indistinctly defined in front, composed of pair of rather | flabelliform diductors enclosing small oval raised adductor scar ; ovarian areas p indefinite. Brachial valve more uniformly convex than opposite valye, occasio with a weak median depression; beak small, inconspicuous; hinge-area nam than that of pedicle-valve, nearly at right angles to plane of valve, with triang la delthyrium. Interior of brachial valve with prominent thick anterior walls to dent sockets, and stout cardinal process composed of pair of closely placed triangular le deeply grooved on posterior face ; hinge-plate thick, produced anteriorly int broad rounded median ridge dying out anteriorly, and with small median tuber¢ situated immediately in front of cardinal process ; adductor muscle-scars weak] pressed, indistinct, apparently consisting of two pairs; ovarian areas strongly p indefinite. Surface of valves covered with numerous regular closely placed radi small rounded or subangular ribs, slightly curved back on each side, with rather finer ones intercalated rather irregularly between the primaries, us at one-third and again at one-half to three-fourths their length, and with ver concentric striation over all. Ri: Dimensions.—Leneth, 18-24 mm; width, 20-27 mm. Horzon.—Maulloch Hill Group. Localities.—Mulloch Hill, Craigens, Kast Threave Farm. . Remarks.—The crenulations of the inner part of the hinge-line are rarely - owing to the state of preservation of the specimens. But the bilobed ¢ process at once separates it from the true Orthis sagittifera, M‘Coy,* to Davupson referred some examples, and the coarse ornamentation as well as the ii features render it difficult to see how Davrpson was led to label some other spe “ Strophomena” expansa, Sowerby. The reference to the subgenus Brach * M‘Coy, Syn. Brit. Paleoz. Foss. Woodw. Mus., 1852, p. 227, pl. ix, figs. 15-19. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 897 is indicated by the limitation of the cardinal crenulations to a portion of the hinge- line, and the internal characters agree in a general manner with Str. (Br.) varistriata (Conrad) * and Str. (Br.) profunda (Hall),+ of the American Silurian. Stropheodonta (Brachyprion) Walmstedti (Lindstrém). (Plate XVII, fig. 3.) 1860. Strophomena Walmstedti, Lindstrom, Proc. Roy. Acad. Stockholm, pl. xiii, fig. 16. 1868. Strophomena Walmstedti, Lindstrom, Davidson, Trans. Geol. Soc. Glasyow, vol. i, p. 18, pl. iil, figs. 5, 6. 1870. Strophomena Walmstedti, Lindstrom, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 290, pl. xl, figs. 6-8. 1883, Strophomena Walmstedti, Lindstrom, Davidson, zbid., vol. v, Stlur. Suppl., p. 226. This species is recorded by Davipson { from the Wenlock Shale of the Pentland Hills and from the Upper Llandovery of Cuddystone Glen, Girvan. The two specimens from the latter locality in Mrs Gray’s collection labelled by Davripson are poor ; one is the internal cast of.a pedicle-valve, measuring about 40 mm. in length, and shows ob- securely the muscle-scars and pittings and coarse oblique denticulations on the hinge-line close to the beak which were not mentioned or depicted by Davinson but are typical of the subgenus. A specimen from Penkill occurs in the Hunterian collection at Glasgow. Horizons.—{1) Camregan Group; (2) Penkill Group. Localities.—(1) Cuddystone Glen; (2) Penkill. Stropheodonta (Brachyprion) sp. (Plate XVII, fig. 20.) Shell transversely semicircular, widest along hinge-line ; cardinal angles slightly alate and pointed at about 45°-60°. Pedicle-valve gently convex, lateral portions somewhat flattened; beak small, acute ; hinge-area narrow, triangular, transversely striated, steeply lied to plane of wahees with triangular delthyrium ; hinge-plates semiova] near delthyrium, elongate, extending for a short distance along hinge-line close to beak, marked with 12-14 transverse oblique parallel straight grooves: remainder of hinge-line not denticulate. Diductor scars large, flabelliform, indistinct, very weakly impressed. Surface ornamentation unknown. Dimensions.—Length about 12 mm.; width about 24 mm. Horizon. —Camregan Group. . Locality._Camregan Wood. Remarks. —The more transverse shape of this shell and the apparent absence of the coarse radial ornamentation distinguish this imperfectly known shell from Str. columbana. Only two internal casts of the pedicle-valve are known, so that the material is quite insufficient for a satisfactory comparison. * ConraD, Journ. Acad. Sc. Philadelphia, viii, 1842, p. 255, pl. 14, fig. 6 + Hatt, Palecont. New York, vol. ii, 1852, p. 61, pl. 21, figs. 4, 5. $ Davipson, op, cit., vol. iii, p. 290, pl. xl, figs. 6-8. 898 DR F. R. C. REED ON THE Stropheodonta? Nicholsoni, sp. nov. (Plate XVII, fig. 21; Plate XVIII, figs. 1-8.) 1883. Strophomena imbrex, var. semiglobosina, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 195, pl. xv, figs. 9, 10 (non fig. 11), ‘ Shell concavo-convex, compressed, subquadrate to semielliptical, widest along hinge-line, about one-third wider than long; cardinal angles subrectangular to sv acute, less than right angles, sometimes slightly alate and pointed at 60°% [hinge-line furnished along whole length with coarse projecting denticles not clos placed]. Pedicle-valve inflated, strongly convex, with slightly flattened later l angles; beak obtuse, somewhat swollen, rather high and incurved; hinge-a triangular, transversely striated, lying in plane of valve, with large triangular thyrium, Interior of pedicle-valve with pair of stout triangular teeth and weal impressed subcircular muscle-scar less than one-fourth length of valve and composed of pair of short blunt flabelliform diductors separated by narrow oval adduc Brachial valve concave, usually with broad shallow indefinite median depression ; _ small, not swollen, inconspicuous; hinge-area narrower than that in opposite v. and nearly at right angles to it; interior with strong short crural plates inel at 15°-20° to hinge-line; cardinal process imperfectly known; muscle-scars faint; weak concentric submarginal rugose and coarsely pustulose internal thic ing sometimes present. Surface of valves covered with 60-100 straight subang or rounded radiating subequal riblets, mostly primaries, closely placed, with sh and usually rather smaller secondary ribs occasionally intercalated at irregu intervals towards margins ; interspaces equal in size to ribs, subangular to roun all ribs and interspaces crossed by fine regular concentric lineation ; ribs appe on interior of both valves. Dimensions.— (1) (2) Length . ' ; ¢ 2h 110 Width : : sea) 15°5 mm. Horizon.—Balclatchie Group (including conglomerate). Locality.—Balclatchie. : Remarks.—The cardinal process of this shell is unfortunately not well preserved in any of the specimens; its apparent simplicity and absence of bifurcation is prob due to the fact that only the stalk is preserved in the few imperfect casts or im: sions which are available. The cardinal denticulation also is somewhat uncert The shell has altogether a certain Orthoid appearance which renders its true posi doubtful, though its general characters seem to point to Stropheodonta. The specimens from the Balclatchie conglomerate referred by Davison to “ Str phomena imbrex, var. semiglobosina,” were inaccurately figured, for in one (op. fig. 10) the cardinal angle on the left side is omitted and the long straight hinge-li therefore unduly shortened, In the other specimen (op. cit., fig. 9) the cardinal angles ar ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 899 imperfect and broken, but the flattening in this region is apparent. In both the orna- mentation is completely different from that of the true Raf. semiglobosina (Dav.) (see p. 869). The ribbing and shape of Str. Nicholsoni recall some species of Strophomena from the American Ordovician (e.g. Str. planoconvexa, Hall,* and Str. sulcata, Verneuil +), but the strongly swollen beak, the apparently denticulated hinge-line, and the reversed convexity of the valves mark it off from all of these species and from this genus. It is named in honour of Professor ALLEYNE NicHoLson, whose monograph on the Girvan fossils, in collaboration with R. ErHEriper, jun., is well known. Stropheodonta? Waltoni (Davidson) ? 11847. Leptena Waltoni, Davidson, London Geol. Journ., vol. i, p. 56, pl. xxvi, fig. 3. 11848. Leptena Waltont, Davidson, Bull. Soc. Geol. France, ser. 2, vol. v, p. 317, pl. iii, fig. 6. 1883. Strophomena Waltoni, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 195, pl xvi, fig: I. There is only one poor external cast of one valve of this species, and Davrpson’s figure is more or less a restoration. Davinson’s identification with Str. Waltonz is uncertain, though the shape of the shell is similar to the type of Str. Walton: from the Wenlock Shale of Falfield,t but the ornamentation is not clearly preserved. The hinge-line is apparently denticulate, and the shell may be probably referred to the genus Stropheodonta. It is possible that this specimen may belong to the species Stroph. arenacea, Salter, which has been above described from Camregan. Horizon.—Camregan Group. Locality.—Cuddystone Glen. Stropheodonta sp. (Plate XVIII, figs. 9, 10.) There are some fragmentary shells from Mulloch Hill in Mrs Gray’s collection which seem to possess many of the characters of Str. corrugatella (Dav.) of the Upper Ordovician, but on account of certain differences it does not seem possible to include them in this species. Thus the shell is subquadrate in form, with small pointed ears, and it is rather suddenly arched down at about two-thirds its length, though not provided with a flattened dise or distinct geniculation. The character- istic ornamentation is mainly limited to the portion. of the shell which is not bent down, the latter portion being only covered with the radiating primary riblets with the wide interspaces showing 2-3 finer radial lines. These characters resemble those of Str. Patersoni, Hall,§ of the Devonian of North America rather than the typical Str. corrugatella (Dav.). One internal cast of a pedicle-valve, somewhat broken and * Forrste, Bull. Scient. Lab. Denison Univ., vol. xvii, 1912, p. 54, pl. i, figs. 1A—B, 2A, B. + Ibid., p. 102, pl. i, figs. 4a-c ; pl. xi, fig. 2. { Davrpsoy, op. cit., vol. ili, p. 310, pl. xii, fig. 11, § Haut, Tenth Ann. Rep. New York State Cab. Nat. Hist. 1857, p. 114, figs. 1-5; Haun and Crarky, op. cit. Brach., i, 1892, pl. xiv, fig. 15. TRANS. ROY, SOC. EDIN., VOL. LI, PART IV (NO. 26). 129 900 DR F. R. GC. REED ON THE crushed, shows a short subcircular muscle-scar, more deeply sunk and more sh margined than in the last-mentioned species, but otherwise the internal charaet are not preserved. Dimensions.—Leneth about 22°0 mm. ; width (on. hinge-line) oben 26°0 mm, — Horizon.—Mulloch Hill Group. B Locality.—Mulloch Hall. Genus STROPHONELLA, Hall. Strophonella euglypha (Hisinger). 1819. Leptena euglypha, Hisinger, Anteck, pl. vi, fig. 4. 1870. Strophomena euglypha (Hisinger), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, Pe 38 pl. xl, figs. 1-5 (for further references). This species seems very rare in the Girvan district. There is one zood ai of a pedicle-valve from Woodland Point in Mrs Gravy’s collection, showing characteristic large diductor scars and other features of the species, and it | -indistinguishable from bypial Wenlock examples. Davripson does not record species from Girvan. Horizon.—Saugh Hill Group. Locality.—Woodland Point. Strophonella penkillensis, sp. nov. (Plate XVIII, figs. 11-13, 14 2) 1883. Strophomena antiquata, Sowerby, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. p. 193 (non pl. xv, figs. 12-14), Shell transverse, triangular, maximum width along hinge-line ; concavo-¢ strongly resupinate; cardinal angles flattened, acutely pointed at about 30° weak indefinite median fold present near anterior margin; disc flattened, some obscurely marked off from anterior half, which is arched down; hinge-line fi crenulated. Pedicle-valve strongly concave; beak small, pointed, not pro inconspicuous ; hinge-area linear, narrow ; interior with pair of small teeth an transversely subcircular completely circumscribed muscle-scar, about one-fo less) the length of shell, composed of two flabellate broad short diductors, divergent, with narrow ridge between them anteriorly. Brachial valve with fl disc and cardinal angles; margin arched down and swollen into low fold in mi beak inconspicuous ; hinge-area very narrow, linear ; interior of valve with pro elevated large bilobed cardinal process, each lobe oval and grooved on inne crural plates long, straight, thick, at 30°-45° to hinge-line; hinge-plate posterior adductors small, deeply sunk, subcircular ; anterior adductors longer, | parallel, suboval, undefined in front, but laterally with thickened edges. Su shell marked with regular fine equidistant primary radii about 20-30 in number, interspaces occupied by 5-7 much finer thread-like equal radii, all crossed by deli concentric lineation. Occasionally a shorter secondary radius equal in stren; ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN. DISTRICT. 901 to the primaries is intercalated at half their length, but all the primaries and secondaries are equidistant on the margin. Dimensions.—Leneth, 14 mm. ; width, 24 mm. Horizons.—(1) Penkill Group; (2) Camregan Group. Localities.—(1) Penkill; (2) Bargany Pond Burn. Remarks.—This rare shell must be referred to Strophonella, and it agrees in many respects, such as ornamentation and general internal characters, with Stroph. euglypha,* but it is more transverse and the muscle-scar in the pedicle-valve is smaller. The difference in the ribbing distinguishes it at once from Str. antiquata, var. woodland- ensis, and the shell is also shorter, but more transverse, and the cardinal angles more pointed. Some examples of it were labelled by Davipson Strophomena antiquata. Strophonella undata (M‘Coy). | (Plate XVIII, figs. 15-18.) 1846. Orthis undata, M‘Coy, Syn, Stlur. Foss. Ireland, p. 36, pl. iii, fig. 21. 1852. Strophomena deltoidea, var. 8, wndata, M‘Coy, Syn, Brit. Pal. Foss, Woodw, Mus., p. 234, pl.1H, figs, 38, 39. 1870. Strophomena deltoidea, var. 8, undata, M‘Coy, Davidson, Mon, Brit. Foss. Brach., vol. iu, pt. vii, p. 295, pl. xxxix, figs. 23, 24. 1883. Strophomena deltoidlea (Conrad), Davidson, zbid., vol. v, Silur, Suppl., pp. 199, 226. The type of this species was obtained from Ireland, and both M‘Coy and Davipson came to regard it as merely a variety of Raf. deltoidea, but apparently they were ignorant of its internal characters. Amongst the specimens from the Starfish Bed which I would refer to this species, are casts of the interior. The pedicle-valve is concave and geniculated, not convex as in &. deltoidea; the disc is slightly convex and is as long as wide, and the closely placed rather narrow rugz on it meet the hinge-line at right angles; the valve is geniculated at about two-thirds its length ; the hinge-area is strongly striated and the hinge-line coarsely denticulated ; the muscle- scar is well defined, transversely subcircular, about one-third the length of the valve, and the large flabellate diductors are radially ridged; the ovarian areas are coarsely pustulate. In the brachial valve the disc is usually depressed and slightly concave and the beak is inconspicuous ; internally the cardinal process shows two lobes widely diverging and grooved on their posterior face ; the thickened rhomboidal hinge-plate is produced as a low pointed ridge passing into a thin low median septum between faintly impressed adductors. The radial ornament on both valves is almost the same as described and illustrated by M‘Coy, but the primary radii are less distinct. The true generic reference of this shell must be to Strophonella rather than to Strophomena, on account of the denticulation of the hinge-line, and a comparison with M’Coy’s figured specimens indicates that it should be placed in his species Str. undata. lxternally it is practically indistinguishable from Raf. deltoidea, but the internal characters and reversed convexity of the valves forbid us to associate * Davipson, op. cit., ili, p. 288, pl. xl, figs. 1-5 (especially figs. 4, 5). 902 DR F. R. C. REED ON THE it with this species or with Lept. rhomboidalis, with which fragmentary specim might be confused. a Horizon.—Drummuck Group (Starfish Bed). a Locality.—Thraive Glen. Genus Curistianta, Hall and Clarke. Christiania tenwicincta (M‘Coy). : (Plate XVIII, fig. 19.) 1846. Producta tenuicinecta, M‘Coy, Syn. Silur. Foss. Ireland, p. 25, pl. iil, fig. 4. 1852. Leptwna tenwicincta, M‘Coy, Syn. Brit. Pal. Foss. Woodw. Mus., p. 239, pl. in, fig. 40. ia 1870. Leplena tenuicincta, M‘Coy, Davidson, Mon. Brit. Foss, Brach., vol. iii, pt. vil, p. 326, p. fig. 7-18. ae 1883. Leptena tenuicincta, M‘Coy, Davidson, zbid., vol. v, Silur. Suppl., p. 168, pl. xi, fig. 28 ; pl figs. 17-21. gt Davinson’s figures and description of this species give a good idea of its ext and internal characters, but he omitted to mention the presence of the foramei the beak of the pedicle-valve. ; This foramen is frequently well seen in the specimens from Balclatchie witl shell preserved, and even in internal casts the little projecting peg is seen 3 cast of the beak indicating its presence. It should also be mentioned that th interior of the shell is sometimes ornamented with rather coarse longitudina broken up in places near the margins into elongated pustules. The Craighead shells are narrower and longer than the others, widening ant and are usually bilobed by a shallow rounded continuous depression extendi the beak to the anterior margin of the pedicle-valve ; the ears are also more dept and more sharply marked off from the body of the hell, I am doubtful if al Girvan shells from different horizons belong to the same species, and the Craig] form may at any rate be regarded as a variety for which the designation biloba suggested. Han and CLarke™ referred the species to their genus Christian Horizons.—(1) Stinchar Limestone Group; (2) Balclatchie Group; (3) V house Group; (4) Drummuck Group. A Localities.—(1) Craighead; (2) Balclatchie, Ardmillan; (3) Shalloch Mil Thraive Glen. Genus STROPHOMENA, Blainville (sens. vestr.). Strophomena antiquata (Sowerby), var. nov. woodlandensis. (Plate XVIII, figs. 20, 21; Plate XIX, figs. 1-5.) 21870. Strophomena antiquata (Sowerby), Davidson (pars), Mon. Brit. Foss. Brach., vol. iii, pt. W p- 299, pl. xliv, fig. 5 (non cet.), 1883. Strophomena antiquata (Sowerby), Davidson (pars), ibid., vol. v, Silur Suppl., p. 193, pl figs. 12-14, Shell subtriangular to semioval, rather wider than long, widest along hit more or less produced anteriorly into a tongue-like fold or bluntly pointed * HALL and Ciarxg, op, cit., Brach., i, p. 300. Wi ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 903 somewhat flattened disc, subgeniculate, concavo-convex ; cardinal angles slightly flattened, pointed, and produced. Pedicle-valve concave, generally produced in front into short blunt tongue ; interior of valve with short subcircular well-defined muscular area with sharp raised edges, extending about one-fourth the length of valve, composed of a pair of contiguous flabelliform diductors. Brachial valve gently convex, with more or less flattened ill-defined dise and bent-down margins ; cardinal process bilobed, deeply cleft, set on thick hinge-plate ; crural plates short, stout, inclined at about 30° to hinge-line; muscle-scars well marked, deep, the posterior adductors subcircular, rather wide apart, and the anterior pair longer, elongate-oval. Surface of valves covered by rather coarse angular straight or slightly curved ribs of unequal and irregular size varying in number and increasing by intercalation at irregular intervals of smaller ribs of unequal size, all rugose and crossed by con- centric striations, more or less coarse. Dimensions.— (1) (2) (3) Length : : ; ene 29 26 22, mm. Width : ; pee Re 27 mm. Horizon.—Saugh Hill Group. Locality.—W oodland Point. Remarks.—This form is a well-marked variety and is abundant at Woodland Pomt. It approaches most nearly the Wenlock variety scabrosa, Davidson,* but the ribs in the Girvan variety are not so numerous nor so regular, and are also coarser and more variable in size, while the scabrous concentric lamelle are less developed. The presence of a more or less definite disc and of the tongue-like projection in front are additional points of difference. The internal characters, how- ever, are essentially the same as Davipson figured for his var. seabrosa, so that it can only be regarded as a local variety. Strophomena deficiens, sp. nov. (Plate XIX, figs. 6-10.) Shell semielliptical to subquadrate, concavo-convex to plano-convex, much com- pressed, widest at hinge-line; cardinal angles rectangular. Pedicle-valve gently concave to nearly flat, slightly swollen near beak; beak small, pointed, rather retrorse, not incurved ; hinge-area of moderate width, triangular, steeply inclined to plane of valve, with large broad triangular delthyrium covered by convex plate ; interior of valve with pair of short small teeth on thin dental plates, posteriorly defining transverse subcircular short muscle-area, about one-third length of valve, composed of a pair of contiguous flabellate short radially ridged diductor scars feebly defined in front; weak submarginal concentric ridge occasionally present near edge of valve. Brachial valve. gently convex; beak small, inconspicuous ; * Davipson, op. cit., vol. iii, p. 299, pl. xliv, figs. 10-13. 904 DR F. R. C. REED ON THE bilobed cardinal process (the oval faces of the lobes facing inwards) rising small thick transverse hinge-plate ; crural plates long, thin, slightly arched very divergent, making angle of 10°-20° with cardinal line. Surface of sg covered with very numerous fine straight rounded closely placed subequidista secondary riblets shorter than the primaries ; all covered by close concentric stri Minute closely set pustules generally distributed over interior, fusing into sinuous radial lines near cardinal angles. Ye Dimensions.—Length, 21 mm. ; width, 33 mm. Horizons.—(1) Balclatchie Group ; (2) Stinchar Limestone Group ? 4 Localities.—(1) Balclatchie, Ardmillan ;.(2) Craighead. “4 Remarks.—This shell is especially allied to Strophomena incurvata, Shepard * (=Str. filitexta, Hall) of the Trenton Formation in America. But the Bri Strophomena simulans, M‘Coy,t is also a near relative, though the muscle-sea ii the pedicle-valve is of a different character, and the brachial valve is more conv and there are concentric ruge near the hinge-line. The Stinchar Limestone spec seem to have the pedicle-valve rather flatter than those from Balclatchie, and u all the radii are of subequal size, but occasionally the median primary rib enlarged and conspicuous; the muscle-scars also in the pedicle-valves from ( head are less strongly marked. The Balclatchie examples, which are taken as y pic consist of internal casts and external impressions of both valves and are abundant. The relative strength of the primary and secondary radii vary, and they may be almost equal in size. | Strophomena shallochensis, Davidson (emend.). (Plate XIX, figs. 11, 12.) 1883. Strophomena shallockiensis, Davidson (pars), Mon. Brit. Foss, Brach., vol. v, Silur. Suppl, pl. xi, figs. 20, 21 (non pl. xii, fig. 30); pl. xvi, fig. 8. This species was somewhat briefly described by Davripson, and though he mw doubtedly regarded the Shalloch Mill form as the type of the species, yet he fi first (op. cit., pl. xi, figs. 20, 21) a shell from Corwen, North Wales. In the She Mill type specimen (op. cit., pl. xvi, fig. 8) it should be remarked that in both \ the hinge-areas are steeply inclined to the plane of the valve, but in the case of th concave pedicle-valve the hinge-area is at right angles and larger; there is also a lar * Sueparp, Amer. Journ. Sc., vol. xxxiv, 1838, p. 144, figs. 1,2; Hann, Paleont, New York, vol. i, 184 pl. xxxip, fig. 3; Hann and Cuarkg, Paleont. New York, vol. viii, Brach., i, pl. ix, figs. 1-7, pl. ixa, fig Scuucuert, Bull. U.S. Geol. Surv., No. 87, 1897, p. 431. + M‘Coy, Syn. Brit. Palzoz. oss. Woodw. Mus., p. 403, pl. in, figs. 38, 34,735; Davipson, op. cit. vol p- 297, pl. xlii, figs. 9, 9a-c, 710. ‘ ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 905 triangular delthyrium in each valve. In this specimen there are about 50-60 ribs on / each valve, all of equal strength on the margin and equidistant, except near the cardinal angles, where they are more crowded; the ribs are rounded, straight, simple, and separated by rather wider interspaces ; about half the number of ribs are primary, the others arising by intercalation usually close to the beak; all are concentrically lineated. In some of the other specimens from Shalloch Mill, perhaps attributable to this species, there are 3-4 considerably finer radial lines between the larger ribs, as are shown in the shell from Corwen. The interior is very inadequately known, but Davinson’s figure of the interior of a supposed ventral valve (op. cit., pl. xi, fig. 21) looks more like the interior of a brachial valve. One imperfect interior of a pedicle-valve from Shalloch Mill in Mrs Gray’s collection shows traces of weakly impressed short flabelliform diductors. Dimensions (Davipson’s type, pl. xvi, fig. 8).—Length, 21 mm. ; width, 35 mm. Horizon.—W hitehouse Group. Locality.—Shalloch Mill. Strophomena valens, sp. nov. (Plate XIX, figs. 13-17.) Shell semicircular to transversely semielliptical, nearly twice as wide as long, compressed, slightly resupinate, concavo-convex; maximum width of shell along hinge-line ; cardinal angles rectangular or bluntly pointed at about 75°. Pedicle- valve gently concave, with rather prominent retrorse pointed beak rising a little above hinge-line ; hinge-area triangular, steeply inclined to plane of valve, with rather narrow delthyrium. Interior with strong short dental plates diverging at right angles, and large transversely subcircular muscle-scar about one-third the length of the valve, composed of a pair of radially ridged flabellate diductors not deeply sunk nor sharply defined in front. Brachial valve gently convex, with small inconspicuous beak; hinge-area narrower than in opposite valve. Interior of brachial valve with strong broadly oval bilobed cardinal process, the inner faces of the lobes elongated, sublanciform, parallel, closely placed, lying in plane of valve, with posterior face grooved ; crural plates thick, short, shghtly curved, diverging at more than right angles; muscle-scars faint, weakly impressed, separated by low broad median ridge. Surface of shell covered with very fine numerous straight radiating closely placed radii, increasing by intercalation, numbering about 350-400 on margin, with every 6th, 7th, or 8th one rather larger, all crossed by fine concentric strize. (The ribs show strongly round the margins in internal casts.) Dimensions.—Length, 22 mm.; width, 36 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Remarks.—This species is only found in the Starfish Bed, but some of the 906 DR F. R. C. REED ON THE triangular specimens were labelled by Davipson Sir. shallochensis, though he 7 il recorded this species (op. cit., p. 192) from Shalloch Mill in the Girvan area. It undoubtedly resembles the Corwen shells also referred to Str. shallochen valens ave usually more pointed, and the difference in the strength of the radii marked. Our Str. deficiens from Balclatchie is a near relative, and so is Str. si M‘Coy, while several American species (e.g. Sti. filiterta, Hall) appear to be all Genus SCHUCHERTELLA, Girty. Schuchertella applanata (Salter). (Plate XIX, figs. 18-20.) 1846. Orthis? applanata, Salter, Appendix to M‘Coy’s Syn. Silur, Foss, Ireland, p. 72, pl. v, fig. 1. 1865. Strophomena applanata (Salter), Haswell, Silur. Form. Pentlands, p. 34, pl. ii, fig. 18. 1868. Strophomena applanata (Salter), Davidson, Vrans. Geol. Soc, Glasgow, vol. i, p. 19, figs. 19, 20. 1870. Strophomena applanata (Salter), Davidson, Mon, Brit. Foss. Brach., vol. iii, pt. vii. p. 3 pl. xlin, figs. 12-14. ; 1883. Strophomena applanata (Salter), Davidson, ibid., vol. v, Silur. Suppl., p. 226. This species was recorded by Davrpson in his Supplement from Penkill, Pond Burn, and Cuddystone Glen in the Girvan district. These specimens | small, not exceeding 22 mm. in width, and they agree with the typical f the Wenlock Shale of the Pentland Hills, but the species was originally des SatTER from Cong in Ireland as Orthis, with a query. The projection of ribs on the margin is not marked, and a few very oblique short corruga frequently present along the hinge-line, meeting it at very acute angle: internal characters, so far as they are preserved, indicate its close resem Sch. pecten, Linn. . os Horizons.—(1) Penkill Group; (2) Camregan Group; (3) Saugh Hill Group. Localities.—(1) Penkill; (2) Bargany Pond Burn, Cuddystone Glen? ; (3) Newla Fo Schuchertelle ? imexpectata, sp. NOV. (Plate XIX, figs. 21, 22.) Shell semielliptical, compressed, flattened; hinge-line equal to or shig than maximum width of shell; cardinal angles rectangular or slightly ro Pedicle-valve nearly flat; beak small, scarcely rising above hinge-line; h very narrow; interior of valve with small short triangular teeth and indistinet s1 flabelliform contiguous diductor impressions. Brachial valve nearly flat ; beak small; hinge-area linear; imterior with thin laminar short transverse cardi process continued at sides into long thin crural plates making a very small a (15°-20°) with hinge-line. Exterior of valves covered with 30-32 regular distant strong rounded straight radiating primary ribs of equal size, with ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 907 interspaces holding 2-3 smaller and rather shorter ribs, of which one usually becomes larger than the others towards the margin; ribs and interspaces crossed by strong close concentric lineation. Dimensions.—Length, 8°0 mm. ; width, 12°5 mm. - Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Remarks.—This small shell seems to belong to the same genus as Sch. applanata and Sch. pecten, the former of which it much resembles in its ribbing and general aspect. The cardinal process and crura, so far as they are preserved, appear to possess the typical structure of this genus, and the whole character of the shell suggests its reference to Schuchertella rather than to Strophomena, though its occurrence on an Ordovician horizon is unexpected. Schuchertella pertinax, sp. nov. (Plate XIX, fig. 23; Plate XX, figs. 1-3.) 1883. Strophomena pecten (Linné), Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 194, pl. xv, fig. 15 (non Str. pecten, Davidson, zbid., vol. il, p. 304, pl. xliii, figs. 1-11). Shell semielliptical to subquadrate, large, flattened, compressed ; hinge-line equal to or rather less than maximum width; cardinal angles rectangular or slightly obtuse. Pedicle-valve flatter than brachial valve or very slightly concave, except near beak, where it is gently convex; beak small, pointed, not projecting above hinge-line’; hinge-area narrow, steeply inclined, longitudinally striated, with small triangular convex deltidium and inner “secondary area.” * Interior of pedicle-valve with large short stout triangular teeth, and pair of large indistinct flabelliform diductors about one-third the length of the shell; genital areas coarsely and radially pustulate. Brachial valve very gently convex, flattened ; beak small, inconspicuous ; hinge-area linear, narrower than in opposite valve. Interior of brachial valve, with cardinal process low, sessile, broad, stout, fused with crural plates, arising at right angles to floor of valve, divided into a pair of low rounded subparallel oval lobes closely placed, highest in front, facing posteriorly, grooved (?); crural plates short, strong, slightly curved, making angle of 30°-40° with hinge-line ; low weak broad median ridge running forward from base of, cardinal process between very faintly impressed adductor scars. Surface of shell ornamented with about 30-40 primary riblets, regular, rounded, or subangular, equidistant, equal, and straight, except those on lateral portions near hinge-line, which are very gently curved back; interspaces wide, flat, holding 2-6 much finer thread-like radii increasing in number by intercalation, and 1 or more of them becoming stronger at one-third to one-half their length, and as strong on margin as primaries, so that 70-90 strong equal equidistant riblets are found around * Tuomas, “Brit. Carb, Orthotetine ” (Mem. Geol. Surv., vol. i, pt. ii, 1910), p. 97, fig. 2. TRANS. ROY. SOC, EDIN., VOL. LI, PART IV (NO. 26). 130 908 ~ DR F. R. C. REED ON THE the edge. Riblets and interspaces crossed by very fine inconspicuous concentric striation. Substance of shell finely punctate. Dimensions.— : (1) (2) (3) (4) (Davrpson’s type.) Length ; ; ; = pede 39° 21 33 mm. Width ‘ : ae 51 OE 45 mm. Horizon.—Saugh Hill Genin: Localities.—W oodland Point, Newlands. Remarks.—Davipson referred these shells to “ Strophomena” pecten (Linn.),* but they differ from this species firstly and principally by their distinct. ornamentati and secondly by the more rectangular shape of the cardinal angles, shorter hing line, and absence of strong concentric lineation. The species seems more to resemb! Sch. tenuis (Hall) + of the Niagara Group. Genus TripLecta, Hall. Triplecia crargensis, sp. Nov. (Plate XX, figs. 4, 5.) 1883. Rhynchonella balcletchiensis, Davidson (pars), Mon, Brit. Foss. Brach., vol. v, Silur. Ss p. 200, pl. xi, figs. 23a-d. Non 1883. Rhynchonella balcletchiensis, Davidson, tbid., p. 160, pl. x, figs. 15, 16. Shell transverse, widest across middle, broader than long, subellipiicalll strong angular median fold on brachial valve and corresponding sinus on opp valve, and deeply sinuated anterior edge, usually emarginate; hinge-line str short, about one-third width of shell. , Pedicle-valve gently convex, shallow, with angular median sinus arising at abou “half the length of valve and rapidly increasing in depth and width to front m edges of sinus sharply angular; lateral lobes flattened; beak small, acute, above hinge-line, with apical foramen ; hinge-area small, triangular, slightly co steeply inclined, with median triangular delthyrium. Brachial valve swollen, carinat with strong median fold rising a little in front of beak and rapidly becoming: sul angular or angular, rising high above lateral lobes on front margin, and s marked off from them ; lateral lobes gently swollen, dependent ; beak small, rou incurved, lower than that of opposite. valve. Surface of lateral lobes of brachial valve marked with about 15 fine faint equidistant thread-like radial lines, sl curved hack on each side; fold of brachial valve and whole of pedicle-valve cove with very fine concentric lineation. Interior unknown. . Dimensions.— (1) (2) Length . ; Soli 11 mm, Width : f : eae ks) 14 mm. * DAVIDSON, op, cit., vol. iii, p. 304, 61 xliii, figs. 1-11. t+ Hatt, 11th Ann. Bon State Geol, Indiana, 1882, p. 287, pl. xxiii, figs. 11-13, ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 909 Horizons.—(1) Stinchar Limestone Group; (2) Balclatchie Group (conglomerate) ? Localities.—(1) Minuntion, Craighead ; (2) Balclatchie. Remarks.—It is difficult to understand the reasons which led Davinson to include this well-marked shell in his Rhynchonella balclatchiensis (see p. 923), for it is more transverse and more convex, the fold and sinus are much stronger and more angular; there is no short groove in the middle of the anterior end of the fold, and there are more or less distinct radial lines on the lateral lobes. A comparison with Davipson’s type-specimens of Rh. [Camarella] balclatchiensis proves that their association is inadmissible. Our shell is more like J. extans, Emmons,* and it also resembles Mimulus contrarius, Barrande,t and M. waldron- ensis, Miller and Dyer,{ but the two latter occur in the Silurian. It differs from T. nucleoides, sp. nov. (see below), of the Stinchar Limestone, in its more transverse shape, flatter pedicle-valve, and the more angular fold on the brachial valve, as well as in its ornamentation. In general shape it bears a superficial resemblance to Pugnax acuminata (Martin) of the Carboniferous. Triplecoa insularis (Kichwald). 1842. Terebratula insularis, Eichwald, Urwelt. Russl., vol. ii, p. 49, pl. ii, figs. 6a-c. 1870. Orthis insularis, Eichwald, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vil, p. 273, pl. xxxvil, figs. 8-15. Davinson does not record this species from Girvan, but there are good examples from the Starfish Bed and Whitehouse Beds, showing the characteristic external features and the peculiar muscle-scars, etc., in internal casts. Some of these from Shalloch Mill were labelled by Davipson Porambonites intercedens, Pander, with a “query; but they neither show the external ornamentation of that species nor its other characters. The form of 77. insularis occurring in these Girvan localities is suboval in shape and rarely transverse, and those from the Starfish Bed especially resemble the specimen from the Chair of Kildare figured by Davipson (op. cit., iii pl. xxxvii, figs. 8, 12, 138, 14) and the shell named Orthis galea, M‘Coy,§ from the same locality and horizon. Horizons.—(1) Whitehouse Group ; (2) Drummuck Group (Starfish Bed). Localities—(1) Shalloch Mill; (2) Thraive Glen. Triplecia ? nucleoides, sp. nov. (Plate XX, figs. 6, 7.) Shell subtriangular, rounded to subcireular, widest a little in front of middle, strongly biconvex, inflated; anterior margin strongly sinuated; hinge-line about - one-third the width of shell. Pedicle-valve with wide rounded sinus arising at half the length of valve, rapidly widening to more than one-third the width of valve on * Hatiand CLARKE, op. cit., Brach., i, p. 269, pl. xic, figs. 1-7. _ + Barranve, Syst. Silur. Bohéme, vol v, pl. ix, figs. via, E. _ ~ Hat and Crarz, op. crt., pl. xic, figs. 23-28. § MCoy, Syn. Silur. Foss. Irel., p. 30, pl, iii, fig. 12. SiG. =, DR F. R. C. REED ON THE front margin; lateral portions high, swollen, subangular; beak small, pointed scarcely incurved, with small triangular hinge-area below it. Brachial valve in- flated, swollen, more convex than opposite valve, with broad rounded fold arisin; about half its length, and rising highest at front margin; lateral lobes slig swollen, dependent ; beak swollen, rounded, strongly incurved; interior with pairs of subcircular muscle-scars, and median septum nearly half the leneth of valve. Surface with regular close fine concentric lines (with traces of minute p between them ?). 2 Dimensions.—Length, 10 mm. ; width, 11 mm. Horizon.—Stinchar Limestone Group. Locality. Craighead. Remarks.—Some of these small shells were labelled by Davinson “ Tripl maccoyana, Dav.,” but they are quite distinct from that species, the type of whi came from Brynbedwog Quarry near Bala,* and Davipson does not record it Girvan in his list of species.t The nearest ally of our species appears to be nucleus, Hall,{ from the Trenton Limestone. Jr. poloi (Martelli)§ from the Orc vician of China seems to be allied. Marrexii|| put the latter species, which is 1 attributed by WELLER to Triplecia, in the genus Porambonites, and the occurren small pits on the Girvan shell would also suggest this generic reference. Triplecia woodlandensis, sp. nov. (Plate XX, figs. 8-16.) 1883. Pentamerus undatus, Sowerby, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. 8 pl. ix, figs. 16-18 (non figs. ]0-15, 19-20, nec Pentamerus undatus, auctt.). Non 1839. Atrypa undata, Sowerby, in Murchison’s Silur. Syst., p. 637, pl. xxi, fig. 2. Shell subquadrate, rounded to transversely elliptical; anterior margin subtru or lobate; hinge-line less than width of shell; cardinal angles rounded, o Pedicle-valve flattened on lateral portions, slightly swollen near beak, middle occupied by broad concave sinus, rapidly widening anteriorly, with straight angular margins, and concave floor produced as rounded tongue; beak small, slightly incurved, scarcely rising above hinge-line, with minute apical for hinge-area narrow, at right angles to valve, with wide triangular delthyrium ; tor muscle-scars separate, flabelliform, strongly divergent. Brachial valve ¢ swollen, with beak much inflated and incurved, rising high above hinge-lin nearly touching beak of opposite valve ; well-defined broad median rounded f surface increasing anteriorly in height and width and rising steeply from lobes ; muscle-scars very faint. * Davipson, op. cit., vol. iii, p. 199, pl. xxiv, fig. 29. + Davrpson, op. cit., vol. v, Silur. Suppl., pp. 224, 225. : t Haun, Palxont. N.Y., i, 1847, p. 188, pl. 33, fig. 2; Hann and CiarKg, op. cit., Brach., i, pl. xic, figs. 8, 9. § Weer, Research in China, vol. iii, 1913, p. 281, pl. 25, figs. 7-9. || Marrenwt, Boll. dell. Soc. Geol. Ital., vol. xx, 1901, p. 304, pl. iv, figs. 13-16. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 911 Surface of shell smooth, except for a few widely spaced fine concentric growth- ridges near margins. Dimensions.— (1) (2) (3) Length : ‘ : ana: 15) 25 mm. Width : ; ; aro) 2, 32 mm. Horizon.—Saugh Hill Group. Locality.— Woodland Point. Remarks.—tThe affinities of this shell are with 7. wenlockiensis, Davidson,* and T. insularis, Kichw.,+ but it differs from the former by the better defined fold on the brachial valve, and from the latter by the sharp edges to the sinus on the pedicle- valve. It bears a closer resemblance in these respects to the Clinton species T. Ortoni, Hall,{ but is more transverse and less quadrate. Davipson’s figure § of a Llandovery shell referred to T. insularis from Minsterley bears more resemblance to our Girvan species, and it seems indistinguishable from the Kashmir shell referred to T. insularis by the author.|| It is difficult to understand how Davinson confused some of the better specimens belonging to this new species with the true Penta- merus undatus (Sowerby), which also occurs at Woodland Point associated with it, though crushed and fragmentary shells are not easily distinguished. Genus Mimu.vs, Barrande. Mimulus? incertus (Davidson). (Plate XX, figs. 17-23.) 1870. Atrypa incerta, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vil, p. 203, pl. xxiv, fig. 30; pl. xxv, figs. 7, 8. 1883. Triplesia incerta, Davidson, zbid., vol. v, Stlur. Suppl., p. 145, pl. viii, figs. 24-28, ? 29. In this species the convexity and fold belong to the pedicle (ventral) valve, and the interior of this valve was figured by Davipson (op. cit., pl. viii, fig. 28) and correctly described as such in the text, but referred to as a “dorsal” valve in the explanation of the plate. The fold is often undifferentiated from the general swollen subcarinate rounded surface or is only developed near the front margin; the beak is very high, acutely pointed, and more or less incurved, with the apex perforated by a small foramen ; the hinge-area is triangular and very high, and the wide flattened — deltidium seems to have a narrow median slit or impressed line, if Davripson rightly interprets the markings on the hinge-area. Internally there seem to be small teeth | in the pedicle-valve, but the interior of the beak is apparently filled up by a pair of thick plates completely fused together except in front, where they form thick short * Davison, op. cit., vol. v, Silur. Suppl., p. 144, ple viii, fig. 23; Hatt and Cuarxe, Pal. N.Y., Brach., i, pl. xic, fig. 22. + Davipson, op. cit., vol. iii, p. 273, pl. xxxvii, figs, 8-15. { Hatt and Crarkg®, op. cit., pl. xic, figs. 12-20. § Davinson, op. cit., vol. v, Silur. Suppl., p. 134, pl. viii, fig. 18. || Ruxp, Rec. Geol. Surv. India, vol. xlii, pt. i, 1912, p. 27, pl. ix, fig. 14. ae _ DR F. R. CG. REED ON THE free triangular projections ; these Davipson called the dental plates, but it is doubtful if they are homologous with true dental plates, as they seem to be within the cavity of the beak and away from the teeth. These plates project into the base of the short flabelliform muscle-scars (?diductors) which are in contact in the middle line and form a triangular transverse scar about one-third the length of the valve ; the scars are separated by a narrow groove, and each scar is also longitudinally divided into: two subequal portions. A pair of narrow short obscure adductors (?) lies immediately in front of them. The brachial valve has a low inconspicuous beak, and is marked anteriorly (or fo most of its length) by a wide indefinite shallow sinus, having a broad rounded anterior tongue bending up into the marginal fold on the opposite valve. The interior of the ‘esata valve is unknown. With these corrections and additions, Davinson’s ; description holds good. This species may be compared with Rhynchonella apiculata, Salter and Forbes MS.,* from the Chair of Kildare. The fold being in the pedicle-valve and the sinus in the brachial recalls Mimulus perversus, Barrande,t the type of the genus Mimulus, rather than any species of Triplecia, and it seems more probably / to belong to Mimulus than to any other genus. Horses: —(1) Penkill Group ; (2) Camregan Group; (3) Saugh Hill Group? Localitves.—Penkill, Sia lae Glen; (3) Bargany Pond Burn; (3) Wood- land Point. | Genus Srrepris, Davidson. Streptis Horner, sp. nov. (Plate XX, figs. 24, 25.) 1883. Triplesia monilifera, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., Ps uy, pl. vii, fig. 31; p. 225. Non 1870. Triplesia monilifera, Davidson, op. cit., vol. iti, p. 200, pl. xxv, figs. 3-5. “ei Al n ie Shell small, transversely elliptical, symmetrical, biconvex ; hinge-line less tha maximum width of shell; cardinal angles rounded. Pedicle-valve moderately convex, with broad Malloy median sinus widening anteriorly, with rather sha edges defining it laterally ; beak small, acute; hinge-area triangular. Brachial valve - more convex, with regular median rounded fold extending from beak to margin, widening anteriorly ; beak rounded, incurved, inconspicuous. Interior of pedicle valve with pair of thin divergent dental plates following the marginal ridges of sinus and extending about one-third the length of valve. Surface of shell mar with 8-9concentric sharp lamellae, having scent frilled edges, not nodulated. interspaces with fine concentric lines. * Davipson, op, cit., vol. iii, p. 202, pl. XXV, fig. 6. + BaRRANDE, Syst. Se Bohéme, v, pl. i, fig. 3,C, E; Hann and CrarKe, op. cit., Brach., i, p. 272, P xi figs, 31, 32. > - - ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 9138 -Dimensions.—Length, 6°5 mm. ; width, 8°5 mm. - Horizon.—Saugh Hill Group. I ocality.—W oodland Point. Remarks.—This species differs from Str. monilifera of the Chair of Kildare tone in the frilled instead of nodulated edges of the concentric lamellee and ‘more sharply bounded.sinus and better defined fold. From Str. Grayi, * of the Wenlock Limestone it is distinguished by its symmetry and by the e of the fold on the brachial valve. It is with pleasure I dedicate this s; to Dr Jonn Horne, F.R.S. Genus Cuiirtonia, Foerste. Cliftoma Grayex (Davidson, emend.). (Plate XX, figs. 26, 27.) Triplesia Grayiz, Davidson, Mon, Brit. Foss. Brach., vol. iii, p. 198, pl. xxiv, figs. 31, 32, (tpl. xxv, figs. 9, 10, 11). _‘Triplesia? Grayi, Davidson, op. cit., vol. v, Salur. Suppl., p. 144, pl. viii, fig. 32. description of this species was given by Davipson in a sufficiently complete , but his figures make the length of the hinge-line too great, the prominence larity of the cardinal angles too marked, and the strength of the radial notation somewhat exaggerated. There seem to be also minute pits between dial lines, a feature which he «lid not mention. There is a small foramen at the the beak, as he suspected ; but the deltidial plate covering the delthyrium is oad nor elevated along its middle, as his figures indicate, being high, narrow, le, as in typical examples of Triplecia and its congener Cliftonia, which has blished by Forrste jf to contain the radially striated members which Hany ARKE} doubtfully retained in Triplecia. The original type-specimens of rany ze described by Davripson came from Craighead (op. cvt., pl. xxiv, figs. 31, ), and other examples were stated by him to occur in a green breccia in Bare ale Glen, but the latter statement | am unable to substantiate. zon.—Stinchar Limestone Group. wlity.—Craighead. a Cliftonia Andersoni, sp. nov. (Plate XXI, figs. 1-4, 75.) . Triplesia’ spiriferoides, M‘Coy, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 146, pl. viii, fig. 30 (non Strophomena spiriferoides, M‘Coy, Ann. Nat. Hist., vol. viii, 1851, p. 402). 83. Orthis biforata, Schlotheim, Davidson (pars), zbid., p. 181. . Rhynchonella nasuta, M‘Coy, Davidson (pars), ibid., p. 160, pl. x, tig. 21. ell transversely elliptical to subquadrate, biconvex, more or less inflated ; line less than width of shell; cardinal angles rounded. Pedicle-valve slightly * DAVIDSON, op. cit., p. 141, pl. xiii, figs. 14-22. + Eastman-ZittTe., Text-book of Palxont. (2nd edit. 1913), p. 387. { Haut and Cxiarkp, op. cit., Brach., i, p. 271. 914 DR F. R. C. REED ON THE swollen near beak, with strong median sinus, having flat floor and steep sides, increasing in width to margin and ending in rounded short obtuse tongue mor less strongly bent up to meet fold in opposite valve; beak rather swollen, ri above hinge-line, incurved; hinge-area small, steeply inclined, triangular ; lat lobes of valve somewhat swollen; teeth small, supported on thin divergent dental plates; muscle-scars very faint. Brachial valve swollen, slightly emarginate in front, with abruptly elevated rounded median fold increasing in width to margin and having steep sides and flattened top; lateral lobes rounded, dependent; beak swollen, less high but more incurved than that of opposite valve ; interior of valve with stout cardinal process (base only known) and pair of short rod-like crura; two pairs of faint subcircular large adductor scars, the posterior ones better marked a1 separated by narrow sharp transverse curved ridges from anterior pair. Surface of valves ribbed, having 6-7 low subequal or equal subangular or rounded closely placec ribs both on the fold and in the sinus, and 8-13 rather larger simple similar ribs of uniform size on each lateral lobe slightly curved back ; occasionally the ribs on fold and in sinus increase in number by division. Dimensions.—- (1) (2) Leneth : ae ee BS 13 mm. Width ~ . 4 21 17 mm. forizons.—(1) Stinchar mneskene om (2) Brleleeiae Group. Localities. —(1) Craighead ; (2) Ardmillan, Dow Hill, Balclatchie. Remarks.—This shell is seen by its internal and external characters to belong t the genus Cliftonia, Foerste, and it is allied to Triplecia plicata, Wiman,* from Lepteena Limestone, and 7. Ulrichi, Winch. and Schuch.,t from the Lorraine Gr It is not always easy to distinguish specimens of it from Orthis (Platystrop. biforata, but the ribs in the sinus are always more numerous and smaller tha: that species. The Balclatchie examples usually have fewer ribs on the fold on the lateral lobes than the Craighead specimens, and the outermost ribs on fold are sometimes smaller than the others. We name it in honour of Mr THOMAS ANDERSON, who was a well-known collector of fossils in Girvan. a Cliftoma spiriferoides (M‘Coy). 1851. Strophomena spiriferoides, M‘Coy, Ann. Mag. Nat. Hist., p. 402. r 1870. Orthis spiriferoides (M‘Coy), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 275, pl. 3 figs. 3-7. 1883. ee spwriferoides (M‘Coy), Davidson (pars), 2bid., p. 224. Non 1883. Triplecia spiriferoides (M‘Coy), Davidson, ibid., vol. v, Silur. Suppl., p. 146, pl. viii, fig. 3 This is a rare species in the Girvan area, but two well-preserved internal and two external impressions occur in Mrs Gray’s collection from the Starfish Davipson records it from Thraive in his list of Girvan brachiopods (op. cit., p. 224) * Wiman, Arkiv f. Zool. Stockholm, Bd. iii, No. 24, 1907, p. 12, t. ii, figs. 13-17. + WincHELL and Scaucuurt, Paleont. Minnesota, iii, 1893, p. 409, text-figs. 344—n. = ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 915 but it does not occur at Ardmillan also, as he stated, for the shell thus quoted belongs to the other species here described under the name Cl. Andersoni. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. we Choneles Genus Cuoneres, Fischer de Waldheim. Chonetes advena, sp. nov. (Plate XXI, figs. 6-11.) Shell concavo-convex, transversely semielliptical, widest along hinge-line, twice e than twice) as wide as long; cardinal angles rectangular. Pedicle-valve and quite uniformly convex; beak small, acute, scarcely incurved, not ent nor swollen, but rising above hinge-line ; hinge-area triangular, narrow, | several (4-5) oblique tubules on each side of beak running inwards through ance of shell from upper‘edge of hinge-area; interior of valve with pair of sub- ar pointed muscle-scars, in contact posteriorly but strongly diverging at 20° anteriorly, extending about one-third the length of valve; each scar y and unequally divided lengthwise by weak median ridge, and separated tly by stronger shorter ridge. At the anterior points of the muscle-scars there ort wide vascular trunks, each of which divides into two divergent smaller , splitting up into numerous smaller vessels which run forwards subparallel 2 margin; the rest of the interior is covered with elongated radial lines of ustules, most developed near the margins. Brachial valve with broad widely nt subtriangular low crural plates (ending in fine points), fused at base so as transverse cardinal plate, deeply and widely cleft in middle ; cardinal process ll, of three (or four) very short thin lobes; muscle-scars large, very faintly d; pair of thick septa slightly divergent, extending forwards for more than ne leneth of valve from base of cardinal process, dying out anteriorly and ntly dividing the contiguous large oval faint muscle-scars into outer and inner ; rest of interior covered with radial lines of elongated pustules. ce of shell ornamented with fine close thread-like radiating lines of which urth or fifth one is slightly stronger than the others. ensions.— Width, 14-18 mm.; length, 7-9 mm. zon.—Drummuck Group (Starfish Bed). cality.—Thraive Glen. marks.—The peculiar tubules through the substance of the shell in the cardinal are characteristic of the group of shells which includes Chonetes, Chonostrophia, ypectus, Chonetina, and Chonetella, and show in casts of this species as small but they do not seem to correspond with external spines, as there are no of the latter in any of the Girvan specimens. It is doubtful if the so-called r ridges in the brachial valve are homologous with the true “ brachial ridges” TRANS, ROY, SOC. EDIN., VOL. LI, PART IV (NO. 26). 131 916 DR F. R. C. REED ON THE of Chonetes proper, and the muscle-scars of the pedicle-valve are more like those o' ( Plectambonites. It seems probable that at least a subgenus should be set up for 1 reception, and the name Hochonetes may be suggested. According to a the earliest known appearance of the genus Chonetes is in zone 3 of the Charleton Formation, Richmondian Series (=Upper Ordovician) of Anticosti Island,t but the species there represented is referred to the subgenus Hodevonaria, See Professor O. T. Jones has attached a MS. name to some examples of this species i Mrs Gray’s collection, but his description has not been published. Genus CLITAMBONITES, Pander. Clitambonites [ Orthisina] ascendens (Pander). (Plate XXI, figs. 12-18.) 1830. Clitambonites (Pronites) ascendens, ete., Pander, Beitr. Ore Russ, Reich., p. 72, pl. ; vii, figs. 2-6. 1870. Orthisina ascendens, Pander, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 278, pl. xi 3 figs. 27-29. ; 1883. Skenidium Grayix, Davidson, ibid., vol. v, Silur. Suppl., p. 175, pl. xi, figs. 3, 4, 5. 1883. Orthisina ascendens, Pander, Davidson, ibid., p. 175, pl. xvi, figs. 16-18. It is curious that while Davipson described and figured Orthisina ascen from Welsh localities,t he did not recognise it in the Girvan area, where it is abundant on one horizon, but he described some imperfect specimens of it as a distinct speci which he referred to the genus Scenidium as Sc. Grayix. These specimens, the numerous other better ones from the Starfish Bed which are now available, show all the typical external and internal characters of the genus Clitambonites § and of t species Cl. ascendens as interpreted by Davipson. The average length is 20 mm. — Horizon.—Drummuck Group (Starfish Bed). a Locality.—Thraive Glen. Clitambonites complectens (Wiman), var. nov. albida. (Plate XXI, figs. 19, 20.) Pedicle-valve very deep, conical, sharply pointed at apex, with rather broad rounded median sinus on surface widening to anterior margin; cardinal an acute ; beak small, pointed, not incurved, forming apex of cone. Hinge-area large, triangular, nearly at right angles to plane of valve, with wide triangu delthyrium nearly one-third the width of valve, having its lower portion coveret convex pseudodeltidium, leaving large oval foramen in upper half; edges of fora elevated to form basal tube for stalk. Surface of valve covered with regular strong * TwENHOFEL, Canadian Geol. Surv. Museum, Bull. No. 3, 1914, p. 26. + ScnHucHERT and TwenHoFEL, Bull. Geol. Soc. America, vol. xxi, 1910, p. 697. { Davinson, op. ett., iii, p. 278, pl. xlix, figs. 27-29 ; ibid., Suppl., p. 175, pl. xvi, figs. 16-18, § Hatt and Crarkz, Brach., i, p. 233, pl. vii, figs. 23, 24, 2 ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 917 equidistant thick raised rounded concentric lines; lateral margins of sinus marked one fine strong straight rib, with traces of 2-3 very faint smaller ribs on lateral slopes of valve. ‘Dimensions.—Length, 3°25 mm.; width, 6°00 mm.; depth at beak, 4°00 mm. Horizon.—Whitehouse Group. Locality.—Shalloch Mill. Remarks.—One of the two small specimens of this peculiar little shell from the above description is drawn up was found mounted on the same tablet with vipson’s types of Cl. shallochensis described below. It is certainly quite distinct that species, and is closely similar to Cl. [Orthisina] complectens, Wiman,* a the Leptena Limestone of Sweden. From the latter it differs only in having radial lines scarcely developed at all, and the cancellation of the surface therefore so noticeable. Clitambonites shallochensis (Davidson, emend.). (Plate XXI, figs. 21-23.) 1 883. Skenidium Shallockiense, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 174, pl. x1, figs. 6, 7. Shell transversely semielliptical to semicircular; hinge-line straight, equal to m width of shell; cardinal angles usually less than a right angle and more s acute and subalate. Pedicle-valve very deep, conical, with a broad rounded w median sinus; beak forming apex of cone, small, pointed, not incurved ; rea very large, Puen cular at right angles to plane of valve, with large wide ar delthyrium covered in lower part by convex plate leaving large oval foramen ~ beak. Brachial valve very gently convex, more or less flattened, with median S bilobing it; beak small, inconspicuous, not elevated ; hinge-area very narrow, broad Berenlar delthyrium. Surface of valves covered with 40-50 low rounded ly placed subequal ribs curving back slightly near cardinal line, and sometimes ‘shorter rather smaller ribs intercalated at about one-fourth their length, all by close concentric rather coarse striz. Hi eons. aaa (1) (Davipson’s fig. 6.) (2) (Davinson’s fig. 7.) (3) Sie ath MS a ori oc 0 Peles 10°5 mm. Mitch -. ane 8°00 17°5 mm. a Depth of pedicle- Eyalve at beak . 6°5 6°00 90 mm. forizon.—W hitehouse Group. Locality.—Shalloch Mill. Remarks.—Davinson’s description was so inadequate that a fresh diagnosis of the species based on the type and on the other specimens now available is necessary. He Vas Ely acquainted with three pedicle-valves, so that his description was unavoidably * Wiman, Arkiv f. Zool. (Stockholm), Bd. iii, No. 24, 1907, p. 11, t. i, figs. 1-4. 918 DR F. R. C. REED ON THE incomplete. The true generic position of this shell is certainly in Cltambonite: although we are still ignorant of the internal characters. , ; The sharply conical shape of the pedicle-valve, the median sinus on its surf instead of a carina, the inclination of the hinge-area and the wider delthyrium, more numerous and less regular ribbing distinguish it from Se. Greenough (s p. 920) from the same horizon and locality. Genus Hemrpronites, Pander. Hemipromtes carrickensis, sp. nov. (Plate XXI, figs. 24-26.) 1883, Strophomena retroflexa, Salter, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Stlur. Sup pl, p. 195, pl. xv, figs. 7-8. 2 Non 1865. Strophomena? retroflexa, Salter MS., Cat. Foss. Mus. Pract. Geol., p: 8. Shell semielliptical to subquadrate, unequally biconvex; hinge-line rather | than maximum width; cardinal angles subrectangular to obtuse. Pedicle-valve deepest at beak, conical, with somewhat flattened surface; beak forming ape: cone, not incurved, the margins of hinge-area meeting at beak at very obtuse (about 130°-140°); hinge-area large, flat, triangular, steeply inclined to plane valve, with narrow median delthyrium. Brachial valve gently convex, with median depression usually present ; beak small, inconspicuous, not rising above h line ; hinge-area narrow, steeply inclined. : or fourth line rather stronger than the others, and all arched back slightly on la a portions of valves; 3-4 strong concentric lamellose growth-ridges usually pres near margin, and fine concentric striation over rest of surface. Dimensions.— Pedicle-valve. Brachial valve. (1) (2) (3) (4) Length. . 5 N60) 160 mam: 12°0 ~125 mm Width = ‘ . 20°0 16°5 mm. Tye) 17°0 mm. Horizons.—(1) Stinchar Limestone Group; (2) Balclatchie Group. _ Localities.—(1) Craighead ; (2) Balclatchie (conglomerate). Remarks.—The external characters of this shell, in spite of our ignorance } internal structure, are such as to make its reference to the genus a by PaHLen,t and to H. Giraldi, Martelli.t The type of “Strophomena” retrofleca, Salter, to which species Dav * Haut and CLARK, op. cit., Brach., i, p. 238. ~ + Panten, Mem. Acad. Tou Se. st Petersburg, ser. vii, vol. xxiv, 1877, No. 8, p. 28, t. iii, figs. 1, 2 (cet Lamansky, Mém. Vom, Géol. St Pétersbourg (N.S.), xx, 1905, pp. 73, 176. { Marvextt, Boll. Soc. Geol. Ital., vol. xx, 1901, p. 305, pl. iv, figs. 7-12. § Davrpsoy, op. cit., vol. iii, p. 298, pl. xlii, figs, 15-17. _ ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 919 tt ‘ibuted these Craighead shells, came from Kildare, and is certainly distinct in racter, having mucronate wings and different ribbing; and the Ardmillan which he also referred to this species, belong to the genus Leptella or mbonites, and are described in this memoir as L. pseudoretroflewa (see p. 874). flattened conical shape of the pedicle-valve, its obtusely pointed and lower surved beak, the flat (instead of concave) and more inclined hinge-area, and s reeular and rather coarser riblets seem sufficient to distinguish this species — SE Henupromtes Thomsona, sp. nov. (Plate XXI, figs. 27, 28.) 3. Orthis turgida, M‘Coy, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 187, pi. xiv, fig. 18 (non figs. 17, 19, 20). j _ Orthis turgida, M‘Coy, Davidson, ibéd., vol. iii, pt. vii, p. 258, pl. xxxii, figs, 12-20. 2. Orthis turgida, M‘Coy, Syn. Brit. Pal. Foss. Woodw. Mus., p. 229, pl. in, fig. 20. Shell subcircular to subelliptical, strongly biconvex ; hinge-line rather less than mum width ; cardinal angles obtuse or rounded. Pedicle-valve convex, rounded, ‘inmate, much longer than brachial valve, with high elevated pointed beak, slightly ,and with cardinal slopes meeting at 80°-100° at beak ; hinge-area large, high, r, lying nearly in plane of valve or slightly sloping back, concave, with rz open delthyrium. Brachial valve transversely elliptical, with traces of a | median depression generally present ; beak small, inconspicuous, obtuse ; hinge- "a steeply inclined. Surface of valves covered with very numerous closely t 150 on edge of valves, gently curved back on lateral Ratiene of shell, and ed by fine concentric striz, with 2-3 lamellose growth-ridges near margins. ior unknown. Pedicle-valve. Brachial valve. 2 (1) (2) (1) (2) Sueneth . , a Wo 4 F116 Ol mm, 1°62") 13°6 mim, Jidth | ee NEO), = 1820 mm. 1650. 9 48"°0 ‘mim! rizon.—Stinchar Limestone Group. ocality. —Craighead. narks.—This shell does not seem identical with M‘Coy’s Orthis turgida* from deilo of Llandeilo; the ribbing is different, and M‘Coy’s type, which is a 1 valve, is more convex and has a more marked rounded sinus. M‘Coy’s and . i are restorations, and inaccurate. The other pase s shell, ich much resemble H. injflexa, Pander,t in shape, ribbing, high * M‘Coy, Syn. Brit. Palzoz. Foss. Woodw. Mus., 1852, p. 229, pl. in, fig. 20 (? 21-24). _ + Von Pauten, Mém. Acad. Imper. Sc. St Pétersbourg, ser. vii, vol. xxiv, 1877, p. 28, pl. iui, figs. 1-5. 920 DR F. R. C. REED ON THE hinge-area to the pedicle-valve, and concentric lamellose growth-ridges. This species is named after Professor WyviLLe THomson. =. 1G Hemiprontes Thomsont, var. nov. nconstans. (Plate XXI, figs. 29-31.) 1883. Orthis elegantula, Dalman, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl, pp. 178, 226. Shell subcircular, subquadrate, or transversely elliptical, unequally biconmemm hinge-line rather less than maximum width of shell; cardinal angles obi e, rounded or subrectangular. Pedicle-valve convex, subcarinate in young, deepest near middle; beak large, elevated, rising high above hinge-line, acute, slightly incurved ; nee area large, high, triangular, slightly concave, nearly half the length of shell in height, lying nearly in plane of valve, with narrow triangular delthyrium Brachial valve much shorter and less convex than pedicle-valve, transversely elliptical in shape, more or less bilobed by a sharp median depression which extends from beak to front margin, increasing in width and causing sinuation of edge of valves; beak small, obtuse, rising slightly above hinge-line, slightly incurved, with narrow hinge-area below, set at right angles to that of other valve. = Surface of valves covered by numerous small rounded somewhat nhegimel close: placed riblets curving back a little on each side and numbering about 60 on - margins, about one-third of which are primaries, the others (usually rather smal arising somewhat irregularly by intercalation at about one-third or one-half their length. Concentric growth-ridges usually few, lamellose, strong, submarginal. Dimensions.—Average leneth, 9-11 mm. ; thickness, 6-8 mm. Horizons.—(1) Stinchar Limestone Grows ; (2) ? Balclatchie Group. Localities. (1) Craighead ; (2) Balelatchie (conglomerate). Remarks.—This small shell, of which only a few specimens from Craighea positively known, appears to belong to the group of Hemipronites | Orthisina] infl Pander.* The internal characters are unknown. It differs from H. carrickensis the strongly marked bilobation of the brachial valve, by its sinuated anterior ma. by the decided carination of the pedicle-valve, and by its relatively higher The ornamentation is quite unlike that of Scenidiwm Lewist, var. crargense, thor the somewhat similar shape of the shell might suggest a comparison. The semblance to H. Thomsoni is, on the other hand, so close that we must regard a variety of this species, distinguished only by its marked carination and bilobation. My J Genus Scenipium, Hall. Scenidium Greenought, sp. nov. = (Plate XXI, figs. 32-34.) aa Shell semielliptical to subtriangular, widest along hinge-line; cardinal angl less than right angles. Pedicle-valve convex, deep, swollen, subcarinate ; beak high, * LaMANSKY, Mém. Com, Géol. St Pétersbourg (N.S.), vol. xx, 1905, pp. 73, 176. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 921 pointed, small, slightly incurved, not swollen; hinge-area large, high, triangular, neave or flat, steeply inclined, with large triangular delthyrium. Brachial valve ow, very slightly convex or flattened, deeply bilobed by rounded median sinus, lateral portions of valve somewhat swollen; beak small, inconspicuous; hinge-area ular, much lower and narrower than in opposite valve, steeply inclined. Surface es covered with 24-30 simple rounded equal closely placed low ribs, with those s of brachial valve rather smaller than rest ; fine concentric striation over all. Dimensions.—Length, 8-9 mm. ; width, 10-11 mm. Horizon.—Whitehouse Group. i Locality. —Shalloch Mill. marks.—lt is impossible to regard this shell as merely the young of Clitam- s [Scenidiwm] shallochensis, Davidson, as might at first sight be suspected, subecarinate contour of the pedicle-valve and the absence of the median sinus it are distinctive. Further differences are mentioned in the description of that ies (see p. 917). Sc. Greenoughi more resembles Sc. Lewisi, var. asteroidea, from ve Glen (9.v.), and Sc. Halli, Safford,* from the Ordovician of the United States. species is named in honour of J. B. GREENOUGH, one of the founders of the gical Society of London. Scenidium Lewisi, Davidson, var. nov. asteroidea. i: (Plate XXII, figs. 1-3.) There are a few specimens of a Scensdium from the Starfish Bed in Mrs Gray’s stion which must be regarded as a variety of Sc. Lewis; though not precisely ing with any of the other varieties here described. The shell is less sharply .and the pedicle-valve is less carinate and more rounded than in the variety yense, and agrees better with var. woodlandense. The ribs also are coarser and ar than in the Craighead form. There are about 20 simple equal angular primary ith only 1 or 2 short marginal intercalated ones near the sinus in the brachial this regularity and simplicity of the ribbing being a noticeable feature ; and all ribs are crossed by strong imbricating striz. The internal casts, which are well sserved, show a spondylium in both valves and in front of it a short stout low lian septum in the pedicle-valve about one-fourth (or less) its length, while the in the brachial valve is long and thin, being about three-fourths the length valve. In the brachial valve the muscular scars are as represented in son’s figure of the interior of Sc. Lewisi.t This new variety does not appear ye very closely allied to the variety Hughesi, Dav.{ from the Coniston Grits, ‘h largest specimen measures 8°5 mm. in width. A forizon.—Drummuck Group (Starfish Bed). Locality. —Thraive Glen. * Hatt and CraRk&gE, op. cit., Brach., i, pl. vii, figs 33-39, + Davipsoy, op. cit., vol. iii, pl. xxvi, fig. 9a. { Davipson, ibid., p, 254, pl. xxxviii, fig. 26. 922 DR F. R. C. REED ON THE Scenidium Lewisi, Davidson, var. nov. craigensis. (Plate XXII, figs. 4-9.) Shell small, transversely triangular to semielliptical, sharply folded down mid line ; hinge-line equal to width of shell; cardinal angles subacute to rectangt Pedicle-valve deeply conical, strongly carinate, with dependent lateral lobes mak a wide angle; hinge-area large, triangular, nearly at right angles to plane of val beak small, prominent, forming apex of cone, pointed, elevated, and sligh incurved. Brachial valve bilobed by deep angulated median sinus increasing width anteriorly, with gently convex lateral lobes; anterior margin more or _ emarginate in middle; beak inconspicuous; hinge-area narrow ; interior with | linear cardinal process, median septum about half the length of valve, and pair o: short converging plates uniting with septum and forming spondylium. Surface of valves covered with 30-40 simple straight subangular closely placed regular ec fine ribs; concentric strize very faint or absent. a Dimensions.—Length, 5-6 mm. ; width, 7-8 mm. Horizons.—(1) Stinchar Tenet one group; (2) Balclatchie Group. 3 Localities. —(1) Craighead ; (2) Ardmillan, Dow Tall, Balclatchie couglaa te. ~ It differs from var. asteroidewm in the greater fineness and number of the ribs a absence of imbricating striz, and from var. woodlandense in its more triangular s. | and finer regular ribs. In shape the shell much resembles Orthis melmerbiensis, Ri from the Dufton Shales. Scenidium Lewisi, Davidson, var. woodlandensis, Davidson (emend.). 1883. Scenidiwm Lewist, var. woodlandiense, Davidson, Mon. Brit. Foss, Brach., vol. v, Silur. , p. 174, pl. xi, figs. 1, 2. As Davinson remarks, this shell “ partakes of all the essential characters of t Wenlock type, differing from it in size only, being nearly twice as large.” It at a width of 13-14 mm. The interior is imperfectly known. This variety differs from the Craighead variety by its larger size and more transverse shape, by the.pedicle-valve being less strongly carinate and without any distinct m¢ fold; the beak also is higher and more incurved, the hinge-area lies nearly 1 mm | plane of the valve, and the ribs are coarser and less regular, increasing by interce tion and rarely by division. It seems more allied to the variety Hughesz, Davidsor (op. cit., p. 254, pl. xxxviii, fig. 26). al , Dimensions (type, op. cit., pl. xii, fig. 1).—Length, 6°8 mm. ; width, 8°5 mm. Horizon.—Saugh Hill Group. Localities.—W oodland Point, Newlands. * DAVIDSON, op. cit., vol. iii, p. 208, pl. xxvi, figs. 4-9. + Reep, Geol. Mag., dec. v., vol. vii, 1910, p, 296, pl. xxiii, figs. 4-8. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 923 Genus PoramponiTEs, Pander. Porambonites acutiplicata, sp. nov. (Plate XXII, figs. 10, 11.) Shell subcircular to subelliptical, biconvex, deeply folded; hinge-line less than idth of shell; cardinal angles obtuse. Pedicle-valve with very deep sharply angular us starting from near beak and rapidly deepening and widening to front margin, by sharp angular edges, and with floor produced into acutely pointed tongue, y folded along middle line; beak swollen, risimg above hinge-line, incurved so than opposite valve) ; hinge-area steeply inclined, triangular, with median ar delthyrium. Brachial valve strongly convex, swollen, and inflated, with inded fold corresponding to sinus occupying middle third of valve on anterior which is acutely emarginate; lateral lobes gently convex, not strongly off from fold; beak large, high, swollen, incurved, rising above beak of ite valve, with Botta triangular hinge-area lying in plane of valve. Surface hell ornamented with small closely set circular pits in radial lines, about ten radial to every 4 mm., and about ten pits of a row in every 3 mm. yoy Dimensions.— e : (1) (2) (Crushed specimen.) weenie eC 18 220 mm. Width . : : : felts) 24 mm. Depth . ’ 16 16 mm. orizon. —Stinchar Famiestone Group. | ocality. —Craighead. emarks.—There are only two specimens of this interesting shell in Mrs Gray’s ion, and neither show the internal characters. It is quite distinct from mbomtes wntercedens, var. filosa, M‘Coy, of Co. Waterford,* and from those ed by Davipson from Wrae Quarry, which have a broad flat sinus. In re- to the valve carrying the fold as the brachial valve I have followed Noeriine.{ Genus CaMARELLA, Billings. Camarella balclatchiensis, (Davidson). (Plate XXII, figs. 12-15.) 1883 . Rhynchonella balcletchiensis, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., a p. 160, pl. x, fig. 15 (mon fig. 16, nec pl. xi, fig. 23). 1883 Beet onehonelis ezmula, Salter, Davidson (pars), ibid., p. 160 (non pl. x, fig. 17). Shell subtriangular to subpentagonal, broadest anteriorly, rounded at sides, ually as wide as long or rather wider than long, biconvex. Pedicle-valve with dian sinus commencing at from one-third to one-half the length of the valve, asing rapidly in width to anterior margin to more than one-third the anterior * Davison, op. cit., vol. iii, p. 195, pl. xxv, fig. 16; pl. xxvi, figs. 1, 2. + Davipson, ibid., pl. xxv, figs. 17-19 ; pl. xxvi, fig. 3. - { NoEruine, Zectschr. deut. geol. Gesellsch., Bd. xxxv, 1883, p. 535. RANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 26). 132 924 DR F. R. CG. REED ON THE width of shell, shallow, rounded, with floor more or less produced into tongue ; lat margins of sinus angular, more or less strongly elevated on front margin; floo sinus with single narrow median rib of variable length and sometimes ne: obsolete ; lateral portions of valve gently convex, with traces of one or more ang short marginal plications, often obsolete, forming zigzag junctions of margin; beak small, acute, elevated, very slightly incurved, higher than that of opposite valve ; triangular delthyrium below beak. Brachial valve usually more convex than pedicle-valve, with a more or less developed fold towards anterior margin composed of a pair of rather closely placed small angular or subangular plications of varia length, becoming highest on anterior margin; lateral portions of valve more or convex and dependent, occasionally with weak marginal plications; beak obtuse, low, small. Surface of valves smooth. Interior of pedicle-valve with median septum fully one-third the length of t the valve. Interior of brachial valve with shorter median septum. Dimensions.— (1) Type. (2) (3) (4) Length Oss 9°2 12°3 = 12°2 oo Width : ‘ ~ ~L0%5 10°0 1358 L2°3) nae Thickness . ‘ Mateo: 52 © 180 9°4 mm. Horizon.—Stinchar Limestone Group. Locality.— Craighead. Remarks.—This species varies to a considerable extent in certain directions it always retains the double plication on the fold, and never has more than a si plication in the sinus. The width relative to the length, the degree of inflation convexity of the valves, the prominence and length of the plications on the fold of the one in the sinus, and the number and development of the lateral marg plications are variable. The forms from Craighead with more accentuated plicat and greater anterior width Davipson referred to Rh. xmula, Salter,* which co typically from the Chair of Kildare. The shells from the Balclatchie conglome figured by Davipson{ as Rh. balclatchiensis differ in their greater length, more elongated shape, and obsolete median plication in the sinus, and constitute a fa well-marked variety (see next page). The specimen from Minuntion { attribute him to this species is completely distinct, and is described in this memoir as Tripl ~ 6 craigensis, sp. nov. (see p. 908). The small specimen from Penkill attributed to Kh. wmula by Dane figured as such is too poorly preserved for satisfactory identification, and the f gure is Incorrect. f ‘a C. balclatchiensis is undoubtedly closely allied to C. Thomsona (Davidson), , a * Davinson, op. cit., vol. 1ii, p. 188, pl. xxiv, fig. 21. + Davinson, op. cit., vol. v, Silwr. Suppl., pl. x, fig. 16. { Davinson, op. cit., vol. v, Silur. Suppl., p. 200, pl. xi, fig. 23. § Davipson, op. cit., vol. v, Silur. Suppl., v, pl. x, fig. 17. = a ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 925 here described and restricted (see p. 927), but the former differs in the fold being _ posed of a pair of plications and in the broader outline of the shell; but the ation of young or imperfect specimens is difficult. y show the spondylium, owing to their imperfect preservation, but otherwise ¢ in all essential characters. bs _ Camarella balclatchiensis (Davidson), var. (Plate XXII, figs. 16-18.) 1883. Rhynchonella balcletchiensis, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., =a p. 160, pl. x, fig. 16 (non fig. 15), 83, Rhynchonella portlockiana, Davidson (pars), 2bid., p. 159, pl. x, fig. 12 (non figs. 13, 14). 68. Rhynchonella 1 portlockiana, Davidson, ibid., vol. iii, pt. vii, p. 189, pl. xxiv, figs. 23-25. ny. This variety is longer, and more elongated than the type; it is widest across ch dle instead of anteriorly, and is subpentagonal in shape, with longer sloping houlders which meet at the beak of the pedicle-valve at about 45°-50°. In these it differs from the Craighead type. The median plication in the sinus pedicle-valve always seems to be obsolete and the fold in the opposite valve / ungrooved, but the specimens are poorly preserved and not satisfactory, a distinctive name is not applied to them. orizon.—Balclatchie Group (conglomerate). ality.—Balclatchie. Camarella? Conybearet, sp. nov. (Plate XXII, figs. 19-24.) hell small, elongate, oval to subtrigonal, posteriorly pointed, anteriorly rounded, long sloping shoulders more than half the length of shell; anterior margins of / meeting in deep zigzag junction. Pedicle-valve with small beak acutely at about 30°; surface of valve occupied by two pairs of large rounded or lar nearly equidistant plications, the middle pair larger than the lateral ¢ all continuous from beak to margin). Brachial valve with smaller lower btuse beak than in opposite valve; surface occupied by a median rounded ion arising in front of beak, and a pair of larger more elevated lateral plications to those on other valve and continuous from. beak to margin. General e of both valves marked by regular rather sublamellose concentric striz. ’mensions.—Leneth, 4-5 mm. ; width, 2°5-3°5 mm. Horizons.—(1) Balclatchie Group ; (2) Stinchar Limestone Group. Localities—(1) Balelatchie (conglomerate) ; (2) Craighead. Remarks.—This small shell was simply labelled “‘ Rhynchonella’ sp.” by * Bruuines, Geol. Canada, 1863, p. 143, fig. 77; Hatu and Cuarkz, op. cit., Brach., ii, p. 220, pl. Ixii, figs. 11-18. 926 DR F. R. C. REED ON THE Davinson, but it somewhat suggests an alliance with Camarella Thomsom a d C. balclatchiensis ; the more elongated and narrower shape of the shell and stror continuous plications help to distinguish it. The ornamentation also is a ni able feature, well seen in the Balclatchie examples. The true generic reference uncertain. It is named after the famous geologist of last century, J. J. Convnnare, Camarella? cuneatella (Davidson). (Plate XXII, figs. 25, 26.) 1883. Rhynchonella cuneatella, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 200 pl, x, ig: 11. In this shell the long sloping flattened shoulders with false areas and the general shape resemble Rhynch. ardmillanensis (see p. 941), but the folds are co fewer, and only marginal; the beak of the pedicle-valve also seems lower. Davipson’s poor type-specimen from the Balclatchie conglomerate there are ; plications on the brachial valve, and they do not curve outwards on each si Davipson incorrectly shows. In other specimens from the same locality (mon and named by Davipson with the type) there are 5 or 6 folds on the bracl valve, the 2 middle ones being usually rather larger and closer, and on the Be li valve there are 3-4 on each side of the shallow narrow sinus. E There are many poorly preserved small shells from Craighead which may. be a variety of this species, but the marginal plications are usually 6 or 7, and the 2 median sinus or fold, the front edge being straight. A shell compared with . species has been described by the author * from the Glensaul Limestone, Co. - Horizons.—(1) Balclatchie Group ; (2) Stinchar Limestone Group ? Localities —(1) Balclatchie (conglom.), Ardmillan ; (2) Craighead. Camarella Peachi (Davidson). * (Plate XXII, figs. 27, 28.) ay 1883. Rhynchonella Peacht, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Supria Pp x pl. xi, fig. 25. re This species has been sufficiently described by Davipson, but he was mi thinking that it resembled his Rhynchonella scotica from Craighead, for the | is a Parastrophia with the characteristic generic features (see p. 929). C. 5 P as above described. It is thus also related te C. Volbortha, Billings,t fl n tl Black River Limestone, the genotype of Camarella. Horizon.—Stinchar Limestone Group. Locality.— Craighead. * Keep, Quart. Journ. Geol. Soc., vol. Ixvi, 1910, p. 278. + Bruxtnas, Geol. Vanada, 1863, p. 143, fig. 77; Hat and Ciarke, op. cit., Brach., ii, pl. Ixii, figs. 13-18. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 927 Camarella Thomsoni (Davidson). | (Plate XXII, figs. 29-32.) 1868. Rhynchonella Thomsoni, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 186, =. pl. xxiv, fig. 18. 1883. Rhynchonella Thomsoni, ibid., vol. v, Silur. Suppl., pp. 148, 225. presence of 2-3 marginal plications in the sinus of the pedicle-valve, the | low fold with 2-3 plications on it in the brachial valve, the presence of lateral ions, and the absence of such sharp. well-defined edges to the sinus help to lish it from C. balclatchiensis, but it is a variable species. In the more verse varieties with accentuated fold and sinus we see a resemblance to zacht, and the flattened more elongated examples are difficult to distinguish the Craighead variety of C. cwneatella. Davipson’s type of the species sont was found at Craighead. The internal characters are unfortunately vn. The more transverse swollen shells may possibly be the adult shells, flatter, more triangular ones, like Davipson’s figured one, may be immature ality.— Craighead. Genus Parastropata, Hall and Clarke. Parastrophia emerita, sp. nov. (Plate XXII, figs. 33, 34.) . Rhynchonella Saltert, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 225. Rhynchonella Saltert, Davidson, ibid., vol. iii, pt. vii, p. 188, pl. xxiv, figs. 19, 20. hell transversely subpentagonal. Brachial valve moderately convex, with broad ; fold composed of 4 plications consisting of a pair of closely placed small ledian ones only developed near the margin, and a pair of considerably larger Interior of brachial valve with minute spondylium and straight n septum nearly half the length of the valve. Surface of shell ornamented a few concentric lamellose growth-ridges and lines. Di nensions.—Leneth, 8 mm; width, 10°5 mm. Be rons—~(1) Stinchar Ginesone Group ; (2) Balelatchie Group. Localities —(1) Craighead ; (2) Ardmillan Brae. -Remarks.—These small shells were labelled by Davipson Rhynchonella Salteri,* : * Davipson, op. cit., iii, p. 188, pl. xxiv, figs. 19, 20. 928 DR F. R. GC. REED ON THE ; but the type of this species which comes from Penwhapple Glen has only 2 « 3 plications forming the fold, and they are coarse and equal, and the shell is transverse in shape. Our shell much more resembles P. thraivensis, from the Starfish Bed (see p. 930), and P. angulosa (Térnquist),* but it shows some variation an irregularity in the plications on its surface (e.g. fig. 34). Parastrophia rotunda (Sowerby). (Plate XXII, figs. 35-41.) 1839. Atrypa rotunda, Sowerby, in Murchison’s Silur. Syst., p. 629, pl. xiii, fig. 7. 1868. Pentamerus rotundus (Sowerby), Davidson, Mon. Brit. Foss, Brach., vol. iti, pt. vii, p. 150, pl. xv figs. 9-12. 1883. Pentamerus rotundus (Sowerby), Davidson, op. cit., vol. v, Stlur. Suppl., p. 165% pl. ix, figs, 6-9. > 211883, Pentamerus galeatus (Sowerby), Davidson, 2b7d., p.§164, pl. ix, figs. 25, 25a (non Pentamerus galeatus, Davidson, op. cit., vol. iii, p. 145, pl. xv, figs. 13-23). The diductor muscles are not shown in Davrpson’s fig. 7 of a pedicle aaa fr Penkill, though they are well seen in the actual specimen, of which the figure i poor restoration. These diductors are flabellate, triangular, lie close up against septum, are radially striated, and are about one-fourth (or rather less) the len, the valve. The fact that the ‘sinus is on the pedicle-valve distinguishes this from Pentamerus galeatus, as well as the low small and less conspicuous bea Davinson’s figure (fig. 9) of the internal cast of the brachial valve from Penkill should be shown two slightly convergent nearly vertical septa, very closely p and making a very narrow elongated spondylium extending about half the len the valve, beyond which the septa seem to unite into one and be slightly prolor as a single vertical plate. The adductors also in this valve are not as he sho form a regularly oval area bisected by the spondylium and composed of two p rather small scars in contact, the anterior ones being the smaller, but not transver as he depicts. The dental plates in the pedicle-valve are short, convergent at a 45°-60°, making a small short spondylium, and the median supporting sept : shorter oe that in the opposite valve. It is probable that the shell from Penkill referred by Davipson to Penta: galeatus (op. cit., pl. ix, figs. 25, 25a) belongs to this species. As Hatt and Crarke pointed out,t Atrypa rotunda, Sowerby, must be placed in the genus Parastrophia, its internal characters, position of fold and sinus external features much resembling P. reversa (Billings) { of the Clinton Gr P. rotunda was first described as Atrypa by Sowrrsy§ from the Wenlock Gi and Davipson (op. cit., ili, p. 151) was perplexed about its proper generic refe rence Horizon.—Penkill rau a Locality.—Penkill. jl * Lainpstrom, Fragm. Silur. (1880), p. 23, t. xiii, figs. 14-19. + Hawt and Criarke, op. ctt., Brach., ii, p. 221. t Haut and Crarkg, cbid., pl. xiii, figs. 8-14. § Sowersy, in Murchison’s Silurian System, pl. xiii, fig. 7; DAvipson, op, cit., vol. iii, p. 150, pl. xv, figs. 9-1 ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 929 Parastrophia? (Camarella’) Salteri (Davidson). (Plate XXII, fig. 42.) 1868. Rhynchonella Salteri, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 188, pl. xxiv, fig. 19 a ‘fig. 20 11907. Pe. oe (Davidson), Wiman, Arkiv /. Zool. (Stockholm), Bd. iii, No. 24, p. 6, t. ii, figs. 4-5. Shell subpentagonal, widest across middle. Brachial (?) valve moderately convex ; obtuse, incurved, somewhat swollen ; shoulders long, about half length of valve, . somewhat flattened slopes forming filke areas ; median fold composed of a pair ge angular subparallel plications arising at about one-third the length of the e e and i increasing in height and strength to anterior margin ; sides of fold steeply ied ; laterai portions of shell narrow, smooth, gently convex, without plications. r with thin median septum, about one-third the length of valve. Horizon.—‘ Caradoc” [? Balclatchie Group]. Locality.—Penwhapple Glen (Jermyn Street Museum). R meurks. —Davipson’s description of this species is incorrect, and his figure t., fig. 19) of the type from Penwhapple Glen is not reliable. The appear- £3 plications on the fold is due to the flattening out of one by crushing, (0 an oblique chip broken off its anterior end. It is doubtful if the Boduan en (Davinson, op. cit., fig. 20) is referable to the same species. The strength rior elevation of the plications, and the presence of only 2, separates the Pen- form from P. Youngi (p. 930), and it is quite distinct from the Ardmillan and d so-called Rh. Saltert (=P. emerita, mihi). It may be a distorted variety balclatchiensis, but it cannot be regarded as an altogether satisfactory and the precise horizon on which it occurs is somewhat doubtful. Parastrophia scotica (Davidson). (Plate XXII, fig. 43.) VIDSON’S ee of the Pe aca not Bit that out of ae 5 incase on ow fold on the brachial valve the 3 middle ones are rather smaller than the ost ones and are more closely placed together. The outermost plication on de of the fold is equal in size to the plications on the lateral lobes. In the -valve there are 4 plications in the sinus. It should also be remarked that ll is more globose than Davipson showed in his figure. The investigations of Miss Witson* show that there is much variation in the number of plications on the shell of the allied P. hemiplicata (Hall).t Th all its essential characters P. scotica much resembles P. hemiplicata (Hall), * Wirsoy, Geol. Surv. Canada Museum, Bull. No. 2 (1914), pp. 131-140, pl. iv. Haut, Palzont. N.Y., vol. i, 1847, p. 144, pl. 33, fig. 10; Hann and Cuarge, op. ctt., Brach., ii, p. 221, pl. Ixiti, figs. 1-3. 930 DR FF: RB: C. REED*ON THE from the Trenton Limestone, and also is much like the Swedish “ Cam Tornquisti, Wiman.* Hatt and Ciarket recognised that Davipson’s Rh. belonged to Parastrophia. . Horizon.—Stinchar Limestone Group. Locality.— Craighead. Parastrophia thrawensis, sp. nov. (Plate XXII, figs. 44, 45.) Shell strongly biconvex, with deeply re-entrant valves, subcireular ; beal obtuse, approximate. Pedicle-valve with deep sinus increasing in width and anteriorly, with sharp prominent angular edges, and with flattened floor produc blunt tongue, and bearing in anterior half 2 small closely placed plications of « size ; sides of sinus steep, high; lateral lobes of valve bearing 3-4 coarse subar plications arising at about half the length of valve, those bordering the being the longest, sharpest, and most prominent. Interior of pedicle-valve short narrow spondylium and median septum extending from it about one-th length of valve. Brachial valve strongly convex, with median fold composed of a pair of b: lateral subangular plications and a smaller narrower median one, all continuo : arising nearer margin. Surface of shell marked 7 few strong concentric g ridges. Dimensions.— ; ; (1) (2) Length ; oe oe BOO 9°0 mm. Width . ; LOG 10°5 mm. Horiwzon.—Drummuck Grou (Starfish Bed). Locality.—Thraive Glen. Remarks.—This shell is very closely allied to “ Camerella” angulosa, To: from the Leptena Limestone. Parastrophia reversa (Billings) § also bear: siderable resemblance to it. ; Parastrophia Youngi, sp. nov. (Plate XXII, figs. 46, 47.) Shell more or less globose, subcircular; beaks approximate, subequal, swoll obtuse. Pedicle-valve convex, with broad shallow median sinus arising distance from beak, with flattened floor continued as obtuse tongue bent do * Wiman, Bull. Geol. Instit. Upsala, No: 10, vol. v, pt. ii, 1900, p. 175, pl. vil, figs. 22-25. + Haut and Crarge, op. cit., Brach., ii, p. 223. { Linpstrro6m, Fragm. Silur., p. 23, pl. xiii, figs. 14-19. § Hatt and Crarkg, op. cit., Brach., ii, pl. lxiii, figs. 8-14. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 931 en ntering other valve, and holding 2 equal angular plications arising at about half th e length of valve ; anterior flee of sinus steep, with angular edges; 2 or 3 short sub- 1 ginal angular plications of decreasing strength on each ee lobe. Brachial lve very convex, swollen, with low road median fold arising some distance in of beak and composed of 3 coarse equal angular plications; 2 or 3 similar ones ch lateral lobe arising at about half the length of valve. Beak rather higher that of opposite valve. Interior of brachial valve with elongated narrow lylium nearly one-third length of valve, and thin median septum extending nt of it. Interior of pedicle-valve with minute spondylium and short median m about one-third the length of valve. Surface of valves marked with con- c growth- -lines. rvensions.—Length, 8°75 mm. ; width, 9°50 mm. ; depth, 8°25 mm. Hovizons.—(1) Stinchar Limestone Group ; (2) Balolatchie Group. Localities.—(1) Craighead ; (2) Balclatchie. Remarks:—This species differs from P. scotica by its fewer and coarser plications, ticularly on the fold and sinus. It seems more closely allied to “ Camerella” ngusti, Wiman,* from the Ordovician of the Baltic provinces. Our species is ated to the late Professor Youne of Glasgow. —————S Genus STRICKLANDINIA, Billings. ——_——$_ eee Stricklandinia lens (Sowerby), var. _ (Plate XXII, fig. 48.) il 883. Stricklandinia lens (Sowerby), Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 165, pl. ix, figs. 4, 5 (non figs. 2, 3). 339. Atrypa lens, Sowerby, in Murchison’s Silurian System, p. 637, pl. xxi, fig. 3. 66. Stricklandinia lens, Sowerby, Davidson, op. cit., vol. ili, p. 161, pl. xix, figs. 13-21. he transverse shells figured by Davinson as Sér. lens from Woodland Point differ the typical subcireular Str. lens of English and Welsh localities principally by r shape, being much wider than long. The beaks, moreover, of the valves are yin contact, and there is no cardinal area exposed below the beak of the pedicle- as in the typical form. The fold and sinus are usually more sharply defined D ic lateral portions of the shell than in the type, but the internal characters wr to be identical. We may observe some variation in the shape of the shell, for s between transversely elliptical and subcircular examples occur. The general e of radial plications and the more broadly rounded cardinal angles and shorter line serve to distinguish this form from the Mulloch Hill species described slow as Strickl. mullochensis. But some subcircular specimens with faint radial narkings prove the intimate relationship of the various Girvan forms. -Horizon.—Saugh Hill Group. Localities. Woodland Point, Newlands, Shalloch Forge. * Wiman, Bull. Geol. Instit. Upsala, No. 10, vol. v, pt. ii, 1900, p.°175, pl. vii, figs. 22-25. "TRANS, ROY. SOC. EDIN., VOL. LI, PART IV (NO. 26). 133 932 DR F. R. C. REED ON THE Stricklandinia lirata (Sowerby), var. nov. scotica. (Plate XXIII, figs. 1-4.) 1883. Stricklandinia lirata (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. v, Silur, Suppl., p. 168 5, pl. ix, fig. 1: 1883. Stricklandinia lens (Sowerby), Davidson (pars), ‘bid., p. 165, pl. ix, figs. 2, 3 (non figs. 4, 5). q This variety is distinguished from the type-form* by possessing low rounded — radial plications, fewer in number and less distinctly marked. In shape it is oy. subcircular and not transverse, the hinge-line is always considerably less than the maximum width of the shell, and the cardinal angles are well rounded. The var grows to a large size, some shells measuring over 50 mm. in length. Davinson figured some of these shells as Strickl. lens, and in some specimens the radial plications are SO faint that the resemblance is great. The shape of the shell, moreover, is that of the typical Str. lens. The interior is not well seen in any of the specimens. There are usually fairly well-marked concentric growth-ridges on the surface with finer co centric lines between them, but these concentric markings are subordinate to the r plications. From Str. mullochensis this variety differs in the stronger radial plic tions as well as in the shape of the shell and length of the hinge-line. The doubt: examples from Newlands furnish a connecting link. The species described by GacEL + as Strickl. Schmadti, Kiesow, is very closely allied, but seems to have no radial plications on the fold. Horizon.—Saugh Hill Group. Localities. —W oodland Point, ? Newlands. Stricklandima mullochensis, sp. nov. (Plate XXIII, figs. 5-8.) Shell transverse, wider than long, widest at hinge-line ; cardinal angles less than right angles or subrectangular, very slightly rounded. Pedicle-valve gently com beak small, pointed, slightly incurved, rising above hinge-line ; hinge-area narr surface of valve with narrow rounded shallow median sinus widening slightly tows the front ; interior with small spondylium supported on short septum extendii little in front of it. Brachial valve gently convex, with small pointed incurved b and narrow triangular hinge-area ; surface with narrow rounded median fold ; inter with very short closely placed socket-plates produced into small crura, and wi narrow linear faint median muscle-scar ; ovarian areas coarsely pitted and extendi nearly to cardinal angles. Surface of both valves covered with strong low concen sublamellose growth-ridges, and generally with faint low irregular radial undulati 1S. Dimensions.— (1) (2) (3) Length ./ 28 21 18 mm. Width 4 { i 20 24 24 mm. * Davipson, Mon. Brit. Foss. Brach., at lii, pt. vii, p. 159, pl. xx figs, 1-13. + GacuL, Brach. cambr. silur. Gesch. Ost-West Preussens (Konigsberg, 1890), p. 61, t. iv, fig. 9. ~ em gat Bree bd i ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 933 Horizon.—Mulloch Hill Group. Locality.—Mulloch Hill. _ Remarks.—The transverse shape of this shell and long hinge-line and the surface mentation distinguish it from the typical Strickl. lens. The transverse variety of latter, above described, from Woodland Point and Newlands, has not the coarse icentric ridges nor the radial undulations. Pentamerus nacrocamerus, M‘Coy,* é ely similar in shape. Genus PEnTaMERUS, Sowerby. ‘ : Pentamerus oblongus, Sowerby. 1839. Pentamerus oblongus, Sowerby, in Murchison’s Silurian System, p. 641, pl. xix, fig. 10. . Pentamerus oblongus, Sowerby, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 151, pl. xviii, figs. 1-12; pl. xix, figs. 1, 2. . Pentamerus oblongus, Davidson, op. cit., vol. v, Silur, Suppl., pp. 161, 225. pical examples of this well-known species occur. at Cuddystone Glen, Pen- le Glen, and Penkill, and attain a length of 50-60 mm. I have not seen any mens from the Middle Llandovery of Newlands, as stated by Davipson. orizons.—(1) Penkill Group; (2) Camregan Group. calities. (1) Penkill, Penwhapple Glen ; (2) Bargany Pond Burn, Cuddystone ee ee Se * Subgenus BARRANDELLA, Hall and Clarke. Pentamerus (Barrandella) undatus (Sowerby). 1. 839. -Atrypa undata, Sowerby, in Murchison’s Silurian System, p. 637, pl. xxi, fig. 2. 866. Pentamerus undatus (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 155, pl. xix, : figs. 4-9. smorus undatus, Davidson (pars), op. cit., vol. v, Stlur. Suppl., p. 162, pl. ix, figs. 10-15, 19, 20 (non figs. 16-18). considerable number of specimens of a large subglobose transverse smooth neroid, having the sinus on the pedicle-valve and the fold on the brachial valve, in the beds at Woodland Point and Newlands, and are certainly referable to By S Pentamerus undatus,t sens. str., for they possess all the typical internal rnal features. Some of the specimens measure as much as 40 mm. in width. e swollen incurved beak of the pedicle-valve rising high above that of the Ive and with a concave false area below it crossed by an open triangular rium are characters which serve to distinguish it from the transverse variety of ichl, lens, which occurs associated on the same horizon ; but crushed and imperfect cimens, especially of brachial valves, are difficult to separate. Davipson figured lexamples of Pent. undatus from Woodland Point, but unfortunately included specimens (figs. 16-18) of Triplecia woodlandensis under this name (see p. 910). Horizons.——-(1) Saugh Hill Group; (2) Mulloch Hill Group. Localities (1) Woodland Point, Newlands; (2) Mulloch Hill. - : * M‘Coy, Ann. Mag. Nat. Hist., ser. 2, vol. viii, 1851, p. 290. c. + Davipson, og. cit., vol. iii, p. 155, pl. xix, figs. 4-9. 934 DR F. R. C. REED ON THE Pentamerus (Barrandella) undatus (Sowerby), var. nov. penkillensis. (Plate XXIII, figs. 9-13.) 1883. Pentamerus undatus (Sowerby), Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur, S2 pp. 161, 225 (Penki!l and Penwhapple Glen). oy. 1883. Pentamerus globosus (Sowerby), Davidson (pars), tbéd., p. 163, pl. ix, figs. 21-24 Gon Atrypa globosa, Sowerby). 1883. Whitfieldia tumida (alman), Davidson, zbid., p. 224, A Shell transversely elliptical, rounded,subglobose, very unequally biconvex. Pedicle- valve swollen, uniformly convex, rarely with faint broad median flattened sinus nea anterior margin ; beak large, swollen, incurved over concave false area ; floor of sit ‘ produced anteriorly into short rounded tongue. Interior of pedicle-valve with short median septum about one-fifth to one-sixth length of valve, and small sh narrow spondylium beneath beak; straight subparallel vascular markings radiating over whole internal surface. Biecuinl valve very gently convex or flattened, with promi nent rounded median fold increasing rapidly in height and width to anterior margin defined in front. Interior with pair of very short crural (?) plates diverging ata 60°. Surface of shell smooth, except for a few concentric submarginal growth-lines. 8. Dimensions.—Length, 15-17 mm. ; width, 23-25 mm. - Horizons.—(1) Penkill Group; (2) Camregan Group. Localities.—(1) Penkill, Penwhapple Glen ; (1) Bargany Pond Burn. Remarks.—This shell seems more allied to P. undatus than to P. globosus, but differs from the typical P. wndatus in the almost complete absence of a sinus on pedicle-valve. The true P. globosus (Sow.),* has a subcircular shape and a muel longer rene septum. Genus METACAMARELLA, nov. Shell oval, biconvex. Pedicle-valve with low median fold near antenae ont anf posed of several longitudinal plications chiefly developed towards front end; heal high, incurved, with open delthyrium ; small false area on each side; interior y small subumbonal spondylium and short median septum. Brachial valve with lo beak than opposite valve; low median fold near anterior end, composed of severa longitudinal plications ; interior with pair of long recurved crura, pair of long paral median septa, and muscle-scars as in Parastrophia. Shell thick, fibrous, punet externally. Metacamarella balclatchiensis (Davidson). ~ bel (Plate XXIT, figs. 14-18.) ; = 1883. Stricklandinia 1 balcletchiensis, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., Rl 166 pl. ix, figs. 27, 28, 29. a. 1894. Camarella ? balcletchiensis (Davidson), Hall and Clarke, Paleont. New York, vol. viii, Br ach., 11 p. 221. . Shell elongated, oval to subpentagonal, widest across middle, somewhat ni in front, strongly biconvex. Pedicle-valve rounded, convex, more or less inflated * DAavipson, op. cit., vol. iii, p. 156, pl. xix, figs. 10-12. ke - ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 935 with median group of 6-8 low rounded closely-placed longitudinal plications forming broad weak flattened fold, arising at about half the length of the valve and most. pene close to anterior margin, scarcely elevated above smooth lateral se : h aS parte off to ‘jefe epecrocsly a slightly es false area ; lange en ete iene below vee Interior of valve with small Reis and Te eee Spe erura long, stout, recurved, diverging at about 60°, close to hinge-line ; posterior etors large, oval, lying against septa and extending to their extremities ; anterior ors small, deep, subtriangular, situated immediately in front of posterior pair middle line. Shell substance thick, fibrous, and externally punctate. Brachial valve. Pedicle-valve. Length . ster ae 27 e. 35 mm. Width . z 25 c. 28 mm. 0 Be, —Balclatchie. arks.—DaviDson figured (op. cit. nk ix, fig. 27) 0} one pedicle- suiles showing the is specimen shows the median septum through the shell and a shallow groove the low median fold on the surface. The third figure (fig. 28) represents a ul valve, and shows the two median septa to some extent. But the best in- of a brachial valve, mounted on the same tablet, was not figured by him, his cast shows the crura, septa, and muscle-scars (Pl. XXIII, fig. 18). does not seem possible to keep this pentameroid species in the genus Camarella, Hatz and CrarkE* suggested as its probable place, for the strong elongated in the brachial valve are peculiar, and, moreover, in Camarella there is no ‘of median septa in the brachial valve. The muscle-scars of the brachial valve closely similar to those of Parastrophia.t But the high beak of the pedicle- and the open delthyrium, as* well as the spondylium in this valve, forbid us ng it in the last-mentioned genus. The very thick fibrous and punctate shell is ther distinctive feature of our Balclatchie species, and on the whole it seems to nerit a new genus for its reception, and the name Metacamarella is proposed as n tion of its affinities. "4 Earn and CiLarkE, Palzont. New York, vol. viii. Brach., ii, p. 221. + Ibid., p. 387, pl. 1xiui, fig. 21. S35GnN . DR F. R. C. REED ON THE Genus Ercuwa pi, Billings. Kichwaldia personalis, sp. nov. (Plate XXIII, fig. 19.) rounded shoulders, somewhat contracted near umbo and reaching nearly two- the length of the valve; beak high, prominent, inflated, rounded, rising hinge-line and incurved, nearly touching beak of opposite valve; apex of | truncated by large circular foramen with ill-defined small concave false area’ it; interior with pair of long subparallel muscle-scars about half the — of valve. Brachial valve strongly convex, subcircular; beak swollen, not elevated, incurved; interior with median septum extending about three-fou the length of valve. Surface of both valves smooth; interior of shell co thickly with very numerous minute puncte arranged in regular quincunx m in curved intercrossing closely placed lines (like the engine-turning of a watch-case about 10-12 punctze to a millimetre. ea Dimensions.—Length of pedicle-valve, 17°0 mm.; length of brachial : 14°5 mm.; width of shell, 15°0 mm. ; depth of shell, 13° 0 mm. Horvzon.—Saugh Hill Group. Locality.—Woodland Point. Remarks.—Only one specimen of this Tatarestine shell is known to me, an in the Sedgwick Museum. The generic reference to Hichwaldia seems prol ba and our species much resembles E. gibbosw (Hall),* from the Niagara Format The ornamentation is internal and much finer than in H. Capewellr (Davidso the only British species hitherto known, and the shape of the shell is differer median fold and sinus being present. These two Silurian shells are usually in the genus Hichwaldia, Billings,t but the original species of this genus, Es trigonalis, Billings, has a smooth exterior, and ScHucHERTS refers all those a pitted surface to Dictyonella, Hall. | * Haw, 20th Rep. New York State Cab. Nat. Hist., 1867, p. 278; Hien? and CLARE op. cit., Brach h, 1 pl. lxxxiii, figs. 6, 7. ns + Davipson, op. cit., vol. iii, p. 193, pl. xxv, figs. 12-15. ers | t Brings, Geol. Sure: Canad Rep Progress, 1857-58, pp. 190,192, fig. 24. g ScuucHeEr?, Bull, U.S. Geol. Surv., No. 87, 1897, pp. 210, 220 ; Hasruan-Ztrvmn, Tent book of Pala eOn (2nd edit. 1913), p. 396. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 937 Order TELOTREMATA. Genus ProtoruyncuHa, Hall and Clarke. Protorhyncha nasuta (M‘Coy). 4 (Plate XXIII, figs. 20-23.) 18: : 2. Hemithyris nasuta, Moy, Syn. Pal. Foss. Woodw. Mus., p. 203, pl. it, figs. 5, 5a. 1868. Bhynchonella nasuta (M‘Coy), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 173, pl. xxiii, figs. 19, 19a-). 3. Rhynchonella nasuta (M‘Coy), Davidson (pars), op. cit,, vol. v, Silur. Suppl., p. 160, pl. x, fig. 20 (non fig. 21). ll longitudinally oval to subquadrate, either longer than wide or about as wide cardinal angles and sides rounded; anteriorly more or less produced into tongue and marginally sinuated; unequally biconvex; hinge-line curved. valve less convex than brachial valve, but with more prominent higher im- u e (?) beak rising above hinge-line, incurved, with triangular hinge-area and : riangular open delthyrium below it; surface of pedicle-valve with broad v flat-bottomed sinus increasing in width anteriorly, and having steeply sloping d angulated edges towards front margin ; floor of sinus occupied by 6-7 equal ribs; sides of sinus with usually one similar rib in posterior half but smooth y ; lateral portions of valve convex, covered with 12-16 somewhat smaller er straight ribs; interior of valve with short median septum in umbonal Brachial valve more convex than opposite valve, with weak rounded median surface increasing in height and width anteriorly ; beak much smaller and han that of pedicle-valve, obtuse, incurved ; surface of fold bearing 5-7 ribs, ral lobes bearing 12-16 similar or slightly smaller ribs. (1) (M‘Coy’s type.) (2) (3) (Davipson’s figured specimen, pl. x, fig. 20.) Ped.-valve. Brach. valve. Ped.-valve. Brach. valve. 29°0 26°0 Cy 220 25°5 mm. We Dons e. 21:0 c. 23°0 mm. yon.—Stinchar Limestone Group. ocality.—OCraighead. cemarks.—This species appears to be related to the so-called Porambonates vensis, Billings,* which Hatt and CiarkeE f believed might be referred to either us Protorhyncha or Orthorhynchula. Our shell seems to suggest its refer- 0 Protorhyncha, for the hinge-line is not straight as it is in Orthorhynchula. a dubia, Hall,t is chosen as the type of Protorhyncha, and this species occurs Chazy Limestone, but Hatt and Ciarke (op. cit., pl. lvi, figs. 7-9) figure ther species (Pr. zquiradiata) from the Clinton Group which seems more like * Biuiines, Geol. Canada, Palzoz. Foss., vol. i, p. 140, figs. 117a-g. + Hatt and Crarkz#, op. cit., Brach., ii, pp. 180, 181. ~ Hatt, Palzont. New York, i, 1847, p. 21, pl. iv bis, fig. 5 938 DR F. R. C. REED ON THE our Craighead shell. The definition of Pr. nasuta given by M‘Coy was incomplete and has been here revised; but his original specimen still is the best-known examp of the species. | Protorhyncha sp. (Plate XXIII, figs. 24, 25.) There is another shell from the Stinchar Limestone which closely resembles Protorhyncha altilis (Hall)* of the Chazy Limestone. It possesses a very globose shell like Walsonia Wilsona; there are about 24-25 equal rounded radii, on th surface, 7 or 8 of which lie on a scarcely perceptible broad undifferentiated fold the brachial valve, and a corresponding number in an equally weak sinus of opposite valve. The incurvature of the beaks, the impressed lateral false areas, ribbing and general aspect of the shell so closely agree with Pr. altilis tha’ specific identity may even be suspected; but the specimens are too poor for s: factory identification, and only one brachial valve from ili and a pedicle valve from Minuntion are known to me. ; Dimensions.—Brachial valve, length, 13'0 mm. ; width, ¢. 13° 0 mm. : Horizon.—Stinchar Limestone Group. Localities.—Craighead, Minuntion. Genus ORTHORHYNCHULA, Hall and Clarke. Orthorhynchula sub-borealis (Davidson). (Plate XXIII, figs. 26, 27.) 1883. Rhynchonella sub-borealis, Davidson, Mon. Brit. Foss. Brach., vol. v, Sdlur. Suppl., p. 149, pls figs. 5, 6. Shell transversely elliptical, gently biconvex, ee swollen ; hinge-line str about three-fourths the width of shell. Pedicle-valve with rather pron elevated slightly incurved beak, with triangular concave hinge-area below (? ar delthyrium covered by pair of deltidial plates); median sinus moderately deep t shallow, widening anteriorly, with sides sloping and floor bearing 2-4 suban subequal ribs; lateral lobes with 5-7 rather larger or subequal subangula: straight or gently curved back. Brachial valve deeper than pedicle-valve, smaller low inconspicuous incurved beak and very narrow hinge-area; median 1 rounded, more or less distinctly elevated, rising anteriorly, less than one-third wid of shell on front margin, composed of 3-4 subangular equal ribs ; lateral lobes gently convex, bearing 5-7 straight or slightly curved equal subangular ribs. | Dimensions.— 1a (1) (Davipson’s fig. 5.) (2) (Davipson’s fig. 6.) Length . : ee 1A 12°0 mm. m Width . : ; ; pert Ban | 15°0 mm. Thickness . ; ; Sew) 8°5 mm. * Hau, Palxont. New York, i, 1847, p. 23, pl. iv bis, figs. 9a—-d : Haut and CuaRKE, op. cit., Brach., ii, p 181. fe ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 939 _ Horizon. —Stinchar Limestone Group. Locality. —Craighead. -Remarks.—This shell must be referred to the genus Orthorhynchula, Hall and ,* on account of its straight hinge-line and the presence of a hinge-area. It patie clear if the delthyrium is closed by plates or open. The first specimen d by Davinson (fig. 5) is an unusually thick and subglobose example, and his represents the common type, the other being quite exceptional. We may are this species with O. Linneyi (James),+ from the Lorraine Group, the only bed species of the genus; this species also varies in its degree of globosity. Genus RHyncHOTREMA, Hall. Rhynchotrema Lapworthi (Davidson). (Plate XXIII, figs. 28-31.) 1883. Rhynchonella Lapworthi, Davidson, Mon. Brit. Foss. Brach., vol. v, Stlur, Suppl., p. 154, pl. x, fig. 7. is species has 4 ribs on the low fold in the brachial valve, 3 in the better- d sinus in the opposite valve, and 3-4 on each lateral lobe, with 1 or 2 smaller near the hinge and the lateral false areas. The ribs are all equal in size, ey pt t the marginal ones bounding the sinus, which become rather higher anteriorly, 1 the 1 or 2 cardinal ones which are smaller than the rest. Davrpson’s figure does ike the ribs on the fold straight enough, nor the top of the fold sufficiently flat ; ides of the sinus are also not shown as steep as they are in reality. This species L h resembles Rh. inequivalve (Castelnau),} from the Trenton Formation of North ome specimens of Rh. Lapworthi are considerably more globose and _ less erse than the type, but cannot be separated specifically. rizon.—Stinchar Limestone Group. pe? :; ocality.—Craighead. Rhynchotrema shallochense (Davidson, emend.). ‘ (Plate XXIV, figs. 1, 2.) 1883, Rhynchonella shallockiensis, Davidson, Mon. Brit. Foss. Brach., vol. v, Silur. Suppl., p. 154, pl. x, fig. 19. This is a badly founded and badly defined species, for undoubtedly the shell rich Davipson figured as the type (measuring 7 mm. long and 9 mm. wide) is an ure individual, and on the same tablet there were mounted by him three larger nore normally rhynchonelloid shells and one specimen referable to Platystrophia ata. The shell next in size to the type, and probably more advanced in age, is * Haut and CLARKE, op. cit., Brach., ii, p. 181.' t Nerretroru, Mem. Kentucky Geol. Surv., 1889, p. 41, pl. 34, figs. 7-18; Han and CLARKE, op, cit., p. 181, vi, figs. 10-13, 19 ; Forrste, Bull. Denison Univ., vol. xvi, art. 2 (1910), p. 24, pl. iii, fig. 10. WINCHELL and eevee Palzont. Minnes., “oll ili, p. 459, pl, xxxiv, figs. 9-25. TRANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 26). 134 940 DR F. R. C. REED ON THE a transverse globose shell measuring about 8 mm. long, 10 mm. wide, and 65 mm in depth ; the valves are deeply re-entrant, the tongue of the sinus being lon strongly bent down, and the union of the valves forms deep zigzags ; there are 2 ri in the sinus and 3 on the fold, with 3 on each lateral lobe, the lateral rib bou ing the sinus on each side being more elevated and prominent at its anterior The second larger shell is poorly preserved, but seems similar to the above desc one. The third and largest, measuring 11 mm. long, 15 mm. wide, and 9°5 deep, is similarly transverse and globose, with deep re-entrant valves, but there to be 4 ribs in the sinus, and they are of smaller size than the 5 ribs on each lat lobe; the characters of the brachial valve are not well preserved. Probabh shell is the adult form of the species. It may be compared with Rh. inxquivalve, var. laticostata, Winchell and Schuchert.* , ; Horizon.—Whitehouse Group. Locality.—Shalloch Mill. _ Rhynchotrema? gurvanense (Davidson, emend.). ‘* ' (Plate XXIV, figs. 3-6.) ‘3 1883. Rhynchonella girvaniensis, Davidson, Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., pa 168 pl. x, fig. 26. 11883, Rhynchonella nucula, var.%, Davidson, ibid., pp. 157, 225, pl. x, fig. 30; pl. xi, fig. 24. Shell subcircular to transversely elliptical, subglobose. Pedicle-valve with hig} prominent slightly incurved acutely pointed beak; shoulders sloping at about 90° median sinus shallow, rapidly increasing in width to front margin, where it m more than one-third the width of valve ; floor of sinus flat, carrying 3 equal sub ribs ; sides of sinus steep, low ; lateral lobes bearing 4-6 subequal ribs, of which tl bordering sinus are somewhat elevated at anterior ends. Brachial valve con of 3 subangular ribs; lateral lobes bearing 4-5 similar subangular ribs. with median septum about one-third length of valve. Dimensions.— . (Davipson’s fig. 26.) Length . ; : : . 7°00 mime Width . ; : é , : . 875 mE Thickness. ; : . ¢. 6°00 mm. Horizon.—Balclatchie Group oanptortaraten Locality.—Balelatchie. Remarks.—Davinson’s description, consisting of less than two lines, is quite inadequate as a specific diagnosis, and his figure is a restoration of the sp None of the specimens available are well preserved. The form which he Rh. nucula, var. ?, differs in having rather finer and more numerous ribs, 4 * WincHELL and ScuucHeEr?, Paleont, Minnes., vol. iii, p. 461, pl. xxxiv, figs. 26-29, ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICY, 941 1 the fold, and it is rather more transverse in shape; but it seems doubtfully able, and the material is very poor. Rhynchotrema sp. (Plate XXIV, fig. 7.) re is one pedicle-valve of a rhynchonelloid from the Balclatchie conglomerate ; Gray's collection which may be compared with WINcHELL’s Rhynchotrema * from the Trenton Shales. It has 5 regular equal ribs in the shallow sinus, apidly widens anteriorly, and 7-8 ribs on each lateral lobe, gradually 12 in size to the cardinal edges. The two small subparallel dental plates, one-sixth the length of the valve, are well seen. The number of ribs at once es it from Rh. Lapworthi and Rh. girvanense. nensions.—Length, 16 mm.; width, 17-18 mm. "1z0on.—Balclatchie Group. ety. —Balclatchie (conglomerate). Genus RHYNCHOTRETA, Hall. Rhynchotreta ardmallanensis (Davidson, MS.). (Plate XXIV, figs. 8-17.) , Rhynchonella cuneata, Dalman t, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silwr. Suppl., _p. 152, pl. x, figs. 9, 9a, 9d, non fig. 10 (non Rh, cuneata, Davidson, op. cit., vol. iii, p. 164, pl. xxi, figs. 7-11). Rhynchonella Weaveri?, ibid., p. 158, pl. x, fig. 24. 8. . Rhynchonella Weaveri, Daridscn; op. cit., vol. iii, p. 185, pl. xxiv, fig. 14. cuneiform, subtrigonal, longer than wide, broadest and subtruncate ly, gently biconvex, somewhat flattened, without fold or sinus (or only races of them); long sloping shoulders with elongated more or less flattened vated false areas which extend three-fourths to four-fifths the length of the Pedicle-valve gently convex, with rarely very weak broad shallow sinus ation in front marein; beak high, pointed, acute, slightly incurved, per- yy small foramen ; interior with two thin subparallel long vertical dental Brachial valve gently convex, less so than opposite valve, sometimes or with slightly increased median convexity but no definite fold; beak inted, lower, less acute and less prominent than in opposite valve, scarcely d; interior with minute spondylium and single long straight median septum ing nearly half the length of the valve. Surface of both valves covered with small low subangular or rounded equal plications, strongest near front margin, straight except outermost lateral ones, which curve back very slightly. mensions.—Length (average), 7-9 mm. ; width (average), 5-8 mm. * WINCHELL and ScuucHeERt?, Palxont. Minnes., vol. iii, p. 459, pl. xxxiv, figs. 1-8. 942 DR F. R. C. REED ON THE Horizons.—(1) Balclatchie Group ; (2) Stinchar Limestone Group ? Localities.—(1) Ardmillan, Dow Hill, Balclatchie ; (2) Craighead. Remarks.—This small shell, of which some specimens from Ardmillan Mrs Gray’s collection were labelled by Davipson Rh. cuneata, var. ardmillanensis (MS.), is abundant in the Balclatchie Beds. In the Craighead limestone there is a variety with a weak incipient sinus and coarser ribs. A shell much like it rather more numerous plications (24-26) was described by Davipson as— Weaveri?, but it cannot be associated with the Silurian species of this name wl occurs at Tontavont and most probably it is only a variety of Rh. ardmuallanensis. The first-named variety from Craighead was figured by Davipson as Rh. cuneata’ (op. cit., pl. x, fig. 9), and has rather coarser larger and fewer ribs than the common Ardmillan type, and it seems rather broader and less elongated ; but it can scar be regarded as a distinct species. Rh. [Camarella| cuneatella, Davidson,* from Balclatchie conglomerate, has fewer and broader plications and they are only develo near the margin (see p. 926). The typical Wenlock Rh. cuneata, Dalm., differs fi ours in possessing fewer and coarser ribs, a non-truncate anterior end, and ¢ traces of fold and sinus. Rh. ardmallanensis more resembles in general appearance Camarophoria subcuneata, Hall,t from a much higher stratigraphical horizon, Rhyuchotreta cuneata (Dalman), var. (Plate XXIV, fig. 18.) 1866. Rhynchonella cwneata, Dalman, Davidson, Mon. Brit. Foss. Brach., vol. iii, p. 164, a figs, 7-12. Non 1883. Rh nee cuneata, Dalman, Davidson, ibid., vol. v, Silur. Suppl., p. 152, pl. x, figs. 0,1 10, In this species Davison states that there are 10-14 ribs, of which 4 or 5 lie in fold and 3 or 4 in the sinus. The Girvan examples sometimes have as few as 8 occasionally as many as 14 ribs, and there is no marked fold in the brachial val sinus in the pedicle-valve. The shells also are flatter and have longer strai shoulders. It should also be noticed that there is always a marked interspace not wider or deeper than those between the other ribs), down the median line ¢ brachial valve, and that along it lies the median septum extending for half the k of the shell. There is considerable variation in the degree of coarseness of the The form from the Lower Ordovician beds, attributed to this species by Davips (op. cit., p. 153), is now separated off as distinct under the name Rh. ardmillanensis. Some small rare shells from Thraive Glen are very doubtfully referable to Rh. cuneata Dalman. Horizon.—(1) Mulloch Hill Group ; (2) Drummuck Group (Starfish Bed) ? Localities.—(1) Mulloch Hill, Craigens ; (2) Thraive Glen. * DAVIDSON, op. cit., vol. v, Silur. Suppl., p. 200, pl. x, fig. 11. + Hau and CraRrkg, op. cit., Brach., ii, pl. xii, fig. 35. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 943 Genus Camarota@cnia, Hall and Clarke. Camaroteechia decemplicata (Sowerby). a 1839. Terebratula decemplicata, Sonerbel in Murchison’s Silurian System, p. 641, pl. xxi, fig. 17. 1883. Rhynchonella decemplicata, Davidson, Mon. Brit. Foss. Brach., vol. v, Stlur. Suppl., p. 202, pl. xi, fig. 27. The one poor specimen from Mulloch Hill, of which Davinson’s figure is a liberal nd incorrect restoration, is the only example with which I am acquainted from . The two median ribs are shown too close together in his figure, and ides of the fold should be steeper and lower; there are only 5 plications on the lobes, and they diverge more from one ane than represented, and the le shell is more transverse and not so subcircular. Recently TweNHoFEL* has fied this species in the fauna of the Silurian of Anticosti, and it appears also cur in the Llandovery beds of the Christiania district in Norway. Camarotechia landoverrana (Davidson), var. nov. diversiplicata. (Plate XXIV, figs. 19-22.) 1883. . Rhynchonella Uandoveriana, Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur, Suppl., p. 157, pl. x, figs. 36, 37. [DSON figures a specimen of this shell from Camregan Wood, but the figure t bring out the difference between the size of the ribs in the sinus and those e lateral lobes. For in these Camregan shells this is a constant feature ; there 3-5 strong angular ribs in the sinus, corresponding to 3-4 similar ones on the fold, e on each lateral lobe there are 9-10 smaller less prominent and more closely ribs. Otherwise in shape and general characters our Girvan shells agree fairly with Davrpson’s types of Rh. Wandoveriana, though he says + that in the latter are 7-8 ribs on the fold and in the sinus. ymensions.—Average leneth, 11 mm. ; average width, 13 mm. orizons.—(1) Camregan Group ; (2) Mulloch Hill Group. Localities.—(1) Camregan Wood; (2) Craigens. ... * Camarotechia nucula (Sowerby) ? 1839. Terebratula wucula, Sowerby, in Murchison’s Stlwrian System, p. 611, pl. v, fig. 20. 1868. Rhynchonella nucula (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 181, : pl. xxiv, figs. 1-7 ? Some small shells in a poor state of preservation from Mulloch Hill high-road, nted and labelled with Davipson’s figured specimen of Cam. decemplicata, appear * TWENHOFEL Geol. Surv. Canada, Museum Bulletin, No. 3, 1914, p. 28. + Davipson, op. cit., vol. iii, p. 184, pl. xxiv, figs. 8-13. 944 DR F. R. C. REED ON THE to be referable to C. nucula (Sowerby) in the wide acceptation of the name, but species requires revision, as the term has been loosely applied and several varieties or species have been included under this name. Horizon.—Mulloch Hill Group. ' Pi, Locality.—Mulloch Hill. ; 5 Camarotechia sp. (Plate XXIV, fig. 23.) 11883. Rhynchonella Wilsoni (Sowerby), Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. p. 225 (uon Terebratula Wilsoni, Sowerby). establish a separate species. “ sane more resembles the specimen called a t C. llandoveriana by Davipson,* from the Upper ‘Llandovery of Minsterley, equal in size to those on the lateral lobes, which number 13-18 on each side. As median septum extends nearly half the length of the brachial valve. Dav labelled some specimens of it Rh. Wilsona with a query. Dimensions.—Length, 9 mm. ; width, 11 mm. Horizon.—Camregan Group. Locality.—Camregan Wood. Genus Zycosprrra, Hall. Zygospura orbis, sp. nov. (Plate XXIV, figs. 24-27.) Shell subcircular, flattened, very gently biconvex. Pedicle-valve uniformly com with small elevated slightly incurved beak, apical foramen, and false area. Br valve with small inconspicuous and not elevated beak, and with traces of bi shallow median depression holding 4-5 ribs. Surface of valves bearing 20-22 s straight equal regular equidistant rounded ribs separated by interspaces of equal s Dimensions.—Length, 7 mm. F Horizons.—(1) Stinchar Limestone Group ; (2) Balelerenie Group ? Localities.—(1) Craighead ; (2) Ardmillan, Balclatchie (conglomerate). Remarks.—This small shell seems comparable to the more finely plicated exa of Zygospira recurvirostra (Hall) from the Trenton Limestone, figured by Hat Crarke.+ In the Balclatchie specimens, which are doubtfully attributed to the san species as those from Craighead, the median interspace in the brachial valve is d * DAvInsoN, op. cit., vol. v, Stlur. Suppl., pl. x, fig. 32. + Haut and Crarke, op. cit., Brach., ii, p. 154, pl. liv, figs. 4, 5. - ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 945 wider and the interior shows a short median septum. It is somewhat allied to Hicksi, Reed,* from the Slate Beds of Haverfordwest. Zygospira ? sp. (Plate XXIV, figs. 28, 29.) internal cast and external impression of a pedicle-valve from the Starfish rs a considerable resemblance to the species Zygospira cincinnatiensis, | from the Lorraine Group. Our specimen measures just over 9 mm. in length same in width, being subcircular in shape ; there is a low indistinct median ) a narrow deep depression along its middle holding 2 ribs which arise from urcation of one near the beak; on each side of this depression the fold 3 rather broader rounded ribs, with a very short narrower one intercalated the margin alongside the depression. The lateral portions of the valve ry 8—9 similar rounded ribs, of which half arise by intercalation at rather an one-third their length. The short thin dental plates are visible in species appears to lie between Z. cincinnatiensis, Meek, and Z. putilla, Hall rke.} rizon.—Drummuck Group (Starfish Bed). Be Theaive Glen. Genus ProtozyGa, Hall and Clarke. Protozyga carrickensis, sp. nov. (Plate XXIV, figs. 30, 31.) ° ubpentagonal, unequally biconvex, widest across middle, rounded, anteriorly ate, and sinuated. Pedicle-valve convex, swollen, subcarinate, highest iddle line, with short broad shallow groove on keel near anterior margin ; 1, pointed, incurved, rising above hinge-line, with apex perforated, and small broad fissure below. Brachial valve weakly convex or flattened, bilobed n row median straight impressed line and by broad indefinite sinus near margin ; beak small, inconspicuous. Surface of valves smooth. Interior ensions.—Length, 11 mm.; width, 11 mm. izon.—Stinchar Limestone Group. lity.— Craighead. : arks.—This interesting little shell, of which I have only seen two specimens s Gray’s collection, closely resembles Protozyga exiqua, Hall,§ of the Trenton BED, Geol. Mag., dec. v, vol. ii, 1905, p. 452, pl. xxiii, figs, 17-19. [eEK, Palzont. Ohio, 1, 1873, p. 126, pl. 11, fig. 5 ; BEG and CLARKE, op. ctt., Brach., ii, pl. liv, figs. 13, 14. ALL and CLARKE, op. cit., pp. 157, 365, Pi liv, Aan 35-37 ; pl. Ixxxiii, figs. 29, 30. [ALL and CLARKE, op. cit., Brach., ii, p. 151, pl. liv, figs, 47, 48. 946 DR F. R. CG. REED ON THE Limestone, but differs by its more rounded shape and by the median. groove the anterior end of the keeled pedicle-valve. Perhaps Rhynchonella? nana, Sal MS.,* from Tyrone, is allied. - Genus Arrypina, Hall and Clarke. 7 Atrypina Barrandei (Davidson) ? q 1848, erebratula Barrandit, Davidson, Bull. Soc. Géol. France, vol. v, sér. 2, p. 382, pl. iii, fig. 3: ; 1866. Retzia? Barrandii, Davidson, Mon. Brit, Foss. Brach., vol. iii, pt. vii, p. 128, pl. xiii, figs. 10 13 A few small shells, 3-4 mm. long, are probably referable to this species, but the specimens are not good. Hatt and Crarke + referred Davipson’s species Retzia? Barrander to the genus Atrypina, of which the type Leptocelia imbricata, F occurs in the Lower Helderberg Group. Horizon.—Saugh Hill Group. Locality.—Newlands. Genus Guassta, Davidson. Glassia compressa (Sowerby). 1839. Atrypa compressa, Sowerby, in Murchison’s Silurian System, p. 629, pl. xiii, fig. 5. 1886. Athyris compressa (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol, iii, pt. vii, p. 122, pil. xii figs. 16-18. = fee casts and external impressions of a smooth subcircular shell from lands may be attributed to Glassia compressa, Sowerby, the internal characte which do not seem to have been described. In the pedicle-valve the teeth are sm short slightly curved, thickened plates, nearly parallel to the hinge-lin diductors form very narrow elongated flabelliform weak impressions divers about 30°, with a triangular area between them for the adductors, but not en In the brachial valve the hinge-plate has the curious recurved form deseri Hat and oe in Nucleospira, and stated by them (op. cit., p. 152) to extending chone one-fourth ne leneth of the valve between the flabelli c elongated and radially striated adductors; the crura are short, sharp-po and recurved. Dimensions.—Length, 13°0 mm. ; width, 14°5 mm. Horizon.—Saugh Hill Group. Localities. Newlands, Woodland Point. * Davipson, op. cit., vol. ili, p. 192, pl. xxiv, figs. 26, 26a-c. + Haut and CLarkg, op, ctt., Brach., ii, p. 161. { Haut and Cnarkg, op. cit., Brach., ii, p. 142, pl. xlviii, ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 947 a ais Glassia obovata (Sowerby). 1839. Atrypa obovata, Sowerby, in Murchison’s Silurian System, p. 618, pl. viii, fig. 9. +1866. Athyris obovata (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 121, pl. xii, en fig. 19; pl. xili, figs. 5, 5a, 1883. Glassia obovata (Sowerby), Davidson, 2b7d., vol. v, Silur. Suppl., pp. 116, 224, pl. viii, fig. 12. on.—Penkill Group. : ty.—Penkill. Genus CLInToNELLA, Hall and Clarke. Clintonella cf. vagabunda, Hall and Clarke. (Plate XXIV, fig. 32.) 1 regularly oval. Brachial valve flattened, marked by deep well-marked dian sinus extending from beak to margin, rapidly increasing in width and with its floor occupied in anterior third by one simple gently rounded lateral lobes of shell with inner portion occupied by 3 straight subequal unded closely placed ribs on each side of sinus, but outer posterior part of oth and not ribbed ; beak low, inconspicuous. ensions.—Length, 3°4 mm. ; width, 2°8 mm. izon.—Saugh Hill Group. lity.—Newlands. ks.—No species except Clintonella vagabunda, Hall and Clarke,* of the ‘ormation of the United States, appears to bear any resemblance to this 1, of which I know only one specimen in Mrs Gray’s collection, represented vternal cast and the external impression of the same brachial valve. Genus Atrypa, Dalman. a Atrypa expansa, Lindstrém, var. A shell indistinguishable from the variety of A. expansa, with few ribs, which was | d by Linpstrém + from Dalecarlia, occurs at Shalloch Mill, but is rare. It is llied to A. ombricata, Sow. ensions.—Lenegth, 17 mm.; width, 19 mm.’ et * Haut and Crarkg, op. cit., Brach., ii, p. 159, pl. li, figs. 1-11. ‘— + Liypstr6m, Fragm. Silur., p. 22, t. xii, figs. 17-19. TRA 'S, ROY. SOC, EDIN., VOL. LI, PART IV NO. 26). 135 948 DR F. R. C. REED ON THE Horizon.—Whitehouse Group. ~ Locality.—Shalloch Mill. Atrypa imbricata (Sowerby). 1839. Terebratula imbricata, var, Sowerby, in Murchison’s Silurian System, p. 631, pl. xiii, fig. 1866. Atrypa imbricata (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 135, figs. 3-6, 8 (non 7). 1883. Atrypa imbricata (Sowerby), Davidson, zbid., vol. v, Selur. Suppl., pp. 115, 224. Typical examples of this shell at Woodland Point; in these the deep sinus | pedicle-valve, with sharp angular margins, is well marked. Flattened examples’ difficult to separate from the coarser-ribbed forms of A. reticularis, which also’ to occur associated with it. Davrpson records it from Penkill and Bargany P Burn as well as Woodland Point, but I have seen no examples from the two former localities. 4 Horizon.—Saugh Hill Group. Locality.—W oodland Point. Atrypa reticularis (Linné). a 1767. Anomia reticularis, Linné, Syst. Nat., ed. xii, p. 1152. 1866. Atrypa reticularis, Linné, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 129, - : figs, 1-22. 1883. Atrypa reticularis, Linné, Davidson, op. cit., vol. v, Silur. Suppl., p. 225. _ Penkill and Bargany Pond Burn; one example from the former locality was fi by Davinson (op. cat., pl. xiv, fig. 14). The coarser-ribbed variety oce Penwhapple Glen, and approaches A. aspera in character. A larger and flatte covered with numerous closely imbricated ribs, is found at Woodland Point. (4) Penkill Group. F Localities.—(1) Mulloch Hill, Rough Neuk; (2) Woodland Point; (3) Bo Pond Burn; (4) Penkill, Penwhapple Glen. Genus Dayia, Davidson. Dayia cymbula (Davidson), var. nov. girvanensis. (Plate XXIV, figs. 33-35.) 1868. Merista? cymbula, Davidson, Mon. Brit. Foss, Brach., vol. iii, pt. vii, p. 204, pl. xxii, figs, 28, 2 1883. Merista cymbula, Davidson, op. cit., vol. v, Silur. Suppl., p. 138, pl. viii, figs. 6-9. The type-specimen of Dayia cymbula came from Hendre wen, Cerrig y dr The Girvan specimens from Drummuck differ in certain particulars from this they are rather broader and more subcircular, and the beak of the pedicle-valve + Sowmrsy in Murchison’s Silurian System, p. 637, pl. xix, fig. 3. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 949 rather higher and more inflated, and this whole valve is more swollen. Internally there are some differences, and the figures of the interior of pedicle-valves from mmuck given by Davipson are inaccurate, for the strongly marked short flabelli- rm muscle-scars on each side of and slightly behind the crescentic median callosity ere not shown, nor was the thickening of the shell inside the beak clearly indicated. he brachial valve there is also a strong median septum proceeding from the small kened hinge-plate, and extending from one-fourth to three-fourths the length of alve ; the crura are short and rod-like, and there are two pairs of faintly marked e adductors. These differences are sufficient to separate off our Drummuck a variety. The Whitehouse ecco are less satisfactorily preserved and more like the Welsh form. not think that there is a “ shoe-lifter” process in the pedicle-valve, as Davip- esents in his restorations of the interior, and our shell seems rather to possess ral characters of Dayia navicula (Sow.)* than of Merista, sens. restr.,t median crescentic thickening in the floor of the valve has merely its anterior slig shtly excavated, and it thins off on each side into lateral horns. Ey the ae sneaes Dayia pentagonalis, Reed,{ is closely allied to the Dayia, sp. ind. (Plate XXIV, fig. 36.) small subcircular shells from Bargany Pond Burn seem to possess the characters of Dayia and resemble D. cymbula rather than the Silurian cule (Sow.). The material is, however, too poor for a SarUCLeN yg identifica- Davinson labelled them Merista cymbula. 2ensions.—Width, 3°5-4°0 mm. 2on.—Camregan Group. Locality.—Bargany Pond Burn. Genus Cyctosprra, Hall and Clarke. e occurrence of this genus in the Girvan area rests on doubtful evidence. The chosen by Hatt and Crarke is Orthis bisulcata, Emmons,§ of the Trenton sone. Externally this shell is very similar to Dayia navicula (Sow.) of the ek and Ludlow Beds, but internally it differs essentially. se AVIDSON, op. cit., vol. iii, p. 190, pl. xxii, figs. 20-28 ; 2bid., vol. v, Selur. ice. p. 96, pl. v, figs. 1-4. +t Hann and Ciarkg, op. cit., Brach., 1, p. 70. { Rexp, Quart. Journ. Geol. Sock al liii, 1897, p. 75, pl. vi, figs. 5, 5a-c. Exarons, Geol. New York, ep: 2nd Diss 1842, p. 395, fig. 4; Hani and Cuarke, op, cit., Brach., ii, p. 146, pl. liv, figs. 38-40, 950 DR F. RB. OC. REED ON THE Cyclospira ? diversa, sp. nov. (Plate XXIV, figs. 37-43.) 1883. Rhynchonella portlockiana, Davidson (pars), Mon, Brit. Foss. Brach., vol. v, Silur. Suppl. pl. x, fig. 13 (non figs. 12, 14). . Non 1868. Rhynchonella portlockiana, Davidson, op, ctt., vol. iii, pt. vii, p. 189, pl. xxiv, figs. 23-25, ¥ Shell subpentagonal to subtriangular, widest towards front, unequally bico Pedicle-valve convex, highest along middle, subcarinate, with short angular : angular median sinus more or less developed and widening anteriorly ; beak elevated, pointed, slightly incurved, with small false area below; interior o with pair of short thin slightly divergent dental Plates. Brachial valve “ m length of valve; lateral edges of valve somewhat flattened at right an plane of valves; interior of valve with small thick hinge-plate, cleft down the middle, and with long thin median septum extending about | length of the shell. Surface of shell marked by rather strong lamellose conce erowth-lines. : Dimensions.—Leneth, 5-6'0 mm.; width, 4-5°5 mm. _Horizon.—Balclatchie Group. Localities. —Ardmillan, Balclatchie. Limestone. The internal median rae in the brachial valve and fen characters agree better with the first named, but the hinge-plate with th mentioned species. 7 In all the Ardmillan specimens the pedicle-valve is scarcely at all carinat d the median groove is only developed in the anterior portion of this valve. | Davipson’s specimens (op. cit., fig. 13) from the mudstone of Balclatchie brachial valve precisely similar to the type from Ardmillan, but his other specimen (op. cit., fig. 12) is quite distinct in character, having a much longer median fold in this valve, and its true reference is doubtful; it occurs in the clatchie conglomerate. The typical form of this new species is totally distin the Kildare“ Rhynchonella” portlockiana, Davidson. Crushed and imperfect sp may be confused with Camarella cuneatella, C. Conybearei, C. balclatchien C. Thomsoni. The generic reference must be considered extremely doubtful. — * HAL and CiarKg, op. cit., Brach., ii, p. 146, pl. liv, figs. 38-40. t+ Haun, Palexont. N.Y., vol, i, p. 141, pl. xxxiii, figs. 6a-d. { Haun and CrarKg, op. cit., Brach., pp. 149-151, pl. liv, figs. 47, 48. Genus Sprrirer, Sowerby. Spirifer plicatellus, var. radiata, Sowerby. 1825. Delthyris lineatus (radiatus), Sowerby, Miner. Conchol., vol. v, p. 493, figs. 1, 2. 866. Spirifera plicatella, var. radiata (Sowerby), Davidson, Mon. Brit. Foss, Brach., vol. iii, pt. vil, p. 87, pl. ix, figs. 1-6. , Sptrifera plicatella, Davidson, op. cit., vol. v, Silur, Suppl., p. 137, pl. viii, figs. 2, 3. In these shells the lateral lobes are devoid of folds, as is the case in SOWERBY’S * type from the Wenlock Limestone. It appears to be rare in the Girvan area. 7zons.—(1) Penkill Group; (2) Camregan Group. ities. (1) Penkill, Penwhapple Glen ; (2) Bargany Pond Burn. Genus Cyrtra, Dalman. Cyrtia exporrecta (Wahlenberg). . Anomites exporrectus, Wahlenberg, Nov. Act. Reg. Soc. Scient., vol. viii, p. 64, No. 3. 66. Cyrtia exporrecta (Wahlenberg), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 99, pl. ix, 4 figs. 13-24. 33. Cyrtia exporrecta (Wahlenberg), Davidson, ibéd., vol. v, Silur. Suppl., p. 137, pl. viii, fig. 4. is a rare shell in the Girvan district, but the few specimens observed show haracteristic features, and the internal structure is seen in casts to agree with examples of the species. zons.—(1) Penkill Group; (2) Camregan Group. littes—(1) Penkill; (2) Bargany Pond Burn. Genus Ruyncnospira, Hall. Subgenus HommospiRa, Hall and Clarke. Ri ynchospira (Homeospira) Bouchardi (Davidson), var. nov. mullochensis. = (Plate XXIV, figs. 44-46.) ell transversely oval to subtrigonal, biconvex. Pedicle-valve uniformly convex ; sutely pointed, high, slightly incurved; surface ornamented with 1 median ht rib and 8-9 similar or rather smaller low rounded ribs on each side, closely. and curving slightly outwards. Brachial valve shorter than pedicle-valve, with small low inconspicuous obtuse beak and shallow median sinus holding 1 fine ht median rib (or none ?), with 9-10 rather stronger ribs (similar to those on alve) closely placed and curving slightly outwards on the gently swollen lateral ; Interior with median septum extending about half the length of the valve. Dimensions.—Length about 5 mm. orizon.—Mulloch Hill Group. Localities—Mulloch Hill, Craigens ? Remarks.—This small shell may be regarded as a variety of Rh. Bouchards Day.),t of the Wenlock Beds, but it differs in its more transverse shape, the presence J * Davison, op. cit., vol. iii, pl. ix, fig, 1. + Davinson, op. cit., vol. ili, pt. vil, p. 127, pl. xii, figs. 26-80. 952 DR F. R. C. REED ON THE of a median rib instead of a groove on the pedicle-valve, and of a single rib in the sinus of the opposite valve. If the material was better it would deserve a sep specific designation. There are some coarser-ribbed forms, probably of the sam an allied species, which have only 6-7 ribs on each side of the brachial valve and the ribs are more angular, more prominent, and wider apart. , a Rhynchospira (Homeospira) camreganensis, sp. nov. (Plate XXIV, figs. 47-49.) 11883. Rhynchonella cuneata (Dalman), Davidson (pars), Mon. Brit. Foss. Brach., vol. v, Silur. Si p. 152 (non Rhynchonella cuneata, Dalman, sens, str.). Shell oval, longer than wide, widest across middle, unequally biconvex. Ped valve convex, subcarinate, with sloping shoulders inclined at about 30°-45° the length of the valve. Surface of valves covered with 27-31 small low row closely placed ribs of equal size, except median one on pedicle-valve, which is sligh larger ; lateral ribs on brachial valve curve gently outwards near anterior end, — Dimensions.—Leneth, 7°0 mm. ; width, 5°5 mm. Horizon.—Camregan Group. Locality.—Camregan Wood. Remarks.—This small shell, which was labelled in Mrs Gray’s col Rhynchonella cuneata, resembles Rhynchospwra Bouchardi, var. mullochensi has more numerous ribs and is more oval in shape. It may be compare Rh. (Homcospira) evax (Hall),* of the Niagara Group, of which some of the ex have finer and more numerous ribs than others. Genus Nucieospira, Hall. Nucleospira pisum (Sowerby) ? 1839. Spirifer? pisum, Sowerby, in Murchison’s Silurian System, p. 630, pl. xiii, fig. 9. 1866. Nucleospira pisum (Sowerby), Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 106, ph figs. 16-20. 1883. Nucleospira pisum (Sowerby), Davidson, op. cit., vol. v, Silur. Suppl., pp. 91, 224, pl. i fige, 15-18. Two internal casts of brachial valves in Mrs Gray’s collection may probably be referred to this species, but the evidence for its occurrence is poor. As Da remarked, it can scarcely be specifically distinguished from N. pisiformas, Hall, « the Niagara Group. Horizon.—Camregan Group. Locality.—Bargany Pond Burn. * Haun, 28th Rep. New York State Mus. Nat. Hist., 1879, p. 160, pl. 25, figs. 13-21. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 953 Genus WHITFIELDELLA, Hall and Clarke. Whitfieldella angustifrons (M‘Coy). (Plate XXIV, fig. 50.) . Hemithyris angustifrons, M‘Coy, Ann. Mag. Nat. Hist., ser. 2, vol. viii, p. 391. . Terebratula angustifrons, Salter, Quart. Journ. Geol. Soc., vol. vii, pl. 9, fig. 10. . Hemithyris angustifrons, M‘Coy, Syn. Brit. Pal. Foss. Woodw. Mus., p. 199, pl. ix, figs. 6-8. . Meristella angustifrons, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, p. 111, pl. x, figs. 21-27, 885. Meristella? angustifrons, Davidson, op. cit., vol. v, Silur. Suppl., p. 133, pl. viii, fig. 1; p.$224. M‘Cov’s type-specimens came from the “sandstone of Dalquorhan,’ Mulloch ry, and all Davinson’s figured specimens came from Mulloch Hill, but he also ords the species from the “head of Thraive Glen, Auld Thorns, and Thraive.” I nly seen specimens from Mulloch Hill, Auld Thorns, and Craigens. The al characters are excellently displayed in internal casts, and were partly ibed by M‘Coy and Davinson. It should be mentioned that the hinge-plate is sd into a pair of triangular portions by a narrow median slit; the crura seem to and sharply pointed, and there is no definite median septum, but only a thin w median ridge. The pedicle-valve shows a circular apical foramen with the um below it, and the diductor scars are more marked and flabelliform than VIDSON represented in his fig. 26a, pl. x, while there is only a narrow low median vhich is not continued in front of the diductors; the beak also has the shell ened internally. We may especially compare this species with Whitfieldella drica, (Hall),* of the Clinton and Niagara Formations. lorizon.—Mulloch Hill Group. ocalities.—Mulloch Hill, Auld Thorns, Craigens. Whitfieldella nitida (Hall), var. ? y. (Plate XXIV, fig. 51.) there are several specimens of a broad flattened subcircular shell associated with eldella angustifrons at Mulloch Hill which seem closely to resemble Whit- a nitida, var. oblata, Hall,t Whitf. intermedia, Hall,f and Whitf. oblata, Hall.§ terior of a shell of similar shape and external appearance shows the spiral cones outwards, as is typical in the genus, and they are composed of 6-8 turns, pair of parallel closely placed supporting bands. The broadly obovate shape as a very broad scarcely perceptible median sinuation. * Hatt, Palzont. N.Y., ii, 1852, p. 76, pl. 24, fig. 2; Haun and CuarKzE, op, cit., Brach., ii, p. 58, pl. xl, 16-22. ; + Hatt, Palzont. N.Y. ii, 1852, p. 269, pl. 55, fig. 2. + Ibid., p. 77, pl. 24, figs. 3, 4, 26. _§ Ibid., p. 9, pl. 4, figs. 4, 5. 954 DR F..R. C. REED ON THE Dimensions.—Length, 18°5 mm.; width, 17°5 mm.; thickness, 10°0 mm. Horizon.—Mulloch Hill Group. Locality.—Mulloch Hill. | Whitfieldella? sp. (Plate XXIV, fig. 52.) One internal cast of a brachial valve of a simple subcircular moderately ot shape is doubtfully referred to the genus Whitfieldella. There is no median fold, the surface is uniformly convex. The hinge-plate is very large and stout, triangul in shape, thickened anteriorly, and fills the whole umbonal region and is cle the middle by a narrow slit. The anterior edges of these halves into wh: divided are furnished with short sharply pointed thin laminar crura curved i There is a weak low rounded ridge between the narrow elongated parallel mt scars which extend more than half(?) the length of the shell. The hinge-plat crura seem to resemble those of Hyattidina | =Hyattella, Hall and Clarke, n rather than Whatfieldella or any other genus, but it is possible that the Hindella, of which the internal characters of the brachial valve are incomp known,* should receive it. : Dimensions.—Length about 10°5 mm. ; width about 11°5 mm. Horizon.—Drummuck Group (Starfish Bed). Locality.—Thraive Glen. Genus MeristE.ia, Hall. Meristella cf. Circe, Barrande. (Plate XXIV, figs. 53, 54.) Cf. 1847. Terebratula Circe, Barrande, “‘Silur. Brach, Bohéme,” Nat. Abhandl. vol. i, p. 37, pl. xvi, fig. 6. y, » Shell oval to elongated subpentagonal; margin sinuated. Pedicle-valve conve with broad shallow sinus arising at about half to two-thirds the length of the and produced anteriorly into short rounded tongue; edges of sinus formed b narrow folds, slightly divergent or subparallel ; narrow rounded median fold 1 weakly developed near beak but becoming stronger anteriorly ; interior of va pair of thin dental plates converging at first inwards and then running f nearly parallel for less than half the length of valve and enclosing between small low boss or thickening of the shell at their point of convergence, with ridged diductor scars in front of it. Surface of shell marked with a few strot concentric growth-ridges and striz. Brachial valve unknown. Dimensions.—Leneth, 90 mm. ; width, 7°5 mm. Horizon.—Mulloch Hill Group. | -* Hawn and CLARKE, op. cit., Brach., ii, p. 65, - ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 955 - Locality.—Mulloch Hill. Remarks.—Only three specimens of this shell are known, and they are in pedicle-valve. In shape and external features it much resembles the shell . sd by Davipson* to Meristella Circe, Barrande, and its internal characters seem f Meristella? sp. (Plate XXIV, fig. 55.) internal casts of small oval or subcircular smooth shells, measuring only ig and enclosing between their anterior ends the beginning of the long leeply impressed and longitudinally striated diductor scar. The pedicle- also shows a strong muscle-scar. These features are typical of Meristella,t some extent recall M. crassa, Sowerby,{ from the Llandovery. Wiman§ has a species of Meristella? from the Leptzena Limestone which may be correlated Starfish Bed, and LrypstR6m || figures a shell which he refers to M. crassa Leptzena Limestone and from a Lower Ordovician horizon. on.—Drummuck Group (Starfish Bed). ity.—Thraive Glen. Genus Caospira, Hall. Celospira hemispherica (Sowerby). Atrypa’? hemispherica, Sowerby, Davidson, Mon. Brit. Foss. Brach., vol. iii, pt. vii, pl. xiii, figs. 23-30 (1 fig. 26). . Leptocelia hemispherica, Davidson (pars), op. cit., vol. v, Silur. Suppl., pp. 147, 225, jical examples of this species occur at Camregan Wood and at Penkill. These regular simple equidistant rounded ribs, as Davipson well depicts (op. cit., his figures of May Hill specimens (fig. 23). He also figured examples from fig. 27), Penwhapple Glen (fig. 29), and Mulloch Hill (fig. 26), and from illan (fig. 25).1 But I have not seen any typical specimens from Mulloch Hill, the species C. scotica is found. The true C. hemispherica, sens. str., seems 1 to the Upper Llandovery Beds. Some examples from Penkill in Mrs Gray’s tion were named by Davipson Atrypa reticularis. Horizons.—(1) Camregan Group ; (2) Penkill Group. Localities.—(1) Camregan Wood; (2) Penkill. VIDSON, op. cit., vol. iii, pt. vii, p. 116, pl. x, figs. 33-35 ; ibid., Suppl., v, p. 108, pl. iv, fig. 5. t L and CLARKE, op. cit., Brach., ii, p. 73, pl. xliv. t Davinson, op. cit., ili, p. 117, pl. xiii, figs. 1-3. § Wiman, Arkiv for Zool., Bd. iii, No: 24, 1907, p. 6. I LINDSTROM, Hragm. Seine: 1880, p. 21, t. xiii, figs. 4-8, 23-25. There must be some mistake in this Teeality, as no Diendokee Beds occur at Ardmillan. RANS. ROY SOC. EDIN., VOL. LI, PART IV (NO. 26). 136 \ 956 DR F. R. C. REED ON THE Celospira scotica (M‘Coy). 1851, Atrypa hemispherica, Sow., var. (Hemithyris scotica, M‘Coy MSS.), Salter, Quart. Journ. Soe., vol. vii, p. 178, pl. ix, fig. 12a-c. ; 1852. Boren hemispherica, var. scotica, M‘Coy, Sy yn Brit. Pal. Foss. Woodw, Mus., y pl. in, fig. 10. “a 1866. Atrypa? scotica (M‘Coy), Davidson, Mon. Brit. Foss. Brach., vol. ili, p. 140, pl. xiii, fig. 31s 1883. Atrypa? scotica Se Davidson, op. cit., vol. v, Silur. Suppl., p. 224. ; The original locality for this species is Mulloch Hill. The differences bet and C. hemispherica were pointed out by M‘Coy (op. cit.). We may add th division of the ribs is somewhat irregular, that there is an incipient fold on the pedicle-valve marked off by rather deeper grooves on each side than those bet the ribs and corresponding to the shallow sinus on. the other valve. The int characters seem to be identical. Davipson, however, in his Supplement (0 p. 224) does not put C. scotica in the same genus (Leptocelia) as C. hemusphe Horizon.—(1) Mulloch Hill Group. Localities.—(1) Mulloch Hill, Craigens, Auld Thorns. Celospira scotica (M‘Coy), var. (Plate XXIV, figs. 56-58.) The Middle Llandovery examples of C. scotica seem to constitute a fairly m variety of the Mulloch Hill type, for there is always a distinct simple single | rib on the pedicle-valve, and the 2 or 3 ribs on each side of it divide into ¢ : smaller ones at about half their length, while the remaining lateral 4 or outside are simple. In the brachial valve there is a median groove in th corresponding to the median rib. Otherwise the shells seem to be identic: the typical form. Some of them reach a length of 12-13 mm. - Horizon.—Saugh Hill Group. Localities.—Newlands, ? Woodland Point. List or SPECIES DESCRIBED, WITH Davrpson’s EQuivaLENTS. ‘ [The numbers refer to the collections in which the species is represented ; see page 2.) (1), (7) Méieromitra (Paterina) Davidsoni, sp. nov, [Acrotreta Nicholsoni (pars)]- (1), (2), (3), (7) Obolus audax, sp. nov. [Lingula Ramsayi (pars) ]. * (1), (2), (7) 4, Maceullochi, sp. nov. [Lingula Ramsayi (pars) ].° (1), (2), (3), (4), (5) Paterula balclatchiensis, Dav. |Paterula balcletchiensis]. (1) » ‘albida, sp. nov. (1) » ¢ Jamesoni, sp. nov. [Discina? balcletchiensis (pars) ]. (1) Dinobolus Davidsoni (Salt.) [Dinobolus Davidsoni]. (1) Rhinobolus ? balclatchiensis, sp. Nov. (1), (2), (3), (4) Lingula amabilis, sp. nov. [Lingula attenuata (pars) ]. (iia brevis, Portl., var. nov. angustior | Lingula brevis (paral (1), (2), (3), brevis, Portl., var. nov. carrickensis [Lingula brevis (pars). “ORDOVICIAN AND SILURIAN (5) Lingula? crumenoides, sp. nov. (1) ,, — Lewisi (Sow.). (1) ,, = mediostriata, sp. nov. (1), ovata, M‘Coy ? [Lingula ovata (pars)]. (1), (3) ,, ~—_psewdoquadrata, sp. nov. (1),(2) ,, — quadrata (Eichw.) [Lingula quadratal. (1) 4, ~— solvens, sp. nov. (1) 4, — Symondsi, Salt. [Linguli Symondsi]. (1) Lingulasma ? ardmillanense, sp. nov. CIs Fs ? canadense (Billings) {Lingula canadensis]. at: A) Feet 1 penultimum, sp. nov. i (1), (2%) ‘3 1 pulcherrimum, sp. nov. * (a sees 1 questor, sp. nov. [Lingula granulata (pars)]. (1), (2) bi 1 tena, sp. nov. (1), (3) Stphonotreta micula, M‘Coy [Siphonotreta micula (pars) |. (1) 9 1 thratvensis, sp. nov. (1), (2), (3) Schizambon scoticus (Dav.) [Siphonotreta scotica (pars). (1), (3) vhs an eT! 7 \é ( q (2), (3), (4), (7) Acrothele medioradiata, sp. noy. oP (1) a ‘sp. (a) [Acrotreta Nicholsoni (pars)]. (1) 75 (Redlichella 2) sp. (b) [Siphonotreta micula (pars)]. (1), (3) Acrotreta ? sp. (a). (6) tsp. @) , (1) ‘3 (2%) costata, Dav. [ Acrotreta costata]. (1), (7) Conotreta conoidea, sp. nov. [Acrotreta Nicholsoni (pars) |. a (1) Trematis craiyensis, sp. nov. E (1) » melliflua, sp. nov. , (1) » siluriana, Dav. | (1) Schizocrania asteroidea, sp. nov. | Discina crassa]. (1), (7) Orbieuloidea Armstrongi, sp. nov. [Discina perrugata (pars) ]. eee V5, (Schizotreta) Forbesi (Sow.). e (1) Philhedra drummuckensis, sp. nov. : (1); (2) ” mullochensis, sp. nov. [Crania siluriana (pars). (1) » | penkillensis, sp. nov. . (1) 4, ° Playfatri, sp. nov. pi), (2), (3) - 4, silurtana (Davy.) [Crania siluriana (pars)]. (7) ” Sp. = (1) Pholidops antigua (Schloth.) ? 00) are implicata (Sow.) [Pholidops implicata]. (1) Pseudocrania divaricata, M‘Coy. oD) Wer multifilosa, sp. nov. (1) css aff. depressa (Hichw.). (1), (7) ” »” q var. mo (2), (3), (5) ,, Playfatri, sp. nov. [Orthis calligramma (pars)]. (1), (2) ,_-— (Plectorthis) ardmillanensis, sp. nov. | (pars)] | (1), (3), 4), ( is ) rustica, Sow. [Orthis rustica (pars)]. BRACHIOPODA OF THE GIRVAN DISTRICT. z ty shallochensis, sp. nov. [Discina perrugata (pars) |. ~ (1), (2), (3), (7) : _stincharensis, sp. nov. [Discina perrugata (pars) ]. 957 of ~ (1), (2), (3), (5) Orthis calligramma, Dalm., var. nov. eraigensis [Orthis calligramma (pars). (1), (2), (5), (6), » var. nov. subplicata {Orthis calligramma (pars) ]. CU) sips ” ) duftonensis, Reed, var. [Orthis valligramma, var. plicata 958 DR F. R. C. REED ON THE (1), (3) Orthis ( Plectorthis) rustica, var. nov. paucicostata [Orthis rustica (pars , Orthis (1), (3) (1) (1), (3) _() (1), (3), (7) (1), (2), (3), (4) (5), (6) (1), (2), (7) (1), (2), (3); (5) (1), (5) (1) (1), (3) (1), (2), (3), (4), (5), (6) (1), (2), (3) (1) (1), (2), (3), 5), (6) (1), (2); (3), (5) (1) (1), (2), (5), (6) (1), (2), (3); (4) (1), (2), (4), (6) (1), (2), (3), (5) @ (1), (2), (3), ) (2), (3), (4), (5) (1) (1), (2) (1), (2), (3), (5) (1) (1) (1), (4) (1) ee) (2), (3), (4), (6), (7) (1) (1) (1), (3), (4), (5) (1) (1) Rajinesquina concentrica (Portl.) [Strophomena expansa (pars) ]. (1), (2), (3), (5) (1), (2), (5) (1) (1) (1) (1) (1), (2), (3), (5) tal eRe Ss calligramma, var. plicata (pars)]. ( ; ) 5, var. rigida, Dav. [Orthis rustica (pars)]. ( > ) 4, var. walsalliensis, Dav. [Orthis rustica, var. wa salliensis]. ( “ ) subfissicosta, sp. nov. [Orthis inberaos aa ( j ) subplicatella, sp. nov. ( » ) thraivensis, sp. nov. [Orthis calligramma (pars) }. (Dinorthis) flabellulum, Sow., var. nov. carrickensis [Orthis flabe (pars) (4, ) poreata, M‘Coy [Orthis sowerbyan (pars), Om Alabellulur (pars) J. ay (Hebertella) balclatchiensis, Dav. [Orthis balcletchiensis]. | “= (4, _ ) bellatrix, sp. nov. [Orthis sowerbyana (pars) ]. ) erispa, M‘Coy [Orthis crispa (pars) ]. (4, ) Lapwortht, Dav. [Orthis Lapworthi]. ) scotica, M‘Coy [Orthis calligramma, var. scotica]. (Platystrophia) biforata, Schl, [Orthis biforata (pars)]. ( » _ ) dorsata (His.) ? [Stricklandinia ? shallockiensis]. (Heterorthis ?) confinis, Salt. [Orthis confinis]. i (Bilobites) biloba, Linn. ( ) 5, var. [Orthis biloba]. (alain basalis, Dalm [O. elegantula (pars) ]. - ) crassa, Lindstr. ? [O. elegantula (pars)]. sf ) elegantula, Dalm. [O. elegantula (pars) ]. i. ) * var. nov. drummuckensis [O. elegan’ ala 5 ) federata, sp. nov. [O. testudinaria (pars) ]. - %) girvanensis, Dav. [O. girvaniensis]. 2) Rankini, Dav. [O. Rankini (pars)]. oe ) testudinaria, Dalm., var. nov. gracilis. a ) * var. nov. shallochensis [O. testudina: % ) 5 var. ind. [O. argentea]. » ) ef. visbyensis, Lindstr. [O. turgida (pars)]. Rhipidomella ?) polygramma, Sow., var. pentlandica, Dav. [O. poly gramma, var. ponklanaeey O. reversa (pars) |. I NN OT a eA SN oma ~ ey ( ” ) prematura, sp. nov. (Schtzophorella, subg. nov.) fallaa, Salt. [O. mullochiensis (pars) ( 53 » __) mullochensis, Dav. [O. mullochiensi (Wicolella, subg. nov.) actoniz, Sow. (has, * ) 4 var. nov. asteroidea [O. actonie 1]. (Harknessella, subg. nov.) vespertilio, Sow. [O. vespertilio]. ¢ nina, Dav. [O. nina]. expansa (Sow.), var. nov. Macallumz [Strophomena expe . (Playfairia, subg. nov.) deltoidea (Conr.) [Strophomen (pars) : ” ( ) _ var, nov, multicorr » ( at Fe ) ~ var. nov, tenuicorrus ” ( - i 2) felix, sp. nov. 5 (ae aes a8 ) Richardsoni, sp. nov. ” ( » s ) semiglobosina (Dav.) [Strophomena brex, var, semiglobosina (pars) ]. d ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 959 ~ (1) Rafinesquina ? brumalis, sp. nov. 1), (2), (3), (4), (8 i 1 subarachnovdea, sp. nov. [Strophomena corrugatella (pars Pp g P (), (2), (3), (4), (5), (7) Leptena rhomboidalis (Wilek.) [Strophomena rhomboidalis]. (1), (2), (3), (4), (5) Leptella Grayz (Dav.) [Leptena Grayz ; L. llandeiloensis (pars)].. (1), (5) 4, % pseudoretroflexa, sp. noy. [Strophomena retroflexa (pars)]. (1), (5) Plectambonites Ktheridgei (Dav.) [Leptena Etheridgei]. (1), (3), (52) s llandeiloensis (Dav.) [Lept. llandeiloensis (pars) ]. (1), (3), (5) Me quinquecostata (M‘Coy), var. nov. balclatchiensis [Lept. quin- ‘ quecostata J. : (1) #. Be 4 var. nov. cylindrica [Lept. quinque- . ‘ costata]. F / (1) P eT.) iy var. nov. depressa [Lept. quinque- costata]. = ae a(t) of 5 . » var. nov. sublobata [Lept. quinque- = costata]. Gyr) aa ruralis, sp. Nov. 7, (1) o scissa (Salt.) [Lept. scissa (pars)]. : (1), (3) 3s segmentum (Ang.), var. nov. woodlandensis [Lept. segmentum ; (pars)]. | > (1) i sericea (Sow.), var. nov. albida [Lept. sericea (pars)]. (1), (3), (4), (5) s 5 5» Var. nov. conspicua [Lept. sericea (pars)]. Ch) 2. 7 ‘3 » Var. nov. craigensis [Lept. sericea (pars) ]. (1) =o hi »» var, rhombica (M‘Coy) |Lept. sericea (pars) ]. (1), (3) as " » var. semirugata [Lept. sericea (pars) ]. (1) 5 ms » var, nov. thraivensis | Lept. sericea (pars). .~ (1) ¥ . subcorrugatella, sp. nov. [Strophomena corrugatella (pars) ]. = (1), (3), (5) - transversalis (Lvalm.) [| Lept. transversalis (pars)]. ae (1), (3), (5) if es var. Duvali (Dav.). rs (1) Ry e var. nov. mullochensis. J (1), (4), (5) cs ¥ var, nov, penkillensis {Lept. transversalis (pars). (1), (3) Bs var. nov. tricostata [Lept. segmentum (pars)]. (1) % a) Teds (Day.) [Lept. youngiana]. (1) Stropheodonta arenacea (Salt.) [Strophomena arenacea]. (1), (2) 3 corrugatella (Day.) [Strophomena corrugatella Gri. ‘ rs donax, sp. nov. mal), (2), (3), (5) A Jamesoni, sp. nov. [Strophomena corrugatella (pars). ), (3), (4), (5), (7) , (Leptostrophia) filosa (Sow.), var. nov. mullochensis [Strophomena expansa (pars) |. (1), (3) i ( i ) subfilosa, sp. nov. (1), (2), (5), (7) 55 (Brachyprion) columbana, sp. nov. [Orthis sagittifera (pars), Stroph. expansa (pars)]. (1), (4) a wl te , ) Walmstedti (Lindstr.) [Strophomena Walmstedti]. = (1) ” ” Sp. (1) 33 ? Nicholsont, sp. nov. |Stroph. imbrex, var. semiglobosina (pars) ]. (1) 53 % Waltoni (Dav.)? [Strophomena Waltoni]. (1) ” ps (1) Strophonella euglypha (His.). (1), (2) x penkillensis, sp. nov. [Strophomena antiquata (pars)] a Cite +, undata (M‘Coy) [Strophomena deltoidea (pars)]. (1), (3), (5) Christiania tenwicincta (M‘Coy) [Lept. tenuicincta]. (1) » var. nov. belobata, (1), (2), (8), (5) Strophomena antiquata, Sow., var. nov. woodlandensis [Stroph. antiquata (pars). 960 | DR F. R. C. REED ON THE (1), (2), (3), (5) Strophomena deficiens, sp. nov. (1) », shallochensis, Dav. [Stroph. shallockiensis]. (1) Fe valens, sp. nov. ' (1), (2), (4), (5) Schuchertella applanata (Salt.) [Strophomena applanata]. (1) a 1 ineapectata, sp. nov. (1), (2), (3), (5) 4 pertinax, sp. nov. [Stroph. pecten (pars)]. (1), (2), (3) Triplecia craigensis, sp. nov. [Rhynchonella balcletchiensis (pars) |. (1), (4) » imsularis (Kichw.). (1) 5, | nucleoides, sp. nov. ae a (1) 5, woodlandensis, sp. nov. [Pentamerus undatus (pars) ]. + (1), (2), (4) Mimulus Yincertus (Dav.) [Triplesia ? incerta]. (1) Streptis Hornet, ap. nov. [Triplesia? monilifera]. (1), (2), (4) Cliftonia Grayx (Dav. em.) [Triplesia? Grayie]. - (v (1), (3) », Andersoni, sp. nov. [Triplesia spiriferoides (pars), Orthis biforata | Rhynchonella? nasuta (pars) |. (1), (7) » spirtferoides (M‘Coy) [Triplesia? spiriferoides]. (1), (7) Chonetes advena, sp. nov. (1), (7) Clitambonites ascentens, Pander [Skenidium Grayz]. (1) Ss complectens (Wiman), var. nov., albida, (1), (2) 4s shallochensis (Dav.) [Skenidium shallockiense ]. (1) Hemipronites carrickensis, sp. nov. [Strophomena retroflexa (pars). (1), (3) " Thomsoni, sp. nov. [Orthis turgida (pars)]. (1) 8 *. var. nov., 7nconstans. ¢ (1) Scenidium Greenoughi, sp. nov. . (1), (7) 5 Lewisi (Dav.), var. nov., asteroidea, (D2), MB) (4) We S53 5 var, NOv., craigensis. (1), (3), (6) is » var. woodlandensis (Dav.) [Skenidium Lewisi, var. woodlandier (1) Porambonites acutiplicata, sp. nov. zemula (pars) J. (1) A ii var. [Rhynchonella balcletchiensis (pars). (1) 4 t Conybearei, sp. nov. (1) » 2 cuneatella (Dav.) [Rhynchonella cuneatella]. (1), (2) A Peachi (Dav.) [Rhynchonella Peachi]. 4 (1), (3), (2) a Thomsoni (Dav.) [Rhynchonella Thomsoni]. (1) Parastrophia emerita, sp. nov. [Rhynchonella Salteri (pars)]. (1), (4) 3 rotunda (Sow.) [Pentamerus rotundus]. L (1) - 1 Saltert (Dav.) [Rhynchonella Salteri (pars) ]. 3 (1) = scotica (Dav.) [Rhynchonella scotica]. ; (1) - thraivensis, sp. nov. (1) 4 Youngi, sp. nov. (1), (3) Stricklandinia lens (Sow.), var. [Strick]. lens (pars)]. (1), (3) » lirata, var. nov. scotica (Strick. lirata, Strickl. lens (pars]. (1), (3) ) mullochensis, sp. nov. (1), (4) Pentamerus oblongus, Sow. |Pentamerus oblongus]. (1), (3) a (Barrandella) undatus, Sow. [Pentamerus undatus (pars). (1), (4) Zz ( Fe ) 4 -var. nov., penkillensis [Pent. undatus (pars), Pent, globosus (pars) |. . (1) Metacamarella balclatchiensis (Dav.) [Stricklandinia? balcletchiensis]. (3) Hichwaldia personalis, sp. nov. (1), (2), (3) Protorhyncha nasuta (M‘Coy) [Rhynchonella? nasuta (pars) ]. (1) 3 sp. f ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 961 (1) Orthorhynchula sub-borealis (Dav.) [Rhynchonella sub-borealis]. (1), (2), (3) Rhynchotrema Lapwortht (Dav.) [Rhynchonella Lapworthi]. (1) 5 shallochense (Day.) | Rhynchonella shallockiensis]. (1) “f ? girvanense (Day.) [Rhynchonella SSUES Rhynchonella nucula?, var. | (1) rr sp. (Os (2) Rhynchotreta ardmillanensis (Dav., MS.) [Rhynchonella cuneata (pars), ? Rhynchonella Weaveri (pars) |. ' (), (2) ae cuneata (Dalm.), var. [Rhynchonella cuneata (pars) |. (1) Camarotechia decemplicata (Sow.) [Rhynchonella decemplicata (pars) ]. (1), (2), (3) - llandovertana (Dav.), var. nov. diversiplicata [Rhynchonella llandoveriana (pars) ]. (1) at nucula (Sow.)? [Rhynchonella decemplicata (pars). (1) Ms sp. [Rhynchonella Wilsoni (pars)]. (1) Zygospira orbis, sp. nov. (1) Fae © 18D (1) Protozyga carrickensis, sp. nov. (1) Atrypina Barrandei (Dav.) 4 (1) Glassia compressa (Sow.). (), (2) ,, obovata (Sow.) [Glassia obovata]. (1) Clintonella, cf. vagabunda, Hall and Clarke. (1) Atrypa eapansa, Lindstr., var. (1), (3) ,, tmbricata (Sow.) [Atrypa imbricata]. (1), (4), ~— reticularis (Linn.) [Atrypa reticularis]. @) (4) Dayia cymbula (Dav.), var. nov. girvanensis [ Merista cymbula]. p(t) Dayia sp. (1) Cyclospira 4 diversa, sp. nov. [Rhynchonella portlockiana (pars) ]. (1) Spirtfer plicatellus, Linn., var. radiata, Sow. (Sp. plicatella, var. radiata]. (1) Cyrtia exporrecta (Wahl.) [Cyrtia exporrecta]. (1) Rhynchospira (Homeospira) Bouchardi (Dav.), var. nov. mullochensis. (1) sp ( s ) camreganensis, sp. nov. ), (3) Nucleospira piswm (Sow.) [Nucleospira pisum ]. ), (7) Whitfieldella angustifrons (M‘Coy) [Meristella angustifrons]. Oh ae nitida (Hall) var. ? (1) Meristella, cf. Circe, Barr. ett) ” Sp. ah 8), (4) Ceelospira hemispherica (Sow.) [Leptoccelia hemispherica]. ), (3), (4) ” scotica (M‘Coy) [Atrypa? scotica]. (1) »” ” var. STRATIGRAPHICAL DISTRIBUTION OF SPECIES. ORDOVICIAN. — Ballantrae Group. (5) Lingula crumenoides, sp. nov. (5) Paterula balclatchiensis (Dav.) (5) Acrotreta (or Acrothele) sp. ' Stinchar Limestone Group. (2) Obolus Maccullocht, sp. nov. (1), (2) Lingula quadrata, Kichw. (1) Lingulasma ? canadense (Billings). © 962 DR F. R. CG REED ON THE (1), (2%) Lingulasma t pulcherrimum, sp. nov. (1) Acrothele? sp. (a). (1) Acrotreta? sp. (a). (1) 4, Psp. (8). (1) Conotreta conoidea, sp. nov. (1), (2), (38) Schizambon scoticus (Dav.). (1) Zrematis craigensts, sp. nov. (1), (2), (8) Orbiculoidea stincharensis, sp. nov. (1), (2), (3), (5) Orthis calligramma, Dalm., var. nov. craigensis. (1), (2), (3), 6) 4, Playfairé, sp. nov. (1) ,, (BPlectorthis) subfissicosta, sp. nov. . (1), (2), (3), (4), (5), (6) 5, (Dinorthis) flabellulum, Sow., var. nov. carrickensis, (1), (5) ,, (Hebertella) balclatchiensis, Dav. ae (LC) (3) e(6)> Se as C5 ee) scotiea, M‘Coy. as (1), (2), (8), (Platystrophia) biforata, Schloth. (1), (2), (3), (5), (6) =, (Heterorthis %) conjinis, Salt. (1), (2), (3), (5) =, (Dalmanella) federata, sp. nov. (1), (2), (3), (4), (5) 4, ( 4, 2) gtrvanensis, Dav. (Gl eet +, ) testudinaria, var. nov. gracilis % (1), (2), (3), (4), (5) =, (Harknessella) vespertiiio, Sow. : _ (1) ,, (Nicolella) Actoniz, Sow. - (1) Scenidium Lewisi, Dav., var. nov, craigensis. (1), (2), (5) Rafinesquina expansa (Sow.), var. nov. Macallumi. s (1), (2), (5) 7 (Playjairia) deltoidea (Conr.). _- (1), (2), (3), (5) (y+) semaiglobosina (Dav.). (1) Leptena rhomboidalis (Wilck.), var. a. (1), (2), (3), (4), (5) Leptella Graye (Dav.). (1), (2), (3), (5) Stropheodonta (Leptostrophia) subfilosa, sp. nov. (1) Christiania tenuicincta (M‘Coy), var. nov. bilobata. (1), (5) Plectambonites Etheridget (Dav.). (1), (3), (4), (5) 5 sericea (Sow.), var. nov. conspicua. a (1), (3) i ta og » var, nov. craigensis. ¢ (1), (3), (4), (5) 3 youngiana (Dav.). a (1), (2), (5) Strophomena deficiens, sp. nov. (1), (2), (3) Triplecia craigensis, sp. nov. (1) » | nucleoides, sp. nov. (1), (2), (4) Cliftonia Grayx (Dav.). (1), (3) 3 Andersont, sp. nov. (1) Hemepronites carrickensis, sp. nov. (1) . Thomsont, sp. nov. (1) ” ed var. nov. inconstans. (1), (2) Camarella balclatchiensis (Dav.). (1) » 4 Conybearet, sp. nov. (1) » + cuneatella (Dav.) ? (1), (2 2) a Peachi (Dav.). (1), (2), (3) i Thomsoni (Dav.). (1) Parastrophia emerita, sp. nov. (1) 5 scotica (Dav.). (1) > Youngt, sp. nov. (1) Porambonites acutiplicata, sp. nov. (1) Orthorhynchula sub-borealis (Dav.). (1), (2), (3) Protorhyncha nasuta (M‘Coy). ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 963 Se (1) Protorhyncha sp. (1), (2), (3) Rhynchotrema Lapwortht (Dav.). (1) Rhynchotreta ardmillanensis (Dav., MS.) var. (1) Zygospira orbis, sp. nov. (1) Protozyga carrickensis, sp. nov. > Balclatchie Group. (1), (2), (3), (4) Obolus audazx, sp. nov. y (1) Micromitra (Paterina) Davidsoni, sp. nov. (1) Rhinobolus? balclatchiensts, sp. nov. (1), (2), (3), (4) Paterula balclatchiensis, Dav. (1), (2), (3), (4) Lingula amabilis, sp. nov. (1), (2), (3) ,,_~— brevis, Portl., var. angustior. (1), (3) ,, pseudoquadrata, sp, nov. (1) Lingulasma’? ardmillanense, sp. nov. - (1) ‘ ? canadense (Billings). (1) Fe 1 puleherrimum, sp. nov. Qe 1 questor, sp. nov, (1), (2), (3), (4) Acrothele medioradiata, sp. nov. @y 5. Aap. (2), ),Q) tsp. (8). (1) Acrotreta 4 sp. (a). (1), (2) Conotreta conoidea, sp. nov. (1), (3) Stphonotreta micula, M‘Coy. (1) Trematis melliflua, sp. nov. : (1) Orbiculoidea sp. a (1) Philhedra Playfairi, sp. nov. a (1) Pseudocrania, aft. depressa (Eichw. ). 4 (1) Orthis Playfairi, sp. nov. (1), (2) ,, (Plectorthis) ardmillanensis, sp. nov. rans Ge 5 ) duftonensis, Reed, var. (1) Bae tenes ) subplicatella, sp. nov. a “s — (1) 4, + (Dinorthis) flabellulum (Sow.), var. nov. carrickensis, . (1), (2), (5) 4, (Hebertella) balclatchiensis, Dav. eB . Cya(D)) Ga, 0) bellatrix sp.nov. / Cy i CO pee 2) Lapworthe. Dav. P Cyn, A meee) scalscan MiCoy? (1) ,, (Platystrophia) biforata (Schloth. ). (1), (2%, (3), (Dalmanella) federata, sp. nov. var. y - GQ) UP ay |) +) girvanensis, Dav. sy ye ee (| » 2) Rankini, Dav. (1), (2) ( * ) testudinaria (Dalm.), var. nov. gracilis, (1) Bein concentrica (Portl.). (1), (2), (5) ms (Playfairia) deltoidea (Conr.). (1) » (Gs aner Ser var. nov. multicorrugata. (1) ” ( ij ) - var. nov. tewutcorrugata. (1), (29), (3), (5) ” ( i ) semiglobosina (Dav.). » 'brumalis, sp. nov. (1); (2), (3), (4), (5) » 4 subarachnoidea, sp. nov. (1) Leptzna rhomboidalis (Wilck.), var. a. (1) Leptella Grayx (Dav.). (1) Leptella? pseudoretroflexa, sp. nov. S. ROY. Soc. EDIN., VOL. ye PART IV (NO. 26). 137 964 DR F.R.. CG) REED ON THE (1) Stropheodonta Nicholsoni, sp. nov. (1), (3), (5) Christiania tenuicineta (M‘Coy). (1) Plectambonites Etheridgei (Dav.). (1), (3), (5) b llandetloensts (Dav.). (1), (3), (5 2) 3 quinquecostata (M‘Coy), var. nov. balclatchiensis. (1) * “A var. nov. sublobata. (1), (5) a sericea (Sow.), var. nov. conspicua ? (1), (3) be » var, noy. semirugata. (1) = t youngiana (Dav.). (1) Strophomena defjiciens, sp. nov. (1), (4%) Cliftonia Andersont, sp. nov. (1), (2), (3), (4) Scentdium Lewisi, var. nov. craigensis. (1) Hemipronites carrickensis, sp. nov. (1) rf, Thomsont, var. nov. inconstans ? (1) Camarella, balclatchiensis (Dav.), var. 4 (1) » | Conybearet, sp. nov. (1) » § cuneatella (Dav.). (1) Parastrophia emerita, sp. nov. (1) i. 1 Saltert (Dav.) ? (1) ” Youngt, sp. nov. (1) Metacamarella balclatchiensis (Dav.). (1) Rhynchotrema girvanense (Dav.). (CD =A sp. (1), (2) Rhynchotreta ardmillanensis (Dav., MS.). (1) Zygospira orbis, sp. nov. ? (1) Cyclospira? diversa, sp. nov. Whitehouse Group. (1) Paterula’? albida, sp. nov. (1) Lingula brevis, Portl., var. nov. carrickensis. (1) ,, + ovata, M‘Coy ? (1) Orbiculoidea shallochensis, sp. nov. (1) Pholidops antiqua (Schloth.)? (1) Schizocrania sp. (1), (2), (3), (5) Orthis Playfairt, sp. nov. ? (1), (3) ,, (Plectorthis) thraivensis, sp. nov. ? (1) ,, (Hebvertella) crispa, M‘Coy % GQ) C4, ~ 2) Lapworth, Dav.2 (1), (Platystrophia) biforata, Schloth. (Le a a dorsata, His. (1) ,, (Bilobites) biloba, Linn., var. (1), (2), (3), (5) 4, (Dalmanella) testudinaria, var, nov. shallochensis. (1) ” ” ” var, (1) 4, (Nicolelia) actonix, Sow. (1) ,, ? nina, Dav. (1) Rafinesquina (Playfairi) Richardsont, sp. nov. (1) Leptena rhombotdalis (Wilck.), var. B. (2) Stropheodonta corrugatella (Dav.)? ; (1) Christiania tenuicincta (M‘Coy). (1) Plectambonites quinquecostata (M‘Coy), var. nov. cylindrica. (1) e i var. nov. depressa. (1) " sericea (Sow.), var, nov. albida. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 965 (1) Plectambonites sericea, var. rhombica (M‘Coy). Py a or subcorrugatella, sp. nov. (1) Strophomena shallochensis, Dav. (1), (2) Triplecia insularis, Eichw. (1) Scenidium Greenoughi, sp. nov. (1), (2) Clitambonites complectens (Wiman), var. nov. albida. (1), (2) age shallochensis (Dav.). (1) Rhynchotrema shallochense (Dav.). (1) Dayia cymbula (Dav.), var. nov. girvanensis. (1) Atrypa expansa, Lindstr., var. / ; Drummuck Group. (1) Paterula? Jamesoni, sp. nov. (1) Lingula mediostriata, sp. nov. (1) ,, ~~ solvens, sp. nov. (1) Lingulasma’ penultimum, sp. nov. (1) Stphonotreta? thraivensis, sp. nov. @), (3) Schizambon scoticus (Dav.), var. (1) Trematis siluriana, Dav. (1) Schizocrania asteroidea, sp. nov. (1) Orbiculoidea Armstrongi, sp. nov. (1) Phithedra drummuckensis, sp. nov. (1) Pholidops antiqua (Schloth.) ? (1) Pseudocrania divaricata, M‘Coy. (1) 3 multifilosa, sp. nov. (1) Orthis calligramma, Dalm. (1) ,, (Plectorthis) thratvensis, sp. nov. (1), (2) ,, (Dinorthis) poreata, M‘Coy. (1) ,, (Hebertella) crispa, M‘Coy. (1) ,, (Platystrophia) biforata, Schloth. “a, (2), (3), (5) 4, (Dalmanella) elegantula, Dalm., var. nov. drummuckensis. ee Ce ac sonal 6 ) testudinaria, Dalm.? i . (1) ,, (Rhipidomella) prematura, sp. nov. (1) ,, (Schizophorella) fallax, Salt. (1) ,, (Wicolella) actoniz (Sow.), var. nov. asteroidea. (1) Rafinesquina (Playfairia?) felix, sp. nov. (1) Leptxna rhomboidalis (Wilck.), var. y. (1) Stropheodonta corrugatella (Dav). (1) y donax, sp. nov. (1) Strophonella wndata (M‘Coy). : (1) Christiania tenuicincta (M‘Coy). (1) Plectambonites ruralis, sp. nov. 2 (1) “ sericea (Sow.), var. nov. thraivensis. 4 (1) Sérophomena valens, sp. nov. (1) Schuchertella? inexpectata, sp. nov. - (1) Chonetes advena, sp. nov. (1) Triplecia insularis (Kichw.). (1) Cliftonia spiriferoides (M‘Coy). (1) Clitambonites ascendens, Pander. (1) Scenidiuwm Lewisi, Dav., var. nov. asteroidea. (1) Parastrophia thratvensis, sp. nov. (1) Rhynchotreta cuneata (Dalm.), var. ? 966 DR F. R. C. REED ON THE (1) Zygospira? sp. (1), (2), (3), (4) Dayia cymbula (Dav.), var. nov. girvanensis. (1) Whitfieldella? sp. (1) Meristella? sp. SILURIAN. — Mulloch Hill Group. (1) Dinobolus Davidson (Salt.). (1) Lingula Lewisi, Sow. (1) Orbiculoidea (Schizotreta?) Forbesi (Dav.)? (1), (2) Phithedra mullochensis, sp. nov. (1) - siluriana (Dav.)? (1), (2), (5%), (6) Orthis calligramma, Dalm., var. nov. subplicata. (1) ,, (BPlectorthis) plicata, Sow. CD) eas P rustica, Sow., var. rigida, Dav. % (1), (2%), (3), (4) 4, (Dalmanella) crassa, Lindstr. ? (1), 2)" . ) elegantula, Dalm. (1), (2), (3), (4), (6) = ,, (Schtzophorella) mullochensis, Dav. (1), (3), (5) Leptena rhomboidalis (Wilck.), var. 8. (1), (2), (3), (4). (5), (7) Stropheodonta (Leptostrophia) filosa (Sow.), var. nov, mullochensis. (1), (2), (5), (7) 35 (Brachyprion) columbana, sp. nov. ee (1) » sp. : (1) Plectambonites scissa (Salt.). _ (1) s transversalis (Dalm, ), var. nov, 7, amalloclieresiss (1) Stricklandinia mullochensis, sp. nov. Py (1) Pentamerus (Barrandella) wndatus (Sow.). <2 (1) Rhynchotreta cuneata (Dalm.), var. (1) Camarotechia decemplicata (Sow.). (1), (3) ms Mandoveriana (Dav.), var. nov. diversiplicata. (1) re, nucula (Sow.) ? (1) Rhynchospira Bouchardi (Dav.), var. nov. mullochensis. (1), (3), (4) Whitfieldella angustifrons (M‘Coy). ; (1), (2) ss nitida (Hall), var. : > (1}, (3), (4) Coelospira scotica (M‘Coy). (4) Atrypa reticularis (Linn.) ? (1) Meristella, cf. Circe (Barr.). Saugh Hill Group. (1) Dinobolus Davidsoni (Salt.). (1) Lingula Lewisi, Sow. (eras Symonds, Salt. (1) Lingulasma ? tenax, sp. nov. (1) Orbiculoidea (Schizotreta) Forbesi (Dav.). (1) Philthedra mullochensts, sp. nov. (1), (2), (3) ‘3 siluriana (Dav.). (1), (2) Pholidops implicata (Sow.).. (1), (3), (4) Orthis (Plectorthis) rustica (Sow.). ) (113) aera F ) 4 var. nov. paucicostata. (LG) horn _ ) 45. Varerigida, Day, (1) ,, (Platystrophia) biforata, Schloth. (1), (2), (3), (5) ,, - (Bilobites) biloba, Linn. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. (1), (2), (5), (6) Orthis (Dalmanella) basalis, Dalm. (1 (CG) eae Gey ) elegantula, Dalm. WDE xrae wth ; -), ef. vtsbyensis, Lindstr. (1) ,, (Rhtpidomella) polygramma, Sow., var. pentlandica, Dav. 1 (1), (3), (6) Scenidium Lewitsi, Dav., var. woodlandensis, Dav. (1), (3), (5) Leptena rhomboidalis, Wilck., var. e. (1) Stropheodonta arenacea (Salt. MS.). (1), (2), (3), (5) - Jamesoni, sp. nov. (1) Strophonella euglypha (His.). (1), (3) Plectambonites segmentum (Ang.), var. nov. soodlandensis, (1), (3), (5) r transversalis (Dalm.). (1), (3), (5) 5 3 var. Duvali (Dav.). (1), (3) zy is var. nov. tricostata. et) (2), (3), (5) Strophomena antiquata (Sow.), var. nov. woodlandensis. (1) Schuchertella applanata (Salt.). (1), (2), (3), (5) 3 pertinar, sp. nov. (1), (2) Triplecia woodlandensis, sp. nov. (1) Mimulus incertus (Dav.) ? ed : (1) Streptis Hornet, sp. nov. _ (1), (3) Stricklandinia lens (Sow.), var. (1), (2), (3) n lirata (Sow.), var. nov. scotica. (1), (3) Pentamerus (Burrandella) undatus, Sow. (1) Atrypina Barrandei (Dav.) % (1) Glassia compressa (Sow.). (1) Clintonella, cf. vagabunda, Hall and Clarke. (1), (3) Atrypa imbricata, Sow. (1) ,, - reticularis, Linn. (1) Celospira scotica (M‘Coy), var. (3) Hichwaldia personalis, sp. nov. Camregan Group,* _ (1), B. Lingula Symondst, Salt. (1); B. Orbiculoidea sp. ‘ (1), (2) B. Pholidops implicata (Sow.). a (1) Orthis (Plectorthis) rustica, Sow., var. nov. paucicostata. (1), (2), (5) B. ,, (Bilobites) biloba, Linn. (1), (2% B. ,, (Dalmanelia) crassa, Lindstr. ? (1), (5) B. Leptena rhomboidalis, Wilck., var. &. . (1) Stropheodonta arenucea (Salt. MS.). Ch). 7 1 Waltoni (Dav.) ? (1) 7 (Brachyprion) Walmstedti (Lindstv.). (1) ” ( ” ) sp. — (1) B. Strophonella penkillensis, sp. nov. (1) B. Plectambonites scissa (Salt.). (1) Be ra transversalis, var. nov. penkillensis. (1), ie); (5) B. Schuchertella applanata (Salt.). (1) B. Mimulus’ incertus (Dav.). (1) B. Pentamerus oblongus (Sow.). (CL SR eer (Barrandella) undatus. var. nov. penkillensis. (1), (2), (3) Camarotechia landoveriana, var. nov, diversiplicata. (1) ” Sp. * All the species marked B occur at Bargany Pond Burn. 967 968 DR F. R. C. REED ON THE (1) B. Atrypa reticularis, Linn. (1) B. Dayia sp. (1) B. Spirifer plicatellus, var. radiata, Sow. (1) B. Cyrtia exporrecta, Wahl. (1) Rhynchospira (Homeospira) camreganensis, sp. nov. (1) B. Nucleospira pisum (Sow.)? (1), (3) Celospira henvispherica (Sow.). Penkill Group. (1) Lingula Symondsi, Salt. (1), (2) Lingulasma? tenax, sp. nov. 4 (1) Philhedra? penkillensis, sp. nov. - (1) Pholidops implicata (Sow.). (1), (2) Orthis (Dalmanelia) crassa, Lindstr. ? OD ete ee ( s ) elegantula, Daim. Clr seek me ) ef. visbyensis, Lindstr. ; (1), (2) ,, (Rhiptdomella) polygramma, Sow., var. pentlandica, Dav. (1) ,, (Plectorthis) rustica, var. walsalliensis, Dav. y (1), (4) Leptana rhomboidalis (Wilck)., var. & : (1), (4) Stropheodonta (Brachyprion) Walmstedti Gidea) (1), (2) Strophonella penkillensis, sp. nov. (1), (4) Plectambonites transversalis, var. nov. penkillensis, (1), (2), (4%) Schuchertella applanata (Salt.). (4) Triplecia insularis, Kichw. ? (1), (2), (4) Mimulus? incertus (Dav.). (1), (4) Pentamerus oblongus, Sow. (1), (4) ‘5 (Barrandella) undatus, Sow., var. nov. penkillensis. (1), (4) Parastrophia rotunda (Sow.). (1), (2) Glassia obovata (Sow.). (1), (4) Atrypa reticularis, Linn. (1) Spirifer plicatellus, var. radiatus, Sow. (1) Cyrtia exporrecta, Wahl. 4 (2) Nucleospira pisum, Sow. ? ~ (1) Celospira hemispherica (Sow.). CoMPOSITION OF THE FAUNA. In the foregoing pages over 230 species or varieties have been described as occurring in the Ordovician and Silurian rocks of the Girvan area. Amongs' there are over 70 named species new to science and a large number of named va as well as several shells whose specific individuality or relations are not determi vb Most of the others have been previously found in different parts of the British Isles though some have not hitherto been recorded from Girvan. Davipson’s * Girvan brachipods showed only 121 species, but in many cases he included wu one specific name several different forms which have now been separated off. The distribution of the brachiopods on the various stratigraphical horizons sl some interesting features, the abundance in the Stinchar Limestone and Bale Groups being in strong contrast to the poverty in the Camregan and Penkill Groups * Davrpson, op. cit., vol. v, Silur. Suppl., pp. 224-227. ( ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 969 RELATIONS OF THE SuccEssIVE BracHiopop Faunas. I. The Ballantrae rocks have so far afforded such poorly preserved material and SO o scanty a number of brachiopods that it is impossible in the present state of our wledge to give a satisfactory survey of their characteristics. We may note, swever, that all the forms belong to the Atremata, and with the possible exception species none occur in higher beds. The Stinchar Limestone Group has yielded, on the other hand, a plentiful y of species and individuals. Sixty species and varieties are now recorded. of these are peculiar to this horizon or to the Girvan area, but about 20 of the number range up into the overlying group. | . The richness of the Balclatchie Group in brachiopods is remarkable, about 70 pecies or varieties being recognised. The majority are limited to this group of eds, and none of the peculiar local forms pass up into the Whitehouse Group. IV. Only 34 different forms occur in the Whitehouse Group, and of these about e positively limited to it. A few of the others are found in ‘the overlying muck Group, but practically none range down into the underlying beds. The Drummuck Group contains over 40 species and varieties at present nised, and a few forms such as Dayia cymbula, var. gurvanensis, Christiania meta, and Triplecia msularis have come up from the Whitehouse Group. . The Mulloch Hill Group ushers in the Silurian fauna, and its connection with older faunas is slight. Hven such long-ranging forms as Lept. rhomboidalis are "ese ented by distinct varieties. But new genera, especially pentameroids, rhyn- loids, and spire-bearing species make their first appearance. Twenty-nine and varieties of brachiopods from this horizon have here been described. . The Saugh Hill Group has furnished 42 different species and varieties, 1 of which have occurred in the underlying group. But new forms preponder- nd these ally it more closely with the succeeding Camregan Group. (II. In the Camregan Group 27 different brachiopods have here been recorded, ost of these come from the locality termed Bargany Pond Burn and are ll species, so that the individuality of the brachiopod fauna of this group is ked, and the occurrence of some of the species rests on doubtful evidence. . The Penkill Group has yielded 25 species or varieties. Several of these up from the Saugh Hill Group, or occur in the Camregan Group. Parastrophia tunda, Philhedra penkillensis, Strophonella penkillensis, and perhaps Orthis poly- ramma, var. pentlandica, seem to characterise it and to be restricted to it. : AFFINITIES OF THE Fauna. Tn the Stinchar Limestone and Balclatchie Groups we have been led to note the e relations of many of the species with American rather than with English or orth European forms.* In the Whitehouse and Drummuck Groups these affinities * This feature is also observable in other zoological groups, ¢.g. the Bellerophontacea. 970 DR F. R. C. REED ON THE are practically absent or to a great extent unrecognisable ; and though many spec: are local yet others occur in English, Welsh, and Ivish localities in Bala beds; general facies is thus predominantly British. North European relations are also not wanting in the Drummuck Group, especially in the Starfish Bed. The Silurian beds have a marked British stamp, and the American elements seem to have disappeared. In common with northern Europe, we may note the exister of a widespread and dominant facies of a fairly uniform brachiopod fauna accompanied by the development of local species or varieties. oO wt x, + CoNCLUSIONS. — . The Ordovician may be recognised as possessing two main brachiopod faunas, the bee one being limited to the Stinchar Limestone and Balclatchie Groupee the newer one to the Whitehouse and Drummuck Groups. 2. In the case of the Silurian the brachiopod faunas of the Mulloch Hill Saugh Hill Groups form one diviison, while those of the Camregan and Pe Groups are on the other hand linked together; but the distinctive characters these two divisions are not very strongly marked. = 3. Each stratigraphical group of beds has a certain individuality in. thes com position of its brachiopod fauna, and usually possesses some peculiar species or varieti 4, The assemblage of species in the Stinchar Limestone and to a less extent the Balclatchie Group exhibits distinct American affinities. But these are los the succeeding stratigraphical groups, and in the Silurian beds the European fi is practically unmixed with alien elements. 5. The predominance of local and peculiar species and varieties in the ¢ the brachiopod fauna of the Girvan district on every horizon of the Ordovicia Silurian formations is a distinctive feature, and gives to the whole a remarks individuality suggesting a considerable degree of isolation of this marine area. PRINCIPAL LITERATURE. 1848. Satter (in paper by Nicol), Proc. Geol. Soc. Lond., vol. iv, pp. 205, 206. —- 1849. SaurEeR, Quart. Journ. Geol. Soc., vol. v, p. 15, pl. i, fig. 4. “a 1851. Saurer, “List of some of the Silurian Fossils of Ayrshire,” Quart. Journ. Geol. Soc., vol. pp. 170-177, pls. viii, ix, x 1851. M‘Coy, Syn. Brit. Palzxoz. Foss. Woodwardian Museum. = 4 1865. Gray, “Silurian Brachiopoda from near Girvan,” Proc. Nat, Hist. Soc. Glasgow, i, p. 124. 1867. Murcuison, Siluria (4th edit.), pp. 148, 163. ; 1866-70. Davinson, Mon. Brit. Foss. Brachiopoda, vol. iii, pt. vii (Palewont. Soc.), pp. 1-397, pls. i-l (pa 1870-71. Youne, Proc, Nat. Hist. Soc. Glusgow, ii, pp. 21, 166. 1873. Davinson, Trans. Geol. Soc. Glasgow, Paleont. Ser., pt. i. 1876, Armstrone and Youna, Catal. West Scottish Fossils (List of Brachiopods, pp. 17, 18). 1882-1883. Davinson, Mon. Brit. Foss. Brachiopoda, vol. v, Supplement to the Brit. Silur, pp. 64-242, pls. iv-xvii (passim). , Orthis cata, Porambonites Chonetes cea, Stropheodonta . litam bonites Schizocrania . , Camarella , Metacamarella ensis, Orthis sis, Rhinobolus trypina bee ise, Lingulasma xensis, Hemipronites ensis, Protozyga nensis, Bhynchotreta : PAGE 813, 814 815, 816 860, 861 923 915 800 803 913 953 825 902 906 808 831 941 891 820 916 819 947, 948 946 798 923-925 934 840 800 802 946 933, 934 849 841 845 848 848 951 896, 897 803, 804 870 827, 828, 829 923-926 943, 944 952 808 918 945 INDEX OF GENERA AND SPECIES. Chonetes Christiania Circe, Meristella Cliftonia Clintonella Clitambonites Ceelospira ‘ columbana, ciropheodont complectens, Clitambonites compressa, Glassia . concentrica, Rafinesquina confinis, Orthis conoidea, Conotreta Conotreta Conybearei, Camarella corrugatella, Stropheodonta costata, Acrotreta . craigensis, Trematis craigensis, Triplecia crassa, Orthis erispa, Orthis crumenoides, Lingula cuneata, Rhynchotreta cuneatella, Camarella Cyclospira cymbula, Dayia Cyrtia Dalmanella Davidsoni, Dinobolus Davidsoni, Micromitra decemplicata, Camarotcechia deficiens, Strophomena deltoidea, Rafinesquina depressa, Pseudocrania Dinobolus Dinorthis divaricata, Pseudocrania diversa, Cyclospira donax, Stropheodonta dorsata, Orthis drummuckensis, Philhedra duftonensis, Orthis Etchwaldia elegantula, Orthis . emerita, Parastrophia ANS. ROY. SOC. EDIN, VOL. LI, PART IV (NO. 26. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. O7al PAGE 915 902 954 913, 914 947 916, 917 955, 956 896 916 946 864 847 817 817 925 891 815 818 908 849 842 804 942 926 949, 950 948 951 849-856 801 197 943 303 866, 867 826 801 837-840 826 950 892 846 822 832 936 850 927 138 S72 DR F. R. C. REED ON THE. PAGE Etheridgei, Plectambonites - 4 . 875 | mediostriata, Lingula euglypha, Strophonella . : F 5 . 900 | melliflua, Trematis . expansa, Atrypa. : . 947 | Meristella expansa, Rafinesquina. 5 . : . 865 | Metacamarella exporrecta, Cyrtia ; : j : . 951 | Micromitra micula, Siphonotreta fallax, Orthis ; ‘ 2 . : . 858 | Mimulus ; federata, Orthis . : j : . . 851 | mullochensis, Orthis felix, Rafinesquina ; : ; . 867 | mullochensis, Philhedra . filosa, Stropheodonta . : : : . 894 | mullochensis, Stricklandinia flabellulum, Orthis ; : ‘ : . 837 | multifilosa, Pseudocrania Forbesi, Orbiculoidea tee 2 . 821 nasuta, Protorhyncha | girvanense, Rhynchotrema é : . 940 | Nicholsoni, Stropheodonta girvanensis, Orthis ‘ : . 852 | Nicolella Graye, Cliftonia . : : d . -913 | nina, Orthis Gray, Leptella_ . : : : 5 . 873 | nitida, Whitfieldella Greenoughi, Scenidium . ; : : . 920 | nucleoides, Triplecia ‘| Mucleospira Harknessella . 3 eu . ; cj fi 862 nucula, Camarotcechia Hebertella : ; ; : . 840-844 Hemipronites , ; ; 911829900 Onoias , hemispherica, Coelospira . : : ; - 955 | oblongata, Orbicula Heterorthis . 4 : 1 ic ree 847 oblongus, Pentamerus Homeospira . : f ; : - 951, 952 | obovata, Glassia Hornei, Streptis . : : 5 . 912 | Orbiculoidea . 28 | orbis, Zygospira imbricata, Atrypa . ; : : : . 948 | Orthis implicata, Pholidops : : ‘ - «825 | Orthorhynchula incertus, Mimulus . : 3 ; : - 911 | ovata, Lingula inexpectata, Schuchertella ; ; 4 . 906 insularis, Triplecia ; : : - . 909 | Parastrophia . Paterina Jamesoni, Paterula F : , : . 801 | Paterula : Jamesoni, Stropheodonta é : f . 893 | Peachi, Camarella . penkillensis, Strophonella Lapworthi, Orthis . 3 é : A -. 843 | Pentamerus ? Lapworthi, Rhynchotrema eu et - 939 | penultimum, Lingulasma lens, Stricklandinia : . : - _. 931 | personalis, Eichwaldia Leptena : : : Ditiets . 871-873 | pertinax, Schuchertella . Leptella : é : : ‘ . 873-875 | Philhedra Leptostrophia : cee its : . 894, 895 Pholidops Lewisi, Lingula. , : . : . 805 | pisum, Nucleospira . Lewisi, Scenidium, - : : - 921,922 | Platystrophia Lingula sea : : . . 803-807 | Playfairi, Orthis Lingulasma . : : . . - 808-811 | Playfairi, Philhedra lirata, Stricklandinia : : ; . 932 | Playfatria llandeiloensis, Plectambonites . 4 : . 876 | Plectorthis llandoveriana, Camarotechia . : : . 943 | plicatellus, Spirifer polygramma, Orthis Maccullochi, Obolus , ; : : . 799 | Porambonttes medioradiata, Acrothele . ; ; i . 813) porcata, Orthis - PAGE natura, Orthis. . . 1 Figaro k TT BBY R 937 945 7 ; : : : . 826 drata, ieingula 2 . . 806 oflexa, Leptella SS erke: : . 874 imum, Lingulasma : : Pcs) Lingula . ct. : - 807. ingulasma : k 810 tata, Eertassbonites 877- 879 a. . A : : . 863-870 SMe AT es yl. A gg9 aes : ; ; . 814 Atrypa. er . 948 ee Rag, Y= ht . 802 — : : “ . 857 Leptena . pe: : Sil ra. eo. ; eee. DL 9b: 939-941 ee 8, +841 942 i, Rafinesquina 3 s . 868 istrophia_. war . 928 ambonites .. * F ; 879 is 832- 834 trophia (or Camarella) : 12929 ae f 920-922 812, 813 819 858, 859 821 , Saale : ‘ seegUG O07 ubonites . : ; . 880 Tae. 5 . é : . 956 ee 844 maropiia® =. °°.* Pee e OG meembon = 5 pelle Plectambonites i, . 881 Rafinesquina . ¢ ; 2) G69 lectambonites pepe OO4- S80 Rhynchoirema . ; Cem ekooo Clitambonites © . 4 eee SG, Orbiculoidea ae: . 820 nsis, Strophomena pte . 904 hilhedra oe eter ry oLd ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. siluriana, Trematis . Siphonotreta . solvens, Lingula Spirifer spiriferoides, Gliitond stincharensis, Orbiculoidea Streptis . Stropheodonta Strophomena . Strophonella . : subarachnoidea, Aasneaduine : sub-borealis, Orthorhynchula . subcorrugatella, Plectambonites subfilosa, Stropheodonta . subfissicosta, Orthis subplicatella, Orthis Symondsi, Lingula . tenax, Lingulasma . tenuicincta, Christiania . testudinaria, Orthis Thomsoni, Camarella Thomsoni, Hemipronites thraivensis, Orthis . thraivensis, Parastrophia thraivensis, Siphonotreta | transversalis, Plectambonites . Trematis Triplecia undata, Strophonella undatus, Pentamerus -vagabunda, Clintonella valens, Strophomena vespertilio, Orthis . visbyensis, Orthis . Walmstedti, Stropheodonta Waltoni, Stropheodonta . Whitfieldella . woodlandensis, Triplecia Youngi, Parastrophia youngiana, Plectambonites Zygospira 973 PAGE 819 811 807 951 914 821 912 890-899 902-905 900, 901 870 938 886 894 834 835 807 811 902 854- 856 927 919, 920 836 930 811 886-889 818, 819 908-910 901 933, 934 947 905 862 856 897 899 953 910 930 889 944, 945 974 DR F. R. C. REED ON THE ° EXPLANATIONS OF THE PLATES. - Reference Letters. G. = Mrs Gray’s Collection. E.= Geol, Surv. Museum, Edinburgh. M.P.G. = Museum of Practical Geology, Jermyn Street, London. B.M. = British Museum (Nat. Hist.), London. H.G. = Hunterian Museum, Glasgow. S.M. = Sedgwick Museum, Cambridge. Puate I. Fig. 1. Micromitra (Paterina) Davidsoni, sp. nov. Pedicle-valve, apical view, x 10. la. Side- x10. 16. Posterior view, x 10. TET eu Group, Balclatchie. G. (Figured, Davipson, St pl. xvi, fig. 22, as Acrotreta Nicholsoni.) ; , Fig. 2. Ditto. Pedicle-valve, apical view, x10. 2a. Side-view, x10. 20. Posterior view, «10 Same horizon and locality. G. ; Fig. 3. Obolus audaa, sp. nov. Pedicle-valve, x 24. 3a, Outline side-view, x 24. Balclatchie Group Balclatchie. G. (? Figured by Davrpson.) a Fig. 4. Ditto. Pedicle-valve. x24. Same horizon “and locality. G. (? Figured by Davinson.) — Fig. 5. Ditto, ditto, 24. Same horizon and locality. G. ; ae Fig. 6. Ditto. Brachial valve. x3, Same horizon and locality. G. oa Fig. 7. Ditto. Pedicle-valve, internal cast, 23. Same horizon and locality. [13159.] M.P.G. | Fig. 8. Ditto. Pedicle-valve, interior. 23. Same horizon and locality. [4052] M.P.G. Fig. 9. Obolus Maccullochi, sp. nov. Pedicle-valve. 3. Stinchar Limestone Group, Craig G. (? Figured by Davipson.) ; Fig. 10. Ditto. Brachial valve. x2. Same horizon and locality. G. Fig. 11. Ditto. Impression of exterior of pedicle-valve? 3, Same horizon and locality. G. Fig. 12. Ditto, Internal cast of same specimen. x3. Same horizon and locality. G. | Fig. 13. Ditto. Brachial valve, interior. 23. Same horizon and locality. [118/IIL] E. Fig. 14. Paterula balclatchiensis, Dav. Pedicle-valve, 6. Balclatchie Group, Balclatchie (Figured, Davipson, Suppl., pl. xvii, fig. 41.) » Fig. 15. Ditto, ditto. - 6. Same horizon and locality. G. (? Figured by Davidson.) Fig. 16. Ditto, ditto. <6. Same horizon and locality. G. — Fig. 17. Ditto. Impression of exterior of pedicle? valve. ™»6. Same horizon and locality. G. Fig. 18. Ditto. Internal cast of pedicle-valve. <5. Same horizon, Dow Hill. G. ; Fig. 19. Ditto. Pedicle-valve. 8. Same horizon, Balclatchie. [13109.] M.P.G. _ Fig. 20. Ditto. Brachial valve. x10. Ballantrae Group, Benuane Head. [4337.] E. ng Fig. 21. Paterula? albida. Internal cast of pedicle-valve. x5. Whitehouse Group, Whitehous Bay. G. i Fig. 22. Ditto. Brachial valve. 5. Same horizon and locality. G. 3 y Fig. 23. Ditto, External impression of brachial valve. x5. Same horizon and locality, G. — Fig. 24. Ditto, Brachial valve of same specimen, with shell preserved. 5, Same horizor locality. G. E Fig. 25. Ditto. Internal cast of pedicle-valve. 5. Same ireneen and locality. G. Fig. 26. Paterulat Jamesoni, sp. nov. Internal cast of brachial? valve. x 10. Starfish Drummuck Group, Thraive Glen. G. SS Fig. 27. Ditto. Interior of same specimen. x10. G, e ; Fig. 28. Ditto. Internal cast of brachial? valve. Same horizon and locality, G. Fig. 29. Rhinobolus? balclatchiensis, sp. nov. Internal cast of pedicle-valve, 23. Balelatel Group, Balelatchie. G. it + - ‘ on - ORDOVICIAN AND SILURIAN ‘BRACHIOPODA OF THE GIRVAN DISTRICT. 975 Prare II. é ‘I: Rhinobolus? balclatchiensis, sp. nov. Internal cast of brachial valve. x 24. Balclatchie , Balclatchie. G. ie, 2. Ditto, Brachial valve. x2. Same locality and horizon. G, 3. Lingula amabilis, sp. nov. Pedicle-valve. x 2}. Balclatchie Group, Ardmillan. G. | 4, Ditto, ditto. <2. Same horizon and locality. G. (Figured, Davipson, Suppl., pl. xvii, g.5, Ditto. Brachial valve, showing partly internal cast. x 3. Balelatchie Group, Balclatchie. G. 6, Ditto. Pedicle-valve. 3. Same horizon and locality. G. 7. Ditto. Internal cast of pedicle-valve, with portion of shell attached. x4. Same horizon lity. (13155.] M.P.G. . 8. Ditto? Pedicle-valve. <3. Same horizon and locality. G. ). Lingula brevis, Portl., var. nov. angustior, Pedicle-valve. 4.. Balclatchie Group, eG. : . Ditto, ditto. 4. Same horizon and locality. G. . Lingula brevis, Portl.,. var. nov. carrickensis. Pedicle-valve. <4. Whitehouse Group, . Ditto, ditto. x5. Same horizon and locality. G. : . Ditto. Brachial valve. 5. Same horizon and locality. G. . Lingula’? crumenoides, sp. nov. FPedicle-valve. x14. Ballantrae Group, Bennane Head. 174.) £E. ; ‘14a. Ditto. Outline side-view of same specimen. x 14. 15. Lingula mediostriata, sp. nov. Pedicle-valve. 3. Drummuck Group, Thraive Glen, G. s. 15a. Ditto. Portion of surface of same specimen. x 8. . Ditto, ditto. Internal cast. x23. Starfish Bed, Drummuck Group, Thratve Glen. G. a Lingula pseudoquatrata, sp.nov. Pedicle-valve. 5. Balclatchie Group, Dow Hill. G. . Ditto, ditto. 5. Same horizon and locality. G. . Lingula solvens, sp. nov. x13. Drummuck Group (Starfish Bed), Thraive ee ‘Ga. a. Ditto. Portion of surface of same specimen. x 6. ° Lingulgsma’ ardmillanense, sp, nov. Internal cast of pedicle-valve. 2}. Balclatchie illan, “Go . Ditto. Internal cast of brachial valve. x24. Same horizon and locality. G. . 22. Ditto. Interior of shell of same specimen. 24. . 23. Ditto. Portion of surface of shell. <8. Same horizon and locality. G. me canadense (Billings). x14. Balelatchie Group, Balclatchie. G. (Figured a. Ditto: Pontion of surface of same specimen. x 10. 25. Lingulasma’ penultimum, sp. nov. Pedicle-valve. x3. Drummuck Group (Starfish Bed), . Ditto. Portion of shell showing lamination and striation of interior. x8. Same horizon 26a. Ditto. Portion of external surface of shell of same specimen, showing granulation. x 15. . Lingulasma’? pulcherrimum, sp. nov. Brachial valve. <2. Balclatchie Group, G. 27a. Ditto. Portion of external surface of same shell, showing ornamentation. x 15. ig. 28. Impression of surface of another specimen (figured Pl. III, fig. 1) showing ornamentation. 15. Same horizon and locality. G. {FT &, 73968 976 DR F. R. C. REED ON THE Puate III. Fig. 1. Lingulasma? pulcherrimum, sp. nov. Brachial valve (with pedicle-valve displaced behind it), x2. Balelatchie Group, Ardmillan. G. | . ; SS Fig. 2. Lingulasma’? questor, sp. nov. Internal cast of brachial valve. x 2. Se Ardmillan Brae. G. (Figured, Davipson, Suppl., pl. xvii, fig. 20.) . Fig. 3. Ditto, ditto. <2. Same horizon and Heeahy G. (Bigured Davinson, Suppl.; fig. 21.) — Fig, 4. Ditto. Pedicle-valve. x2. Balclatchie Cao, Balclatchie. G., Fig. 4a. Ditto. Portion of surface of same specimen, showing ornamentation, x 10,” Fig. 5. Lingulasma ? tenax, sp. nov. Pedicle-valve. x2. Saugh Hill Group, Woodland Point, Fig. 5a. Ditto. Portion of surface of same specimen, showing ornamentation. | “x 20. ' tea Fig. 6. Ditto. Brachial valve. 2. Same horizon and locality. G. Fig. 7. Siphonotreta? thratvensis, sp. nov. Pedicle-valve. x BL Drammick | a Glen. G. a" 7 < ae Fig. 8. Ditto. Internal cast of brachial valve. 5. Same horizon and locality. G. es Fig. 9. Schizambon scoticus (Dav.). Pedicle-valve. - x 23. Stinchar Limestone bees Craighead (Figured, Davipson, Suppl., pl. xvi, fig. 32.) Fig. 10. Ditto. Brachial valve. x24. Same horizon and ae G. (Figured, Davinson, pl. xvi, fig. 33.) . : ili ah Wo he Fig. 11. Ditto, ditto. x3. Same horizon, Minuntion. G.— ee Fig. 12. Ditto, ditto. 3. Same horizon, Craighead. G. . a Fig. 13. Ditto. Pedicle-valve. x24. Same horizon and locality. G. IP RSENG a pl. xvii, fig. 6, as Lingula Ramsayi.) 1% ‘ ie Fig. 14. Ditto. Umbonal region of pedicle-valve. x4. Same horizon and locality.” a “é Fig. 15. Schizambon scoticus, Dav., var. Internal cast of brachial valve. x 24. - Starfish, Drummuck Group, Thraive Glen. G. . Fig, 16. Ditto. Impression of exterior of same specimen, x 23. ; ; Fig. 17. Ditto. Impression of exterior of pedicle-valve. 24. Same horizon and ibeallieas Bre: . Fig. 18. Acrothele medioradiata, sp. nov. Internal cast of pedicle-valve. x 5, Balclatehi Balclatchie. G. : aan Fig. 19. Ditto, Impression of exterior of same specimen. ~~ 5. i ’ Fig. 20. Ditto. Internal cast of pedicle-valve. 3. Same horizon and locality. (13105. 7 ™. 3 Fig. 20a. Ditto. Outline side-view of same specimen. x 3. i Fig. 21. Ditto. Internal cast of pedicle-valve. x3. Same horizon and locality. [15104] | w Fig. 22. Ditto. Interior of brachial valve. 4. Same horizon and locality. G. Fig. 23. Acrothele 1 sp.a. Brachial valve. 5. Same horizon and locality, G. — see Fig. 24. Ditto? Brachial valve. 5. Same horizon and jopalaty & Le a ae Fig. 25. Ditto? ditto. 5. Same horizon, Ardmillan. G. OM a Fig. 26. Acrothele (Redlichella?) sp. b. «8 Same horizon, Balelatchie. G. Fig. 27. Ditto. 8. Same horizon and locality. G. ; Fig. 28. Acrotreta? sp. a. Brachial valve. x 8. ‘Stinchar Limestone Group, crsighetan ‘@ Fig. 29. Acrotreta? sp. b. Interior of brachial valve. x 5. Same horizon and locality. G..* Fig. 30. Ditto. Counterpart of same specimen. 5. er vt Fig. 31. Ditto, Interior of brachial valve. x5. Bennane Burn, R. Stinchar. [2809. 1 iE Puata 1V.;° * bo eed 20) to on Fig. 1. Conotreta conoidea, sp. nov. Pedicle-valve, posterior view. x5. Stinchar Limes Group, Craighead. G. (Figured, Davipson, Suppl., pl. xvi, fig. 21, as Acrotreta Nicholsont.) Fig. 2. Ditto. Interior of pedicle-valve (partly exposed by removal “of shell). x4. Same and ipeality, G. ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 977 _ Fig. 3. Ditto. Side-view of pedicle-valve. x5. Same horizon and locality. G. *‘Fg. 4. Ditto. Anterior view of pedicle-valve. 5. Same horizon and locality. G. od ‘Fig. 5. Ditto. Side-view of pedicle-valye. 5, Balclatchie Group, Balclatchie. G. .6. Trematis craigensis, sp. nov. Interior of brachial valve. x2. Stinchar Limestone Group, ad. G. ; , tig. 7. Ditto, ditto. 23. Same horizon and locality. G. Fig . 8. Ditto. Portion of surface of shell. 15. Same horizon and locality. G. 9. Ditto. Impression of portion of surface of shell, x15. Same horizon and locality. G. - Trematis melliflua, sp. nov. Brachial valve. 4, Balclatchie Group, Balclatchie. G. ~ 10a. Ditto. Portion of surface of same specimen, x 24, “in Trematis siluriana, Dav. Imperfect pedicle-valve. 14. Starfish Bed, Drummuck Group, fie G. ~ | . Schizocrania asteroidea, sp. nov. Brachial valve. x3. Starfish Bed, Drummuck Group, mn. G. a. Ditto. Portion of surface of same specimen. x 8. . Ditto. (Shell partly eroded.) 22. Same horizon and locality. G. . Ditto. Internal cast of brachial valve. 23, Same horizon and locality. G, 15. Orbiculoidea Armstrongi, sp. nov. Interior of pedicle-valve. x3. Starfish Bed, Drum- roup, Thraive Glen. G. a. 16. Ditto. ‘Impression of exterior of pedicle-valve. 3, Same horizon and locality, G. . Ditto. Brachial valve. x3, Same horizon and locality. G. : Orbiculoidea shallochensis, sp. nov. Pedicle-valve. x5. Whitehouse Group, Whitehouse Fi 19. Ditto. Brachial valve. x6. Same horizon and locality. G. 20. Ditto. Interior of pedicle-valve. x24. Same horizon, Shalloch Mill. G, 4 Orbiculoidea stincharensis sp. nov. Pedicle-valve. x3. Stinchar Limestone Group, Craig- P, 2. Ditto, ditto. 38. Same horizon and locality. G. 3. Ditto. Interior of pedicle-valve. 3. Same horizon and locality. G. 4, Ditto, ditto. x3. Same horizon and locality. G. g. 25. Orbiculoidea (Schizotr eta 1) Ere: Dav. Brachial valve. x24. Saugh Hill Group, Wood- g, 26. ‘Philhedra dr Eoiuickensis, sp. NOV. Upper valve, internal cast. x3. Starfish Bed, Drummuck rou} Thraive Glen. G. 27. Ditto. Counterpart of same specimen. Bs g. 28. Ditto. Interior of valve. x5. Same horizon and locality. G. 29. Ditto. Outline side-view of upper valve. 3, Same horizon and locality. G. go. 0. Philhedra. -mullochensis, sp. nov. Internal cast of upper valve. x24. Mulloch Hill group, B 31. Ditto, dito. 13x 21, Te 6 horizon and locality. G. g. 32. Ditto, ditto. x2. Same horizon and locality, G. 32a, Ditto. Outline side-view of same specimen. x 2. g. 33. Ditto. ‘Internal cast of upper valve, «2. G. 33a. Ditto. Portion of surface of same. x 10. 34. Philhedra' penkillensis, sp.nov, x4. Penkill Group, Peniail. G. 34a, Ditto. Outline side-view of same specimen, x 4. ; 35, Philhedra pyar 3 nov. Internal cast of upper valve. x2. Balclatchie Group, Dow z pe: 36. Ditto, piaipicscen of eolere one of same sell ye So Ge ane. ‘eae Fig. “317. Philhedra siluriana (Dav.). Internal cast. x 22. Saugh Hill Group, Woodland Point. G. Fig. 38. Philhedra sp. Internal cast of upper valve. 3. Starfish Bed, Drummuck Group. B.M. _ Fig. 38a, Ditto. Outline side-view of same specimen. x 3. 978 DR F. R. C. REED ON THE Puate V. Fig. 1. Pholidops antiqua (Schloth.)? Interior of upper valve. x6. Whitehouse Group, Shalloch Mill. G. + Fig. 2. Ditto. Internal cast. <6. Same horizon and locality. G. 7 Fig. 3. Ditto. Exterior. x6. Same horizon and locality. G. Fig. 4. Pholidops implicata (Sow.). Internal cast. x6. Saugh Hill Group, Newlands. G, Fig. 5. Pseudocrania multifilosa, sp. nov. Upper valve. x3. Starfish Bed, Drummuck Group, Thraive Glen, G. a Fig. 5a. Ditto. Outline side-view of same. x 3. 4 Fig. 5), Ditto. Portion of surface of same specimen. x 12. i Fig. 6. Pseudocrania aff. depressa- (Kichw.). Internal cast of brachial valve. 23. Balclatchie Group, Ardmillan. x G. 3 Fig. 6a. Ditto. Portion of surface of same internal cast. x 8. ¥ Fig. 7. Orthis calligramma, Dalm., var. nov. craiyensis. Complete shell, from brachial side. x 1}, Stinchar Limestone Group, Craighead. G. Be Fig. 8. Ditto. Brachial valve. 1}. Same horizon and locality. G. Fig, 9. Ditto. Posterior view of complete shell. 14. Same horizon and locality. G. r Fig. 10. Orthis calligramma, Dalm., var. nov. subplicata. Impression of exterior of brachial valve. x1}. Mulloch Hill Group, Mulloch Hill Wood. G. f, Fig. 11. Ditto. Internal cast of brachial valve. 12. Same horizon and locality. G. Fig. 12. Ditto. Impression of exterior of brachial valve. x1}. Same horizon and locality, [418 5.] M.P.G. 3 Fig. 13. Ditto. Internal cast of pedicle-valve. 14. Same horizon and locality. G. Fig. 14. Ditto. Impression of exterior of same pedicle-valve. «14. Same horizon and locality. Fig. 15. Ditto. Internal cast of pedicle-valve. 14. Same horizon and locality. G. Fig. 16. Orthis calligramma, Dalm., var.? Internal cast of brachial valve. x3. Starfish Drummuck Group, Thraive Glen. G. Fig. 17. Ditto. Impression of exterior of same shell. x3. G. Fig. 18. Ditto. Internal cast of brachial valve. x3. Same horizon and locality. G. Fig. 19. Ditto. Impression of exterior of same shell. x3. G. Fig. 20. Orthis Playfatri, sp. nov. Pedicle-valve. x2. Stinchar Limestone Group, Craighead. eG Fig. 21. Ditto, ditto. x2. Same horizon and locality. G. 4 Fig. 22. Ditto. Brachial valve. x24. Same horizon and locality. G. ae Fig. 23. Ditto. Interior of brachial valve. x24. Same horizon and locality. G. Fig. 24. Ditto. Pedicle-valve. x2. Same horizon and locality. G. Fig. 25. Ditto. Impression of exterior of brachial valve. x2. Same horizon and locality. G. Fig. 26. Ditto. Outline side-view of complete shell. x2. Same horizon and locality. G. Fig. 27. Ditto. Internal cast of pedicle-valve. x2. Balclatchie Group, Balclatchie. G. Fig. 28. Ditto. Internal cast of brachial valve. x24. Same horizon and locality. G. Fig. 29. Ditto, ditto. x2. Same horizon and locality. G. Fig. 30. Ditto. Pedicle-valve. x14. Whitehouse Group, Shalloch Mill. G. Fig. 31. Ditto, ditto. x14. Same horizon and locality. G. Fig, 32. Ditto, ditto. x14. Same horizon and locality. G. ° “a Fig. 33. Orthis (Plectorthis) ardmillanensis, sp. nov. Pedicle-valve. x2. Balclatchie Group (con- glomerate), Balclatchie. G. ‘ : ‘ Fig, 33a. Ditto, Median transverse section of same. x 2. Fig, 34. Ditto. Internal cast of pedicle-valve. x3, Balclatchie Group, Ardmillan. G. Fig. 35. Ditto. Impression of exterior of brachial valve (with hinge-area of pedicle-valve). Same horizon and locality. G. Fig. 35a. Ditto. Portion of surface of same specimen, showing ribbing. x 6. , Fig. 36. Ditto. Brachial valve of complete shell, x2. Stinchar Limestone Group, Craighead. — f _ ORDOVICIAN AND SILURIAN BRACHTOPODA OF THE GIRVAN DISTRICT. B78) Fig. 36a. Ditto. Outline side-view of same shell. x 2, _ Fig, 37. Ditto, var. Impression of exterior of brachial valve. x24. Balclatchie Group, Ardmillan, G. Fig. 38. Ditto, ditto. x2. Same horizon and locality. G. Fig. 39. Ditto. Internal cast of same shell. x 2. tig. 40. Orthis (Plectorthis) duftonensis, Reed, var. Impression of exterior of brachial vaive. . x 2. atchie Group, Ardmillan. G. ig. 41. Ditto. Internal cast of same shell. x2. G. — Puate VI. 1. Orthis (Plectorthis) duftonensis, Reed, var. Internal cast of brachial valve. x2. Balclatchie Ardmillan. (Figured, Davinson, Suppl., pl. xiii, fig. 26.) G. ig. 2. Ditto. Internal cast of pedicle-valve. x2. Same horizon and locality. G. Fig. 3. Orthis (Plectorthis) rustica, Sow. Interior of pedicle-valve. x1}. Saugh Hill Group, Woodland 4. Ditto. Interior of biachial valve. x1}. Same horizon and locality. G. Ditto. Interior of pedicle-valve. x14. Same horizon and locality. G. Ditto. Internal cast of pedicle-valve. x14. Same horizon and locality. G. Ditto. Internal cast of brachial valve. x14. Same horizon and locality. G. Ditto. Internal cast of pedicle-valve. x14. Same horizon and locality. G. Orthis (Plectorthis) rustica, var, nov. paucicostata. Interior of brachial valve. x1}. Saugh Woodland Point. G. g. 10. Ditto, ditto. «14. Same horizon and locality. G. 11. Ditto. Pedicle-valve. x14. Same horizon and locality. G. 12. Ditto, ditto. x14. Same horizon and locality. G. 3. Orthis (Plectorthis) rustica, var. rigida, Dav. Internal cast of pedicle-valve. x2. Saugh Hill odland Point. G, 14. Ditto, ditto. x2. Same horizon and locality. G. 5. Ditto. Interior of brachial valve. x2. Same horizon and locality. G. 16. Ditto, ditto. 2. Same horizon and locality. G. . Ditto. Exterior of brachial valve. x2. Same horizon and locality. G. . Ditto, ditto. x2. Same horizon and locality. G. . Ditto. Pedicle-valve. x14. Same horizon and locality. G. . Orthis (Plectorthis) rustica, var. walsalliensis, Dav. x1}. Penkill Group, Penwhapple . Orthis (Plectorthis) subfissicosta, sp. nov. Pedicle-valve. x2. Stinchar Limestone Group, G. , Portion of surface of shell (from squeeze), x5. Same horizon and locality. G. . Internal cast of brachial valve. x24. Same horizon, Ardmillan. G. 5. Ditto. Same shell (from squeeze of impression). x24. G., 26. Ditto. Internal cast of pedicle-valve. x2. Same horizon, Balclatchie. G. 27. Orthis (Plectorthis) thraivensis, sp. nov. Pedicle-valve (from squeeze). x1}. Starfish Bed, uck Group, Thraive Glen. G. ig. 28. Ditto. Internal cast of pedicle-valve. x14. Same horizon and locality. G. 29. Ditto. Internal cast of brachial valve. x14. Same horizon and locality, G. 30. Ditto, ditto. x1}. Same horizon and locality. G. 31. Ditto, Posterior view of internal cast of complete shell, x2. Same horizon and locality. G, 3. 32. Ditto. Portion of surface of shell (from squeeze). x4. Same horizon and locality. G. S. ROY SOC. EDIN., VOL. LI, PART IV (NO. 26). 139 980 DR F. R. C. REED ON THE Puatre VII. Fig. 1. Orthis (Plectorthis) thraivensis, sp. nov. Brachial valve (from squeeze of impression). x 1}. Starfish Bed, Drummuck Group, Thraive Glen. G. ; . Fig. 2. Orthis (Dinorthis) flabellulum, Sow., var. nov. carrickensis. Pedicle-valve. Nat, size. Stinchar Limestone Group, Craighead. G. (Figured, Davipson, Suppl., pl. xi, fig. 5.) . Fig, 3. Ditto. Brachial valve. Nat. size. Same horizon and locality. G. (Figured, zid., pl. xiii, fig. 3.) ; Fig. 4. Ditto. Internal cast of pedicle-valve. x14. Balclatchie Group, Ardmillan. G. > Fig. 5. Ditto. Impression of exterior of pedicle-valve. x14. Same horizon, Balelatchie. G, Fig. 6. Ditto. Internal cast of same shell. x14. G. 2 Fig. 7. Orthis (Dinorthis) porcata, M‘Coy. Internal cast of brachial valve. x14. Starfish Bed, Drummuck Group, Thraive Glen. G. a Fig. 7a. Ditto. Posterior view of same specimen. x 1}. . Fig. 8. Ditto. Internal cast of cardinal and muscular area of brachial valve, x25, Same horizon - and locality. G. ; 4 Fig. 9, Ditto. Internal cast of pedicle-valve. x14. Same horizon and locality. G. Fig. 10. Ditto. Same pedicle-valve (from squeeze of impression), x1}. G. Fig. 11. Ditto. Left half of brachial valve (from squeeze of impression). x14. Same horizon and locality. G. oe Fig. 12. Ditto. Median vertical section through complete internal cast. x14. Same horizon and locality. G. = Fig. 13. Orthis (Hebertella) balclatchiensis, Dav. Pedicle-valve. x2. Balclatchie Group (conglom ) Balclatchie. G. . Fig. 13a, Ditto. Outline side-view of same specimen, x 2. Fig. 14. Ditto. Pedicle-valve. x2. Same horizon and locality. G. . Fig. 15. Ditto. Brachial valve (from squeeze of external impression). x2. Balclatchie Group, Ardmillan, G. ee a Fig. 16. Ditto, ditto. x24 (from squeeze). Same horizon and locality. G. (Figured, Davipson, Suppl., pi. xiii, fig. 8.) : eee — Fig. 17. Ditto. Brachial valve. x2. Stinchar Limestone Group, Craighead. G. 7 Fig. 18. Ditto. Internal cast of posterior portion of pedicle-valve. x2}. Balclatchie Group, Balclatchie. G. a Fig. 19. Ditto. Internal cast of pedicle-valve. x24. Same horizon, Ardmillan. G. : Fig. 20. Ditto. Internal cast of median cardinal region and muscular area of brachial valve, x 2h Same horizon and locality. G. 4 Fig. 21. Orthis (Hebertella) bellatri: ED sp. nov. Impression of exterior of brachial valve, x1} Balclatchie Group, Dow Hill. G. j Fig, 22. Ditto. Median transverse section of complete internal cast of same shell. x li. G. Fig. 23. Ditto. Brachial valve. x1}. Balclatchie Group (conglom.), Balclatchie. G. Fig. 24. Ditto. Complete shell, pedicle-valve. x14. Same horizon and locality. G. Fig. 24a, Ditto. Outline side-view of same specimen. x 14, Fig. 25. Ditto. Impression of exterior of brachial valve. x 14. Balclatchie Group, Balelatchie. Fig. 26. Ditto. Internal cast of same shell. x14. G. Fig. 27. Ditto. Impression of exterior of pedicle-valve. x14. Same horizon and locality, G. Fig. 28. Ditto, Internal cast of pedicle-valve. x14. Same horizon and locality. G. > Puate VIII. Fig. 1. Orthis (Hebertella) bellatria, sp. nov. Internal cast of pedicle-valve. x14}. Balclatchie Gro up, Balclatchie. G. Fig. 2. Ditto. Internal cast of brachial valve. x1}. Same horizon and locality. G. : ORDOVICIAN AND SILURIAN BRACHIOPODA OF THE GIRVAN DISTRICT. 981 a Pig. 3. .Orthis (Hebertella) crispa, M‘Coy. Brachial valve (from squeeze of impression). x2. Starfish Bed, Drummuck Group, Thraive Glen. G. qg Vig. 4. Ditto. Internal cast of brachial valve. x2. Same horizon and locality. G, Fig. 5. Ditto. Internal cast of pedicle-valve. x2. Same horizon and locality. G. _ Fig. 6. Ditto. Internal cast of brachial valve. x2. Same horizon and locality. G. - Fig. 7. Ditto. Internal cast of pedicle-valve. x2. Same horizon and locality. G. Fig. 8. Orthis (Heberlella) Lapworthi, Dav. Pedicle-valve- x3. Balclatchie Group (conglom.), G. (Figured, Davipson, Suppl., pl. xiii, fig. 9.) . 8a. Ditto. Outline side-view of same specimen. x 3. . 9. Ditto. Brachial valve. x3. Same horizon and locality. G. (Figured, Davipson, LS In ordinary circumstances it would not be necessary, in what is mainly a paleo- botanical paper, to enter into a detailed geological description of the district. But as the Wyre Forest Coal Field, though mapped on the one-inch scale by the Geological Survey over fifty years ago, has not been described in any official memoir, it has ‘ e een thought advisable to lay before the student a brief account of the strati- graphical sequence and geographical distribution of its various Carboniferous divisions. The reader will thus more readily appreciate the ae that still await solution. The Forest of Wyre Coal Field { commences at Brdehorhs on the Severn, in * Quari. Journ. Geol. Soc., vol. lxi, p. 320, 1905. + See GoprErt, Floram Frasier igemtetsans Transitionis = Die fossile Flora des Ubergangsgebirges, Breslau, 1852. t The coal etl: is included in the Old Series One-Inch Geological Maps, Sheets 61 S.E. (1852-55, revised 1868) Bind 5B N.E. (1853-55), and in the New Series One-Inch Oniirenee Maps, Sheets 167 and 182, It covers parts of _ Shropshire, Worcestershire, and Staffordshire. 1002 DR R. KIDSTON, MRT. C. CANTRILL, AND MR E. E. L. DIXON. Shropshire, where the Coalbrookdale Coal Field reaches its southern limit. There is no complete break between the two, but the outcrop of the Coal Measures is reduced to a narrow band between the Old Red Sandstone on the west and the so-called Permian rocks on the east. From Bridgnorth the Coal Field extends southward, broadening out till it reaches the latitude of Bewdley, in Worcestershire, beyond which it is cut off on the south by the emergence of the Lower Old Red Sandstone of Heightington. In the south-west a narrow neck of Coal Measures, half a mile broad, connects this main northern area with a peninsular portion, which extends in a south-eastward direction from Bayton as far as the Silurian ridges of the Abberley Hills. ; To these two portions must be ‘added the anticlinal tract of Trimpley, where a a broad fold of the Old Red Sandstone is bounded on each side by a narrow strip of Coal Measures. These strips run from near Bewdley in a north eastward direction for about 5 miles, and unite in the neighbourhood of Compton, where the Coal — Measures of the Forest of Wyre are separated from those of South Staffordshire by a distance of only 5% miles. | The broad outlines of the geological structure are shown on the accompanying sketch-map (text-fig. 1), which is based on that of the Geological Survey. The out- — crops of the coals and Spirorbis-limestone have been taken partly from that authority — and partly from the maps published by Mr Danten Jones in 1894 and by myself in| 1895. As the base of the Sulphur Coal Group has not yet been defined, the line on — the map has been taken at or not far below the Main Sulphur Coal. From~ Billingsley to Kingswood and thence to Button Oak the line is purely diagrammatic, | and must remain so till the district is properly surveyed. j On the west and south the Coal Field is bounded by the Lower Old Red Sandstone, on which the Coal Measures rest unconformably. The Carboniferous Limestone is absent. The so-called Millstone Grit, shown on the Geological Survey map at one spot only, near Bagginswood, is believed by Mr EH. K. L. Dixon to be merely a white — sandstone in the Coal Measures. LHastward, the Coal Measures pass under, or are faulted against, various subdivisions of the Trias. Between the Triassic rocks on the east and the Old Red Sandstone on the west the following descending sequence has been established :— “Permian” and Keele Beds, : Sulphur Coal Group, with pyritous coals, and Spirorbis-limestone. The chief seams are the Brock Hall and the Main Sulphur Coals. Sweet Coal Group, with non-pyritous (“sweet”) coals and good ironstones. — Coat Maasures At the present time the sweet coals are being raised in the Highley region at the Billingsley, Kinlet, and Highley Collieries. In the Mamble region the sulphur coals are worked at three or four collieries between Bayton and Abberley. In the past the coals of both groups have been worked over most of the Coal Field, but t generally in a small way, and not far from their outcrops. ee, (BRIDGNORTH. Celel.:/ SEPA is St Oe oe DONG: Sy hs 3 Oe ea Nea OS) “WeQuatford ~-.°, °°. as EOP Rs IP Chae DE +Kinlet Beek t+Earnwood HIGHLEY REGION a7 Were! eh a ee eee ee er en eee ee RMIANIB le 92 SCS Se (as are ace =a] ge ‘PERMIAN’ & KEELE eae BEDS ee J.C. SULPHUR COAL GROUP + += Staffordian »L. | with coals & Cpsere | Spirorbis Limestone SWEET COAL GROUP with coals Peet eect eh aera a Be ceca PRE-COAL-MEASURE 5 - . 4 oncois Ss eres TFC.C. ~ Westphalian Text-FIc. 1.—Geological sketch-map of the Wyre Forest Coalfield. 1004 DR R. KIDSTON, MR T. C. CANTRILL, AND MR E. E. L. DIXON. According to memoranda made by Mr DanieL’Jones, the Main Sulphur Coal being worked in 1869 at the Harpsford or Harpswood Colliery, at the northern end of the Deuxhill outlier, where the seam lies about 2 yards above the Old Red Sand stone floor; it is 4 feet 2 inches thick, and the Spirorbis-limestone lies 21 — above it. In a deep pit near Eardington, on the main outcrop farther east coal was found to be 10 feet above the Old Red Sandstone, the Brock Hall G being 34 yards still higher. On Chelmarsh Common the latter coal was we close to the farm from which it is named; it is presumably the lower of the t CHELMARSH BILLINGSLEY Keele Beds RADSTOCKIAN [==> Z ; < Brock Hall Coal ay tg ie) LL < SpirorbisLst. iE Main Sulphur Coal n N N x uz SS < Ni < EF “ 6 ‘ i Sweet o N F \ Coals 0) \ i} $ Trxt-FIG. 2,—Diagram showing successicn at Chelmarsh and at Billingsley. upper seams shown on the Survey map; below it the Spzvorbis-limestone crops | in Borle Brook, while still lower the Main Sulphur Coal was proved, close above Old Red Sandstone floor. It will thus be seen that, in this part of the coal the relationship of the Sulphur Coal Group to the Old Red Sandstone is tha _ presented diagrammatically in the left-hand column of text-fig. 2. South of Billingsley, however, a new feature appears. At the Billingsley Pit (a little west of the present Billingsley Colliery), the Main Sulphur Coal, i of being underlain at once by the Old Red Sandstone, is underlain by 48 yards clunches, rocks, binds, and ironstones, below which follow several sweet coals. — conditions here, then, are those shown in the right-hand column of the diagran (text-fig. 2). These sweet coals crop out and have been worked at Billingsl Harcott, Bagginswood, and Baveney Wood. position assigned to it upon the map (text-fig. 1); but little or nothing is known of it till we reach Winwoods and Kingswood, where it was worked at a depth of about 25 yards. The Spirorbis-limestone must have been passed through in the shafts, as pieces of the stone are to be seen at the pit-mouths. The seams worked at Baveney Wood were sweet coals ; and it is significant that they are said to have shown signs _ of having deteriorated in number and thickness as compared with their development at Harcott and Billingsley. A coal -eut through by the Birmingham aqueduct at Silligrove was ‘said to be a sweet coal, and Mr Dante. Jonss noted a coal- crop, which he believed to belong to the Sweet Coal Measures, ‘Ina cutting on the Tenbury railway opposite Furnace Mill. These outcrops may be a southerly extension of the Sweet Coals of Baveney Wood. Still farther south, in the Dowles ‘Valley region, no such sweet coals are known; and their absence may be due to their having become unworkable or to their having thinned out completely, as is suggested in the diagram forming text-fig. 3. Hast of Winwoods and Kingswood, the Main Sulphur Coal was worked below the Spirorbis-limestone by pits at Button Oak, from which point its outcrop is apparently shifted northward by a fault, and what is believed to be this coal _ is exposed in a cutting on the Severn Valley railway a few yards south of Arley Station. It then crosses the Severn and runs through Eymore Wood to Shatterford, the outcrop being marked by numerous old pits. At Shatterford it was worked at a depth of 176 yards; but a deep sinking, “carried down below the sulphur coals in 1850-60, though ‘it proved a great thickness of measures, failed to prove any sweet coals worth getting. It was probably the Main Sulphur Coal that was worked in pits and drifts east of Brettels. Beyond this point nothing is known of it; but it presumably turns round the end of the Trimpley anticline and is then cut off by the faulted margin of the Keele Beds. In the Mamble region the Main Sulphur Coal, which _ appears to be represented by two seams separated by about FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. > 2 S Se Q SS / 1 / / i f ' t ! \ \ A ol Py \ ms al is) ~ Sy (QQ) Bh 3 eS 5S So Ss QB -———-—— S = Re =~ 3 B----- = = a) < S 2 St 8 S 7) a wi a Ww 4 wW uw x > 0 @ ------- 1005 Still farther south, what is probably the Brock Hall Coal was worked at Earnwood ‘and the Birch. The Main Sulphur Coal should crop farther west, somewhere in the Trxt-ric, 8.—Diagrammatic longitudinal section from north to south across the Wyre Forest Coalfield. 1006 DR R. KIDSTON, MR T. C. CANTRILL, AND MR HE. HE. L. DIXON. 7 yards of clod, has been shown on the Survey map; above it the Spzrorbis-limestone was detected at many points by Mr Dante. Jonss.* " MURCHISON in 1839 was aware that in the Coalbrookdale Coal Field two groups of Coal Measures are present: a newer group, containing a thin band of Spirorbis-lime- stone and several sulphur coals similar to those of the Shrewsbury and Lebotwood Coal Fields, and an older group, in which the chief coal-seams and ironstones are contained ; but he appears to have inferred that in the Forest of Wyre Coal Field only the newer group is represented. It has long been known that this inference was incorrect. The researches of Mr Danie Jonzs,t begun in 1868 in connection with the fivatl Royal Commission on Coal Supplies, materially advanced our knowledge. He cor- related the sweet coal-seams with those of the Clee Hills, Coalbrookdale, and South Staffordshire, and traced throughout most of the district the Spvrorbis-limestone, known to be associated with the sulphur coals in the Shrewsbury, Lebotwood, and Coalbrookdale Coal Fields. He was thus enabled to indicate the broad geographical distribution of the Sulphur Coal Group and Sweet Coal Group respectively, and established the fact that their distribution is anomalous. In the north, at Chelmarsh, bs and generally from Bridgnorth to Billingsley, the Sulphur Coal Group, as we have seen (p. 1004), rests directly on the Old Red Sandstone. This is the case in the south-west also, from Bayton to Abberley. But in the intermediate district, from Billingsley to Baveney Wood, it is the Sweet Coal Group that rests directly on the Old Red Sandstone. To explain this anomaly, Mr Jones applied the theory of great wash-out or denudation of Coal Measure date, basing this view on Marc Scorr’s conclusion that in Coalbrookdale the Sulphur Coal Group is unconform to the Sweet Coal Group, the plane of unconformity or wash-out being known the Symon Fault. On this hypothesis, the Wyre Forest district, after the format t of the Sweet Coal Group, was subjected to an extensive denudation, with the result that the Sweet Coal area was reduced to a small isolated patch or outlier, which owed its preservation to the protection afforded by the basaltic intrusion of Ki Over this outlier of Sweet Coal Measures the Sulphur Coal Measures were then laid down unconformably. The outcrop of the Sweet Coal Group between Billingsley and Baveney Wood was thus regarded as the exposed western margin of this outlier, the rest of the outlier being buried under the Sulphur Coal Group; though it w known to extend eastward as far at least as Highley, where the sweet coals had been worked for some years at a considerable depth vertically below the sulphur | coals. That this supposed outlier does not extend as far south as the Dowles Valley was thought to be established by the fact that a series of deep bore-holes there had proved a great thickness of Coal Measures, but had one and all failed to prove work- * Trans. Manch. Geol. Soc., vol. x (1870-71), p. 37. + Geol. Mag., 1871, pp. 200-208, 363-371 ; ibid., 1873, pp. 138, 348. Colliery Guardian, vol. xxii (July—Dee. 1871), pp. 580, 605, 633, 659. Trans. Fed. Inst. Min. Eng., vol. vii (1893-4), pp. 287-301, 577-580; vol. vill (1894-5), pp. 356-360. . FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. 1007 able coals: The Dowles Valley Measures were therefore believed to be an abnormally thickened lower part of the Sulphur Coal Group, directly overlying the Old Red ‘Sandstone. In an eastward direction also, as no workable coals were found in the _ group of measures underlying the Sulphur Coals in the deep sinking at Shatterford, it was concluded that there again the lower beds of the Sulphur Coal Group are abnormally thickened, and rest directly on the pre-Carboniferous floor. If the Sweet Coal Group had ever been deposited in the Dowles Valley and Shatterford regions, the denudation of the Symon Fault had commleuely removed it before the Sulphur Coal Group was formed. But, in 1895, Dr Kinston, on examining the plant-remains collected by myself _ from the surface-beds in the Trimpley area, showed that there the lowest Coal Measures directly overlying the Old Red Sandstone, instead of being of “ Upper,” are of Middle Coal Measure age. Further, a number of plant-remains from ex- posures at Cooper’s Mill, in the Dowles Valley, on a horizon probably not more - than 200 or 300 feet below the Main Sulphur Coal of Button Oak and Kingswood, _ also proved to be of Middle Coal Measure age. I was thus led to express the opinion that the Sweet Coal Group is of Middle Coal Measure age, and corresponds to the productive measures of Coalbrookdale and South Staffordshire, and that the - Sulphur Coal Group, with its pyritous coals and Spzrorbis-limestone, is therefore of “Upper” Coal Measure age, 2.e. equivalent to the barren measures that in South Staffordshire lie between the top of the productive measures and the base of the Permian,” and equivalent also to the Coal Measures of the Shrewsbury and Lebot- wood Coal Fields. I stated also* that the “‘ Upper” Coal Measures of the Wyre Forest, are “of great similarity to the upper parts of the Halesowen Sandstones of South Staffordshire.” As to the relationship between the two groups of measures in the Wyre Forest district, I was at first disposed to accept the view that it is one of strong unconformity, as seemed to have been proved to be the case in Coalbrookdale by Marcus Scorr and Mr Dantex Jones. But the discovery, based on the fossil evidence, that in the Dowles Valley and around the Trimpley anticline the Middle Coal Measures are not washed out, but are present in great thickness, led me subsequently to doubt this conclusion, and to regard the abnormal distribution of the two groups as due to overlap, accompanied perhaps by a slight unconformity. It seemed to follow that the Sweet Coal Group of the Highley region, instead of forming an isolated patch or outlier, restricted to the region lying north of the Dowles Valley and west of Trimpley, extended into these districts in force; and I explained the absence of workable coals as due to the conditions having been un- favourable to the formation of the requisite beds of vegetable matter on account of the proximity of a land-mass to the south. In other words, the sweet coals worked at Harcott and Highley, if followed underground southward and south-eastward, * “A Contribution to the Geology of the Wyre Forest Coalfield,” 8vo, 1895, p. 19. See also “The Wyre Forest Coalfield,” Colliery Guardian, vol. Ixxi, 1896, p. 351. TRANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 27). 143 1008 DR R. KIDSTON, MR T. C. CANTRILL, AND MR E. E. L. DIXON. : would be found to deteriorate in quality and thickness, to become unworkable, and nnelly to! die out altogether. This remains to be proved. Bue that the _ deposited ee fig. 3). Certain other results followed from investigations in the type district of the “ Salopian Permian” at Upper Arley,.Highley, and Alveley. The discovery of a ame of Sprrordis-limestone and a thin coal-seam between 200 and 300 feet above the base of these’red rocks led me to advocate their inclusion in the Upper Coal Measures.* The almost complete absence of plant-remains was not a serious obstacle to this proposal, as the fossil flora collected from the same group of rocks in South Stafford- shire during the sinking of the Hamstead shafts had already led Dr Kipstow to determine the same beds at Hamstead to be of Upper Coal Measure age. This view of the systematic affinities of the Lower Permian rocks of Salopian type has since been adopted by the officers of the Geological Survey, who found that the same beds in North Staffordshire are characterised by bands of Spzrorbes-limestone, thin con and Carboniferous plant-remains ; and under the name of Keele Series they are now included in the Upper Coal Measures Since that date, now twenty years ago, few additions to our knowledge hae beer en made. In 1901, while the aqueduct of the Birmingham Corporation Waterw (Elan supply) was under construction across the Wyre Forest Coal Field, I was dep to note the sections exposed along the pipe-trench. The result} showed that Coal Measures, where crossed by the pipe-track, “appear to consist of two groups: a lower, of variegated marls with yellow and green sandstone . . . anda higher gr of grey and yellow sandstones with thin pyritous coals and one or more band Spirorbis-limestone. The lower group has some likeness to the Old Hill Marl South Staffordshire and the Etruria Marls of North Staffordshire ; while the hig r as seen at Winwoods and Kingswood, and again at Woodhouse Farm near Button Oak, may represent the Halesowen Sandstones of the former district and the New= castle-under-Lyme beds of the latter.” It followed from this that the lower group, which rests directly on the Old Red Sandstone, if not the equivalent of the Ktrur Marls, must be Middle Coal Measures that have assumed the barren condition an lithological peculiarities of the Etruria Marls. The plant-remains from the Cooper's Mill exposure in the Dowles Valley had, however, already shown that there at least the surface-beds are of Middle Coal Measure age. I felt, therefore, that of the two alternatives the second was the more probable, though I did not feel justified in| putting this opinion into print, as to do so without further confirmation would have run counter to the experience of my colleagues in North Staffordshire, who had found d * Quart. Journ. Geol. Soc., vol, li (1895), p. 528 ; ‘Colliery Guardian, vol. Ixxiii, 1897, p. 581. + Summary of Progress of the Geol. Survey for 1901, p. 63 (1902). “ FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. 1009 that neither there nor anywhere else in the Midlands do red marls associated with espley rocks set in at so low a level in the Coal Measure sequence. A few years later my colleague Mr E. KE. L. Drxon* found that in the Titterstone Clee Hill Coal Field, a few miles to the west, the ordinary productive measures, characterised by sweet coals and believed by Mr Daniex Jones to belong to the same horizon as the Sweet Coal Group in Wyre Forest, contain “red clays and green ‘sandstones of ‘espley’ type, at intervals from a few feet above the base upward.” Mr Drxon, who extended his investigations into the western borders of the Wyre- Forest Coal Field, leaned to the belief that near Kinlet the Sweet Coal Group underlies the Sulphur Coal Group without unconformity, and therefore corresponds to part of the Ktruria Marls. This view was not, however, supported by the evidence of the fossil flora of the Clee Hill Coal Measures, which suggested to Dr Kinston a horizon lower than that of the Etruria Marls, nor by the presence of workable coals and ironstones, which nowhere else in the Midlands occur in the Etruria Marl Group. Meanwhile the attention of geologists and mining engineers was diverted from the visible coal field to the borderland of red rocks lying to the east, where an experimental bore-hole had been put down in 1903-1905 at Claverley, 5 miles east of Bridgnorth, to test the character of the ground between the Wyre Forest and South Staffordshire Coal Fields. The site lies a quarter of a mile east of Bulwardine Farm, a mile south of the village of Claverley. The chief results of this experiment have been described by Dr Watcor Gipson; +} it will therefore be sufficient to give an abstract of the section (see Table, p. 1010). The Lower Permian Group yielded fossil plants of Upper Coal Measure species, Dy. which Dr Kinston correlates the group with the Keele Beds of North Stafford- shire, 7.¢. the basal beds of the Radstockian subdivision of the Upper Coal Measures (p. 1079). The occurrence of bands of Spzrorbis-limestone forms another link by which these beds are attached to the Coal Measures and not to the true Permian. _ The Halesowen Sandstone Group proved to be 364 feet thick, whereas, 6 miles to the south-west, at Highley Colliery, the shafts of which commence below the ee Keele Beds, the group appears to be at least 394 feet thick. In the boring the Halesowen sandstones contain two thin coals, separated by 109 feet of measures. These may possibly represent the Brock Hall and Main Sulphur coals; but the Spirorbis-limestone of Kingswood, which should come between them, was not found, t hough looked for. In this respect the boring agrees with the sections of the Highley Colliery and the Kinlet Colliery, in which no such limestone is recorded, though it cannot be said that it was absent. The Halesowen Sandstones in the boring yielded abundant plant-remains. Below the Halesowen Sandstones, 193 feet of alternating red clays and espley — * Rep. Brit. Assoc. for 1910, pp. 611-613 (1911). + Summary of Progress for 1904 (Mem. Geol. Surv.), pp. 150, 151 (1905); ibid. for 1905, pp. 172-174 (1906). Also Trans, Inst. Min. Eng., vol. xlv (1912-13), pp. 30-48. 1010 DR R. KIDSTON. MR T. C. CANTRILL. AND MR E. E. L. DIXON. rocks were met with. These beds, which yielded no fossils, Dr Gipson names che Brick-Clay or Espley Group, and correlates on lithological grounds with the Etrur Marls of North Staffordshire. Below them follow the ordinary grey measures, witl coals and ironstones of the usual type, which yielded abundant fossil plants of Mic Coal Measure (Westphalian) species, and contain scarcely any of the red Claverley Bore-hole. Thickness. Dept h ft.) im: Upper “ Permian” Group ; - : : : : ; : . 185 6 Middle “ Permian” Group : ; ' : : 286 8 Lower “ Permian” Group = Keele Beds (768 ft. ): — Red and mottled marl and red and lavender-coloured sandstones, with plant-remains ; bands of ha eG at iy ft., 775 ft., a and 972 ft. No coals recorded . : 768 0 1240 — Halesowen Sandstones = Newcastle Group eee ft. yi ~. Grey and blue shale and sandstone . ; 3 ; : 5 46 0 1286 Coal ‘ : , : ‘ ; , : 3 : 3 1286 Blue shale and grey sandstone . ; : : : ; ; 2 1090 1395 Coal, friable : ‘ : ‘ 3 1395 Grey and blue shale and sandstone, aie plant-r remains - . d 3 208 6 1604 Brick-Clay or Espley Group (193 ft.) :— Chocolate-red: and variegated hard marls, with thick beds of green grit a and breccia (“espley” rock) . ; : : : cree? L930; 1797 24 Productive Measures (393 ft. 4 in.) :— , | Grey, dark-grey, and black shale, with 1 ft. of variegated marl ; Lingula at 1811 ft. 8 in. . : : Sauveurt Crépin. - ef. nummularia Gutbier. iS Marrati Kidston. 2 by Duponti Stur sp. te Jurcata Bronegt. 3 3 Walteri Stur sp. bok 5 Dixon Kidston nu. sp. (See p. 1077, Pl. V, figs. 34 Renaultia gracilis Brongt. sp. Oligocarpia Brongniarte Stur. Zeileria delicatula Sternb. sp. * LapwortH and Warts, “Shropshire,” in Geology in the Field, 8vo, 1910, p. 739. + The Etruria Marl Group of North Staffordshire. t Le., p. 429. ‘ § I unite with this species the Sphenopteris (Diplothmema) Schumanni Stur (Die Farne: Carb. Flora d. Sel Schichten, p. 352, pl. Ixy, fig. 2) and the Sphenopteris striata Gothan (‘‘ Oberschlesische Steinkohlenflora,” i Theil, p pl. v, figs. 3, 3a, pl. vi, figs. 3, 3a, Abhandl. Geol. Landesanstalt, Neue Folge, Heft 75,1913). The distinctive char: of the latter are, I believe, entirely due to the condition of preservation of the specimens. Both forms occur here, See Kipston, Trans. Roy. Soc. Edin., vol. 1, part i, p. 78, 1914. FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. LOS *Zeileria avoldensis Stur sp. Pecopteris Miltona Artis sp. Ss cf. Volkmanni Sauveur. Dactylotheca plumosa Artis sp. Mariopteris muricata Schl. sp. a Neuropteris heterophylla Brongt. % - tenurfolra Schl. sp. | sp. allied to Newropteris Blissi Lesqx. gigantea Sternb. Grangert Brongt. obliqua Brongt. sp. ze Scheuchzert Hoffm. Alethopteris lonchitica Schl. sp. Es Serli Brongt. sp. : ‘ Grandini Brongt. sp. es us aquilina Schl. sp. Lonchopteris rugosa Brongt. :; Bricer Brongt. Calamites Cista Brongt. 2 undulatus Sternb. (C. varians Sternb). t ramosus Artis. verticillatus L. & H. Sachser Stur. schiitzeformis Kids. & Jongm. ai +; Gopperti Ett. Asterophyllites equisetiformus Schl. sp. f cf. longifolius Sternb. sp. *Annularia galioides L. & H. microphylla Sauveur. sphenophylloides Zenker sp. ; radiata Brongt. Calamostachys sp. Palzxostachya Ettingshaust Kidston. Paracalamostachys striata Weiss. Cingularia typica Weiss. (See p. 1042, Pl. I, figs. 1, 2.) 3 Cantrilli n. sp. (See p. 1045, Pl. I, figs. 3-5.) * Macrostachya sp. Sphenophyllum cunerfoloum Sternb. sp. : < Specimens indicated by a * are preserved in the collections made by Dr ARBER, now in the Sedgwick Museum, Cambridge. ka + See Jonamans and Kukok, “Die Calamariaceen des Rheinisch-Westfilischen Kohlenbeckens,” Mededeelingen — -van’s Rijks Herbarium, Leiden, No. 20 (1913), p. 12. > 9 1088 DR R. KIDSTON, MR T. C. CANTRILL, AND MR E. E. L. DIXON: Sphenophyllum cuneifolium forma saxifragefolium Sternb. pro sp. majus Bronn sp. 3 ef. trichomatosum Stur. Lepidodendron ophiurus Brongt. us simile Kidston. i dichotomum (?Sternb.) Zeiller. Lepidodendron Wortheni Lesqx. ee ge: acerosus L. & H. (See p. 1057, Pl. I, fig. 5.) x ef. lariconus Sternb. Lepidophyllum lanceolatum L. & H. *Lepidostrobus variabilis L. & H. Asolanus camptoteania Wood. *Sigillaria scutellata Brongt. e mamillaris Brongt. = Boblayi Brongt. (See p. 1048, Pl. IV, fig. 3.) » vulgaris Artis sp. : » Davreuwsxr Brongt. - Candollet Brongt. (See p. 1049, Pl. IV, figs.1, 2.) Rs Kinletensis Arber. a » nudicaulrs Boulay. > » destans Sauveur. (See p. 1051, Pl. IV, fig. 4; Pl. v, fig. 10.) av » Pringle n. sp. (Seep. 10538, PE Ml tie. 12) , of. Utschneideri Brongt. » transversalis Brongt. (See ped054, eR, IL, figs. 2-4.) . Voltzi Brongt. 5 rugosa Bronegt. , reticulata Lesqx. (See p. 1055, Pl. Il, figs. 1, late). Sigillariostrobus ciliatus Kidston. : rhombibracteatus Kidston. 35 nobilis Zeiller. Stugmaria ficoides Sternb. sp. Cordates borassifolius Sternb. sp. » prinevpalis Germar sp. » palmeformis Gopp. sp. Samaropsis crassa Lesqx.t %5 emarginata Gopp. & Berger sp. = Gutbiert Geinitz sp. %» cf. Meachemi Kidston sp. * * Specimens indicated by a * are preserved in the collections made by Dr sea now in the e Sedgwick Muse 1m my Cambridge. + Coal Flora, pl. cix, fig. 12. ? .- > : FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. 1039 *Car ‘polithes perpussilus Lesqx. s membranaceus Gépp. 5 areolatus Boulay sp. “* Holcosper mum problematicum Arber sp. Trigonocarpus Parkinsons Bronet. Pinnularia capillacea L. & H. No discussion is necessary on this list, as Dr ArBer has shown that the “Sweet Coal Group” is clearly of Westphalian age; but the additional species derived by us from this series adds to our knowledge of the fossil plants of this Coal Field. On later pages the following species are described and figured from this series :— Annularia spicata Gutbier sp. (p. 1039). Cingularia typica Weiss (p. 1042). Cantrilli n. sp. (p. 1045). Sigilaria Boblay: Bronet. (p. 1048). Candollei Brongt. (p. 1049). 2 nudicaulis Boulay (p. 1050). " distans Sauveur (p. 1051). + Pringlei n. sp. (p. 1053). transversalis Brongt. (p. 1054). reticulata Lesqx. @ 1055). Tapidophloebs acerosus L. & H. (p. 1057). Samaropsis emarginata Gopp. & Berger (p. 1058). - Gutbrert Geinitz sp. (p. 1059). Carpolithes membranaceus Gopp. (p. 1060). areolatus Boulay sp. (p. 1062). DESCRIPTION OF SPECIMENS. Annularia spicata Gutbier sp. PIP, figs. 6,96a;,27, 7a: _ 1849. Asterophyllites spicata, Gutbier, Vers. d. Zechstein u. Rothl. in Sachsen, Heft ui, p. 9, pl. ui, figs. 1-3. 1861. p » Geinitz, Dyas, p. 136, pl. xxv, figs: 5, 6. 1871. - spicatus, Weiss, Foss. Flora d. jiingst. Stk. wu. Rothl., p. 128, pl. xviii, fig. 32. 1869. Annularia spicata, Schimper, T’raité d. paléont. végét., vol. i, p. 350. 1892, 55 » GZeiller (pars), Bassin houil. et perm. de Brive, p. 68, (?) pl. xi, fig. 4 1893. As » Potonié, Flora d. Rothl. von Thiiringen, p. 175, pl. xxiv, fig. 7. 1907. ~, » Zalessky, Donetz, Bull. du Comité géol. (St Pétersbourg), vol. xxvi, p. 509, pl. xxvi, fig. 2. 1911. », Jongmans (pars), Anleitung, p. 263. 1869. cf. lanttarda minuta, Wood (non Brongt.), Trans. Amer. Phil. Sov., vol. xiii, p. 347, pl. viii, fig. 2 * Specimens indicated by a * are preserved in the collections made by Dr ARBER, now in the Sedgwick Museum, Cambridge. TRANS. ROY. SOC. EDIN., VOL. LI, PART IV (NO. 27). 147 1040 DR R. KIDSTON, MR T. C. CANTRILL, AND MR HE. E. L. DIXON. Description.—The branches of the first, second, and third degrees are distichoush developed, two arising from the opposite sides of each node, and all seem to lie in th same plane. : 3 The stem (?) attains a thickness of 0°5 cm. or more, with internodes 3°75 em. striated longitudinally. Branches of first order are about 2 mm. thick, with the he internodes about 1 em. long ; branchlets of the second order very slender, with lo internodes about 0°5 cm. long. | Leaves whorled, those on the stem upright or adpressed, and probably uni each other by their bases. Leaves on branches and branchlets verticillate, spr very small, about 2 to 2°50 mm. long, lanceolate, widest slightly above the middl and as many as fourteen may enter into the formation of a whorl on the | branchlets, but on the smaller ones the whorls are formed of a much less nu of leaves. Cones small, oblong, about 0°5 cm. long, borne on the ultimate branchlet probably in whorls. Internal organisation unknown. Remarks.—The collection contains three small examples of Annularia spicate Gutbier sp., which were found at Cooper’s Mill section. The fossils are of a rich broy colour on a buff. shale, which makes it very difficult to photograph them ; but tw the specimens are given on PI. II, figs. 6 and 7. The plants are much broken up, but show a few branchlets, either isolated or springing in pairs from the nodes of a larger branch. The largest leaves on specimens are from about 2 mm. to 2°50 mm. long, lanceolate and sharp-pointec it is only in rare cases that they are perfectly preserved. The leaves, however both in size and form, with the description given by Gursrer, “linear lanz Blattchen,” and there can be no doubt as to the specimens belonging to Gur “ Asterophyllites spicata,” with which they agree absolutely. They are identical those figured by Grrnirz in Dyas, pl. xxv, figs. 5, 6. The description of the s) given here is partly drawn up from the original figures and description of GuTB ZEILLER, in his Bassin houwil. et perm. de Brive, unites to this species Annularia minuta, of which he figures the two specimens originally so namec BroneniaRt, but which had not previously been figured or described.* This, hoy I believe to be specifically distinct from Annularia spicata Gutbier sp. ZE further points out that these specimens of Annularia minuta much resemb Annularia microphylla Sauveur ; + but the Annularia microphylla is a larger spe with sickle-shaped leaves, and the leaf-margins are recurved.} , According to my view, it is with Annularia galioides L. & H. sp. that Annulana spicata must be compared. | Though the leaves of Annularia galioides appear to have had a pilose uy * ZHILLER, l.c., figs. 2, 3. + Veyét. foss. terr. howil. de la Belgique, p. 1xix, fig. 6, 1848. { See Kipsron, Trans. Roy. Soc. Hdin., vol. 1, p. 172, pl. x, figs. 1-8, 1914. eae ee SUSE SESE TS aa acinannneanenauibmammnnents Pa FOREST OF WYRE AND TITTERSTONE CLEE HILL COAL FIELDS. 1041 surface, it is only observable when the specimens are exceptionally well preserved,* and in this villosity Annularia galioides is thought to differ from Annularia spicata, where the leaves are supposed to be smooth.t I am afraid that this character cannot be regarded as of distinctive value, because occasionally one sees a hirsute covering on the leaves of Annularia radiata,| and Porontm has figured Annularia stellata Schl. sp. showing a similar character.{ I have also received from Dr Moysry specimens of this latter species in which the hairy surface of the leaves is beautifully exhibited. It would appear, therefore, that the presence or absence of hairs on the surface of the leaves of Annularia cannot be taken as a character of specific importance, as their presence or absence seems to be due to the state of preservation of the fossil. | The leaves of Annularia spicata are distinctly narrower in proportion to their length than those of Annularia galioides, and this gives a less dense character of growth to the plant, which, taken in conjunction with the smaller size of Annularia spicata in all its parts, offers an easy means of distinguishing the two species. It is therefore with Annularia galioides that I would unite the Annularia minuta Brongt., with which in the size and form of the leaves it agrees perfectly, especially with those on the smaller branchlets of that species. I am a little doubtful if the specimen figured by ZEILLER in his Bassin howil. et perm. de Brive, pl. xi, fig. 4, as Annularia spicata should be referred to that species, as the broader form of the leaves seems to agree more with those of Annularia galioides L. & H. sp., of which it may possibly be a small condition. The differ- ences between these two species, though not easy to put in words, are quite clear when specimens of the two plants are compared. It has also been proposed by ZEILLER to unite the Annularia radiiformis Weiss sp. || with Annularia spicata - Gutbier sp., but here also I am unable to accept this view, as I fail to discover any character by which it seems possible to separate the Annulama raduformis Weiss sp. from Annularia galioides L. & H. sp. The British specimens of Annularia spicata come from a lower horizon than that from which the plant has been previously obtained. Locality.—Roadside, 250 yards N.H. of Cooper's Mill, 13 mile W. of Dowles Church, Bewdley. Horizon.—Westphalian Series. (Pr. 1156-1158.) * K/2066. + Jonemans, Anleitung, i, pp. 258 and 263. t See Jonamans and Krxux, Calamariaceen d. Rhein. W estfil. Kohlenbeckens, p. 42, pl. xvi, figs. 1, 2, 5, 1913. § Flora des Rothl. von Thiiringen, p. 162, pl. xxiv, figs. 4, 5, 1893. || Asterophyllites radiiformas, Weiss, Foss. Flora d. jiingst. Stk. uw. Rothl., p. 129, pl. xii, fig. 3. I Bassin houtl. et perm. de Blanzy et du Creusot, p. 138, 1906. 1042 DR R. KIDSTON, MR T. C. CANTRILL, AND MR E. E. L. DIXON. Cingularia Weiss. 1871. Cingularia, Weiss, Foss. Flora d. jiingst. Stk. u. Rothl., Heft ii, p. 137. 1876. be Weiss, Steinkollen-Calamarien, Heft i, p. 88. - Remarks.—Up till the present, Cingularia typica Weiss was the only sy comprised in this remarkable genus, but a second species is described below. As the generic and specific characters are so intimately connected, these nee be mentioned here, as they will be dealt with when treating of Cingularia typ Cingularia typica Weiss. PIS hes les 2, 26. 1828. cf. Hquisetum infundibuliforme, Bronn, in Bischoff, Krypt. Gewdchse “Deutschl., p. 52, pl. vi 1828. cf. BS 5 Brongt. (pars), Hist. d. végét. foss., p. 119, pl. xii, fig. 16. 1871. Cingularia typica, Weiss, Foss. Flora d. jiingst. Stk. u. Rothl., p. 138, pl. xiv, fig. 4 1873. - 5, Weiss, Zettsch. d. deut. geol. Gesell., p. 263, fig. 5. "i ve daar » Schimper, Zraité d. paléont. végét., tor iii, p. 460, pl. cix, figs, 1-4. 1876. ‘3 » Weiss, Steinkohlen-Calamarien, Heft i, p. 99, sas vi, figs. 5, 6; pl. figs. 1-9; pl. viii, figs. 1-7; pl. ix, figs. 1-7. Z 1874. 4 » Stur, Verhandl. d. k. k. Retchsanst., No. 7, p. 176. rs 7 1882. . » Weiss, Aus d. Steink., p. 11, pl. x, ie 55 (Zweiter Abdruck).— ‘= 1882. “e _ 4, Renault, Cours d. bot. foss., vol. ii, p. 144, pl. xix, figs. 9-1]. 1887. Hs » Stur, Carb. Flora d. Schatz. Schichten: Die Calamarien, p. 218, figs. 2, 3. ~ 1888. 3 Ks Toula, Die Steinkohlen, p. 206, pl. v, figs. 34-36. 7 1891. :: », Solms-Laubach, Fossil Botany, p. 334, fig. 47. r 1893, Mi » Fischer, Naturf. Gesell. in Bern, Jahrg. 1893, p. 1, pl. 1. 1899. ts » Hofmann et Ryba (pars), Leitpflanzen d. palaeoz. Steink., p. 31, pl. i 1-3 (non fig. 16). 1900. o » Scott, Studies in Fossil Botany, p. 62, figs. 26, 27. 1900. a » Zeiller, Hléments de paléobot., p. 168, fig. 119. 1907. - ‘,, Zeiller, Comptes rendus, vol. cxliv, p. 1140. 1908. - 5, Schuster, Geognostischen Jahresheften 1907, xx Jahrg., p. 206, fig. 1 (2), p. 20 1908.