ay a CEE : =f 4 7 mS * . + « i . . ’ = ‘ * LIBRARY OF D. LACOE. For the Promotion of Research in PALEOBOTANY and PALEOZOOLOGY RETURN TO SMITHSONIAN INSTITUTION WASHINGTON, D. C. Ie eee Fie Se Cod = ‘i THE QUARTERLY JOURNAL GEOLOGICAL SOCIETY OF LONDON. EDITED BY THE ASSISTANT-SECRETARY OF THE GEOLOGICAL SOCIETY... - Quod si cui mortalium cordi et cure sit non tantum inventis herere,- atque iis uti, sed ad ulteriora penetrare; atque non disputando adversarium, sed opere naturam vincere; denique non belle et probabiliter opinari, sed certo et ostensive scire; tales, tanquam veri scientiarum filii, nobis (si videbitur) se adjungant. —Novun Organum, Prefatio. VOLUME THE FORT 1891, 4 ° > terion, WS LONDON : LONGMANS, GREEN, AND CO. PARIS: FRIED, KLINCKSIECK, 11 RUE DE LILLE; F. SAVY, 24 RUE HAUTEFEUILLE. LEIPZIG: T.O. WEIGEL. SOLD ALSO AT THE APARTMENTS OF THE SOCIETY. MDCCCXCI. List OF THE OFFICEHRS OF THE GEOLOGICAL SOCIETY OF LONDON... aaeernannrr—r—r—r “~\ Elected February 20, 1891. vevvee~ WrestVent, Sir Archibald Geikie, D.Sc., LL.D., F.R.S. Wice-restvents. W. T. Blanford, LL.D., F.R.S. Prof. T. G. Bonney, D.Sc., LL.D., F.R.S. L. Fletcher, Esq., M.A., F.R.S. W. H. Hudleston, Esq., M.A., F.R.S. Secretaries. H. Hicks, M.D., F.R.S. | J. E. Marr, Esq., M.A., F.R.S. Foretqu Secretary. J. W. Hulke, Esq., F.R.S. Creasurer,. Prof. T. Wiltshire, M.A., F.L.S. COUNEIL. Prof. J. F. Blake, M.A. W. T. Blanford, LL.D., F.R.S. Prof. T. G. Bonney, D.Sc., LL.D., F.R.S. James Carter, Esq. James W. Davis, Esq., F.L.S. John Evans, D.C.L., LL.D., F.R.S. L. Fletcher, Esq., M.A., F.R.S. Prof. C. Le Neve Foster, D.Sc., B.A. Sir Archibald Geikie, D.Sc., LL.D., F.R.S. Alfred Harker, Esq., M.A. J. C. Hawkshaw, Esq., M.A. H. Hicks, M.D., F.R.S. G. J. Hinde, Ph.D. W. H. Hudleston, Esq., M.A., F.R.S. Prof. T. M°Kenny Hughes, M.A., F.R.S. J. W. Hulke, Esq., F.R.S. J. E. Marr, Esq., M.A., F.R.S. 4. W. Monckton, Esq. F, W. Rudler, Esq. J. J. H. Teall, Hsq., M.A., F.R.S. W. Topley, Esq.,. F.R.S. Prof. T. Wiltshire, M.A., F.L.S. H. Woodward, LL.D., F.R.S. Assistant-Secretary, Clerk, Librarian, anv Curator. L. L. Belinfante, B.Sc. Assistants in Oflice, Library, and Museum. W. Rupert Jones. Francis E. Brown. TABLE OF CONTENTS. Page noe Prof. T. G. Note on a Contact-Structure in the Syenite of rrr ycilaal nnn a Re MEE, on Ca k's, Sislaas hae Rive ge tiem aule, «aes 10t , and Major-General C. A. M°Manon. Results of an Examination of the Crystalline Rocks of the Lizard District. (CELE BOD) Sse ea ie a oc 464. _ Cattaway, Dr. CHas. On the Unconformities between the Rock-— systems underlying the Cambrian Quartzite in Shropshire .... 109 Davison, Cuas., Esq. On the Inverness Earthquakes of Noy. 1d to Dec. 14, TEC eo trie Gas Re ee Pa eR 618 Derby, Orvittz A., Esq. On Nepheline Rocks in Brazil—Part II. The Tingua VELOC ae RUS AC heen Ota NE eR 251 Grucory, J. W., Esq. The Variolitic Diabase of the Fichtel- SOV UMO OP ass taeyas sas rid: B18 Ges Cane Se CEE RMSE tne eke ere ar: 45 The Tudor Specimen of Eozoon ......2.0.ee0eeee 7 SH & 348 HARKER, ALFRED, Esq., and J. E. Marr, Esq. The Shap Granite, and the Associated Igneous and Metamorphic Rocks. (Plates Be eer MOU Vee eres eg Fate wis ah Hine dating bl aja te AG 0 Brees 200 Hicks, Dr. Henry. On some Recently-exposed Sections in the Glacial Deposits at Hendon. (Plate XXIL.).......:....00.. 575 Hinu, Rev. Epwin. On Wells in West-Suffolk Boulder-clay .... 585 , and Prof. T.G. Bonney. On the North-west Region of Charnwood Forest, with other Notes ..........ececceeeeces 78 Hosson, B., Esq. On the Igneous Rocks of the South of the Isle ugh aaiaee (bee aaa a tebe Ahead eas wlan se alee a's en 432 Hoianp, T. H., Esq. Notes on Rock-specimens collected by W. . GOWLAND, Esq, 3 TIO NCO) Els GR © 171 Hotu, Dr. Epw. On the Physical Geology of Tennessee and Ad- - joining Districts in the United States of America........ dies > GD et LOR. Ow IV TABLE OF CONTENTS. = Page Jennines, A. V., Esq., and G. J. Witirams, Esq. Manod and the oT) a2 aire 5 368 JuKkEs-Browne, A. J., Esq., and Prot. J. B. Harrison. The Geo- logy of Barbados.—Part I. The Coral-rocks of Barbados and other West-Indian Islands. (With two Appendices by W. Hix, et EMEC MR NI cf, a) cies Seen st ae eal @ > oe ee 197 Lamp.ueu, G. W., Esq. On the Drifts of Flamborough Head. PRC ea AOU La) ee te ene vis Gola oes Fs wise 'soe shoraye G4 ho 384 LapPaRENT, Prof. A. de. On the Porphyritic Rocks of the Island | Fae REE at tte oe satay Sete ete tlane tose et ees''s sessing dearere: dt avake wa ‘sey oe ListER, J. J., Esq. Notes on the Geology of the Tonga Islands. ENE 00. lt] 1) Raat oa ek ee aR AU ERE ar 590) LYDEKEKER, R., Esq. Ona new Species of Trionya from the Mio- cene of Malta and a Chelonian Scapula from the London Clay. 37 On certain Ornithosaurian and Dinosaurian Remains. Bete We ee a a ee 4h On a Labyrinthodont Skull from the Kilkenny Coal- MEP SUES Ue R cheer ole MUON: Taos «c ciayece Majer Plo leto ws Sayleks lol cuts ae 345 On Lower Jaws of Procoptodon. (Plate XXI.) ........ 571. Marten, H. J., Esq. On some Water-worn and Pebble-worn Stones taken from the Apron of the Holt-Fleet Weir on the River RN TM ene oe icc sicha, ois vas: atin 6/e $00! 6 cig!» uiletaye ed Janel ae Whale "acer keen 63. Nicuotson, Prof. H. A., and J. EK. Marr, Esq. The Cross Fell ~ -Tnlier. (With Appendices by ALFRED HarxeEr, Hsq., and A. bieoor, sq.) (Plate XVI)... ele ool ls so ee 500 Pennine, W.H., Esq. A Contribution to the Geology of the monthern Transvaal. (Plate XV.) .0.0.csiecd cee bee ween 451 Prestwicu, Prof.J. On the Age, Formation, and Successive Drift- Stages of the Valley of the Darent; with Remarks on the Paleolithic Implements of the District, and on the Origin of its Chalk Esearpment. (Plates VI., VII., & VIEL). ....).... 126 Raisin, Miss C. A. On the Lower Limit of the Cambrian Series in Mein (OAPRMAEVOUSHITC. 200..6.56 o..cd0ee ie eden cee s aia weiss 6 5 LS an 329 Rutvey, Frank, Esq. Ona Spherulitic and Perlitic Obsidian from Piles, Jalisco, Mexico.. (Plate X VILL.) .......0../06e8 ee 530 . On some of the Melaphyres of Caradoc, with Notes on the Ascociated Velsites.. (Plate XIX.) ..........2002. seen 2. OOF SEELEY, Prof. H.G. On 2 albd oda sels ove 29 Rbiempole tyes Med alltstsa,t..5 -ovcaiva cia cities steve ove) cn: 9 2106 se aaa eineray, 3S Histol bissoy MeEdalistsh sy falc ci pid we snlee velss aN e THe wielelelnae’s 31 _ Applications of the Barlow-Jameson Fund ........... Bins eiapsieyacs 31 Roiurem etal Me PORE Ie eel ter desea seen rae Rls ofS «Recs, 2 duster ogame 32 Para Oe Git Me NICH AIS ACOs seers) a ek aide Pe eelaea's vide deeds bene eee s 38 TIE Vets aie NOR eaaieaee hts pate ics (lca be do vio aie ie vie dv Vicia a oe 48 Donations to the Library (with Bibliography) .................- 176 Vi TABLE OF CONTENTS. Page Brix, ALFRED, Esq. Notes on some Post-Tertiary Marine Deposits on the.South Coast of Wimgland:4 0.) 00.0... os. ae ee n7e Buckman, 8. 8., Esq. Notes on Nautiliand Ammonites ...... <3 hE6g Dawson, Sir J. W. Note on Photographs presented to the Geolo- mieal Socieby «5... ssaapecse ei ee Nete bile sol? as» 00 ol a 169 Hicks, Dr. Henry. On the Rocks of North Devon ............ 7 Irvine, Rev. Dr. A. Further Notes on the Stratigraphy of the Bagshot Beds of the London. Basin (north side)........... ‘ee Note on some Recent Excavations in the Wellington Col- lege District. C2 sestiak a tana en GlemMerg te leben eats clea ene 171 OtpHaAM, R. D., Esq. Account of an Experimental Investigation of the Law that Limits the Action of Flowing Streams ...... 6 “International Geological Congress.” Announcement of the date of meeting at Washington (1).C.) . 23. ae a 164 LIST OF THE FOSSILS FIGURED. AND DESCRIBED IN THIS VOLUME. (In this List, those Fossils the names of which are printed in Roman type have been previously described. | Name of Species. | Formation. Locality. | Page ! PoLyZzoa. Bactridium Hagenowi. PI. i. f. 18,| \ 11 | ok Scru/a st s f 7 Batopora multiradiata. Fig.......... | 32 Catenaria tenerrima. Pl.i.f.11... | Catenicella continua. Pl.i. f.9,10 6 septentrionalis. Pl.i. f. 1-8 4) Cellepora proteiformis. Pl. iv. f. 13, GIR US SRA EES 2 a ee ee ge 30 Cribrilina chelys. Pl.ii.f.10 ...... | 16 Fedora excelsa.. Pl. iv.f.6 ......... ! | 29 Lepralia (?) Jericensis. PI. iii. f. 18) 21 bisulca. Pl. i. f. 16-18, and| | , 12) ba laa (el Re ees cae ee | 18 tmpressaz.. PUOMST TO... 26s 08 | 19 (?) lontensis. - Pl. iii. f. 5 ...... | | 21 nodulifera. Pl. iif. 13,14 ...| | 19 subchartacea: " Blvii.f. 12°... 17 —— (?)syringopora. PI. iii. f. 2, 3, cos ee RNR tee ee a \ Upper Hocene...|North Italy ... < 20 Membranipora appendiculata. PI. | | Uihia sca ecees kone tees ul" | 13 Dumienlit, PL he 4a... | 12 assem (el nd 2s... | 12 Micropora articulata. Pl. ii. f. 5,6 14 Goracealy, pevaie he Qe. 2 he oes.. | 13 paralielaey Pili £28 3 oc ..ct tens | 14 polystichal Plt fi sosec. ik | 14 Monoporella sparsipora. PI. ii. f. 11) | | 17 Onychocella angulosa. Pl. i. f. 20...) | 9 Porella imbricata. Pl. iii. £16, 17 | j apa: marsupium, var. porifera. P1.| itech: Ut, POSS Be esa teeaaceaat Oo | 23 Porina (?) bioculata. PI. iti. f.15... 26 (?) coronata. Pl. iv. f.1-5,15 | 24 (?) duplicata. Pl. iii. f. 14 ... 25 (?) papillosa. Pl. iii. #19 ... | 25 Retepora elegans. PI. iv. f. 9, 10...) | Vw 730 vill FOSSILS FIGURED AND DESCRIBED. Name of Species. | Formation. Locality. | Page Potyzoa (continued). Rhamphostomella brendolensis. PI.| \ { Bere ehs Me eh. Sec ltentenemaaee | 23 Schizoporella Hoernesi. PI. iv. f.8 27 ——ternata. Pl.iv. f.11,12...... | | 20 Scrupocellaria brendolensis. PI. i. PIN Ea ee naka skinke a Meee | 7 ——elliptica. Pl.i. f. 16,17 ...... | — gracilis. Pli.f. 12,13 ...... | 6 montecchiensis. Pl.i. f. 21, 22} | [| ri Smittia coccinea. Pl. ii. f.8 ...... ‘ Upper Hocene...|North Italy ...... < | 21 , var. alifera. Pl. iii. f. 7 | | 21 exaratas” (El: mi, fs! G .keotee ce 22 Landsborovii, var. cheilopora. BRM ds Ler Peet tivatens cauaace eens 22 pentigens.: \ Ply i. f. 905.2 | 22 Stichoporina simplex. PI. iv. f. 16— | (LR MGR Pepe Ste Ma Rena A Sd | 31 Vibracella (gen. nov.) trapezoidea.| | PMS ck aa wot cs Bauer ee ) \ 11 Mo .uvusca. ( Cephalopoda.) Orthoceras pusgillense .............-+ | Ordovician ...... (Cross Fell ..... ee VERTEBRATA. (Amphibia.) Ichthyerpetum hibernicum ......... | Carboniferous .../Kilkenny ...... | 344 ( Reptilia.) MRO ANCE CTISIS 2 oo. soe sawn cence [ ‘Miocene, .. 225s. [Malta 3.20 . ag OUTS Gt Oo ee ea London Clay .../Sheppey ......... es, Calamosaurus Foxi. Pl. v. f. 1-2...) Wealden ......... Isle of Wight ... 43 +e le oa BRN Paid it ‘s Kamendeeen ...|Dorsetshire ...... 42 supra-jurensis. Pliv.f.4...... Kimeridgian ...|Dorsetshire ...... 44 Agrosaurus Macgillivrayi ............ Secondary ...... Australia ......... 164 Saurodesmus Robertsoni. ............. Lan: cia (eRe (Elein oceans 166 (Mammalia.) Procoptodon Goliah. PI. f: fe i ia. f ] i a ‘| | Pleistocene cele, N. S. Wales...... 571 INCERT SEDIS. Eozoon canadense...... este get bvna Huronian.........|\Tudor (Ont.) ..- | 330 EXPLANATION OF THE PLATES, PLATE Pac i II. | Nortu Irarran Bryozoa, to illustrate Mr. A. W. Waters’s may Paper on Lhose fossils eee cu se cle sac. decid wdaleos nee cesieederies TV. Mr. R. Lydekker’s paper on certain Ornithosaurian and DINOSAURIAN AND ORNITHOSAURIAN Bones, to illustrate Vv DUNO SAPIAT, EROWIALTSY © a tele ood ce ais oe be eoiv awe icecatoecer ean AL < VI DIAGRAM-SECTIONS SHOWING THE ReELAtive LEVELS oF a PuaTeAu- AND VALLEY-Drirts ; | VII. Map or tus Darent Basin, ABovE FARNINGHAM; and 1196 Fuint Implements oF THE CHALK Puarnav, to illustrate Vill Prof. J. Prestwich’s paper on the Age, Formation, and Drift Stages of the Darent Valley ..................ssscooses is first Appendix to Messrs. Jukes-Browne and Harrison’s Corau LimestTones oF BArBapos, to illustrate Mr. W. Hill’s IX. paper on the Geology of Barbados ..............-..-ceceeeees 243 Map ILustrating tHe ReLations or THe SHAP Granite \ xX AND THE ASSOCIATED IcNrous AnD MeErAmorpuic Rocks | To THE NEIGHBOURING FoRMATIONS ; XI Sections oF SHap GRANITE AND MurtamorPHosep ANDE- | : SITES ; and 4 266 Sections or Merramorrnosep Rocks NEAR THE SHAP | GRANITE, to illustrate Messrs. Harker and Marr’s paper a on the Shap Granite and the Associated Igneous and Wetamonpiic RGeks i eececcsnctas sodesdces.nmcccaneden- se sccenseee J XIII SEcTIoNS IN THE Drirts or FuAMBOROUGH HEAD, to illustrate Mr. G. W. Lamplugh’s paper on those deposits............ 384 GroLocicat Map or THE SouTH oF THE IsLE oF Man, AND XIV SEecrions THRouGH Scarier Point, to illustrate Mr. B. Hobson’s paper on the Igneous Rocks of the South of the ator olml Vian cesarcatisudeatete ea iomeactsssicsstatleceteinans teeeenees 432 VOL. XLVI. 0 A EXPLANATION OF THE PLATES. PLATE 'GeroLogicAL Map or, AND SECTIONS ACROSS, THE AREA XV BETWEEN KLERKSDORP AND THE Der-KaAap VAuury, to ‘) illustrate Mr. W. H. Penning’s paper on the Geology of [ the Southern Transvaal 5... cecesttee eer a race .nperonss cece ( BanpEpD SerPENTINE FROM PorTHALLA, to illustrate Prof. T. XVI.; G. Bonney and Major-General C. A. M*Mahon’s paper | on the Crystalline Rocks of the Lizard District ............ Map or tue Cross Feit Inuier, AnD Section 1N SWINDALE XVII. Beck, to illustrate Prof. Nicholson and Mr. Marvr’s paper | on the Oross Well Imliers2sns.tteseesrcceceewenaeaean ace eee [ OBSIDIAN FROM PILAS, AND OBSIDIAN FROM VULCANO, to XVIII. illustrate Mr, F. Rutley’s paper on Obsidian from the | former locality... ..c.c2u:.tsecsees eterna eee aaa et nee Mertaruyres oF Carapoc Hitt, anp Peruitic FEsire ABOVE CARADoc Coppice, to illustrate Mr. F. Rutley’s paper on the Melapbyres of Caradoc, with notes on the Associated Felsites ........... eGR, Sista oad coh kw eae XIX. | Rock Secrions FroM THE INFERIOR OoLiITE oF THE CorTEs- XX. woLps, to illustrate Mr. Edw. Wethered’s paper on that | SHLD JOC: sn uiaininn's 5 ds'n so’ sree Sea soe NCEE eee wes hee xxz {dws or Procorropon, to illustrate Mr. R. Lydekker’s ; 1 paper on that genus %..):¢4.22seesecceenen ee eee wae eeeeee XXII Map or true GuacitaL Diposits In AND AROUND Henpon, to Beak illustrate Dr. H. Hicks’s paper on that subject .....,...... ey { Mar or rus Tonea Isuanps, to illustrate Mr. J. J. Lister’s XXITT, : | paper on the Geology of that group...........:.2s0 Be oe: Pace 464 500° 530° 5384 5d CORRIGENDA. Proc, p. 2d, line 7 from bottom, for ‘« Lausaune” read “ Lausanne.” ? Proc. p. 31, last line, for “‘ Harisron” read “ Harrison.” Page 255, line 4 from top, for “ mountains” read “ mountain.” Page 255, line 2 from bottom, for “ sufficient to” read “ sufficiently hard, to.” Page 256, line 20 from top, for “ eruptions” read “ eruptives.” Page 257, line 4 from bottom, for “ stock” read “ boss.” Page 259, line 17 from top, for “ non-existent ” read “ non-resistant.” Page 262, line 9 from top, for “ besides” read “ outside of.” Page 263, line 2 from bottom, for ‘‘ magnetic” read “ magmatic.” Pages 444, 445. The section described as picrite-porphyrite has since been. determined by Prof. Rosenbusch as diabase. Page 447, line 11 from bottom, for “ pilotaxitic” read “ hyalopilitic.” THE QUARTERLY JOURNAL OF THE GEOLOGICAL SOCIETY OF LONDON, Vout. XLVIT. 1. Norrn-Itatian Bryozoa. By Arnraur Wu. Waters, Esq., F.G.S. (Read June 4, 1890.) [Prarss I., II., IIL, 1V.] Cuitostomata.—The Bryozoa dealt with in this paper are, for the most part, from well-known localities in the Vicentine, as Val di Lonte, Montecchio Maggiore, and Brendola. Most of them have been described by Reuss, but at a time when chief attention was paid to the zoarial mode of growth (the shape of the oral aperture and other zocecial characters being considered to be of secondary importance) and when the avicularia and ovicells did not receive the attention now given to them. The first paper treating of North-Italian Bryozoa, of this series, is one by Reuss *, in which he refers to a. number of species as from an unknown locality in the Vienna Basin; but these, he subsequently 7 found, came from the Val di Lonte, called also Val dell’ Onte (but the exact position is Casa Fortuna), in the Vicentine. Although this correction has been made, references are constantly given which show that authors overlook this rectification. From a similar marl in Montecchio Maggiore, a few miles further south, Reuss also described ¢ a number of others, and a couple from Brendola, in the Colle Berici, south of Vicenza; and from Crosaro, some distance to the north-east, a few which appear to be of nearly the same age. In the first three localities the marl contains a very * ‘* Die fossilen Polyparien des Wiener Tertiarbeckens,” Haidinger’s ‘Natur- wissenschaftliche Abhandl.’ vol. ii. t “Foss. Bryozoen des Oest.-Ung. Miocans,” Denkschr. Ak, Wissensch. Wien, wok, Ske)" : ¢ “ Die fossilen Anthozoen und Bryozoen d. Schichtengruppe von Crosaro,” Denkschr. Akad. Wissensch. Wien, vol. xxix. This paper is referred to in the following pages as “‘ Bryoz. von Crosaro.” Ga.G. S.No, 185, B 2 MR. A. W. WATERS ON similar fauna; and, having collected from all, I found the Brendola beds.the most instructive, and there the preservation is the best *. The late Dr. G. B. Gottardi tT has also given a list of species from Montecchio Maggiore; but he follows the generic and specific names given by Reuss, even where Reuss himself had subsequently intro- duced modifications, or reduced the synonymy ; and nothing is added to our knowledge of the characters, so that we do not know what qualifications Gottardi possessed for making the determinations. Besides these already-known localities, I have collected from two in the Veronese which are of considerable geological interest ; and both are new localities. The first is Ferrara di Monte Baldo. The mountain is on the east of the Lake of Garda, and the deposit occurs at about 4670 feet above sea-level and about 1750 feet above the village (by the path leading to Madonna del Neve, past the Austrian frontier) as a thin bed, where, although the Bryozoa are numerous, they are very badly preserved ; and the same remark will apply to the second locality, Ronzo, near Mori, in the Tirol, north of the Lake | of Garda. [Since this paper was read, I have again visited North Italy, and collected from Malo and Priabona, both near Schio in the Vicentine ; and also from near Ferrara di Monte Baldo. I did not, however, find the beds uncovered at the locality mentioned in the paper; but about halfway between the village and the frontier (marked Novezzina in the Austrian maps) a blue marl attains to a consi- derable thickness, containing many Bryozoa. Besides the species mentioned, Microporella distoma, B., occurs at Malo and Novezzina. A. W. W., Dec. 24, 1890.] The Ferrara deposit lies above a series of beds commencing with those containing large Nummulites, as N. Brongniartz, &c., then beds with Cancer punctulaius, &c., then with Serpula spirulea,—in fact similar to the Vicentine series. Many of the species are known from the Lower Tertiaries of other parts of Kurope ; Reuss, in a series of papers, has given descriptions — from several places in the north of Europe, as Sollingen, Latdorf, &c.; and recently Koschinsky + has published an important work on the South-Bayarian Tertiary Bryozoa. Pergens§ also has given lists and some descriptions of species from Hungarian and other localities. The Vicentine fossils are dark in hue, and this makes the study of them difficult and very fatiguing, especially as the marl is often hardened in the cavities, obscuring the characters. The specimens * Grancona, alluded to in the paper, is near Lonigo, in the Colle Berici; Bocca di Sciesa is between the two. t “ Briozoi Fossili di Montecchio Maggiore,” Atti Soc. Trent. di Sc. Nat. vol. ix. pp. 297-308. { “Bryozoenfauna der alteren Tertiarschichten des siidlichen Bayerns,” ‘ Palexontographica,’ vol. xxxii. § ‘Les Bryozoaires du Syst. Montien,’ 1886 (Louvain); “Bryoz. Foss. de Kolosvar,” “‘ Bryoz. de Tasmajdan,” Bull. Soc. Malac. de Belgique, vol. xxii. ; “ Bryoz. von Wola Lu’zanska,” Bull. Soc. Belg. Géol. Hydrol. &e. vol. iii. NORTH-ITALIAN BRYOZOA. 3 have mostly been collected many years ago, and, when returning to the work, it has often been possible to further clean them by soaking in water, then brushing with a camel-hair pencil, drying, and repeating the process. Recently, however, I have obtained very much better results by placing the fossils in a saturated solution of sulphate of soda and allowing it to crystallize. When this is washed out it loosens the matrix within the apertures, and they are thus more thoroughly cleaned than would be possible by ordinary means. Although all of Reuss’s works were beautifully illustrated, the ficures were rather the artist’s than the naturalist’s figures, one zocecium being drawn and then identically repeated several times ; but this geometrical regularity does not often occur in nature. The figures now given are supplementary, to show the characters to which I allude. Though considering a revision now required, I would point out how great an advance Reuss’s work was upon what had been done before; and that he gave more attention to zocecial characters than his predecessors had, often forming groups according -to the nature of the surface and the presence of oral spines or avicularia. It is true that very few of these groups will now stand, but in this way the study of the value of the characters has been made possible without adding to the number of generic names. : There are many cases of a species occurring in both the incrusting and erect form ; some show considerable difference in zoarial shape, and there are interesting instances of great range, so that different parts of the same colony have widely divergent appearance. Perhaps the most noticeable case is Cellepora proteiformis, which commences in a flat Eschara-form, on which, however, single zocecia may be raised (figure 14); but in later stages the zocecia are piled up irregularly, often several layers thick... Lepralia bisulca varies much in appearance; and Porima coronata may either have the - zocecia distinct or scarcely distinguishable... There may be more or less of a peristome in front, and the number of large pores or avicu- laria around the aperture is very variable; further, the central avicularium may be absent, may be moderately large, or developed into a gigantic raised spatulate avicularilum. When the aperture has a closure the appearance is further modified. Porina papillosa, with or without avicularia or prolonged zocecia, also has a very variable appearance. There are forms with Lunulites-mode of growth with the zocecia of Membranipora, Cellepora, and Lepralia. | As we thus obtain more exact acquaintance with the Bryozoa of past times, we shall in some cases be enabled to check the correct- ness of the principles of present classification and gain new ideas as to relationship ; but fresh difficulties are brought before us. Among the points brought forward, the discovery of Catenicella is of special interest as bearing upon the relationship of this genus. Fedora excelsa is interesting, from the way in which it grows from the apex of the colony,—from occurring on both sides of the Alps, —and from the genus being represented by one recent form dredged by the ‘ Travailleur.’ B2 | |s a || |g] is hi = = | oy o s| 2) |f | P| SC} a 2 S| | a -— b VW epee ite als Other Localities. | |OrS| a|c Jelp JAS ho/e| o/S/s ia SPolelelelSigisie EralSlelSlolelole| Ai ale Io er 1. Catenaria tenerrima, Rss. ............{.++ xk KK! 2. Catenicella septentrionalis, sp. nov.|...|.-.|%| | 3. —— continua, ep. Oy een ame Eee Oe be 4. Scrupocellaria elliptica, Z?ss. .........|.-. XIX | | 5. —— gracilisy Rss. ............ccssessecees BP bod Bed ed od ee |.-.|Gaas. 6. —— brendolensis, sp. nov............ weelee ef MEISE eed | 7. ——montecchiensis, sp. nov. ......|-++ lk | 8. Bactridium Hagenowi, Rss. .........|.+- Falke ead eee ee ee .-» Lower Eocene of Mons; Malo. 9. Cellaria Reussi, d@’Orb. ............06 2 ISK F]---|---19)---|--- Miocene of Vienna. 10. Onychocella angulosa, Rss............. IKK HE/K|--- E/E Cretaceous; Miocene of Vienna ; . / | Malo; Priabona. | 11. Vibracella trapezoidea, Rss. .........|-++ MM HMIK.. --|Bocca di Sciesa; Malo 12. Membranipora macrostoma, Rss....|.-.|% K KKK... %|--- Miocene of Vienna; Wieliczka. S55 ——— tenuirostris, -A.....2...cccscecewcees XK aera ——— PINTER, ARG, ‘anc orecvccs cases © de dicots be eed Be ed eo Crag ; Pliocene of Italy. 15. —— Rosselii, Aud. .........cccceecesees SKIKE HE] -- |---| x | 16. —— patellaria, Moll .........cee.e ee KI | 17. —— appendiculata, RSs, .......6.0.0+6|/ KK * | 18. —— Hookeri, Hate .........ccceceeeeleee SA See eee x | 19. Micropora coriacea, Esp. ............ KIX KK | | 20. —— polysticha, F2s6...........ssceseeceee[ees Fac a ed ee kc | 21. —— parallela, Rss. ........ccsseceeeeee[er SeIK/E|...|--./]...].--/ Novezzina. 22. —— articulata, sp. nov. ...........0.2-Jeee|e* x“ | 28, —— oncullata, BSS. ....ceeececees «KIKI %K/K/K!HK| Miocene of Vienna; Latdorf; As- trupp; Pianosa; Malo; Priabona. 24, Cribrilina radiata, Moll...............+ Dd ial ead cad eed eel eee oe % Sollingen; ‘Miocene. Austria, &c.; Pliocene, Italy, Australia. 25. —— chelys, Kosch. SO nae eet SiBuG b> IE Ans [tae %*'% Priabona; Grancona. | 26. crenatimargo, RSS. ..........00+s.|++- KH HK W|---|...]---/ 96 Malo. | 27. Monoporella sparsipora, Rss. ......|-.. KISEKI FE] ---}. feel -- Cretaceous; Priabona; Malo. | 28. Lepralia subchartacea, d’Orb. ......|.-- ris aie ad on pod Be eee ---| Pyrenees; Hungary. 29. —— semilevis, P88. ............cccececee{ees *IKIX SIK...]... --» Galicia aad Hungary ; ; Malo; Pria- 80, =—='bisules ese. WA AL, cis ile. %/% KW) X%/HK|---| Hungary; Malo. ’[bona. 31. —— nodulifera, Ras. ........c.cceseceeee|ee- MK KKK]... 61! Malo; Priabona. 32. —— impressa, Rss. ....02...ceeseccevene|ees Bel SEI caliedieee toes * Bd. —— CXCENUTICA, WES......c.ccccccersccece|ers ne tO es Bee ed ol ee 34. —— syringopora, RSS................02.[++ *%I% Kl %/%/X)--- Priabona; Malo. 35. ——- bericensis, SP. MOV. ......seeeeeeee|eee|ees ici eta es SN ...|.--| Bocea di Sciesa. 36. —— lontensis, Sp. NOV.........seecseecee[ees KIKI)... ‘Malo. 37. Smittia coccinea, Abild. ...........008- Xl... | 38. —— ——, var. alifera, Bes. .......00.2.|-«- MIM KK .|-+-| Hungary; Malo. 39, —— Landsborovii, var. cheilopora, ESS gesete ea cneutccweesr ends swaccctes|ds0|sos|sn« Ke .41 Hee boa) ed Lad eon) Moravia 40. —— porrigens, P88. ..........scsecsceses[eoe|eoe X/%KI...|- Sdllingen A ——— OX ALAPAHA, FUSS. =chvosccnicevcccectseoee|oes XK 3X 42. Porella imbricata, Rss. ...........60..[+--| KI 43, —— marsupium, var. porifera, H. [ME]... SEISEl- .-| Fossil, New Zealand. 44, Rhamphostomella brendolensis, Re IOV epee ey be ecie ks cacetacvbedesces| ace %| 2% . 45. Porina coronata, Rss. ......ceccecesseee [ees % HK HEHEHE 1% Eocene of Hungary; Malo; Pria- 46, —— duplicata, ss. . wseccccce sees [ee el Mele) dele ae eee leee lee Hun ary ; Malo. bona. 47. —— papillosa, Rss. ..........eeeeeeeeeeleee % HK KEN 13¢|4¢]---| Oberburg, Neustift, Styria; Malo. 48, —— bioculata, sp. NOV. .......cececees [eos [eosleo iXt 49. Schizoporella Hoernesi, Rss, ......}... MK K.. ..%€|---| Curdies Creek. 50. —— squamoidea, Rss. ......eeseeeses [eee xe 51, —— unicornis, Johnst. ........ceceeecleeeleee | Selec alee ...| Miocene and Pliocene. 52, —-— serrulata, Bsa. ........cescccerereee/ooe[ore I%E].-.]-» .|...| Eisenstadt. Boe ——-. Om HON, GOCE. ook cescoieavessceneqfees %|G |-%]...]...]...]...]---| Malo. 54, —— poynsropare, a eae, ae ee %...|%|...|...]...]...|...| Lower Eocene of Mons. 55. —— Schreibersi, Rss. ......cecseees.[eee MIKI... .|..-| Hungary. DOs — COLMA, LOSE. Cove cccdescccecvocsccvcs|ese wx x 57. Fedora excelsa, Kosch. .......ccsocec.{es|eosleoe Seale .-|%|---| Brentonico,,Bocea diSciesa, Spiassi ; 58. Retepora tuberculata, Mss, .........|... > Se {Malo. 59. —— elegans, P88. ........sececeececeeelees X|.- AM 60. Cellepora proteiformis, Fss..........]... %'% KIKIKI...]... % 61. —— oligostigma, P88. ..........:.ceeeesJeee|ee Ze |---]eee|-0- % 62. —— pertusa, Sm. .........ceceesseeeeeees XK 63. Stichoporina simplex, Kosch. ...... HKI...|.--/ HK]... 1€e...] Novezzina. 64. Batopora multiradiata, Rss. .........|... x% x MIE)... |...1.. Priabona, Eocene of Bavaria and 65. —— Btoliczkai, Rss. ......seceeereeeee|eee [eee iG i% (Hungary; Malo. 66. Lunulites quadrata, 2s8. ...000..6.2.|.--]02|000]-0-|ees]oes[ece]encfes Between Sarego and Grotte, Colle MR. A. W. List of Species. WATERS ON Berici. NORTH-ITALIAN BRYOZOA, 3) With regard to structure, the closure of Porina coronata and Le- pralia syringopora by a plate with a tubule in the centre is some- what surprising, as this structure was supposed to he exclusively characteristic of the Cyclostomata. The plate is at some distance above the oral aperture. Some of the erect Hschara-forms have the zocecia at the side opposite, others alternate; and as this may be a useful specific character, it has been mentioned in the diagnosis; but it does not appear to have generic value. The position of these Bryozoan beds has been fairly worked out. Suess *, in 1861, placed them above the Priabona beds and below those of Sangonini. They are the “F.” of Bayan t,or Upper Eocene. Professors Hébert and Munier-Chalmas + have also examined the stratigraphical position of the Vicentine beds, and they place the ‘“¢marne ”’ of Brendola, &c., below the Crosaro and above the Gra- nella beds, but as part of a “méme ensemble,” considering the Brendola beds as Upper Eocene and equivalents of those of Biarritz. It will thus be seen that there is agreement in the views, and that the Bryozoa may be considered as of Bartonian age and may be called Upper Eocene. From their position the earlier writers called them Miocene. To some of these points I may have to refer more fully when dealing with the Cyclostomata. 1. ? CArENARIA TENERRIMA (Reuss). (PI. I. fig. 11.) Crisidia vindobonensis, Reuss, Foss. oly: Wien. Tert. p. 54, pl. vu. fig. 25. Unicrisia tenerrima, Reuss, Bryoz. von Crosaro, p. 279, pl. xxxiv. figs, (72 This is clearly not Cyclostomatous, as Reuss supposed, and, although it differs in some respects from any living Catenaria, it seems to find its place among them. It is not articulated, and the connecting tube is broken off in various positions. The aperture is terminal and seems to have a very wide sinus on the proximal border. . Loc. Val di Lonte (#ss.); Brendola; Montecchio Maggiore ; Fer- rara di Monte Baldo. 2. CATENICELLA SEPTENTRIONALIS, sp. nov. (PI. I. figs. 1-8.) The globule are amphora-shaped, perforated on the front, with an acute avicularium at each upper corner. Oral aperture nearly orbicular, with the lower margin emarginate ; the region round the aperture somewhat elevated. The dorsal surface is sometimes very slightly keeled. From Montecchio Maggiore there are several uni- and bi-globule ; but besides there is one with three zocecial cells (fig. 3), and another with four (figs. 4,5). This is most interesting, as showing that * Atti della Soc. Ital. di Sc. Nat. di Milano, 1861 ; and ‘‘ Ueber d. Gliederung des Vicent. Tert.,” Sitz. k. Akad. d. Wissenschaft., 1868. + “Sur les Tert. de la Vénét.,” Bull. Soc. Géol. France, sér. 2, vol. xxvii. ¢ Comptes Rendus de l’Acad. des Sc. vol. lxxxv. p. 259 & pp. 320. 6 MR. A. W. WATERS ON short internodes containing one, two, or even three zocecia must not be made an absolute generic character, although all living forms are built up of these short beads. The range in this case supports such forms as CO. internodia, Waters, and C. continua, W., being placed with the Catenicellide. Fig. 8 is a long internode with very long ovicells, but the structure of the zocecia and of the dorsal surface is the same as in those with shorter nodes, so that we seem to have a series from the uniglobular (fig. 1); though it may be a question whether fig. 8 should not be called a variety. Loe. Montecchio Maggiore. 3. CATENICELLA CONTINUA, sp. nov. (Pl. I. figs. 9, 10.) Zoarium with long internodes, zocecia on the anterior surface distinct ; vittze at each side of the zocecium, and a small triangular avicularium at each outer corner. Oral aperture rounded above, straight below, and there is either a suboral pore or a portion of the wall is thinner below the aperture. On the dorsal surface there is a curved vitta about the middle of the zocecium. In the long inter- nodes and in other respects this is closely allied to Catenicella inter- nodia, Waters (Quart. Journ. Geol. Soc. vol. xxxvii. p. 318, pl. xvi. figs. 78, 79), fossil from Curdies Creek, Australia, a form which I placed under Catenicella on account of the distinct vitte; and the discovery of C. septentrionalis, Waters, with occasional multilocular internodes indicates that I was probably right in not separating the Australian fossil from the Catenicellide. Loc. Fossil: Montecchio Maggiore ; Brendola. 4, ScRUPOCELLARIA ELLIPTICA (Reuss). (Pl. I. figs. 16, 17.) Bactridium ellipticum, Reuss, Foss. Polyp. Wien. Tert. p. 56, pl. ix. figs. 7, 8. Scrupocellaria elliptica, Reuss, Bryoz. von Crosaro, p. 260, pl. xxix. fig. 3; Foss. Bryoz. Oest.-Ung. Mioe. p. 148, pl. xi. figs. 1-9. Bactridium granuliferum, Reuss, Polyp. Wien. Tert. p. 56, pl. ix. fig. 6. The specimen figured from Montecchio Maggiore has an avicu- larium below the aperture ; but other specimens with the character- istic dorsal surface have no suboral avicularium, and it is an open question whether they should be specifically separated. In recent Caberea we find parts of the same colony with an avicularinm to each zocecium, while in other parts they are only occasionally found. I cannot agree with Mr. Hincks in identifying the British form with the fossil, and would suggest that the former stands as S. imermis, Norm. Loc. Val di Lonte (Rss.); Gaas, 8. France; Austrian and Hun- garian Miocene; Montecchio Maggiore. 5, ScrupocELtaRtA GRAcILIs, Reuss. (PI. I. figs. 12, 13.) Serupocellaria gracilis, Reuss, Bryoz. von Crosaro, p. 260, pl. xxix. fig. 4; Reuss, ‘‘ Fauna von Gaas,” Sitzungsber. Ak. Wissensch. Wien, vol. lvi. p. 466. NORTH-ITALIAN BRYOZOA. 7 A specimen from Montecchio Maggiore seems to be of this species. It is without avicularia on either surface ; but at the outer angle of each zocecium there has been a vibraculum (or strong spine), and the vibracular chamber is tubular. There has been a scutum over the front. I do not find in any of my specimens of Scrupocellaria from North Italy that there has been an avicularium at the outer angle. Reuss seems to have taken the vibraculum for an avicu- larium. Loc. Val di Lonte and Gaas (Rss.); Montecchio Maggiore. 6. ScRUPOCELLARIA BRENDOLENSIS, sp. nov. (Pl. I. figs. 14, 15.) The opesial aperture is about one half of the length of the zow- cium ; near the middle on the inner side the aperture is contracted, no doubt by the base of the scutum ; below the aperture there is a triangular avicularium directed outwards; at the upper outer angle there has been a vibracular appendage or a spine. On the dorsal surface there is a vibraculum to each zoccium, with the base attached in a small semicircular chamber near the inner edge of the zocecium. This would seem related to S. scabra, but differs from it in not having any avicularium at the outer upper angle. Loc. San Clementa (Montecchio Maggiore); Brendola. 7. SCRUPOCELLARIA MONTECCHIENSIS, sp. nov. (PI. I. figs. 21, 22.) The specimen figured cannot be identified with any of the others, and it therefore seems advisable to give it a name, though at best the determination of fragments of Scrupocellarze is not very satis- factory. The aperture is very large; there has been no scutum; and the vibracular chamber on the dorsal surface is very large, so that the external prolongation is also seen from the front. Loc. Montecchio Maggiore. 8. Bacrriprum Hacunowr, Reuss. (PI. L. figs. 18, 19.) Bactridium Hagenowi, Reuss, Foss. Polyp. Wien. Tert. p. 57, pl. v. fig. 28; Bryoz. von Crosaro, p. 266, pl, xxx1i. figs. 5, 6. The aperture is Schizoporellidan, with a wide sinus. The groove on the dorsal surface is usually very wide and deep, as figured by Reuss, but at other times there is not more than a divisional line. In the mode of growth this seems nearly related to Urceolipora and to Ichihyaria oculata, B. Loc. Val di Lonte (Reuss g Waters); Brendola; Montecchio Maggiore; Malo; Lower Eocene of Mons (Mun. & Perg.). 9. CELLARIA Revssr, d’Orb. Cellaria marginata, Reuss (pars), Foss. Polyp. Wien. Tert. p. 59, ply yal. fee 29. : Vincularia Reussi, d’Orbigny, Pal. Fr. vol. v. p. 60. Salicornaria Reussi, Reuss, Bryoz. von Crosaro, p. 261, pl. xxix. fig. 5. . 8 MR. A. W. WATERS ON There are only small fragments of this, which are without ovicells or avicularia, so that it is difficult to say which are its nearest allies; and in the absence of these characters there is nothing to distinguish it from C. bicornis, B., and C. tenwirostris, B., or C. ovicellosa, W. At first Reuss united it with C. fistulosa; but he afterwards saw that they were distinct. Manzoni, however, subsequently con- sidered them synonymous ; but in this I cannot agree. - Loc. Val di Lonte and Montecchio Maggiore (Zss.) ; Miocene ot Vienna (fss.): Brendola; Crosaro. ONYCHOCELLA. Several attempts have been made to group the Membranipore into fresh genera, but there has been very little success, as divisions have so often been based, not on fresh characters, but only on variation in degree of common ones—as, for instance, the sloping inwards of the wall of the area, though no doubt fresh characters will ultimately be found permitting of a division of the genus. There is, however, one section which may be separated, even though the limits scarcely admit of exact definition at present. It is only represented by two or three * living species; but was extremely abundant in Cretaceous times. Jullien f first attempted separation, based upon the nature of the large vicarious avicularia, of which the mandible is attached to the membranous cover, and there is no bar across the calcareous avicularian opening; also, as I have pointed out +, there are free chitinous appendages at each side of the base of the mandible. The avicularian opening is, there- fore, simple, and usually oval, or nearly round, and the mandibles, as far as known, are winged, and this also obtains in Membrani- pora permumia, Micropora lepida, H., and Foveolaria faleifera, B. Jullien established a family, of which the generic divisions were based principally upon the outline shape of the zoarium, and partly upon the shape of the opesia ; but these are very uncertain cha- racters for generic divisions, as may be seen from my figure of the recent Membranipora angulosa §, and also in fossil specimens. On this account, I am quite unable to follow Julhen with regard to the other genera into which he divides his family, and only accept the genus Onychocella. Besides the points to which Jullien drew attention, I would add that in Onychocella angulosa there are trabeculz bordering the operculum. This is general in Cellaria, as pointed out by Mr. Busk; and I have shown that it also occurs in Selenaria maculata ||; but when we go back to the Chalk we have in Escharella argus, d’Orb., a form which shows in a somewhat unex- pected way the connection between Onychocella and Cellaria ; for, in a specimen which I collected from Maastricht, there are teeth in the * Membranipora angulosa, Rss., and Melicerita dubia, Busk. t “Nouv. Div, des Bryoz. Cheil.,” Bull. Soc. Zool. France, vol. vi. 1881. { Journ. R. Microsc. Soe. ser. 2, vol. v. p. 106, fig. 42. § Ann. Mag. Nat. Hist. ser. 5, vol. iii. pl. xiii. fig. 3. | Suppl. ‘ Challenger’ Report, p. 37. NORTH-ITALIAN BRYOZOA. 9 aperture, both on the proximal and distal edge. Although the *¢ onychocellaires”’ (vicarious ayicularia) are not quite as simple, and a tongue projects into the avicularian opening, yet in the main they agree with those of O. angulosa. From general appearance we should not have imagined a connection between the now widely separated genera Cellaria and Onychocella; but, having obtained the key in these minute characters, we seem to find the connection in other species as well. Koschinsky * has made a genus, Ahagasostoma, which, it would seem, should be considered a synonym. The clefts (Spalten) shown at the side of the oral aperture in Koschinsky’s figures are more marked than in any specimens which | have examined, and usually in O. angulosa there is nothing of the kind; but it sometimes occurs, and is then caused by the greater development of the lower lip. The slits at the side of the oral aperture in Aspidostoma are filled by the operculum, but this is not here the case. 10. OnycHocrLLa aneuLosa (Reuss), (non d’Orb.). (a) Incrusting. Cellepora angulosa, Reuss, Foss. Polyp. des Wien. Tert. p. 93, pl. xi. fig. 10. Membranipora angulosa, Reuss, Bryoz. von Crosaro, pp. 253, 262, 291, pl. xxix. figs. 9-11; Foram. Anth. u. Bryoz. von Oberburg, p- 30; Bryoz. Oest.-Ung. Mioc. p. 185 (45), pl. x. figs. 18, 14; Foss. Fauna d. Olig. von Gaas, p. 470; Manzoni, Brioz. di Castrocaro, p- 8, pl.i. fig. 1i ; Brioz. foss. Ital. pt. 4, p.9, pl. ii. fig. 10 ; Waters, Ann. Mag. Nat. Hist. ser. 5, vol. ii. p. 122, pl. xin. fig. 3 ; Pergens, Phioc. Bryoz. von Rhodes, p. 16. Membranipora antiqua, Busk, Quart. Journ. Micr. Soc. vol. ‘vi. Pp. 202, pl. xx. figs. 1, 2 ! ? Molha antiqua, Smitt, Floridan Bryozoa, p. 12, pl. ii. fig. 73. ? Membranipora hexagona, Busk, Quart. Journ. Microsc. Soc. vol. iv. p. 308, pl. x11. fig. 4. Onychocella antiqua, yj ullien, Bull. Soc. Zool. de France, vol. Vi. p. 9. Onychocella Marioni, Jullien, loc. cit. p. 7, woodcut. Amphaiblestrum angulosum, Pergens, Tee. von Wola Lu’zanska, Bull. Soc. Belg. de Géol. vol. iii. p. 67. ? Rhagasostoma hexagonum, Koschinsky, Bryoz. ilt. Tert. siidl. Bayerns, p. 30, pl. v. figs. 5-7. (6) In Vineularia stage, Pl. I. fig. 20. Eschara excavata, Reuss, Foss. Polyp. Wien. Tert. p. 72, pl. vill. fig. 36. Biflusira excavata, Manzoni, Brioz. foss. Mioc. d’Aust. ed Ung. p- 67, pl. xiii. fig. 44. Vincularia pare e Orb. Pal. Fr. p. 108, pl. 602. figs. 12, 13, pl. 673. figs. 2 * Bryoz. alt. Tert. siidl. Bayerns, p. 29. 10 MR, A. W. WATERS ON Eschara Lamarcki, Hagenow, Bryoz. Maast. Kreide, p. 74, pl. ix. figs. 2, 3, 4. Vineularia geometrica, Rss. Bryoz. von Crosaro, p. 276, pl. xxxiii. fig. 16. Periteichisma geometrica, Koschinsky, Bryoz. alt. Tert. sidl. Bayerns, p. 25. | Vincularia disparilis, Beissel, Bryoz. Aachener Kreide; Verh. Hollandsche Maat. d. Wetenschappen, pt. xxii. 1865, p. 15, pl. i. figs. 7, 8. Comparative measurements of the zowcia show no difference between the incrusting form called angulosa, and the erect form named excavata ; and Reuss (Bryoz. von Crosaro, p. 79) refers to M. angulosa in the Biflustra-stage. From Brendola and the other localities there are many speci- mens which are at first incrusting, and then free in the geometrica- form; but the almost cylindrical VW. excavata passes by gradations into compressed Bifiustra-like forms. The M. sexangularis, Goldf., of the Maastricht Chalk is of about the same size, and at one time I was inclined to unite them; but the surface of the zoccium is almost flat in sewangularis instead of depressed, and the zocecia being proportionally shorter, there results a more regular hexagonal appearance, Probably, besides the above syncnyms, many others should be added, as Vincularia excavata, d’Orb. loc. cit. p. 69; V. santonensis, @Orb., p. 73; V. leda, d’Orb., p. 88 ; Eschara acmon, d’Orb., p. 115 ; E. allica, VOrd., p. 125; #. arcas, d’Orb., p.127; EH. actea, d’Orb., p. 116; &#. arethusa, @Orb., p. 127. Loc. Cretaceous: Maastricht; Royan, &c. Lower Tertiary: Val di Lonte ; Brendola; Montecchio Maggiore; Crosaro; Priabona; Novezzina; Malo; Oberburg ; Gaas; Nussdorf, near Vienna (Manz.) ; Gotzreuth ; Wola Lu’zanska; Dego; Turin, &c. Pliocene: Italy ; Sicily; Rhodes. Living: Mediterranean ; Madeira; Ile de France ; Florida (?). VIBRACELLA, gen. Nov. The Plustrellaria trapezoidea of Reuss is an extremely interest- ing form, as in respect of the zocecia it seems connected with Ony- chocella ; but, instead of the vicarious avicularia, there are cells scattered among the zocecia, which I should call vicarious vibracular cells. The beak-like prolongation is wanting, and instead one side of the vicarious cell is very much raised, indicating a vibracular appendage, similar to that of Cupularia, in which genus the one side is in the same way usually “auriform.” The J. trapezordea cannot, on account of these vibracula, be placed with Onychocella, as at present defined ; to leave it with Membranipora after removing Onychocella could only be a provisional arrangement; and /lus- trellaria was made to include free-growing forms of the Onychocella type; so that, although reluctant to do so, until we know more of its allies, 1 propose the genus Vibracella for forms in which the zocecia have moderately large opesial openings, and in which there eae. - NORTH-ITALIAN RRYOZOA. LE -are vicarious eared vibracular cells. The only species is V. tra- pezordea (Reuss). J1. VisraceLLa TRAPEZOIDHA (Reuss). (PI. I. fig. 23.) Cellepora trapezoidea, Reuss, Foss. Polyp. Wien. Tert. p. 96, pl. xi. fig. 21. Flustrellaria trapezoidea, Reuss, Bryoz. von Crosaro, p. 268, pl. xxix. fig. 14. The most important points have been dealt with when describing the genus. As a rule, the zoarium is discoid or conical, but from some speci- mens it would seem to sometimes grow in larger flat pieces. Reuss, in his first description, says “ incrusting ;” but in his second he figures and describes it as free. One specimen from Brendola is -incrusting, but all the others are free. Selenaria miocenica, Seguenza, may be this or Onychocella angu- losa; also Flustrellaria hewagona, d’Orb., from the Senonian, seems very closely allied; but in the absence of vibracula or avicularia it is in both cases impossible to say whether they are identical. Loc. Val di Lonte (Reuss and my collection); Brendola ; Mon- teechio Maggiore ; Bocca diSciesa; Ferrara di Monte Baldo ; Malo. 12. MemBRANIPORA MACROsTOMA (Reuss). Cellaria macrostoma, Reuss, Foss. Polyp. Wien. Tert. p. 64, pl. viii. figs. 5, 6. Biflustra macrostoma, Reuss, Bryoz. von Crosaro, p. 274, pl. xxxiil. figs. 12, 13. Flustrellaria macrostoma, Manzoni, Brioz. foss. Mioc. d’Aust. ed Ung. p. 67, pl. xiii. fig. 46. 2 Membranipora macrostoma, Waters, Quart. Journ. Geol. Soc. vol, xxxvii. p. 323, pl. xiv. figs. 18, 19; Koschinsky, Bryoz. Alt. Tert. stidl. Bayerns, p. 22. Vaginopora texturata, Reuss, Foss. Polyp. Wien. Tert. p. 73, fieax. oe. 1. Flustrellaria texturata, Reuss, Foss. Fauna von Wieliczka, p. 119 ; Manzoni, Brioz. foss. Mioc. d’Aust. ed Ung. p. 67, pl. xiii. fig. 45. Biflustra papillata, Stoliczka, Foss. Bryoz. Orakei Bai, p. 154, pl. xx. fig. 14. From Brendola there is a fragment of M. concatenata, Rss., either in a single free layer or with two layers back to back; and, as pointed out by Koschinsky, the zocecia are the same as those of M. macrostoma, both in shape and size. There are also some fragments with squarer zocecia, which should perhaps be called M. Savartir, Aud. Loc. Val di Lonte ; Montecchio Maggiore; Brendola; Ferrara di Monte Baldo; Ronzo; Nussdorf, &c. (Manz.); Wieliczka (Rss.) 5 Gotzreuth (Kosch.). 13. MemBRanipora TENUIROSTRIS, Hincks. Membranipora tenuirostris, Hincks, Ann. Mag. Nat. Hist. ser. 9, 12 MR. A. W. WATERS ON vol. vi. p. 70, pl. ix. fig. 3; op. cit. vol. x. p. 7; Waters, Journ. R. Microsc. Soc. ser. 2, vol. v. p. 755, pl. xiv. fig. 41. Membranipora Flemingzi, Waters, Ann. Mag. Nat. Hist. ser. 5, Wor ii. p. 122, pli xii. fic. 2. Loc. Living: Mediterranean; Madeira; Queen Charlotte Island. Fossil: Val di Lonte ; Montecchio Maggiore. 14, Memsranipora Dumerini (Aud.). (Pl. II. fig. 4.) For synonyms, see Hincks, Brit. Mar. Polyz. p. 156. Membranipora Dumerilit, Koschinsky, Bryoz. alt. Tert. sudl. Bayerns, p. 21. Membranpora bicornis, Manzoni, Form. Terz. di Reggio, p. 80, pl. viii. fig. 10. The Vicentine specimens may be taken as fairly typical, 1. Du- merilit having two avicularia above the ovicell, and zocecia without an ovicell, having either one or two avicularia. Thisis very similar in shape to M. appendiculata, but is much smaller, the opesia being only about 0°35 millim. long. I described a fossil from New Zealand (Quart. Journ. Geol. Soc. vol. xliii. p. 45) as MZ. Dumerilu, although somewhat divergent from the type. Loc. Living: European Seas. Fossil: Crag; Pliocene of Cala- bria; Waipukerau, New Zealand (?); Brendola; Montecchio Maggiore; Val di Lonte; Gotzreuth (Kosch.). 15. Mempranipora Rosser (Aud.). (Pl. II. figs. 1, 2.) Flustra Rosseli, Audouin, in Savigny’s ‘ Egypte,’ p. 240, pl. x. fie itt, aR atibond Rosselii, Waters, Ann. Mag. Nat. Hist. ser. 5, vol, ili. p. 121; Hincks, Brit. Mar. Polyz. p. 166, pl. xxii. fig. 4 (which work see for synonyms). Cellepora deplanata, Reuss, Foss. Polyp. Wien. Tert. p. 96, pl. xi. fig. 20. Membranipora deplanata, Reuss, Bryoz. von Crosaro, p. 263, pl xxix. fig. 12. Periteichisma deplanatum, Koschinsky, Foss. Bryoz. sudl. Bayerns, p. 26. A colony from Brendola has a slightly raised small hood-like ovicell. There are a few zoccia (fig. lw) nearly double the width of ordinary zocecia, and they cannot be considered ovicelligerous, seeing that there are ovicells; but perhaps irregularities on the surface on which they are growing has caused this difference in size. There are, besides, in some specimens raised giant zocecia, occupying about the space of three zocecia. As the living M. Ros- selui is pretty regular, perhaps we should call this var. deplanata. Loc. Living: Britain; Mediterranean. Fossil: Pliocene of Italy and Sicily; Miocene of Vienna Basin; Val di Lonte; Montecchio Maggiore ; Brendola; Goétzreuth (Kosch.). | NORTH-ITALIAN BRYOZOA. 13 16. MempraniporRA PATELLARIA (Moll). Eschara patellaria, Moll, Die Seerinde, p. 74, pl. iv. fig. 20. Diachoris simplew, Heller, Bryoz. Adriat. Meeres; Verh. k. k. zool.-Bot. Gesellsch. vol. xvii. 1867, p. 94, pl. 1. fig. 4. Mollia patellaria, Smitt, Floridan Bryoz. p. 12, pl. 11. fig. 72. Diachoris patellaria, Waters, Ann. Mag. Nat. Hist. ser. 5, vol. iii. p- 120, pl. x. figs. 6-9. The fossil from Montecchio Maggiore is a very typical MW. patel- laria, with six connecting-tubes. Loc. Living: Naples; Florida. Fossil: Montecchio Maggiore. 17. MemsBRaNIPoRA APPENDICULATA (Reuss). | (Pl. II. fig. 3.) Cellepora appendiculata, Reuss, Polyp. Wien. Tert. p. 96, pl. xi. fig. 22. Membranipora appendiculata, Reuss, Fauna deutsch. Oberoligoc. p- 631, pl. ix. fig. 4; Oest.-Ung. Mioc. p. 181, pl. ix. figs. 138-16 ; Foram. Anth. Bryoz. Septarienthones, p. 171; Waters, Quart. Journ. Geol. Soc. vol. xxxviil. p. 504, pl. xxii. figs. 2-5; Seguenza, Form. Terz. Prov. di Reggio, Accad. dei Lincei, vol. cclxxvii. p. 80; Koschinsky, Bryoz. alt. Tert. siidl. Bayerns, p. 23. Membranipora cyclops, Busk, Catal. Mar. Polyz. p. 61, pl. Ixv. Ae. 3. Membranipora monopora, Reuss, Bryoz. yon Crosaro, p. 262, Diexkix. fe) 7; A specimen from Montecchio Maggiore has wide (0°5 millim.) oval opesia, below which there is a single raised avicularium placed laterally. The determination of simple Membranipore is seldom very satisfactory, and in the present case it is doubtful whether this is not the same as M. rhynchota, Busk. Loc. Brendola; Val di Lonte (Ass.) . Montecchio Maggiore. 18. Memsranrpora Hooxeri, Haime. Membranipora Hookeri, Reuss, Bryoz. von Crosaro, p. 252, pl. xxix. figs. 6, 8, and my coll. The opesia are about 0°35 millim. long and 0-2 millim. wide. Loc. Crosaro (fss.); Val di Lonte (#ss.); Montecchio Maggiore (Rss.) ; India (Haime). 19. Micropora cortacea (Esper). (PI. II. fig. 9.) Micropora coriacea, Hincks, Brit. Mar. Polyz. p. 174 (which see for synonyms). Membranipora gracilis, Reuss, Bryoz. von Crosaro, p. 291, pl. xxix. fig. 13. . Besides those specimens with elliptical zocecia, there is one from Val di Lonte with the sides parallel, and this I suppose must also be united with the MW. gracilis (von Minster). Loc. Living: Britain; Florida; Azores. Fossil: Brendola; Montecchio Maggiore; Val di Lonte; Pliocene of Calabria (Manz.). 14 MR. A. W. WATERS ON 20. Micropors potysticua (Reuss). (Pl. II. fig. 7.) Cellaria polysticha, Reuss, Foss. Polyp. Wien. Tert. p. 61, pl. vil. fig. 33. Eschara polysticha, Reuss, Bryoz. von Crosaro, p. 269, pl. xxxii. g. 3. ? Steganoporella similis, Koschinsky, Bryoz. alt. Tert. sudl. Bayerns, p. 34, pl. i. figs. 8-10. The zoaria vary considerably in the number of rows of zocecia. The zocecia are elongate, and about the same size and shape as those of Micropora parallela, but differ from that species in the absence of ayicularia and in having opesiules. The opesiules in some parts are not apparent, and in different specimens vary in position, occurring sometimes about the middle of the zocecium, but more usually only about one quarter of the way down. Although the zocecia of Steganoporella similis, K., are a trifle larger, the characters are similar, and probably they should be united. Loc. Val di Lonte (Rss.); Brendola; Ferrara di Monte Baldo; Montecchio Maggiore ; Goétzreuth (Kosch.). 21. Micropora PARALLELA (Reuss). (PI. IL. fig. 8.) Eschara parallela, Reuss, Bryoz. von Crosaro, p. 272, pl. xxxiil. fig. 2. When properly cleaned a small triangular avicularium near the oral aperture, directed downwards, is distinctly seen. Ovicell not much raised, wider than a zocecium; apparently there has been an area on the front. When sections are being made, the shell just above the avicularium is seen to be thinner than the surrounding parts, often giving the appearance of a peristomial notch, and this can also be seen in some specimens without preparation. Zocecia at the side of the zoarium alternate. The avicularia were overlooked by Reuss, and since they do not occur on all zocecia, and may be covered by the matrix, it often appears as if there was only a pore, as described by Reuss; and Pergens, thinking this was the case, has made the genus Houzeawina for a fossil which he considered the same as the EL. parallela of Reuss. ‘This is probably not identical with the Vicentine fossils, as M. Pergens writes that the hole was “complet,” and that there was no avicularian chamber ; and no doubt I may be allowed to say that, after seeing a specimen from my collection, he agrees with my view. Loc. Val di Lonte; Brendola; Montecchio Maggiore ; Crosaro ; Novezzina. 22. MicROPORA ARTICULATA, sp. nov. (PI. II. figs. 5, 6.) Specimens from Montecchio Maggiore are about the same size as the recent MW. ratoniensis *, Waters, from New Guinea. Although * It should have been atowensis, but the name of the locality was read Raton, and there does not now seem any reason to change it. Mr. Whitelegge informs me that he has found it recent from Singapore. NORTH-ITALIAN BRYOZOA. 15 the zocecia are not always arranged quite as diagonally as in that recent form, yet the diagonal arrangement is usually seen. The fossil differs from the recent species in the shape of the “ special organs,” which, instead of being a small triangular avicularium placed diagonally, seems to have been a vibraculum; at any rate, below the zocecial area there is an eminence with two almost equal pores in the line of the zoarial axis. There is a suboral slit on one side only of the area. The zoccia are usually arranged on the four sides of the axis; but in one case they are only on two sides, in this respect resembling Diplodidy yma complicata, Rss., from the “ Oligociin ” of Gaas, ne the Cellularia diplodidymoides of Meunier and “Pergens from the Chalk. We have thus four closely allied forms, with similar-shaped zocecia, placed diagonally, with a long suboral slit on one side and a special organ below the aperture. 23. Micropora cucuLLATA (Reuss). Cellaria cucuilata, Reuss, Foss. Polyp. Wien. Tert. p. 60, pl. vii. fig. 31. eee cucullata, Manzoni, Brioz. foss. Mioc. d Austr. ed Ung. p. 69, pl. xv. fig. 50, pl. xvi. fig. 53. Eschara costata, Reuss, loc. cit. p. 72, pl. viii. fig. 37. Eschara Reussi, Stoliczka, Olig. Bryoz. von Latdorf, p. 88; Reuss, Fauna des deutschen Oberoligocins, p. 36. Vincularia Haidingert, Reuss, Bryoz. von Crosaro, p. 275, pl. xxxiil. figs. 14, 15. Biflustra sulcata, Gottardi, Brioz. foss. di Montecchio Maggiore ; Atti d. Soc. Veneto-Trentina di Sc. Nat. vol. ix. fase. ii. p. a) pl. xiv. fig. 2. Ste gamoporella elegans, Koschinsky, Bryoz. alt. Tert. siidl. Bayes 35. : ? Eschara elegans, M.-Edw. Ann. Se. Nat. sér. 2, vol. vi. p. (17) 337, pl. xii. fig. 13. | Mier opora cucullata, Pergens, Bryoz. von Wola Lw’zanska, p. 67. Salicornaria (Cellar va) cucullata, Gioli, Brioz. Neogenici dell’ Isola di Pianosa; Atti Soc. Tose. Sc. ‘Nat. vol. x. p- (10). Besides the erect zoaria in the Vincular ia-form, there are in- crusting specimens from both Val di Lonte and Premio In the erect and incrusting zoaria the ordinary zocecia are similar in size and shape; but the zocecia which we conclude are ovicelligerous are somewhat larger and wider in the incrusting specimens ; perhaps, however, this is to be accounted for by the conditions of growth. The large zocecia have a large shelf above the aperture. The proximal edge in the ordinary zocecia is much thinner than the distal border, but is continuous, so that the appearance is entirely that of Micropora; on the other hand, the broad ovicellular cells are like those of recent Steganoporella magnilabris, and I am unable to find grounds for separating these two genera. 16 MR. A. W. WATERS ON A specimen of Eschara Egea, d’Orb., from Royan has similar larger ovicelligerous cells. Loc. Several Miocene localities in Austria and Hungary. Val di Lonte ; Brendola; Montecchio Maggiore; Ferrara di Monte Baldo; Ronzo; Crosaro; Priabona; Malo; Gotzreuth (Kosch.), Latdorf ; Wola Lu’zanska; Astrupp (Miocene), Pianosa (Miocene). 24, CripRitina RADIATA (Moll). There is considerable variation in the appearance of various specimens, caused by difference in the number of coste; and in some cases vicarious avicularia are present; also some specimens from Val di Lonte and Brendola have the lower lip somewhat raised, and at each side of the aperture there is a small avicularian or vibracular opening, giving the appearance of C. puncturata of Busk. The pores between the coste are more numerous than in C. Hauert, Reuss, though, when crushed, they look like Reuss’s figure 16, pl. xxxil., which he calls ZH. Hauerz, but I have only seen this shape of zocecia in incrusting forms; and stems of other Bryozoa are often so completely covered that without careful examination they might be taken for erect specimens. Loc. Living: Cosmopolitan. Fossil: Val di Lonte; Brendola; Montecchio Maggiore; Novezzina; Sollingen; Wola Luzanska; various Austrian and Hungarian Miocene localities. Pliocene: Crag ; Italy ; Sicily; and Rhodes. Recent, Australia. 25. CRIBRILINA CHELYS, Koschinsky. (Pl. I. fig. 10.) Celleporaria radiata, Reuss (non Moll), Bryoz. von Crosaro, M0. Joep. xx, fig. 9, Cribrilina chelys, Koschinsky (nom. nov.), Bryoz. alt. Tert. siidl. Bayerns, p. 56; Pergens, Bryoz. von Wola Lwzanska, p. 70. There is a large oval avicularium at one or both sides of the zocecium, and the large pores around the area are irregularly placed. What Reuss described as the large pore above the aperture is anu avicularium, that of one zocecium often being situated above the aperture of its neighbour. ‘There are also a few very large vicarious avicularia. The ovicell is very large, but slightly raised, and has an irregularly perforated area on the front, and reminds us of the ovicell of Lepralia occlusa, B. The ovicell and vicarious avicula- rium are added to fig. 10 from different parts of the colony. There is considerable difference in the appearance of various parts of the colony, as in some parts the avicularia are very numerous, but elsewhere absent, and there are sometimes irregular pores at the side of the zocecium ; but in no case have I seen the regularity figured by Reuss, and probably a worn specimen was examined, and then the artist used his imagination. JI have specimens incrusting other objects in a single layer; but, as a rule, there are several layers superimposed. Loc. Priabona (Rss.); Brendola; Val di Lonte; Grancona; Gotzreuth (Kosch.); Wola Luw’zanska (Perg.). RSC NORTH-ITALIAN BRYOZOA. U7 26. CRIBRILINA CRENATIMARGO (Reuss). Cellaria Haueri, Reuss, Foss. Polyp. Wien. p. 63, pl. viii. fig. 9. Eschara crenatumargo, Reuss, op. éit. p. 72, pl. viii. fig. 38. Cribrilina crenatimargo, Pergens, Bryoz. von Wola Lu’zanska, 69: : Eschara Hauweri, Reuss, Bryoz. von Crosaro, p. 271, pl. xxxu. figs. 14-16. The zocecial characters are very similar to those of C. radiata, ‘Moll, but the zocecia are more elongate, the coste are nodulated, and the number of pores between the coste# is fewer than in the fossil C. radiata. The front is often depressed, as shown by Reuss, op. cit. pl. xxxil. fig. 15; but this seems to arise from the front wall being thin and pressed in. The erect form of Cribrélina is unknown living; but it is possible that this should be looked upon as erect “C. radiata, Moll. The specific name Hauert has been given to another Cribriluna, the Lepraha Hauert, Rss. (Bryoz. Oest.-Ung. Mioc. p. 170, pl. i. figs. 1, 2), and it therefore seems better to drop the name Havuert. Loc. Val di Lonte; Brendola; Montecchio Maggiore; Ferrara di Monte Baldo; Malo; Wola Lu’zanska (Pergens); Pap-Patak, Hungary (Perg.); Siebenbiirgen (Perg.). - 27. MonoporELia sparsrpora (Reuss). (Pl. IT. fig. 11.) Lepralia sparsipora, Reuss, Bryoz. von Crosaro, p. 263, pl. xxx. ie, Ly Homalostega exsculpta, Marsson, Die Bryoz. der weiss. Schreib- kreide der Insel Riigen, p. 95, pl. x. fig. 2. From Val di Lonte this occurs both in an incrusting and in a compressed Hschara-form ; from Montecchio Maggiore it is in the Eschara-form, and from Brendola it is incrusting. Ovicells plain, very wide, globose, moderately raised. Lepralia nuda, Reuss (Oest.-Ung. Mioc. p. 33), appears closely related. Loc. Priabona (#.); Val di Lonte; Brendola; Montecchio Mag- giore ; Ferrara di Monte Baldo ; Malo; Riigen (Cretaceous, Marsson). 28. Lepratia supcHartacea (d’Arch.). (PI. I. fig. 12.) Eschara subchartacea, @ Arch. Mém. Soc. Géol. France, sér. 2, vol. ii. p. 410, pl. ux. fig. 2; Reuss, Bryoz. von Crosaro, p. 269, pl. xxxii. fig. 4. Eschara chartacea, d’ Arch. op. cit. vol. 11. p. 196, pl. v. fig. 13. Cellaria stenosticha, Reuss, Foss. Polyp. Wien. Tert. p. 64, pl. vi. fig. 10. Eschara stenosticha, Reuss, Bryoz. von Crosaro, p. 269, pl. xxxil. fig. 2. The thin compressed form described by Reuss as subchartacea is fairly common in Brendola, but not in the other localities; and to some zocecia there is a small avicularium within the depression for the oral aperture, and also there is sometimes an avicularium on the middle of the zocecium. The branches spread out in a some- what fan-shaped manner, at first being about 2 millim. wide, and Q.J.G.8. No. 185. c 18 MR. A. W. WATERS ON expanding to about 10 millim. The cylindrical form, which may be also with or without avicularia, is less common, and was described as EH. stenosticha. Lateral teeth are seen within the oral aperture. Zocecia at the side alternate. Loc. Val di Lonte and Pyrenees (fss.); Brendcla ; Montecchio Maggiore; Pap-Falvi-Patak, Hungary (Perg.). 29. Lepratia seMILzvIs (Reuss). Eschara semilevis, Reuss, Bryoz. von Crosaro, p. 270, pl. xxxil. Sco. Eschara larva, Reuss, Foss. Polyp. Wien. Tert. p. 69, pl. Vill. fig. 29; Pergens, Bryoz. von Wola Lu’zanska, p. 70. Eschara Suesst, Reuss, Bryoz. von Crosaro, p. 270, pl. xxxii. fig. 9; Pergens, op. cit. p. 70. Eschara intermedia, Gottardi, Brioz. foss. di Montecchio Mag- giore; Atti della Soc. Veneto-Trentina di Se. Nat. vol. ix. p. 307, pl. xiv. He.) GO. At the side of the aperture there is usually on one or both sides a narrow spatulate avicularium, placed either slightly diagonal or directed straight upwards. The characteristic avicularia and ovi- cells occur in specimens both with the zoarial growth of semilevis and Swess?, showing that Reuss was right in thinking that these two might be only varieties of growth. In one specimen the bases of five spines round the aperture are very distinct, the lower two being the largest, and these two are figured by Gottardi in his E. intermedia. The zocecia seen at the side of the colony are alternate. ioc. Val di Lonte (Rss. g& Waters); Brendola; Montecchio _ Maggiore; Priabona; Novezzina; Ronzo; Wola Lu’zanska in Galicia (Perg.) ; several localities in Hungary (Perg.). 30, Lepratia Bisutca (Reuss). (Pl. I. figs. 16-18, & Pl. III. fig. 1.) . Eschara bisulca, Reuss, Bryoz. von Crosaro, p. 270, pl. xxxii. es LO. ? Schizoporella bisulca, Koschinsky, Bryoz. alt. Tert. siidl. Bayerns, . 49. Eschara macrodonta, Reuss, op. cit. p. 271, pl. xxxii. fig. 13. Eschara fenestrata, Reuss, op. cit. p. 290, pl. xxxii. fig. 5. This is an extremely interesting form on account of the great difference in shape between the ordinary and ovicelligerous zocecia. In the aperture there is usually an avicularium ; and in this respect, and also in the elevation at the side of the aperture, it resembles Smnittia (Porella) cervicornis. Sometimes this avicularium takes a spatulate form, and may be depressed (as in Pl. IT. fig. 17), or may be much Bieeied (fig. 16). In many other species, in the same way, a smail oral avicularium is sometimes replaced by a large spatulate one, as for instance in Schizoporella auriculata and Porella cervicornis, &c. Occasionally, instead of the elevation at the side of the aperture, two teeth are formed within it (Pl. ITT. fig. 1). NORTH-ITALIAN BRYOZOA. 19 The ovicelligerous zocecia, instead of being parallel with the axis of the zoarium, turn at right angles to it, and are much raised in their distal end, with the ovicell somewhat recumbent behind the peristome. In the form of the aperture and the ovicells this is closely related to ZL. nodulifera (Rss.). Pl. Il. fig. 16 is the L.- _ microdonta-form, with the row of fine linear pores round the border, and the separating ridges running in to the aperture, with the avicu- laria and ovicells resembling those of more typical L. bisulca. I do not feel at all sure from the description that Koschinsky had the same thing before him. Loc. Val di Lonte and Montecchio Maggiore (Rss. §& Waters) ; Brendola (Waters); Ferrara di Monte Baldo (W.); Novezzina ; Malo; Ronzo (W.); Crosaro (W.); Goétzreuth ? (Kosch.) ; several Hungarian localities (Perg.). : 31. LepRALIA NODULIFERA (Reuss). (PI. II. figs. 13, 14.) Eschara nodulifera, Reuss, Bryoz. von Crosaro, p. 271, pl. xxxii. fees. Ut, Periteichisma noduliferum, Koschinsky, Foss. Bryo oz. alt. Tert. stidl. Bayerns, p. 27. Amphiblestrum noduliferum, Pergens, Bryoz. von Wola Lu’zanska, p. 67. The aperture is slightly rounded below, and sometimes the thick calcareous growth of the divisional walls extends round the aperture, but sometimes it only occurs above the aperture. In some cells the proximal end is raised, and in one colony there are ovicells behind the aperture of such zocecia. In one case there seems to bea raised avicularium at the border of the zocecium, but I cannot speak with certainty about it. This seems to have nothing in common with the other Perite- chisme of Koschinsky ; and the genus appears based on the nature of the calcification rather than upon characters of much value; in fact some have an opesial opening, but in others, as in nodulifera, the aperture has, no doubt, been closed by an operculum. Loc. Val di Lonte and Montecchio Maggiore (Zss.); Brendola (Waters); Ferrara di Monte Baldo (W.); Ronzo (W.); Priabona ; Gotzreuth (Kosch.); Wola Lu’zanska (Perg.). 32, Lepraria impressa (Reuss). (PI. II. fig. 15.) Vincularva vmpressa, Reuss, Bryoz. von Crosaro, p. 276, pl. xxxiv. fig. 2. Amphiblestrum impressum, Pergens, Bryoz. von Wola Lwzanska, p. 67. | On a specimen from Brendola there is a raised round ovicell above the aperture, perforated with pores somewhat more crowded than those on the front of the zocecium. In some cases there is an avicu- larium at one side below the aperture, and then it looks like an early stage of Cellepora protetformis, Rss. Loc. Val di Lonte ; Brendola; Wola Lu’zanska (Perg.). ; C2 20 MR. A. W. WATERS ON 30. LEPRALIA EXCENTRICA, Reuss. Lepralia excentrica, Reuss, Fauna deutsch. Oberoligoc. p. 28, pl. xv. fig. 4; Fauna des Septarienthones, p. 175, pl. viii. fig. 2; Bryoz. von Grosaro, p. 256. ? Cumulipora angulata, Reuss, Fauna des Septarienthones, p. 179, pl. vil. ne, 12, A cylindrical zoarium from Montecchio Maggiore, with the zocecia distinct, with broad radial grooves round the border, and a raised suboral pore, would seem to be the L. excentrica of Reuss ; but so many species have been described with similar zocecia that no doubt there are many synonyms. ‘The specimen from Montecchio Maggiore has large spatulate vicarious avicularia. Loc. Sollingen; Oberoligociin of Doberg; Crosaro (Rss.) ; Montecchio Maggiore. 34, Leprarra (?) syrrncopora (Reuss). (PI. III. figs. 2, 3, 4.) Eschara syringopora, Reuss, Foss. Polyp. Wien. Tert. p. 68, pl. viii. fig. 23; Bryoz. von Crosaro, p. 269, pl. xxxil. fig. 1. Schizoporella perspicua, Koschinsky, Bryoz. alt. Tert. stidl. Bayerns, p. 49, pl. iv. fig. 3. ? Eschara minor, Reuss, Bryoz. von Crosaro, p. 272, pl. xxxiii. fig. 4. _ ? Eschara polystomella, Reuss (non Manz.), Foss. Polyp. p. 70, pl. vii. fig. 23. ?Eschara semitubulosa, Reuss, Bryoz. von Crosaro, p. 272, pli sco. fig. 3. _ This isan extremely interesting species on account of the peculiar bar across the aperture. Zoarium in the Eschara-form ; branches small, compressed, with the zocecia of the two layers opposite. Zocecia elongate, the upper end slightly projecting, and a row of large pores down each side of the zocecium. In the central cells these rows are near together, but in the outer zocecia there is a considerable space between them. About the middle of the oral aperture on each side there is a pore, and in many cases a bar across, as figured by. Stoliczka in his Escharifora ornatissima (Bryoz. von Latdorf, p. 86, pl. i. fig. 7), but the shape of the zocecia in that species is hexagonal. | Reuss in his second paper figures his H. syringopora with the oral pores lower than the aperture ; and sometimes when the aperture is somewhat broken down it has this appearance. There are sometimes also other large pores on various parts of the surface, usually one about halfway down, and these may be avicularian. . In some specimens there are ‘‘ closures” over the aperture, and these have a tubule in the centre similar to, those of so many Diastopore, &e. This is at a higher level than the operculum. It is abundant at all the stations. Loc. Val di Lonte and Montecchio Migoeiores ‘(Pade Brendola ; NORTH-ITALIAN BRYOZOA. _ oF Ferrara di Monte Baldo ; Ronzo ; Priabona (Kosch. g W.); Gotzreuth (Kosch.); Malo. 35. Lepratta (?) BERICENSIS, sp. nov. (PI. III. fig. 18.) From Bocca di Sciesa, Colle Berici, there are specimens with Cupularia-growth, the disks measuring 10—15 millim. in diameter. The zocecia are ovoid to hexagonal, separated by a distinct ridge, with a row of large pores round the border; and in a few zoccia there is a small avicularium at one or both sides of the aperture. The oral aperture is rounded on the distal edge, and the proximal border curves inwards, reminding us of the aperture of Leprala castanea, Busk. 26. Lepraia (?) LONTENSIS, sp. nov. (PI. III. fig. 5.) Zoarium very thin and delicate, with the zocecia at the side of the colony opposite. Zocecia long, bordered by a row of pores, with the peristome projecting in a tubular form, within the wall of which there is on one side a small tube, which may be avicularian, and reminds us of the avicularian tube of a group of Cellepore represented by C. granum, FH. I am unable to identify this with any of Reuss’s species, though, as it is abundant, it must have come into his hands. At first, I thought it was his semitubulosa. It also looks very much like the. _Eschara syringopora of Reuss in Foss. Polyp. Wien. Tert. pl. viii. fig. 3; and it seems somewhat doubtful whether he then had the E. syringopora of his later paper before him. Lschara jfistulosa, Kss., in the same paper may be this species. Loc. Val di Lonte; Montecchio Maggiore; Brendola; Ronzo ; Ferrara di Monte Baldo ; Crosaro; Malo. 37. Suir cocornna (Abild.). (PI. III. fig. 8.) Several incrusting specimens from Brendola and Montecchio Maggiore show great range in the size of the avicularia; some having a small one at each side, while others have them directed forwards as in fig. 8. Wie-"y distributed, both living and fossil. 38. Suirria coccrnEa (Abild.), var. attrERA (Reuss). (Pl. III. fig. 7.) Eschara alifera, Reuss, Bryoz. von Crosaro, p. 274, pl. xxxili fic: 1h. Mucronella alifera, Pergens, Bryoz. von Wola Lu’zanska, p. 71. The branches from Brendola are pretty uniform in breadth, but those from Val di Lonte are irregular in size. The pores round the borders of the zocecia are much more distinct than is usually the case in MW. coccinea, although they occur in living forms. In most zocecia there is an avicularium at each side, but in others there is only one, and there is considerable difference in size and direction. Ovicell usually very much immersed, sometimes more exposed and recumbent. In the size of the zocecia the living and Eocene forms agree, and pp MR. A. W. WATERS ON the incrusting Mucronella loricata of Koschinsky and MW. prestans, H., are closely allied. It is considered a variety of S. coccinea on account of its erect growth, and may be said to be S. coccinea in the Eschara-form; but it is Mucronella for those who retain that genus. Loc. Val di Lonte; Brendola; Montecchio Maggiore; several localities in Hungary (Pergens); Wola Lwzanska (Pergens); Ferrara di Monte Baldo; Ronzo; Malo. 39. Smirrra Lanpsporovil (Johnst.), var. Cuzrmorora, Rss. (Pl. III. fig. 12.) | Cellepora cheilopora, Reuss, Polyp. Wien. Tert. p. 91, pl. xi. fig. 4. Lepralia cheilopora, Reuss, "Foss, Bryoz. Oest.-Ung. Mioe. P- 168, pl, iv. fig. 1 Peristome considerably raised, and within the aperture a projecting avicularium. ‘There are pores around the border of the zocecium, and on the round raised ovicells there are also a few large pores. This seems closely allied to the recent form from New South Wales which I described (Ann. Mag. Nat. Hist. ser. 6, vol. iv. p. 16, pl. ili. figs. 14 and 15) as S. malleolus, and both are about the same size, but the recent form is punctured over the surface. Loc. Satschan (Moravia); Brendola. AQ. Sumrrrra PorRIGENS (Reuss). (Pl. IL. fig. 9.) Lepraha porrigens, Reuss, Foram. Anth. und Bryoz. des Septa- rienthones, p. 175, pl. vii. fig. 15. There are incrusting specimens from Montecchio Maggiore and Brendola, which are probably the S. porrigens of Reuss, but have perhaps received various other names. It has a wide lyrula plate in the aperture and a round suboral avicularium ; pores round the border of the zocecia very indistinct ; ovicell globular, recumbent or partly immersed, and punctured. Probably this is ae aha Seguenzai, Reuss (Bryoz. von Crosaro, p. 254, pl. xxxvi. fig. 1 Loc. Sollingen ; me eis, Maggiore ; Brendola. 41, Surrrza Exarata (Reuss). (Pl. IIL. fig. 6.) Cellaria exarata, Reuss, Foss. Polyp. Wien. Tert. p. 61, pl. vii. fig. 32. Vincularia exarata, Reuss, Bryoz. von Crosaro, p. 276, pl. xxxiv. fig. 1. | The zocecia are slightly rounded, and there is a row of pores close to the border. The oral aperture is at some distance from the surface, and I have been able to make out a wide denticle within it ; but the proximal edge of the peristomatal aperture also bends inwards, and might be called a mucro. The ovicells are short and usually considerably immersed. This may be allied to Bracebridgia geometrica (Kss.). NORTH-ITALIAN BRYOZOA. 23 Loc. Val di Lonte (Rss. & Waters); Montecchio Maggiore ; Brendola; Ferrara di Monte Baldo. 42. Poretta mpricata (Reuss). (Pl. II. figs. 16, 17.) Eschara imbricata, Reuss. Foss. Polyp. Wien. Tert. p. 69, pl. viii. fig. 26. Zoarium incrusting; but, when completely covering the branch of some other species, it at first appears to be free. In my specimens it is difficult to see the structure of the younger zoccia; but the proximal edge is straight, with a small central avicularium within the aperture ; this, however, is not visible in the older zocecia. In the older zocecia the transverse shape of the aperture is very curious and often triangular, caused by an infolding of the proximal edge. The pores round the border are readily distinguished in some zocecia, but not in all. Ovicell recumbent, distinct, not much raised, punctured. Fig. 16 represents a rather abnormal zocecium, in which there are large pores at the border. Loc. Val di Lonte (Rss. § Waters); Brendola; Montecchio Maggiore. 43, PorELLA MARSUPIUM, MacG., var. portrEra, Hincks. (Pl. III. fig. 13.) Porella marsupium, MacG., var. porifera, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. xiii. p. 24, pl. iv. fig. 4; Waters, Quart. Journ. Geol. Soc. vol. xlii. p. 63. Fossils from Brendola have the central avicularium somewhat lower than in Hincks’s figure. The two “pores” at the side are the ends of tubes; and in a few instances these only are visible, and the central avicularium is sometimes wanting. ‘The ovicell is small and somewhat immersed. The preservation of this species is not good; but I feel little doubt that, even if not quite identical with the var. porifera, it is closely allied. Loc. Living: Victoria; Bass’s Straits; Queen Charlotte Island. Fossil: Waipukerau; Napier (New Zealand); Brendola; Ferrara di Monte Baldo. 44, RHAMPHOSTOMELLA BRENDOLENSIS, sp.nov. (Pl. III. figs. 10,11.) Zoarium incrusting. Zocecia distinct, oval to hexagonal; surface smooth, with large pores round the edge; peristome raised, with a peristomial notch either central or slightly to one side; on one side of the peristome a small triangular avicularium and in some parts of the zoarium also large elliptical avicularia on one side of the zocecium, directed laterally. Sometimes the zocecia are separated by a wall which is considerably raised vertically, and in some cases the subradiate lines on the surface are very distinct. Within the peristome there is an expanding lyrula; but the cardelle are in- distinct. - Ovicells wide, raised or partly immersed, widely open in front, apparently not perforated, 24 MR. A. W. WATERS ON Rhamphostomella is a genus established by Lorenz*, and of which Mr. Hincks7 has given a revised diagnosis; from this the present fossil differs in having pores round the border of the zocecia, and in apparently not having the ovicell perforate. It is, however, so clearly allied to forms placed under Rhamphostomella that the diagnosis of the genus must be altered to include it. It is not, however, yet quite certain that the genus is based upon satisfactory characters. Loc. Fossil: Brendola; Montecchio Maggiore (?); Val di Lonte. 45. Portna (?) cononata (Reuss). (PI. IV. figs. 1-5, 15.) Cellaria coronata, Reuss, Foss. Polyp. Wien. Tert. p. 62, pl. viii. fig. 3. Eschara conferta, Reuss, op. cit. p. 71, pl. vill. fig. 32. Acropora coronata, Reuss, Bryoz. von Crosaro, p. 277, pl. xxxiv. figs. 3-5. Porina coronata, Koschinsky, Bryoz. alt. Tert. siidl. Bayerns, p. 42, pl. iv. figs. 7-9. This is a most variable species, and the study of better-cleaned specimens and the preparation of further sections have led me to alter my views very materially. I have found it extremely difficult to satisfy myself as to the suboral pore, since the tube from the suboral avicularium and the suboral pore end close together, near to the oral aperture; but I have at last obtained sections showing that, contrary to what I thought from earlier preparations, the suboral pore enters the zocecial cavity just below the oral aperture. This is very important, as it shows that, if we are to consider the position of the suboral pore as of first moment, this species must’ be removed from Porina; and as the position of the pore is nearly the same in Tubucellaria cereoides, we may have to remove it to Tubucellaria. There is considerable variation in the shape of the zoarium, it being sometimes cylindrical, at others compressed; and there is also great variation in the pores or avicularia round the aperture, which are sometimes scarcely distinguishable from those covering the front of the zocecia; in other cases they are very distinct; and there is usually a triangular avicularium directed distally just below the oral aperture and above the suboral pore. In some zoaria there are a few large ra’sed avicularia with broadly spatulate openings, and these seem to be merely suboral avicularia modified. Sections of the interior show the tubular connections from zocecia to zocecia, to which I referred in my ‘ Challenger’ Supplementary Report, p. 32. The branches dichotomize; and in a few cases there are at the distal end of the branch openings for the chitinous tubes, showing that there has been articulation. In the specimen fig. 4 a fracture has taken place where the stem becomes thinner, presumably after death. In the slender forms, such as fig. 5, there * “ Bryozoen von Jan Mayen, Internationale Polarforschung,” Akad. Wis- sensch. vol. iii. p. 93. + Ann. Mag. Nat. Hist. ser. 6, vol. iii. p. 424, NORTH-ITALIAN BRYOZOA. DAS) is sometimes a calcareous basal attachment; but in the stouter forms I have seen nothing to indicate how it is attached. In some specimens a calcareous disk forms a closure over the aperture; and from this caleareous growth a tubule rises up in the middle (figs. 2 & 15). A closure with a tubule occurs in several Cyclostomaia, as in Diastopora, &e.; but I had always looked upon a closure of this kind as confined to the Cyclostomata, whereas we now find it in P. coronaia and Lepralia syringopora. There are a number of slender specimens (fig. 5) in which the proximal part of the peristome is often raised; and this appears to me like the Cellaria labrosa of Reuss (Foss. Polyp. Wien. Tert. pl. vii. fig. 38); but Reuss seems subsequently to have considered that this figure represented Porina duplicaia. Loe. Val di Lonte; Montecchio Maggiore (Rss. g Waters) ; Brendola ; Crosaro; Ferrara di Monte Baldo; Ronzo; Malo; Pria- bona ; Gotzreuth (Kosch.); Wola Lwzanska (Perg.); Eocene of Hungary (Perg.). 46. Portna (?) pupricata (Reuss). (Pl. ILI. fig. 14.) Cellaria duplicata, Reuss, Foss. Polyp. Wien. Tert. p. 62, pl. vii. fig. 34, Eschara duplicata, Reuss, Bryoz. von Crosaro, p. 273, pl. xxxiii. figs. 8-10. Eschara heterostoma, Reuss, op. cit. p. 274, pl. xxvi. fig. 5. Most of the specimens in my collection have the zoarium in the E. duplicata shape ; but, according to our present ideas, there is no reason for separating this from LZ. heterostoma. At the side of the peristomatal aperture there is a large triangular avicularium, forming a part of the peristome. The ovicell is recumbent, not much raised, or subimmersed. We are not yet acquainted, with the oral aperture of this species; but Kirkpatrick has shown that Gigantopora lyncoides has a Schizoporellidan aperture. The large peristomial pore occurs in Gigantopora lyn- coides, Ridley*, Hippothoa fenestrata, Smitty, and Porina (?) colum- nata, Watersit, and cannot be looked upon as a good generic character. Loc. Val di Lonte and Montecchio Maggiore (Rss. & Waters) ; Brendola; Ferrara di Monte Baldo (W.); Ronzo (W.); Malo; Pap-Patak, Hungary (Pergens), as HE. heterostoma ; Kolos-Monostor (Pergens). 47. Portna (?) papritosa (Reuss). (Pl. ITI. fig. 19.) Eschara papillosa, Reuss, Polyp. Wien. Tert. p. 68, pl. vill. fig. 22; id. Foss. Foram. &c. von Oberburg, p. 31, pl. x. figs. 7-8 id. Bryoz. von Crosaro, p. 268, pl. xxxi. figs. 11- 17. * Proc. Zool. Soc. 1881, p. 47, pl. vi. fig. 3; Ann. Mag. Nat. Hist. ser. 6, vol. i. p. 77, pl. vii. fig. 5. + Floridan Bryozoa, p. 47, pl. vi. fig. 142. t Quart. Journ Geol. Soc. vol. xxxvii. p. 334, pl. xviii. fig. 88. at reels me 26 MR. A. W. WATERS ON Porina papillosa, Koschinsky, Bryoz. alt. Tert. siidl. Bayerns, p. 37. The zoarium is narrow at the base, and here there are numerous openings, showing that there was an attachment by means of chitinous tubes*. The zocecia, as seen at the side, are alternate. In the best cleaned specimens I have but very rarely been able to find any suboral pore, though certainly occasionally there is a pore larger than the others, and in imperfectly cleaned specimens the lower part of the avicularium might be mistaken for a pore. Neither have sections revealed any such pore. We may, however, remember that in Tubucellaria cereoides the pore can be distinctly seen on the surface in some zocecia, but not in all, though sections show that in all cases there is below the aperture a pore wider than the others and readily distinguished in structure. At the side of the zocecia, resting on the peristomial projection, there is a long acute avicularium. The peristome is often very much raised, forming a long tubular projection, which may entirely curve over (fig. 19, a), and then the opening is wide and slit-like. Neither Reuss nor Koschinsky has fully appreciated this, and they have merely described these cells as closed, which is sometimes the case. A similar prolongation oceurs in various J'ubuceliaric, and has been described and figured by both Busk and myself (see Ann. Mag. Nat. Hist. ser. 5, vol. xx. p. 190, pl. v. fig. 10). Very often one row has all the zocecia prolonged, the next plain, and the following one again raised, and so on; so that the raised rows are easily seen with the naked eye. Sections show that just above the oral aperture the peristome expands internally on the proximal side, as if there had been a small chamber, thus reminding us, on a smaller scale, of tne ovicell of Turritigera stellata, B. Pergens gives this as a synonym with Hschara cervicornis (Plioc. Bryoz. von Rhodes, p. 25); but they are very different things. In the recent Tubucellaria cereoides there is only one of the glands to which I referred in my ‘Challenger’ Suppl. Report, pp. 2 & 27; and to these glands I hope shortly to refer elsewhere, having cut sections of a number of species with the object of finding how they occur. Both this and P. coronaia present us with so many difficulties with regard to characters, showing relationship with Tubucellaria, that the recent Tubuceliaria must be re-examined with this object. Loc. Brendola; Montecchio Maggiore; Val di Lonte; Ronzo; Crosaro; Malo; Gdtzreuth; Oberburg ; Neustift (Kosch.). 48. Porina (?) BrocuLata, sp. nov. (Pl. III. fig. 15.) A small specimen from Brendola is incrusting. Zocecia not much raised, with peristome prolonged into a kind of neck, on the front of which are two large pores. On one side, about halfway down * Koschinsky says, p. 37, that “Sie sind mit verbreiterter Basis festgee wachsen ;” but this I have never found to be the case. NORTH-ITALIAN BRYOZOA. Di the zocecium, there is an avicularium directed inwards; and there is a row of pores around the border of the zoccium. Ovicell recumbent. This is clearly related to Porina larvalis, MacGillivray, which also has two large pores on the front of the peristome, and has an avicularium about halfway down the zocecium. ‘This last occurs fossil in Mount Gambier and Bairnsdale (Australia). 49. ScuizoporettaA Horrnusi (Reuss). (PI. IV. fig. 8.) Eschara Hoernesi, Reuss, Bryoz. von Crosaro, p. 273, pl. xxxiil. - figs. 6, 7. _ Schizoporella Hoernesi, Koschinsky, Bryoz. alt. Tert. siidl. Bayerns, p. 47. Cellaria scrobiculata, Reuss, Foss. Polyp. Wien. Tert. p. 63, . pl. viii. fig. 4. Now that we know the shape of the oral aperture, it would seem that S. Hoernesi is the same as the Schizoporella submersa which I described from Curdies Creek (Quart. Journ. Geol. Soc. vol. xxxvii. p. 340, pl. xviii. fig. 85). Loc. Val di Lonte: Montecchio Maggiore; Brendola; Gétz- reuth (Kosch.); Curdies Creek, S.W. Victoria; Kolos-Monostor (Perg.). 50. ScHIZOPORELLA SQUAMOIDEA (Reuss). Lepralia squamoidea, Reuss, Fauna des deutschen Oberoligocins, p. 19, pl. xv. fig. 5; Fauna des Septarienthones, p. 172, pl. vii. fig. 3; Bryoz. von Crosaro, p. 254. The small zocecia (0°3 mm. long) have small pores over the sur- face, and this sometimes shows indications of slight furrowing ; the small aperture has a wide sinus; ovicells wide, globose. The Lepralia rugulosa, Rss., differs from this in not having pores on the surface, and S. hyalina shows the same difference and a larger ovicell, but no doubt they are closely allied. Loc. Biinde and Sollingen (#ss.); Val di Lonte; Montecchio Maggiore; Brendola; Crosaro (/ss.). 51. ScHIZOPORELLA UNICORNIS (Johnst.). Cellepora tetragona, Reuss, Polyp. Wien. Tert. p. 78, pl. ix. fig. 19. : Schizoporella unicorns, Hincks, Brit. Mar. Polyz. p. 238 (which see for synonyms). Loc. Living: widely distributed. Fossil: Montecchio Maggiore ; Austrian and Hungarian Miocene ; Crag; Pliocene of Italy. 52. ScHIZOPORELLA SERRULATA (Reuss). Cellepora serrulata, Reuss, Polyp. Wien. Tert. p. 85, pl. x. fig. 12. A specimen from Montecchio Maggicre more nearly corresponds with Reuss’s original figure than with those in Bryoz. Oest.-Ung. Mioe. p. 167, pl. 11. figs. 2, 3, pl. iv. fig. 4. In a young zocecium the 28 MR. A. W. WATERS ON oral aperture is seen to have a wide round sinus. In older zocecia there is a plate inside the proximal part of the aperture sloping inwards, as shown in Reuss’s recent figure, pl. iv. fig. 4, called S. crassilabris on the plate. The ovicell is small and hood-shaped, as shown in the first figure, and also in fig. 4, pl. iv. Just above the peristome, or ovicell, there is an opening on one cr both sides, which may have been vibracular. Loc. Kisenstadt (fss.); Montecchio Maggiore. 53. ScHizoporELLA Ompont (Gottardi). Lepralia Omboni, Gottardi, Brioz. foss. di Montecchio Maggiore, Atti Soc. Veneto-Trentina di Sc. Nat. vol. ix. p. 305, pl. xiv. fig. 1. Zoarium incrusting ; zocecia distinct, separated by a raised line ; on the surface there are a few large pores mostly near the border, and there is a triangular avicularium about halfway down the zocecium, sometimes one on each side. ‘The ovicell is not prominent, and looks like a cap to the aperture. The sinus is wide. This is closely allied to S. Hoernesi, Rss., and it is doubtful © whether it should be separated on account of the mode of growth and the distribution of the pores‘over the surface. It differs from Eschara fissimargo, Kss., in having a row of pores. Probably Lepralia monopora, Rss. (Bryoz. von Crosaro, p. 45), is described from a worn specimen of the present species. Loc, Montecchio Maggiore (Gottardi); Brendola; Val di Lonte ; Malo. | 54+, ScHIZOPORELLA PHYMATOPORA (Reuss). Eschara phymatopora, Reuss, Bryoz. von Crosaro, p. 272, pl. xxxiil. fig. 1. From Australia I have described similar cylindrical forms as S. phymatopora. They differ in usually having the avicularium placed much lower, but the position is not constant (see Quart. Journ. Geol. Soc. vol. xxxvu. p. 338, pl. xv. figs. 31,423 vol. xxxviii. p. 510; vol. xli. p. 300). Loc. Val di Lonte (fss.); Brendola; Lower Eocene of Mons (Mumer & Pergens). 55. Scw1zoPoRELLA ScHREIBERSI (Reuss). Celiaria Schreibersi, Rss. Foss. Polyp. Wien. Tert. p. 63, pl. viii. fig. 8; Bryoz. von Crosaro, p. 262, pl. xxiv. figs. 5, 6. The aperture has'a wide round sinus ; the presence of avicularia is not quite constant, but there is usually one on each side, some- times raised, but usually flat with the surface. Closely allied to S. australis, Woods (see Quart. Journ. Geol. Soc. vol. xxvii. p. 341, pl. xiv. fig. 15). The zoarium is only about 0:4 millim. wide. Loc. Val di Lonte and Montecchio Maggiore (#ss.); Brendola ; Pap-Falva and Kolos-Monostor, Hungary (Perq.). NORTH-ITALIAN BRYOZOA. 29 56. ScHizoporeLLa TERNATA (Reuss). (PI. IV. figs. 11, 12.) Lepralia ternata, Rss. Bryoz. Oest.-Ung. Mioc. p. 167, pl. ili. fig. 11, pl. vii. fig. 5 There is a projection at the side of the aperture, frequently supporting an avicularium ; and in one specimen there is also an avicularium below the aperture, rather to one side. The ovicell is small and globular. None of the specimens are well preserved ; and probably many names have been given to this species. Loc. Nussdorf; Eisenstadt; Brendola; Val di Lonte; Montecchio Maggiore. 57. Fepora Excensa (Koschinsky). (Pl. IV. fig. 6.) Kiontdella excelsa, Koschinsky, “ Bryczoenfauna der alt. Tert. sidl. Bayerns;” Paleontographica, vol. xxxii. P. 68; pl. . vil. figs. 5-12. I have one specimen from the Bocca di Sciesa, Colle Berici, and another from Brentonico* , Mt. Baldo, which differ in a few parti- culars from Koschinsky’s description; but the differences do not seem sufficient to separate them specifically. As far as can be seen, the zoarium is throughout a solid cylinder and not tubular; there is a kind of hood above the aperture, and usually a large lanceolate - avicularium at each side. The ovicell is large, wide, raised, and perforated, apparently somewhat like the ovicell of Schizoporella tuberosa; but the state of preservation is not sufficient to permit or a full description. This is no doubt the Fedora of Jullien +; but Dr. Jullien de- scribed and figured it upside down, as the growth is from the tip of the zoarium (the right-hand side of his fig. 39). I give a figure (fig. 7) of the operculum of Fedora Edwardsi, Jullien, showing that the zocecial characters are Lepralian. Specimens from Spiassi (halfway between Caprino and Ferrara di Monte Baldo) have a small conical zoarium (3-4 millim. long); and from the shape I thought it was Batopora conica, Hantk. Loc. Gotzreuth, Bavaria (Kosch.); Brentonico; Bocca di Sciesa ; Colle Berici; Spiassi; Malo. 58. RerEeroRA TUBERCULATA, Reuss. Retepora tuberculata, Reuss, Bryoz. von Crosaro, p. 267, pl. xxxi. figs. 9, 10. There is a pit below the aperture, which may be avicularian; but in the Brendola specimens these pits do not occur with the regularity figured by Reuss, and there are also other small avicularia scattered about in various positions, and occasionally an enlarged avicu- larium. Loc. Val di Lonte (#ss.) ; Brendola. * Brentonico is on the dives part of the Monte Baldo range, a few miles south of Mori. i “Dragages du teas ” Bull. Soc. Zool. de France, vol. vii. p. 17. 30 MR. A. W. WATERS ON o9. Rerepora ELEGANS, Reuss. (PI. IV. figs. 9, 10.) Retepora elegans, Rss. Foss. Polyp. Wien. Tert. p. 48, pl. vi. fig, 38. There are only unbranched fragments from Brendola. The sub- oral pore is at the end of a groove. The ovicells are recumbent ; and there is sometimes a ligulate avicularium on the front of the zocecium. The zocecia are wide at the distal end, but narrow at the proximal; the lateral zocecia are obliquely arranged ; .and this gives the dorsal surface a very characteristic appearance, with the zocecia turned alternately to the right and the left. Loc. Reuss’s description is from a specimen from the Val di Lonte ; Brendola. 60. CrrtEpors PROTEIFoRMIS, Reuss. (PI. IV. figs. 13, 14.) Eschara diplostoma, Reuss (non Phil.), Foss. Polyp. Wien. Tert. p- 71, pl. viii. fig. 34. Celleporaria proteiformis, Reuss, Bryoz. von Crosaro, p. 264, pl. xxx. figs. 2, 6-8. Cellepora diplostoma, Pergens, Bryoz. von Wola Lu’zanska, p. 72. This is one of the holostomatous* Cellepore, with a large trian- gular avicularium at one side below the aperture, and with a large globular ovicell, perforated in the same way as the surface of the zoarium. At present only two Cellepore are known living from the northern hemisphere having the lower edge of the oral aperture straight, namely CO. sardonica, Waters, and C. pertusa, Smitt; but the group is better represented in the southern hemisphere; and it is interesting to find this species very abundant in the Lower Tertiaries. In one specimen I have found a large spatulate vicarious avicularium. In the specimen, fig. 14, on one side all but three zocecia are flat, with large pores on the surface; but the three are raised in a somewhat ovoid shape, and are at once seen to resemble the zocecia of C. proteiformis. ‘Turning the fragment over, we find that on the other side about one-half of the zocecia are raised above the surface, like big ovicells; and there is now no difficulty in recognizing this as a Stage of C. proteformis. It was only, however, at the eleventh hour that the flat, compressed form, and the cylindrical one, consist- ing of several layers, were seen to be stages of the same species. The description and figure of H. diplostoma are so insufficient that * MacGillivray proposes to leave these to form the genus Cédlepora, while he calls those with a schizostomatous aperture Schismopora. As the latter were the earliest known, it might have been better to give a new name to the holo- stomatous group, but this is not a matter of much importance. In 1881 I pointed out that there were a number of Cel/epore with the oral aperture straight below, and that, perhaps, they should form a subgenus. For a long time workers have recognized that Cel/epora, as understood, could not stand ; and, therefore, when Jullien reproaches us for retaining the genus, it shows an imperfect acquaintance with our work; and I still maintain that science has been better served by showing relationship and collecting facts than by pre- mature classification, NORTH-ITALIAN BRYOZOA, 31 Reuss, in my opinion, was justified in giving it another name; moreover, “ diplostoma” had already been used by Philippi. Pergens is mistaken in supposing that it has been found in the Vienna Tertiaries, though, as mentioned, the Val di Lonte material was for a time supposed to come from the Vienna Basin. Loc. Val di Lonte and Montecchio Maggiore (fss.); Brendola; Ferrara di Monte Baldo; Ronzo; Wola Lu’zanska. 61. CELEEPORA oLIcostiemA, Reuss. Lepralia oligostigma, Reuss, Bryoz. von Crosaro, p. 257, pl. xxxvi. fig. 10. ‘There are two incrusting specimens from Montecchio Maggiore; and the ovate zocecia are in one case arranged in a radiate manner. The peristome is raised, and an avicularium at the side forms part of it, just as in Porwma duplicata; but there is no suboral pore, the avicu- larian chamber is larger, and the shape is more distinctly seen from the outside, thus causing greater irregularity of appearance. The ovicells are globular, recumbent, perforated. The oral aperture is, no doubt, nearly round; but it has not been possible to see the exact shape: probably oligostigma belongs to the Schismopora group. Loc. Crosaro (fss.); Montecchio Maggiore. 62. CELLEPORA PERTUSA, Smitt. Orbitulipora lenticularis, Reuss, Bryoz. yon Crosaro, p. 289, pl. xxx. figs. 12-14; Pergens, Foss. Bryoz. von Wola Lw’zanska, p. 72. In a specimen from Val di Lonte, not well preserved, the ovicells are broken down, but the oral apertures are nearly straight below and about 0-15 millim. wide. I have mentioned a subglobular C. pertusa from Aldinga (Quart. Journ. Geol. Soe. vol. xli. p. 305). Marsson (Bryoz. Schreibkreide der Insel Riigen, p. 101) says that O. lenticularis, Reuss, is the Cellepora accumulata of Hagenow; though, from the description and figures, it is impossible to be sure of this. A cylindrical specimen from Montecchio Maggiore also has the pertusa aperture and a suboral avicularium ; but neither is the pre- servation of this satisfactory. 63. SricHoPoRINA SIMPLEX, Koschinsky. (Pl. IV. figs. 16-18.) Stichoporina simplex, Kosch. Bryoz. alt. Tert. siidl. Bayerns, p. 64, pl. vi. figs. 4—7. From Brendola this occurs in a disk form; in Cupularia-form ; and in flat pieces which must have been from a larger growth. The zocecia are raised and rounded, with the oral aperture slightly coarctate, nearly central, somewhat depressed; on the right side of many zocecia there is a large triangular avicularium with a bar across. Out of a number of specimens, I have only found one zoc- cium with an avicularium on the left side. Some pieces have an avicularium to almost every cell, others only to one or two zoccia. ne Se Serre te Tear 1 thew pe MR. A. W. WATERS ON In one small specimen (fig. 18) there is above many zocecia a thick curved bar, somewhat thicker at the base; but, as the cells are a good deal crushed, I am not sure to what extent the appearance may be due to this. As there is matrix between the zocecia, the drawing is in certain parts a restoration. This small specimen shows a relationship to Stichoporina crassilabris, Koschinsky ; but in other specimens the relationship is not so apparent. The dorsal surface shows the base of each zocecium as a much raised rounded area, with a few large pores. There are two specimens (fig. 17) with rather smaller zocecia, but with a similarly shaped aperture, but slightly smaller; and, as a rule, there are no avicularia, though zocecia with ovicells have a small avicularium on one or both sides. The ovicell is not very much raised, partly immersed, and is merely an enlargement of the distal end of the zocecium. ‘This might perhaps be called var. minor. Zocecia seen laterally, as on the free border of the zocecium, show a contraction about a quarter of the height from the base ; and there are here two. rosette plates or pores to each zocecium. This is allied to S. protecta and S. crassilabris from Gdtzreuth ; and the zocecia resemble those of Aiontdella obliquiseriata, Kosch., op. cit. pl. vii. fig. 13 6; and it would seem that the two genera should be united. At first, when I had only examined specimens which were not thoroughly cleaned, I took this for Cupularia biden- tata, Reuss, as the avicularia are very prominent. I have not found O. bidentata. in either locality, and feel in doubt about it *. Loc. Gotzreuth (Kosch.); Brendola; Ronzo; Pap-Falvi-Patak. 64, Batopora MuLtTrRADIATA, Reuss. Batopora multiradiata, Reuss, Bryoz. von Crosaro, p. 265, pl. xxxi. figs. 1-4. The zocecia are barrel-shaped, with a semicircular aperture, straight below (about 0-1 millim. wide), and a recumbent ovicell, which, how- ever, is directed towards the apex of the zoarium. In the shape of the zocecia, the aperture, and the ovicell, this is very similar to the Orbitulipora of the Chalk and Lower Oligocene ; and the two genera are clearly closely allied even if separation is necessary. | In Orbitulipora petiolus (Lonsd.) the ovicell is also directed to- wards the centre of the zoarium. It is difficult to understand how Batopora grew, for it does not seem to start from the large round cell at the apex, as there is a layer of zocecia below that. It appears * [A specimen sent me by Dr. Pergens trom Pap-Falvi-Patak as Cupularia bidentata, Rss., is S. simplex, K. Since this paper was read Mr. R. Kirkpatrick, of the Natural-History Museum, has submitted to me a specimen, from Murray Island (15-20 fath.), of recent Stichoporina, which I should call S. simplex. The aperture is rather wider and rounder than in the fossils, and it may have to be separated as a variety on this account, though the shape of the zoecia, the position of the avicularium, and the structure of the dorsal surface are the same in both. Mr. Kirkpatrick informs me that he has also had it from the Cape of Good Hope and Malacca ; and we may look for a description from his pen very shortly.—A. W. W., December 24th, 1890.] NORTH-1ITALIAN BRYOZOA. 33 more probable that it grows from a central basal cell, gradually form- ing a nearly globular zoarium; and then from the apex of this a second layer is formed, and perhaps a third. Some specimens are thus partially capped with a growing layer. (See woodcut. ) Section of Batopora multiradiata, Reuss. ap., apical cell; ov., ov., ovicells. Loc. Val di Lonte and Priabona (fss.); Brendola (Waters) ; Ferrara di Monte Baldo (W.); Montecchio Maggiore (Gottards & W.); Ronzo (W.); Malo. ocene of Bavaria (Pergens).. Various Hungarian localities (Pergens). 65. Batopora? SroriczKat, Reuss. Batopora Stohezkai, Reuss, Bryoz. deutsch. Unteroligoc. p. 228, pl. u. figs. 2-4. The globular zoaria from Brendola vary from 1 millim. to 3 millim. in diameter, and at first I thought the small ones might be young zoaria of Batopora multiradiata ; but the size of the larger specimens shows that this cannot be the case. The aperture is not round, but flattened on the lower side, so that it does not differ much from that: of B. multiradiata, Reuss. Loc. Unteroligociin of Calbe (Rss.); Montecchio Maggiore (Got- tardi) ; Brendola. 66. LuntLitEes auapRata, Reuss. Cellepora quadrata, Reuss, Foss. Polyp. Wien. Tert. p. 95, pl. xi. fig. 17. . Lunulites quadrata, Reuss (“tetragona” on the plate), Bryoz. von Crosaro, p. 278, pl. xxviii. fig. 18. In a large specimen from between Grotte and Sarego, Colle Berici, the proximal edge of the aperture is straight, and the avicularia are larger than figured. There are no doubt many synonyms for this (see Quart. Journ. Geol. Soc. vol. xxix. p. 442). Gd. Gis... No. 185; D , ’ 34 MR. A. W. WATERS ON NORTH-ITALIAN BRYOZOA. EXPLANATION OF PLATES I.-IV.* The figures are magnified 25 times, except those otherwise marked. Figs. Fig. Fig. Figs. Fig. Figs. Fig. Puate I. 1-6. Catenicellaseptentrionalis, sp.nov. a, aperture, magn. 85 times. (Fig. 5, dorsal surface.) 8. —— , var. (Fig. 7, dorsal.) 9, 10. continua, sp. noy. (ig. 10, dorsal.) . Catenaria tenerrima, Reuss. (a, aperture, magn. about 50 times.) . Scrupocellaria gracilis, Reuss. (Fig. 13, dorsal.) . —- brendolensis, sp. nov. (Fig. 15, dorsal.) : elliptica, Reuss. (Fig. 17, dorsal.) . Bactridium Hagenowi, Reuss. (Fig. 18, dorsal.) 20. Onychocella angulosa, Reuss; form excavata, Reuss. Figs. 21, 22. Scrupocellaria montecchiensis, sp. nov. (Fig. 22, dorsal.) . Vibracella trapezoidea, Reuss. Puate IT, . Membranipora Rosselii, Aud., var. deplanata, Reuss. appendiculata, Reuss. Dumerillii, Aud. . Micropora articulata, sp. nov. polysticha, Reuss. parallela, Reuss. coriacea, Esper. . Cribrilina chelys, Koschinsky. The ovicell and vicarious avicu- larium are added from different parts of the colony. . Monoporella sparsipora, Reuss. From Montecchio Maggiore. . Lepralia subchartacea, @ Arch. nodulifera, Reuss. From Brendola, impressa, Reuss. bisulca, Reuss. From Brendola. Puate III. . Lepralia bisulca, Reuss. From Brendola. syringopora, Reuss. lontinensis, sp. nov. . Smittia exarata, Reuss. coccinea, Abild., var. alifera, Reuss. From Brendola. coccinea, Abild. - porrigens, Reuss. . Rhamphostomella brendolensis, sp. nov. (Fig. 10a, magn. 50 times.) 12. Smittia chilopora, Reuss. 13. Porella marsupium, var. porifera, Hincks. 14. Porina (?) duplicata, Reuss. 15. bioculata, sp. nov. . Porella imbricata, Reuss. 18. Lepralia (?) bericensis, sp. nov. 19. Porina papillosa, Reuss. 4, tubular curved peristome. PuateE LV. 1-3. Porina coronata, Reuss. . Termination, showing openings for articular tubes. . Small form, like /abrosa, Reuss. Edwardsi, Jullien. Operculum. . Schizoporella Hoernesi, Reuss. 4 5. 6. Fedora excelsa, Koschinsky. (Fig. 6a, magn. twice.) a 8 9,10. Retepora elegans, Reuss. (Fig. 9, dorsal.) 11,12. Schizoporella ternata, Reuss. 13, 14. Cellepora proteiformis, Reuss. 15. Diagrammatic section of Porina coronata, Reuss, 16-18. Stichoporina simplex, Koschinsky. — * These have been drawn at the Author’s expense. Va] a YW. Lis Quart. dourn. Geol. Soc. imp. Mintern Bros. NORTH-ITALIAN BRYOZOA. - Imp. Mintern Bros Quart.Journ. Geol. Soc. Vol. XLVI. Pl. Il. NORTH -ITALIAN BRYOZOA. A.W. Waters del. M.P Parker hth. wiv IIT. Vol. XLVII. Pl. spac. art.dourn. Geol Qui tem Spa — AW.Waters del. Mintern Bros . imp. M.P. Parker hth. TALIAN BRYOZOA. = 16, A.W. Waters del. M.P. Parker lith. NORTH-ITALIAN BRYOZOA. Mintern — Dros, imp. ay f a my tf on PicrTn ‘ ie ON THE PORPHYRITIC ROCKS OF THE ISLAND OF JERSEY. 30 2. On the Porpuyritic Rocks of the Istanp of Jersey. By Prof. A. pr Lapparsnt, Foreign Correspondent of the Society. (Read November 12, 1890.) (Communicated by the President.) Havine been supplied some years ago, through the kindness of my friend the Rev. C. Noury, of St. Helier, with a good series of speci- mens of porphyritic rocks from the Island of Jersey, I was struck by the exceptional appearance of some felsitic and globular varieties, in which the spheroidal concretions, instead of being minute globules, as in the usual pyromerides, attained a size of two feet and more in diameter. According to the information which I then possessed, the porphyritic rocks occupied the whole of the north-eastern part of the Island, resting on Cambrian schists, and underlying, from Rozel to St. Catherine’s Bay, a coarse conglomerate, which had been described by Ansted as New Red Sandstone. At that time I strongly advocated the views entertained by my distinguished friend M. Michel-Lévy, on the relation between structure and geological age in eruptive rocks; and, judging from many examples collected in various districts of France, I believed that every felstone and pyromeride would be found to be of Permian age. Accordingly, as this belief seemed to be warranted by the facts in Jersey, I described the whole of the porphyritic rocks of the Island as Permian *. Moreover, I even ventured to lecture my English predecessors for having failed to recognize the true age of the eruptive series, the felsites of Jersey having been described by Mr. Davies as “old rhyolites.” This was a mistake on my part, which I now feel myself bound to confess before the Geological Society of Londen. The first doubts as to the correctness of my theoretical views arose in my mind immediately after the Meeting of the Geological Congress in London in 1888. I had taken part in the excursion to North Wales, under the guidance of Dr. Hicks, and had there observed the intercalation of true felsites in the ‘“‘ Precambrian slates,” while numerous boulders of the same felsites were to be found among the constituents of the conglomerate: which underlies the purple slates of Llanberis. Accordingly I could no longer hesitate to believe that eruptive rocks with true felsitic structure might belong to the earliest geological periods. Some months later, my friends Mr. Hill and M. Bigot entered, each for himself, upon the geological study of the northern group of the Channel Islands. They both were_led to the conclusion that the grits and conglomerates of these Islands, which are not to be separated from the conglomerate of the north-east corner of Jersey, and which also contain pieces of felsite and globular porphyry, ought * Bulletin Soc. Géol. France, 3° sér. vol. xii. pp. 284-289. D 36 ON THE PORPHYRITIC ROCKS OF THE ISLAND OF JERSEY. to be considered as synchronous with the purple conglomerates of the Cotentin—that is, with the very base of the Silurian formations. On this view I could easily see that the porphyritic series of Jersey ‘might represent something equivalent to the oldest felsites of North Wales. But having never visited the ground, and being desirous of correcting my error by actual inspection of the rocks in situ, I paid a visit to Jersey last summer. At the first’ glance I could perceive that the porphyritic series, instead of being discordant with the schists, is intimately connected with that formation and partakes of its general strike and dip. At the base of the eruptive mass, tuffs and breccias prevail, which under the microscope prove to be mainly tuffs of porphyrites. Such are the rocks at Havre-Giffard, the so-called metamorphic schists, with large crystals of felspar, at the Imperial Hotel, and also the rocks which are worked at Stephen’s Mill for road-material. Then comes, at Anne Port, a blood-coloured porphyritic mass, which looks quite like a red quartziferous felsite, but which, microscopic- ally examined, proves to be a porphyrite with a large admixture of iron-bearing quartz, becoming in some places a true red jasper. This mass underlies the reddish felstones, with well-marked columnar structure, of La Créte and Archirondel, while the pyromerides of Boulay Bay form the top of the series, being, at the Téte des Hougnes (as clearly stated by the Rev. C. Noury, in his ‘Géologie de Jersey ’), immediately succeeded by the first purple- coloured layer of the Rozel conglomerate. I do not intend to enter here into a more detailed study, which I am preparing for the Geological Society of France. But I thought it my duty to do justice to the geologists who had long ago recognized the true character of those “old rhyolites,” which, from mistaken theoretical views, I had been led to regard as Permian eruptive rocks, although, in fact, when carefully examined, they exhibit very little in common with the true Permian porphyries of France or of Saxony. Discussion. Mr. E. Hitt considered the Author’s former opinion was the natural consequence of Ansted’s views. Messrs. Davies and Bigot’s work had entirely overthrown these. He agreed with the Author as to the relation between the argillites and the porphyritic rocks, and awaited with interest the further information promised. ‘36 5 ai MR. R. LYDEKKER ON A NEW SPECIES OF TRIONYX. OT 3. On a New Spucres of Trionyx from the Miocens of Matra and a CHEtonran Scaruta from the Lonpon Cray. By R. Lypexxnr, Ksq., B.A., F.G.S. (Read November 12, 1890.) I. Trionyx From Matra. In a paper read before the Society in November, 1885, I described part of a Crocodilian skull from the Miocene of the Maltese Islands, which was referred to the existing Oriental genus Tomistoma * ; attention being at the time particularly directed to the interest of the occurrence, in those deposits, of a genus now confined to one island in the purely tropical Malay Subregion of the Oriental Region. On the present occasion I bring to the notice of the Society evi- dence of Oriental affinities in a member of the Chelonian family Trionychide, of which the remains have been recently obtained from the Miocene of Malta. The specimen forming the subject of this part of the paper is one of a small collection brought from Malta by Dr. John Murray, and presented by him to the British (Natural History) Museum. It con- sists of a portion of the middle and right half of the anterior region of the carapace of a large Chelonian referable to the family Trrony- chide. The specimen, of which a reduced and restored representa- tion is given in the accompanying figure, is embedded in the characteristic buff limestone of Malta, with the sculptured surface exposed. ‘The nuchal bone (nz) is missing, but the greater part of the first four costals (c-c') of the right side are preserved ; and there also remain portions of five neural bones, and the inner extremities of the first, second, and third costals of the left side. The form of the neural (n’) situated between the third costals, with its shorter lateral surfaces placed posteriorly, is alone sufficient to show that the specimen belongs to the anterior half of the carapace. The forward inclination of the fourth costal is, however, apparently due to the flattening which the specimen has undergone. The carapace indicates a species nearly or quite as large as the existing Clitra wdica, the length of the third neural being 28 inches. It also agrees with that species in the coarseness of the sculpture, but this feature is also met with in some species of Zrionyx. The comparative shortness of the ribs and costal plates suggests that the specimen is not fully adult. Before proceeding further, it should be observed that the three Indian species of T’rionya (viz. T. gangeticus, T. Lathi, and 7. hurum) differ from all other members of the family in having two neural bones between the first pair of costals f, this being apparently due to a subdivision of the normal first costal. All the fossil species hitherto described, of which the entire carapace is known, agree * See Quart. Journ. Geol. Soe. vol. xlii. p. 20 (1886). Tt See Boulenger,‘ Catalogue of Chelonians, &c. in Brit. Mus.’ p. 244 (1889). 38 MR. R, LYDEKKER ON A NEW SPECIES OF TRIONYX. with the normal type in having but a single long neural between the first pair of costals ; and no species, so far as I am aware, has been named from the Maltese Miocene. An inspection of the figure of the Maltese specimen will, however, at once show that it agrees with the 7. gangeticus group in having two neurals (n' and n’™) Fig. 1.—Upper surface of the anterior part of the carapace of Trionyx melitensis ; from the Miocene of Malta. (One fourth of the natural size.) between the first costals, and that it is therefore specifically distinct from all fossil species based on specimens sufficiently perfect to exhibit the characteristic features of this part of the carapace. From 7’. gangeticus and its allies it is distinguished by the greater elongation of the second moiety of the divided neural (7*), in con- sequence of which the proper second neural (x*) becomes much shorter, and also by the coarser sculpture. In its coarse sculpture it agrees with 7’. planus, of the Hordwell beds, in which the anterior part of the carapace is unknown, but is of larger dimensions. It has already been mentioned that in its large size and coarse sculpture the fossil approximates to Chitra indica ; and the question naturally arises whether extinct species of Chitra may not, like the Indian species of Zrionyx, have the first neural divided. In the absence of the skull it is almost or quite impossible to say whether MR. R. LYDEKKER ON A CHELONIAN SCAPULA. 39 the fossil species should be referred to Chitra or Trionyx ; but in either case its Indian affinities would be certain, since Chitra, like the species of Zrionya with a divided first neural, is now confined to India. Since we know of the existence of a divided first neuralin Trionyx, and have no evidence of such a condition in Chitra, I propose to refer the species represented by the specimen under consideration to the former genus, with the designation 7’. melitensis. Il. Scapuna oF EosPHARGIS FROM THE LonDON CLAY. A short time ago Mr. W. H. Shrubsole, F.G.S., submitted to my notice three fragments of a large reptilian bone obtained from the London Clay of Sheppey, which I soon recognized as portions of the left scapula of a gigantic turtle. The fragments, which are repre- sented in their approximately relative positions in fig. 2, indicate a larger scapula than has hitherto come under my notice; and the only known turtle, from these deposits, to which they can be referred is the so-called Chelone gigas of Owen, a species which I have made the type of the genus Hosphargis, and classed among Dermochelyide*. The fragments comprise the glenoidal portion of the bone, the distal Fig. 2.— Ventral aspect of the left scapula of Eosphargis gigas, From the London Clay. (One sixth of the natural size.) Se,, scapula; P.Cor., precoracoid ; Cor., coracoidal facet; g/., glenoid facet. extremity of the true scapular bar, and a considerable portion of the precoracoidal bar. Among the specimens of EHosphargis gigas pre- served in the British Museum is a slab of rock, numbered 440897, which contains the imperfect bones of the pectoral girdle. In this specimen there is the glenoidal extremity of a scapula (or, as it may * Quart. Journ. Geol. Soe. vol. xlv, p. 241 (1889). t See ‘ Cat. Foss. Rept, & Amphib. Brit. Mus.’ pt. iii. p. 226 (1888-89). 40 MR. R. LYDEKKER ON A CHELONIAN SCAPULA. be more precisely termed, scapulo-precoracoid), partly concealed by — the matrix. The exact transverse diameter of the “neck” of that scapula cannot be precisely determined, but it was evidently very considerably less than in the present specimen, where it is upwards of 6°7 inches. The especial interest of the present specimen is that it is more like the scapula of Ohelone than that of Dermochelys, this being especially shown by the sharpness of its ridges, notably the one at the junction of the inferior border of the precoracoid with the neck, and the long and triangular form of the facet for the articulation of the coracoid. Indeed, if we had no evidence of the existence of gigantic Dermochelyide in the London Clay, I should have been disposed to refer the specimen to a member of the Chelonide, by the side of which the Chelone Hoffmanni of Maastricht would be a dwarf. Although too much stress should not be laid on the resemblance of this bone to the scapula of the Chelonida, yet it to a certain extent supports the very strong evidence lately brought over by Dr. Baur * as to the intimate affinity between the Dermochelyide and Chelonide; and I may take this opportunity of stating my belief that the writer here mentioned has practically proved his view that the so-called Athecata are nothing more than a specialized offshoot from the earlier Chelonide. In provisionally following the opinion of writers who adopted an opposite view, I always felt it to be a great difficulty how the resemblance between the limb-bones of the Dermochelyide and the Chelonide was to be explained if they had no direct affinity with one another. I am glad to say, in conclusion, that Mr. Shrubsole has signified his intention of presenting the specimen under consideration to the National Collection. * * American Naturalist,’ 1890, pp. 530 ez seg. ON CERTAIN ORNITHOSAURIAN AND DINOSAURIAN REMAINS. 4] 4. On certain ORNITHOSAURIAN and DInNosAURIAN REMAINS. By R. Lypexxer, Esq., B.A., F.G.S8. (Read December 10, 1890.) [Puats V.] I may state, by way of introduction, that I am indebted to my friend’ Professor 0. C. Marsh for the correct determination of the interesting reptilian bones forming the subject of the present com- munication. I. Ornithosaurian Quadrates. When engaged in compiling Part I. of the ‘ British Museum Catalogue of Fossil Reptilia and Amphibia,’ I was considerably puzzled with three imperfect bones from the Kimeridge Clay of Weymouth. Eventually I considered that they represented a peculiarly modified ulnar metacarpal of an Ornithosaurian; and they were accordingly entered at page 41 of the volume cited (Nos. 43034, 44183, and 41179) as the distal extremities of that bone. It was mentioned at page 40 of the same volume that these bones differed from normal specimens of the ulnar metacarpal in having a flat bony plate attached to one of their lateral surfaces, which I considered might have aided in the support of the pa- tagium. The resemblance of the free trochlear extremity of these bones to that of the distal extremity of the ulnar metacarpal of an Ornitho- saurian is, Indeed, very striking; but, on seeing them, Professor Marsh at once said that they were Ornithosaurian quadrates. On comparison with the quadrate of the skeleton of Rhamphorhynchus Cuviert preserved in the Museum*, and also with that of Sca- eee ees Purdoniy (acquired by the Museum since the first part of the ‘Catalogue’ was written), no possible doubt remained as to the correctness of this identification. In Plate V. figs. 3a, 36, 4a, 46, views are given of two of these Ornithosaurian quadrates, one (No. 44183) belonging to a small, and the other (No. 41179) to a large form. From a comparison of these specimens with the quadrates of Sphenodon, Rhamphorhynchus grandis, and Scaphognathus Purdoni, it is quite evident that they belong to the right side of the skull. The distal extremity of each forms a deeply-greoved oblique trochlea, above which there is a nearly quadrangular shaft. To the inner side of this shaft is attached by suture a flattened plate of bone, concave internally and convex externally, which, from the analogy of Sphenodon, must evi- dently represent part of the pterygoid. In the larger specimen part of the anterior free border of the pterygoid is preserved in the upper part of the bone. To the outer surface of the distal trochlea of * Catalogue, &c., p. 33. no. 387002. + Phil. Trans. for 1888, pp. 503-5387, pls. Ixxvii, Lxxviii. A? MR. R. LYDEKKER ON CERTAIN these specimens it is evident that the quadrato-jugal must have been attached. By comparison with the skull of Scaphognathus Purdon, it is evident that the smaller quadrate indicates an Ornithosaurian of the same approximate size as that species, while the other figured quadrate indicates a much larger form. ‘These specimens show conclusively that the flat internal plate described in Mr. Newton’s figures of the skull of Scaphognathus Purdoni as part of the quadrate, is really a portion of the pterygoid, and, conse- quently, that the relation of the quadrate to the pterygoid in the Ornithosauria is the same as in the Rhynchocephdlia, and quite different from that obtaining in the recent Crocodilia, where those two bones are widely separated. So far as I can gather from the figures published by Professor Marsh, many Dinosaurs had the pterygoid ankylosed to the quadrate after the Rhynchocephalian and Ornithosaurian plan. age With regard to the species to which the specimens under con- sideration may have belonged, I find, by comparison with the above- mentioned skeleton of Rhamphorhynchus grandis, from the Litho- graphic Limestone, that the smaller quadrate would agree approxi- mately in relative size with the so-called Pterodactylus Mansel, Owen, founded upon the distal portion of the humerus*. The larger quadrate would agree more nearly with the so-called Ptero- dactylus supra-jurensis of Sauvaget, founded upon a coracoid from the Kimeridgian of Boulogne; and, in the absence of any evidence to the contrary, I propose to refer it provisionally to that species. In the ‘Catalogue’ cited I stated that there was no evidence for referring Pterodactylus Manseli (and the allied or identical P. Pleydelli) to Pterodactylus ; and from the large size of the specimens under consideration, and their marked resemblance to the quadrates of Lhamphorhynchus grandis and Scaphognathus Purdom, I consider it probable that Pterodactylus Manselt, P. Pley- delli, and P. supra-jwrensis are all three referable either to Rhampho- rhynchus or Scaphognathus; and, since the former genus appears to be of commoner occurrence in the Lower Kimeridgian Lithographic Limestones than the latter, I am inclined to provisionally refer all three species to Rhamphorhynchus. Il. Vertebra and Tibia of a Celuroid Dinosaur. In the ‘Geological Magazine’t I described two Dinosaurian vertebre from the -Wealden of the Isle of Wight under the name of Calamospondylus Foxt, being at that time unaware that the generic name had been previously suggested §, although without a defini- tion, for the form subsequently described as Aristosuchus pusillus. Although the description of the original Calamospondylus is vague, it is undoubtedly a preoccupation of the name, and I accordingly * See ‘Catalogue,’ &c., p. 40. + Bull. Soc. Géol. France, sér. 3, vol. i. p. 875 (1878). t Decade 3, vol. vi. p. 121 (1889). § Fox, Geol. Mag. decade 1, vol. iii. p. 883 (1866), ORNITHOSAURIAN AND DINOSAURIAN REMAINS. 43 propose that Calamospondylus Foxi should henceforth be known as Calamosaurus oat. These two associated vertebre, which belong to the cervical region, indicate a Dinosaur referable to the Theropodous family Celuride, all the members of which are characterized by the extremely pneumatic structure of the skeleton, as is especially shown by the cervical vertebre. The cervicals of Oalamosaurus differ from those of Oelurus (represented in the Wealden by C. Daviesi) in being shorter, and also in that all were probably opisthoccelous, the middle and posterior cervicals of the type genus being amphiccelous*. Since the woodcut in which one of the vertebrae of Calamosawrus was figured in the original description is on a reduced scale, and is not altogether satisfactory, two views of this specimen are given of the full size in Plate V. figs. 1a,16. It may be mentioned that these vertebre are of a characteristic brown colour, evidently indicating that they were obtained from a bed of the Wealden different from that which yielded the types of Calurus Daviesi and Aristosuchus pusillus, all of which are stained of a deep black colour. Hitherto the above-mentioned vertebree are all that have been known concerning Calamosaurus. During his recent visit to the Natural-History Museum Professor Marsh called my attention to the right tibia (B. M. No. R. 186) of a small Dinosaur obtained by the late Mr. Fox from the Wealden of the Isle of Wight, which has been incorrectly referred to Hypsilophodont. This specimen was at once identified by my friend as referable to some member of the family Celuride ; and on careful examination it is at once seen to have nothing whatever to do with the Iguanodontide. This bone, of which two full-sized figures are given in Plate V. figs. 2 a, 2b, is of the same brownish hue characteristic of the vertebre of Calamosaurus, and was therefore probably derived from the same bed. Its total length is 6-1 inches. Its Theropodous characters are shown by the highly polished and dense nature of the outer surface, by the long and sharp fibular ridge (f) on the outer borderz, by the extreme flattening of the distal half, and, above all, by the large facet at the distal extremity of the anterior surface (fig. 2 a, as.), for the articulation of the ascending process of the astragalus, the inner border of that facet terminating in avery sharp ridge. There is also the absence of the division of the distal extremity into two ‘« steps.” Where the specimen has been fractured the very large size of the central cavity may be seen. It appears to me probable that this bone is of rather teo small a size to have belonged to the same individual as that to which the i. In Celophysis, of the Trias of New Mexico, all the cervicals were amphi- coelous. t See ‘Cat. Foss. Rept. & Amphib. Brit. Mus.’ part i. p. 194. The whole of the four tibiz mentioned on that page are referred to the wrong side; those marked right being left, and vice versd. { This is absent in the Jguanodontide; see Huxley, Quart. Journ. Geol. Soe. vol. xxvi. p. 19. +5 ON CERTAIN ORNITHOSAURIAN AND DINOSAURIAN REMAINS, type vertebra of Calamosaurus Foxi pertained; but, since it was obtained apparently from the same bed as the latter, there is a probability that it may have belonged to another individual of that species, to which I propose to refer it provisionally. Since Celurus Daviesi is still larger than the type Calamosaurus Fowi, it is unlikely that the present specimen belonged to that species. In his description of the femur of Megalosaurus, Professor Huxley * commented upon its extremely bird-like form. These avian features are, however, still more intensified in the present specimen, which may be compared with the slightly smaller tibio- tarsus of Apteryx Oweni. ‘The whole bone is relatively longer and more slender than the tibia of Megalosaurust, its fibular ridge is much longer and more prominent, and the distal extremity is still more flattened. ven more remarkable is the great upward extension of the facet for the ascending splint of the astragalus, which reaches fully an inch up theshaft. The sharp ridge forming the inner border of the distal part of the suriace for the astragalus corresponds to the ridge bordering the outer side of the extensor groove of the tibia of Apteryx, and thus still further intensifies the avian characters of the bone. The present specimen is, indeed, the most bird-like Dinosaurian tibia that has come under my observation; and, in conclusion, attention may be directed to the very curious feature—that, whereas the Ornithopoda make the nearest approach to the avian plan of organization in the structure of the pelvis, it is among the Thero- poda, so far as regards European types, that we find the most avian characters in the structure of the hind limb. EXPLANATION OF PLATE V. Figs. 1a, 1b. Left lateral and anterior aspects of a cervical vertebra of Calamo- saurus Foxit; from the Wealden of the Isle of Wight. s, neural spine; prz, prezygapophysis ; ptz, postzygapophysis; 7, cervical rib ; Jj, pneumatic foramen. Figs. 2a, 26, 2c, Anterior, distal, and posterior aspects of a right tibia referred to Calamosaurus Foxi; from the Wealden of the Isle of Wight. a, inner condyle ; 6, outer condyle (broken); d, cnemial crest ; /, fibular ridge; as, facet for astragalus. Figs. 3 4,3. Posterior and distal aspects of the right quadrate of Rhamphorhyn- chus Manseli; trom the Kimeridgian of Dorsetshire. g, quadrate; pt, pterygoid. figs. 4a, 4b. Posterior and distal aspects of a right quadrate referred to Rhamphorhynchus supra~jurensis; from the Kimeridgian of Dorset- shire. All the figures are of the natural size. * Quart. Journ. Geol. Soc. vol. xxvi. p. 192 ez seg. t Huxley, op. cit. p. 20, fig. 1. 1. Soc. Vo n. Geo IS Quart.Jou ohalinesoremcnceny, Mintern Bros DINOSAURIAN & ORNITHOSAURIAN BONES. ON THE VARIOLITIC DIABASE OF THE FICATELGEBIRGE. 45 5. The Vartoritic Diasase of the FicurenrceBrrer. By J. Water Grecory, F.G.S., F.Z.8., of the British Museum (Nat. Hist.). (Read December 17, 1890.) ConTENTS. I. Introduction. II. General Features of the Surface. III. The Variolitic Diabase. IV. The “ Pseudocrystallites.” V. The Relations of the Diabase. VI. The Origin of the Variolitic Structure. VII. Summary of Conclusions. I. Lyrropvucrion. Amone the many interesting rocks described in 1874 by Prof. von Giimbel in his ‘ Die paliolithischen Eruptivgesteine des Fichtel- gebirges’* was a variety of diabase crowded with small round, apparently felspathic bodies. As these stood out on weathered sur- faces like peas or pearls, he suggested for it the name of “ Perl- diabas.” Seeing that this peculiar modification occurred around the margins of the diabase masses, close to the contact with the sedimentary rocks into which they were intruded, Prof. von Giimbel regarded these pearl-like spheroids or “ Knollchen” as fragments of the neighbouring beds caught up by the igneous rock, rounded by the rolling and the fusion of the sharp angles, and baked till they acquired the porcellanitic appearance that they uow present. In the following year, Prof. von Zirkel, in his well-known paper “ Die Structur der Variolite” +, described the microscopic structure of this rock, and, recognizing that the rounded spheroids are merely ordinary spherules, he identified it as variolite, and compared it with the typical varieties ; he objected alike to von Giimbel’s name and theory. . Prof. von Zirkel’s conclusions were confirmed by Prof. Rosenbusch in his detailed microscopic examination of the rock +, and he, in accordance with the view taken by the French geologists of the origin of the Savoy variolites, regarded these also as ordinary ‘‘endomorphen Contactformen der Diabase” on the selvage of intrusive dykes. Prof. von Giimbel’s faith in his own theory stood firm against these criticisms and this new evidence; and in 1879, in the third volume of his great work on Bavaria §, he re-enunciated his hypo- * P. 31; and N. Jahrb. 1876. pp. 42, 43. t Berichte k. sachs. Gesell. Wissensch. vol. xxvii. (1875), pp. 209-220. ¢ ‘Mikroskopische Physiographie der massigen Gesteine’ (1877), pp. 358- 366, and ed. 2, vol. ii. (1887), pp. 227-234. § ‘ Geognostische Beschreibung des Koénigreichs Bayern, Abth, iii. Geogn. Beschr. des Fichtelgebirges,’ pp. 213-218. Gotha, 1879. 46 MR. J. W. GREGORY ON THE thesis, and supported it by a chemical and microscopic examination of the rocks ; the criticism and works of both Prof. von Zirkel and Prof. Rosenbusch were, however, wholly ignored. The two opposing theories had one point in common, viz. that they regarded variolite as the contact-product of abasic dyke with the rocks into which it is intrusive. In a paper ‘“ On the Variolitic Rocks of Mont Genévre,” published in this Journal *, it was shown that this, the generally accepted view, did not hold for the rock in its typical locality. The admirable description of M. F. Loewinson-Lessing 7 has shown that the same is the case with the less altered variolite of Yalguba. Prof. Dalmer’s<{ descriptions of the mode of occurrence of the Schénfels variolite in Saxony also suggest doubts as to the applicability of the older theory to that rock, since the diabase here seems to be variolitic throughout, as at Galgenberg, or at least to the extent of fifty yards from the contact-plane; this might, of course, be explained as due to subsequent intrusions of diabase into already consolidated masses, but there is nothing in Prof. Dalmer’s description to show that this has occurred. Such very various materials have been at different times included among the variolites that it seemed quite possible that the Fichtel- gebirge varieties might have had a somewhat different origin. Hence it became advisable to examine some of the South-German localities of the rock, and compare its mode of occurrence at these ~ with that of the Cottians and Liguria. For this purpose the district around Berneck seemed most suitable, as from it had been derived the varieties of which the microscopic structure had been so care- fully described by Prof. Rosenbusch, and for which Prof. Giimbel’s theory had been proposed. Il. GeneraL FEATURES OF THE SURFACE. The little town of Berneck is picturesquely situated in the mouth of the valley of the Velschnitz, at the north-east margin of the Fichtelgebirge. It extends from the confluence of the Oelschnitz with the Weisser Main, for about a kilometre up the deep and narrow valley which the former stream has cut through the great diabase massif to which this locality owes most of its geological interest. The diabase appears in the map of the Bavarian Ober- bergamt (Blatt Mimchberg, No. xi.) as covering an irregular triangle, about 4 kilometres long, with a base of some 24 kilometres. The area occupied by the diabase forms a high rugged plateau, divided into a series of *‘ Leite ” §, on the sides of which the best sections are to be seen; these are, however, never very satisfactory, as the * G. A.J. Cole and J. W. Gregory, Quart. Journ. Geol. Soe. vol. xlvi. (1890), p. 295-382. + T'sch. Min. u. Petr. Mitth. vol. vi. (1884), pp. 281-300, pl. iv.; and a Olonetzkaya Diabazovaya Formatziya,” Trudui St. -Peterburgskagho Obshch. Pstest. xviii. (1888), pp. 165-169. { Erlauterungen zur geol. Specialkarte des K. Sachsens. Section Planitz- Ebersbrunn ; Bl. 124, pp. 25. Leipzig, 1885. § This is a provincial word, meaning “‘ place,” whether plateau, slope, crest, or even valley. VARIOLITIC DIABASE OF THE FICHTELGEBIRGE, 47 junctions of the diabase and the Devonians are, as a rule, hidden either by dense fir-forest or talus. The diabase is bounded on the north-east and south by narrow bands of Devonian and Lower Palseozoic rocks, including some chloritic schists and phyllites; these run from 8. W. to N.E., extend- ing up the valley from Berneck to Metzlersreuth, Zell, and Sparnack. The strike of the rocks is generally parallel to that of the valley, but they have been greatly contorted, and sometimes inverted, by the earth-movements which squeeze them in between the Miinch- berg gneiss plateau on the north, and the granitic and gneissose area of the Fichtelgebirge on the south. Prof. von Giimbel’s map is here very diagrammatic, and in many cases the lines of the outcrop of the diabase are very different from those on the accompanying sketch-map, which is on a larger scale. The floor of the valley along which runs the path from Berneck to Fig. 1.—Sketch-map showing the Distribution of the Variolitic Diabase, near Berneck, Fichtelgebirge, Bavaria. die) Sa BADSLEITE: Kenran oa Mes ont cue ht BARNREUTH A>. oe eee SQ tess ini won ccen ges ay oem Se Saroets ALLUVIUM. === CHLORITIC. SCHISTS. fea] PALZOZO/C iS tli PABASE Wy aNerss. [228 x | anveuine DIALBASE . | / | AULT. Eoene eon FOOTPATHS. Heinersreuth, the lower parts of the wooded slopes of the south bank of the Oelschnitz opposite Stein, much of the meadow-land south of Meyerhof and ofthe fields that run up into the Miuhlleite are occupied by the shales, grits, and slates of the Devonian, instead of diabase, as is shown on yon Giimbel’s map. 48 MR. J. W. GREGORY ON THE III. Tue Varrorrric DraBase. At the north-east end of the town of Berneck, just above the uppermost bridge over the Oelschnitz, is a boss of spheroidal variolitic diabase ; it is situated in the angle between the path that runs beside the river to the “ Colonnade ” and that which traverses the valley between the Badleite and the Miihlleite. Beside the former path there is an excellent section some 65 yards long. The main mass of the diabase is a close-grained decomposed aphanite, jointed into spheroids, which vary in diameter from 80 millim. to more than a metre. The diabase of the spheroids is compact both in the centre and on the periphery, but at alittle distance from this, and running parallel to it, is a band of variolite; the varioles in the centre of this band are usually from 2 to 3 millim. in diameter, but on each side they gradually decrease in size and number, so that it passes into the compact normal diabase. Thus, to quote the figures in one case,— externally there is a layer of compact diabase 30 millim. thick; then a 50 millim. layer of variolite, and within a mass of the aphanite 120 millim. across; below there is the same 50 millim. of the variolitic, and 30 millim. of the compact diabase. Fig. 2.— Section of Variolitic Spheroidal Diabase. Footpath to the Colonnade, Berneck. About 10 feet high. Some of the smallest of the spheroids, however, are variolitic throughout. When examined under the microscope with a low power, the rock is seen to consist of a thick green groundmass, containing a number of round white varioles, sharply marked off from the groundmass. In both occur small circular vesicles filled VARIOLITIC DIABASE OF THE FICHTELGEBIRGE, 49 with chlorite, and a large number of irregular spaces of a slightly dichroic, transparent material. Under a higher power, though the rock is greatly decayed, the principal points in structure can be determined. The groundmass is resolved into three constituents: there is a greenish, isotropic matrix, the transparency of which is diminished by a dust of minute opaque grains; it also contains a crowd of small granules and crystals; in most cases these are certainly augite, and it is probable the whole series are of this mineral. Occasionally a few fine felspar needles occur in the groundmass, and sometimes these are grouped into radial clusters. The small irregular rifts and spaces come out very prominently, owing to the transparency of the mineral that fills them; this is sometimes slightly dichroic and polysynthetically twinned, and shows a strong tendency to crystallize in rhomboidal forms. These characters show that it is dolomite, and it is obviously second- ary in origin; most of the smaller lines, however, are only singly refracting, and in this material occur numerous small inclusions. These structures must be compared with the “ pseudocrystallites ” described and figured by M. Michel-Lévy, and further reference will be made to them in a later section of the paper. The varioles in this rock belong to the variety with radial struc- ture and sharply marked off from the matrix: they thus differ from those figured by von Giimbel* from the opposite side of the Oelschnitz. The outer layer of the varioles is greatly altered and decomposed, and, while the minerals themselves have been destroyed, the external margin has been corroded. Internally they consist of a large number of small, radiating, and often branching fibres of a grey mineral, which is certainly a plagioclase, and apparently oligoclase; intercalated between these are granules of augite, similar to those that are scattered through the groundmass. The varioles are of the compound radial type. Fig. 3 (p. 50) shows one with the plagioclase needles radiating from several centres, while around the margin are several smaller secondary varioles. The extensive alteration that the rock has undergone, indicated both under the microscope and in the rock by the calcareous veins that traverse the diabase, does not encourage one to expect much assistance from chemical analysis. Prof. von Giimbel gave analyses of the groundmass and variolesf ; that of the latter is the more interesting, but it really does not help much :— RTECS fe ah si chie waaccs & aiita wat os, OEE PUT re ae en ees ee 13°46 OPRIGG OL ATONE op mux dt acs 8 8°29 RGEOW traechei en ied Kook twa hs 4°63 fi Oa CT ar 1°58 ORME city SUE nia Awe 5 ok bia es 1:75 PR dn Bl toa vila Mined <.dn, catia 5°36 ; 99°40 * Geogn. Beschr. Fichtelgebirg. fig. 31. t Ibid. p. 217. Q.J.G.8. No. 185. ) E 50 MR. J. W. GREGORY ON THE The paucity of alkaline earths and alumina renders it difficult to see what combination of oligoclase and augite could give such an analysis, especially when we have to deduct a little lime and magnesia for the secondary products in the interspaces. The history of this analysis, however, does not dispose one to place much confidence in it, and, with such intensely altered rocks, the micro- scope is probably a safer guide than the chemical balance. Fig. 3.—Section of the variolitic Diabase, with a large variole ; from the pathway to the Colonnade, Berneck. Some Pseudocrystallates 4 occur in the lower part of the figure. (2A , The ie: Wile numerator represents the original magnification, and the deno- minator the reduction from the size of the field of the micro- scope. ) The microscopic examination, however, fully proves the identity of this rock with that of Mont Genévre, though, being older, it has undergone greater alteration ; the pea-shaped bodies or “‘ Knollchen ” are shown to be true varioles, composed of a mixture of felspar fibres and augite granules and needles similar to that which occurs in the typical variety from the Cottian Alps. ; Here, then, in this boss of diabase we have the variolite of the Fichtelgebirge ; but, instead of being a mere contact-product deve- loped at the junction of the diabase and the rocks into which it is intrusive, it occurs im the centre of the igneous rock, at some places at least fifty yards from any other rock. According to yon Giimbel’s map this knoll is in ‘the middle of the diabase area; but on the opposite slope of the Celschnitzthal there may be seen, sandwiched between coarse ribs of diabase, some sinooth slopes that indicate the presence of the Devonian *. Pebbles of the same formation may be picked up in the gully on the south side of the knoll up which the path rises from the foot-bridge ; and a few yards * These are shown in the woodcut by Prof. von Giimbel, op. cit. p. 525. VARIOLITIC DIABASE OF THE FICHTELGEBIRGE, 51 along the footpath to Heinersreuth the sedimentary rocks may be seen in situ. Hence the diabase was probably close in contact with a band of Devonian, which now crosses the stream obliquely at this point; so, though the variolitic spheroids are in one place no less than sixty yards from the Devonian horizontally, they may have been nearer to it vertically. The gully up which the path rises from the bridge, the river, and a pine-covered slope completely obscure the junction of this mass of spheroidal diabase with the adjacent beds ; hence one turns to the neighbouring Devonians to follow the margin of their outcrop in the hope of finding a section elsewhere. The pathway that goes east to Heinersreuth is at first bounded to the south by steep crags of diabase, while to the north there is a more gentle fir-covered slope in which occasional exposures of the same amygdaloidal diabase are to be seen. The south margin of the Devonian crosses the brook that drains the valley and winds up the slope of the Badleite, at first roughly skirting the wood ; after one or two sharp bends, it turns abruptly to the south, and the Devonians are soon succeeded by the Silurians and Cambrians which rise from beneath them and abut against the diabase. There are no clear sections of the junction, but in a few places the diabase close by it can be seen; in such cases it is somewhat variolitic, though in places the amygdules that also occur obscure the varioles. Returning to the north side of the valley, an outcrop of Devonian in the bed of the stream enables one to get better acquainted with one important member of this formation—a coarse grit composed of fragments of quartz, quartz-mosaic, micropegmatite, and worn cleavage fragments of plagioclase; by the increase in the amount of the fine matrix that frequently occurs between the grains, and the decrease in the size of the coarser constituents, the rock passes into a shale; it is comparatively unaltered, except for a tendency to cleavage. Along the north slope of the valley the junction of the diabase and Devonian is buried in fir-woods; occasionally there is a limited outcrop of an amygdaloidal diabase overlain by a thin-bedded shale. The sections are, however, rather unsatisfactory. A little less than a kilometre along the valley the Devonians project into the diabase and run up the slope to the summit of the Mihlleite; they occupy a slight depression, which at the south end is 100 yards wide. Patches of variolitic diabase occur along the junction, and weathered lumps of if are common upon the surface at the foot of the bank of diabase that bounds the Devonian. The variolite is best seen, in sité, in a tree-covered ridge that forms the eastern boundary of the field occupied by the Devonian shales; a larger section is exposed in a knoll at the end of this ridge, which overhangs the path to Heinersreuth. The section is as follows, in descending order :— - 1. Massive-jointed and bedded grit (dip 30° to N., 25° W., mag- netic), 15 feet. E2 “5p MR, J. W. GREGORY ON THE 2. A thin-bedded finely-jointed shale, with some calcareous nodules and quartz veins ; along its lower margin it is some- what baked and in places brecciated. 3. An amygdaloidal diabase, cutting irregularly across the edges of the shale, which are sometimes curved up along the junction. The exposed surface of the diabase is sometimes variolitic. Fig. 4.—Junction of Variolitic Diabase and Black Shales; in the Crag, north of the pathway to Heinersreuth. Beds 1 and 2 belong to the Middle Devonian, and apparently to the lower part of this division (the ‘“‘ Unterschalstein mit Kalkgeoden”’ of Giimbel). The Lower Devonian can be seen lower down the slope, and especially at a pit in the “ Nereitenschiefer ” in the meadow where the brook that drains the upper part of this valley takes its source. The interpretation of this section appears to be somewhat as follows :—A tongue from the diabase massif has projected for some distance into the Devonian rocks, running approximately E.8.E., and forming a ridge that extends as far as the Heinersreuth footpath. The face of the intrusive diabase has been here exposed by removal of the grits and shales during the denudation of the valley. The microscopic structure of this variolite must be compared with that of the spheroidal knoll at Berneck. The rock consists of a series of altered varioles in a transparent, light-green, slightly VARIOLITIC DIABASE OF THE FICHTELGEBIRGE. 55 dichroic groundmass. An examination with higher magnification shows in this chloropitic groundmass numerous scattered, opaque white granules and large numbers of minute needles of actinolite ; the former frequently occur in lines which curve round the varioles. The varioles themselves are so much altered that their original structure cannot be fully determined. The resemblance of the alteration-products to those of the variolite already described suggests a community of composition. The varioles are now an opaque dusty mass charged with lines of granules and needles of ilmenite, while in places there are transparent areas of a dolomitic mineral traversed by numerous needles of apatite. ‘The opaque dusty material is probably a kaolin resulting from the decomposition of the felspathic microliths or globulites, while the dolomite and ilmenite are the products of the alteration of the pyroxene. The true radial arrangement of the Berneck varioles is absent, and the only approach to it is due to the disposition of the secondary ilmenite needles, The whole structure of these spherulites shows that they are more primitive than the others, and due to a more rapid and imperfect development. Returning along the south side of the tongue of diabase we can trace the junction with the Devonians across the crag already described, and beside the tree-covered bank that separates two fields ot shale and grits, to the summit of the Miuhlleite. Just on the brow of the hill a brecciated limestone crops out across the field, and is succeeded by beds of shale with calcareous nodules (Middle Devonian). This series of sedimentary rocks can be traced away to the north-west down the steep slope to the Oelschnitz, across the stream, and up the opposite bank. It forms a slight depression, bounded on each side by a rib of diabase; that on the north-east side of the Devonians is the more conspicuous, but that on the south-west is spheroidal in one place, a few feet from the junction. A section through this has been cut in making a wood-cutters’ road, which ends abruptly on the Devonian band. ‘The diabase is less amygdaloidal, and is more regularly variolitic, than when in imme- diate contact with the clastic deposits. Along the line of junction the influence of contact-alteration can be seen in ‘the Devonians ; the shales are indurated, baked, and iron-stained, and the limestone brecciated and marmorized. The same contact-phenomena, both in the development of a variolitic structure and the alteration in the sedimentary deposits, can be seen round a similar tongue of diabase that projects from the Miihlleite a little further to the north-west, and runs for some distance towards the south-east into the Devonian. Except for an occasional band of Devonian that stretches across the ridge, the whole of the south bank of the Oelschnitz and the Metzlersreutherbach in this area is of the ordinary amygdaloidal diabase. Having descended from the Mihlleite to the Oelschnitz, one turns to the steep left bank of this river in the hopes of finding a better section than the few scrappy exposures previously noticed. 54 MR. J. W. GREGORY ON THE A dense fir-forest, however, clothes the whole slope and prevents any very satisfactory evidence being obtained. Nevertheless, here and there a knob of rock rises above the soil of pine-needles, the path beside the stream shows an occasional low section, and the bed of the stream affords a little additional information as to the general nature of the diabase and its varieties. ‘The principal types are: (1) a compact non-vesicular diabase; (2) a weathered amygdaloidal diabase; (8) the variolitic diabase, and this is usually associated with spheroidal varieties. The compact diabase varies considerably in coarseness: some is quite microcrystalline, some is full of delicate acicular crystals of felspar ; while in other places it is very coarse, as at about 30 yards on the Berneck side of the drinking-fountain. One of the most in- teresting varieties of this group of diabases may be seen crossing the Metzlersreutherbach at the south end of the meadow which stretches from the Oelschnitz up the course of its tributary. This rock stands out from the slopes on either side as a dyke-like wall that forces the stream into a small cascade. The rock is much altered ; zoisite has been extensively developed in the plagioclase, and the whole of the pyroxene has been converted into a dichroic, chloritic, greenish mass; the most marked feature in this—as in most of these compact diabases—is the great abundance of leucoxene, result- ing from the alteration of ilmenite. Though the specific gravity of this rock is high, and it probably originally contained an excess of iron in the form of primary ilmenite, no doubt most of this was secondary, and produced as a result of the decomposition of the augite. Specimens can be collected which are quite white from the great abundance of the leucoxene. The compact diabases are, however, rather exceptional, and the weathered amygdaloidal varieties occupy the largest part of the area. The commonest type of this is a rock with a bright green, strongly dichroic matrix, crowded with actinolite needles and grains of leucoxene; some of the latter still retain part of the original ilmenite. The vesicles are mainly filled with calcite, but some zeolites are present. The sections are so scrappy that the relations of the two classes of diabase cannot be accurately determined. It is possible that some of the compact diabase may be in the form of dykes intrusive into the amygdaloid, but in several cases a gradual passage between the two can certainly be traced. The relations of the normal and variolitic diabases to one another, and of both to the Devonians, is, however, clearly shown at several places along the Mihlleite. Two large and several small bands of the Devonians can be seen in the bed of the Oelschnitz and traced up the banks on each side; tongues of the diabase may be seen intrusive into these. One of the best cases is a short section cut by the lower path, in which a bed of lydianstone, a baked shale, is irregularly overlain by the amyzdaloidal diabase, and in this is a band of still more altered Devonian. Microscopic examination of this rock shows that it was once a shale with a few layers of grit, and that it VARIOLITIC DIABASE OF THE FICHTELGEBIRGE. 55 has been crushed and minutely faulted. This case is further of interest as there is no variolite along the contact with the Devonian. An equally clear section occurs in a small quarry in line with the waggon-ford that crosses the stream just above the second foot- bridge. The section is composed of a mass of spheroidal variolitic diabase, and across it runs a band of baked shale which thins towards, and forks at, the lower and north end. Here, again, the diabase is neither variolitic nor spheroidal at the actual contact with the shale, but acquires these structures at about one foot from it. Fig. 5.—Section of the normal compact Diabase from the Oelschnitz- ; 80 thal, with a vacuole filled with chlorite. (x. See p. 50.) ri 3 The junctions of this diabase and the neighbouring Devonians are obscured, but the latter occur in mass a few yards away. It is unnecessary to trace the junctions of the diabase and the Devonians further through the area. On the opposite side of the Metziersreutherbach, in the bank of the same stream near Heiners- reuth, in the wood south of Meyerhof, and in other places, the same features are repeated. The rélations of the diabase to the eye-gneiss S.W. of Berneck are, however, worth consideration ; the junction can be seen in the road that rises steeply up the west bank of the valley which leads to Micheldorf, a little south of the quarry. The two rocks are certainly faulted against one another, and hence we search in vain in the crushed and brecciated diabase for any trace of the variolite. The evidence adduced is sufficient to show, at least in the case of the area south of the Oelschnitz and the Metzlersreutherbach, that, though the variolite does occur along the line of the junction of the Devonian and the diabase, when it appears as a true contact-selvage, the varioles are not well developed, and usually lack the radial structure. Further, that it is only in the compact, or but slightly vesicular, spheroidal diabase, usually at some little distance from the actual plane of contact, that the sharply-marked radial 56 | MR. J. W. GREGORY ON THE varioles are typically developed. I did not myself have the good fortune to obtain from the area specimens of all the types of vario- litic structure described by Prof. Rosenbusch from Berneck; but it is probable that a more careful examination, especially to the north of the Oelschnitz, would show that they may be ali grouped into the two classes—of those that are true contact-selvage-products and those formed within the diabase. No doubt occasionally a more perfect variole may be found among the former, and a rudimentary one among the latter, but as a rule this division will probably be found to hold. IV. Tue “ PsrvpocrystaLLiteEs.” In the course of M. Michel-Lévy’s admirable description * of the microscopic structure of the Mont Genévre variolite, he described and discussed the nature of certain enclosures in the varioles which he named “ pseudocrystallites.” In the previously published de- scription of the fine plate ft in the ‘ Minéralogie Micrographique,’ MM. Fouqué and Michel-Lévy attributed these bodies to “ fissures of retreat;” but in the latter author’s more detailed subsequent examination he was led by the peculiar optical properties of these structures to regard them as felspathic pseudocrystallites. He observed tiat when their long axes are parallel to the neighbouring felspar fibres they extinguish at once throughout their entire length, but when they traverse the radial fibres they disappear under crossed nicols, as the felspars seem continuous across them; hence M. Michel-Lévy was forced to regard such pseudocrystallites as composed of a complex aggregate of felspar crystals, differently orientated, so that their long axes were in the same straight line as those of the radiating felspar fibres of the variole. As it did not» seem easy to see how such an arrangement should have taken place if the felspar in these bodies were a secondary product, M. Michel- Lévy was led to this modification of the earlier theory. | In the paper by Mr. Cole and myself, we were led, by noticing how the larger rectangular fissures passed off into the minute branching irregular cracks, to regard them as due to rifts formed during contraction, but we could offer no explanation of the abnor- mal optical properties or arrangement of the felspathic constituents, We concluded { that they were ‘little fissures due to fracture or contraction,” and that “ until a similar structure is found in other rocks, so that ample comparison may be made, the last word cannot be said on these interesting ‘ pseudocrystallites.’ ” At Berneck there is a great development of these structures throughout the whole rock ; they frequently occur in great numbers in bands which pass through variolite and groundmass alike, while * “Mémoire sur la variolite de la Durance,” Bull. Soc. géol. France, 3™¢ sér. vol. v. (1877), ue 238. tT Pl. xxiv. fig. 2. t Quart. Journ. Geol. Soc. vol. xlvi. (1890), pp. 313 & 314. VARIOLITIC DIABASE OF THE FICHTELGEBIRGE, 57 the individual pseudocrystallites can often be traced from the one into the other. There seems no reason to doubt the identity of these bodies with those at Mont Genévre; the difference between them is only in the infilling material. At Berneck the rock is permeated with calcareous infiltration-products which line the joints and form veins that cross the spheroids; hence the pseudo- erystallites are mainly filled with calcareous or dolumitic products : but at Mont Genévre the veins and cracks are mainly occupied by felspathic material, and hence these rifts often contain plagio- clase fibres. At Berneck the secondary material in the spaces is quite different from that which forms the rock on either side; they are therefore sharply marked off from it in any condition of illu- mination. At Mont Genévre, on the other hand, not only is the material that has filled up the rifts similar to that of the surrounding rock, but the felspar has been deposited in optical continuity with the crystalline fibres that were broken and separated by the forma- tion of the pseudocrystallite. Such cases of the restoration of the optical continuity of crystals across cracks by the deposition of secondary minerals in the latter appear to have been fully estab- lished *. If this suggestion be accepted, it would supply a very striking and complex case of this phenomenon, while it would also afford a full explanation of the abnormal optical properties that M. Michel-Lévy has described in these structures. It appears to have been these optical characters almost entirely that led M. Michel-Lévy to his theory of the origin of the ‘‘ pseudo- crystallites”; apart from these, and though in many cases it must be admitted that the regular reticulation of these structures does present the aspect of a crystalline meshwork, the evidence in favour of the origin of these bodies as cracks and fractures subsequent to consolidation seems fairly conclusive. This view is supported by their irregular distribution, both in the varioles and the groundmass, well shown in specimens from Berneck, by the gradual transition that can frequently be traced from them into undoubted cracks, and by the fact that they sometimes pass from a variole into the groundmass. ‘The freshness of the material is an additional argu- ment, and apparently this alone was sufficient to induce Prof. von Giimbel to consider them as secondary. As first figured by MM. Fouqué and Michel-Lévy, there was a certain regularity in the disposition of these cracks, as they occurred along a zone arranged concentrically with the variole; in this case the variole is sharply marked off from the groundmass, and no doubt the shrinkage that caused this distinct separation found out a line of weakness due to one of the concentric zones of glassy matter that often occur in spherulites. Iddings+ has figured a spherulite in which a number of trichites are disposed in a similar circle to that in this variole. Where, however, the spherulite passes gradually * See, for example, the cases figured in this Journal by Miss Raisin, vol. xlv. p- 253, and by Cole and Gregory, vol. xlvi. pl. xiii. fig. 1. + J. P. Iddings “Obsidian Cliff;” 7th Ann. Rep. U.S. Geol. Sury., Wash- ington, 1888, p. 276, pl. xv. fig. 4. 58 ; MR. J. W. GREGORY ON THE into the glass, as in many of the Fichtelgebirge variolites, the con- traction due to the segregation of its constituents has not tended to the production of concentric cracks, but of some which are either irregular or parallel to any structural planes, such as lines of flow, that there may have been in the rock: these primary fissures would of course be connected by cross cracks ; thus would arise such a reticulate series as is shown by Prof. von Giimbel *. V. Tue RELATIONS oF THE DIABASE. In Prof. von Giimbel’s description clear evidence is given of the intrusive nature of the diabase into the Devonian. The long rib- like dykes below the Schloss, the baked shales and marmorized brecciated limestones of the Muhlleite, the irregular nature of the junction, with the long tongues of the diabase running out into the Devonian, the inclusions of the baked sedimentary rocks, and the absence of tuffs, leave no doubt as to the truth of this view. In connection with the last point, however, it is necessary to examine the grit-like rock found at the crag by the path to Heinersreuth, and a similar rock with large shale fragments on the south bank of the Oelschnitz opposite Stein, which are no doubt the schal- steins mentioned by von Gimbel. The latter rock is well exposed in a couple of ribs of rock running down the steep slopes of the bank of the river; it has a coarse gritty matrix, charged with large angular fragments of shale. Microscopic examination entirely dispels the idea that it is a true tuff; it has, however, been made up almost entirely by the decomposition of igneous and volcanic rocks, and thus chemically, and at first sight microscopically, resembles a tuff. The matrix of the rock is composed of frag- ments of a fine-grained diabase, rolled cleavage-fragments of plagioclase, and fragments of quartz and quartz-mosaic, &¢.; in addition there is a good deal of quartz with a remarkably well- developed micropegmatitic structure. The same constituents, but without the coarse shale fragments, occur in the grit of the crag. In both cases the fragments are mostly somewhat rounded and water-worn, and the rocks are no doubt true grits, the materials of which have been derived from the denudation of an area containing such rocks as those which form the highlands of the Fichtelgebirge to the south, or of the Miinchberg gneiss massif to the north; in the latter Prof. von Giimbel has described a micropegmatite very similar to that which is so striking a feature in these grits. As to the date of the intrusion but little can be said. The diabase can be seen intrusive into the Lower Devonian on the flanks of the Badleite, into the Middle Devonian on the Mihlleite, and into the Upper Devonian in several places along the right bank of the * Geogn. Beschr. Fichtelgebirg. fig. 31. + Brauns has recently described a variolite forming a crust over a diabase- lava stream at Homertshansen: ‘ Mineralien und Gesteine aus dem hessischen Hinterland,” Part ii. Section 4; Zeitschr. deut. geol. Gesellsch. vol. xli. (1890), pp: 502-582. VARIOLITIC DIABASE OF THE FICHTELGEBIRGE. 59 Oelschnitz, as at Berneck and near Stein. This gives usa maximum age. The intrusion was apparently before the great earth-movements in which the band of Palaeozoic sediments that now occupies the valley from Berneck through Gefrees and Sparneck to Schwarzen- bach was squeezed in between the Munchberg gneiss and the igneous rocks of the Fichtelgebirge. The normal strike of the Devonian was no doubt originally parallel to that of the Silurian and Cambrian and to the valley; but the mass of the hard diabase has resisted compression better than the more yielding sediments, and these have been bent round the diabase until their strike is often very oblique to their original direction. Further, the fact that the diabase is faulted against the eneiss on the other side of the Oelschnitz suggests that the diabase was consolidated before the great earth-movements took place which impressed upon this area its principal features. VI. Tue OriIGiIn oF THE VARIOLITIC STRUCTURE. In this examination of the diabase to the south of the Oelschnitz we have seen that the variolite in a more or less perfect form constantly tends to appear along the lines of contact with the neighbouring rocks, or where the diabase becomes spheroidal. It was this relation of the variolite to the neighbouring deposits, and a certain resem- blance of the varioles to the baked shales, that led Prof. von Giimbel to the theory that these bodies were fragments of the Devonian rocks caught up at the time of intrusion. The detailed microscopic descriptions by Prof. Rosenbusch clearly demonstrate that, except possibly in one case, the varioles are true spherulites. Nevertheless, as Prof. von Giimbel has again stated his theory in his latest work *, it may be worth while to note what light is thrown upon it by field work. Included fragments of the neighbouring shales in the diabase do occur, as is shown by a good specimen in the Heidel- berg Museum ; but, though baked, these have no resemblance to the true varioles. In all cases in which the varioles were formed of more than mere globulitic accumulations, or which are sufficiently well preserved for their original constituents to be determined, they consist of plagioclase needles, ilmenite, and pyroxene; the baked schists may be sericitic and chloritic, but the above minerals have not been developed. Moreover, the occurrence of the varioles in bands parallel to and at a little distance from the circumference of the spheroids, their regularity, and the gradual passage of the variolitic into the normal compact diabase by the diminution in size and number of the varioles, are features which cannot be satis- factorily explained on Prof. yon Giimbel’s hypothesis. In some cases there seems to have been some confusion, here as elsewhere, between the true variolites and the pseudovariolites or spilites. It. is sometimes not easy, without the aid of the microscope, to dis- tinguish between the variolite and some of the weathering amyg- * “Geologie von Bayern,’ Th. i. Lf. 1 (1884), pp. 78, 79. 60 MR. J. W. GREGORY ON THE daloids. Prof. von Giimbel has described * the diabase sheets of the Labyrinthenberg as coated with variolite; but, though I hammered carefully over the whole section, I could find no true variolite, but plenty of an amygdaloid that somewhat closely re- sembled it *. Though the variolite is thus shown to be a true endomorphic alteration-product, and due, no doubt, to contraction during a some- what rapid cooling, the view that it is an ordinary contact-selvage is not fully adequate, as frequently around some of the Devonian masses in the Miihlleite no variolite occurs at the junction. It is only where the diabase is at the same time spheroidal that the variolitic structure has been fully produced. In the Fichtelgebirge, as at Mont Genévre and in Saxony, the variolite mainly occurs, not as a contact-alteration-product along the junctions of the diabase and other rocks, but on the surfaces of great diabasic spheroids. At Mont Geneévre, in one or two places, a thin spherulite film does occur along the margins of the diabase dykes, but this is rare and always minute. Similarly, in this locality, where the variolite occurs as a contact-product it is thin and inconspicuous, and the varioles are less perfectly developed. In such cases the solidification has apparently been too sudden to allow of the segregation of the felspathic and pyroxenic constituents into spherulites. The variolitic structure has been due to rapid cooling, but is of a less extreme type than that which has produced the amorphous glass of normal basic selvages. y These considerations further suggest the explanation of the rarity of the variolitic amygdaloid, as compared with the compact varieties of the rock. When, owing to diminished pressure or other cause, the solidifying diabase became vesicular by the explosion of its water into steam, the amount of water in the molten magma would be lessened; this would consequently become Jess fluid, and the solidification of the rock be further hastened. That the variolite appears most perfectly in association with spheroidal masses is quite natural if we regard the latter structure as due to processes of contraction during solidification, as shown by Prot. Bonney in his well-known paper ¢. The diabase in this knoll, rapidly cooling by the conduction of its heat to the neighbouring rocks, contracted into great spheroids, which, while semiviscid on the periphery, were still fluid within ; under the pressure of the forces that drove them upwards, these rolled over one another and were drawn out into oval masses. It was during this process that the felspathic and pyroxenic constituents * Geogn. Beschr. Fichtelg. p. 483. t Nor did there seem to be any variolite from this locality in the Museum of the Bavarian Oberbergamt in Munich. This collection contains specimens from several of the other localities mentioned by Prof. von Giimbel, and these are certainly true variolites, as, ¢. g., from Steinbach. For the opportunity of examining the collection there I must express my thanks to Dr. L. von Ammon. { T. G. Bonney, ‘On Columnar, Fissile, and Spheroidal Structure,’ Quart. Journ. Geol. Soe. vol. xxii. (1876), pp. 140-154. VARIOLITIC DIABASE OF THE FICHTELGEBIRGE. 61 began to crystallize out around various points; the plagioclase needles forming radiating clusters, between which theaugite granules were wedged in; as these half-formed varioles were rolled over, still other layers of the variolitic constituents were deposited around them: when the varioles were originally in close contact, these later layers enclosed several and built up the compound varioles. The close connection of the typical variolite with spheroidal struc- ture, which is the case in the Alps, in Italy, in Saxony, and in Wales, reminds one of the recent theory of Prof. de Stefani *, attri- buting the formation of the varioles to secondary decomposition in the outer layer of diabase spheroids. In fact, just as Thomson and other early authorities regarded the spheroidal structure itself as due to weathering, Prof. de Stefani considers these spherulites as due to the same cause. But this theory of the formation of the great spheroids has been generally abandoned since Prof. Bonney’s paper on the subject; while the objections to this hypothesis are still more weighty when it is applied to these smaller structures. The fact that the varioles often occur, not on the extreme edge of the spheroids, but often 20 to 30 mm. from it, in rock which is comparatively unaltered, while, where decomposition has gone on along cracks or fissures, it has not produced any such structure, would be conclusive evidence against this view, even were not the analogy between these Paleozoic varioles and the spherulites of recent lavas sufficiently exact to demonstrate their community of origin. VII. Summary or ConcLustons. 1. That the variolitic diabase of Berneck, which may be taken as a type of those of the Fichtelgebirge, is intrusive into the Devonian. 2. That the variolitic structure occurs in two different arrange- ments :-— (a) On the surfaces of spheroidal masses of compact diabase, which are comparable to those of the eruptive rock of Mont Genévre. | (6) As a true contact-product on the selvage of the diahase. The latter are comparatively rare, and the varioles less perfectly developed. 3. That the varioles are true spherulites, and not included frag- ments of the Devonian rocks. 4. That, though the varioles be the product of rapid cooling, too sudden a solidification of the diabase may prevent their formation. 5. For a similar reason the amygdaloidal is less variolitic than the compact diabase, the loss of the water that occupied the vesicles having diminished the fluidity of the rock. 6. That the “ pseudocrystallites” are rifts and fissures due to contraction; and that the remarkable optical properties described by M. Michel-Lévy are due to the filling-up of cracks by felspathic * ©. de Stefani, ‘‘ Le rocce eruttive dell’ Eocene superiore nell’ Apennino,” Boll. Soc. geol. Ital. vol. viii. no. 2 (1889), 1890, p. 223. 62 ON THE VARIOLITIC DIABASE OF THE FICHTELGEBIRGE. matter deposited in optical continuity with the crystalline fibres on either side. 7. In regard to the terms used, it should be explained that the word diabase has been retained in Hausmann’s sense; most of this rock, however, is more strictly an augite-porphyrite, probably resulting from the alteration of an augite-andesite. The varioles have been throughout referred to as spherulites, from the belief that the variations in structure which place them among the ‘ pseudo- spherulites” of Prof. Rosenbusch are due solely to secondary alterations. ON SOME WATER-WORN AND PEBBLE-~WORN STONES, 63 6. On some WateR-worn and PEeBBie-worn Stones tuken from the Apron of the Horr-Frrer Weir on the River Severn. By Henry Joun Marten, Esq., M.Inst.C.E., F.G.8., Engineer to the Severn Commissioners, (Read December 10, 1890.) In the year 1844 a weir was constructed by the Severn Commis- sioners across the River Severn at a place called Holt Fleet, about eight miles above the city of Worcester, with the object of impound- ing the water above it, for navigation purposes, to a height of 5 feet 4 inches above previous low summer-level. The weir, a plan and section of which accompany this paper (figs. 7 & 8), is a dam of solid masonry across the river, without any sluices in it; so that the whole of the water passing down the river at that point flows over the crest of the weir and down the slope, or face of the apron, on its lower side. The crest of the weir is 300 feet in length, and the drainage from 1870 square miles (nearly 1,200,000 acres) of country above it, extending to the Plinlimmon range, is discharged over it. The stones of which the weir is constructed were taken from an adjoining quarry at Holt Fleet, and consist of a soft red sandstone of the Upper New-Red-Sandstone Formation, marked F. 5 on the one-inch map of the Geological Ordnance Survey. In the year 1887, or 43 years after the stones were placed on the apron of the weir, an examination was made of them, as some of the stones had become displaced and others showed signs of decay. The examination brought to light the fact that a large proportion of the stones on the Island side of the central portion of the apron of the weir had been drilled through and through by the action of the current upon small pebbles lodged either in hollows on their exposed surfaces, or between the joints of the stones *, A table is appended (page 67) giving various particulars of each of the photographed stones (figs. 1-6)—namely, the dimensions, cubic contents, and estimated weight of each stone when placed in the weir, and the cubic contents and weight of each stone when removed from the weir, together with the percentage of loss during the period of 43 years, and the average annual loss during that time. In order to estimate the weight of each stone when placed in the weir, a piece of one of the stones, which was an average sample of the whole—and they were throughout of a uniform character—was cut exactly to a six-inch cube; that is, to one eighth of a cubic foot. This six-inch cube, after saturation in water for three hours, * A sample of a portion of one of these stones, accompanying this paper, together with a set of photographs (here reproduced, at the Author’s expense, figs. 1-6) of six of the stones themselves, were exhibited when the paper was read. In nearly all the photographs the small pebbles referred to are shown in sit, as they were found. 64 MR. H. J. MARTEN ON SOME Fig.1.STONENSI se 66 MR. H. J. MARTEN ON SOME weighed 16 Ib, 12 oz., thus giving the weight of a cubic foot of saturated stone, as lying in the quarry and in the apron of the weir, as 134 lb.; and the weight of each stone when placed in the weir was estimated on this basis. The stones taken out of the weir were also weighed in a saturated state. — The pebbles were probably rolled by the current in flood-times up the somewhat steep up-stream face of the weir, and deposited, as . previously described, in the hollows and joints of the stones forming the flat apron or down-stream side of the weir (see figs. 7 & 8). Figs. 7 & 8.—Plan and Section of the Holt Weir on the Severn. TOWING PATH 700 50 0 ~ +00 200 300 400 $00 FEET Osa “se Ss ML an 6a) ah a GRAS Scale for Plan | a Sa ¥ SSS eS oe =e Aa st uae N NSS Ri Soo Pare cs CL ee ALLL LAA Scale or Weir When the weir was first constructed, the current on the up-stream face was so strong that boulders of considerable size were driven over the crest. The average quantity of water passing over each square foot of the surface of the stones composing the apron of the weir is esti- mated at about 2000 gallons a minute; each gallon of water has an average scrubbing-velocity of from 12 to 15 feet per second, and this, acting on the small pebbles, will give some idea of the forces at work for the 43 years during which the stones were in the weir apron. "There are not many instances in which the specific facts relating to the action of water and pebbles upon a certain class of stone can be so accurately ascertained as in the present case; and the writer has therefore ventured to submit a record of them, in the hope that they may be of use to those who may have occasion to investigate the periods of time likely to be occupied in changes resulting from the abrading action of water and pebbles upon the rocky beds of streams and rivers and over waterfalls into ravines. WATER-WORN AND PEBBLE-WORN STONES. APPENDIX. Particulars of six Water-worn and Pebble-worn Stones taken from the apron of the Severn Commissioners’ Weir across the River Severn at Holt Fleet, about eight miles above Worcester. The stones were placed in the weir during construction in 1844 and were removed in 1887, after having been in the weir apron and subject to the action of the water and the pebbles for 43 years. The stone is a soft red sandstone, trom Holt-Fleet Quarry, which is in the rocks marked F. 5 of the Upper New-Red- Sandstone Formation. Cubic contents and esti-{ Cubic contents and Dimensions of stone when mated weight of each weight of each stone Number of stone placed in weir. stone at 134 lb, the cube] when removed from ’ euicok foot when placed in weir. weir. Loss pe Loss per set of six, 43 years. annum. Length. | Breadth. | Thickness.| Contents.| Weight. [Contents.| Weight. eee eee | ee ee No. ft. | ins. | ft, | ins. ft. ins. | ft. | ins. | cwt.| qrs,| lb. | ft. | ins. | cwt.|qrs.| Ib.] per cent. per cent. Der ee RP 8 Be) HB eprB so febde) eB pede hare el Tes al oldies Oa ele 47 1:09 WD, ReAacuexraiveinees ree 0 1 0 | Bele Oe <2) 2 BO ou Seo tea ane hae ie ape 60 1°40 | eee joapeddsntguevevsiveceh 40) O°) DSBs) 5 Bal 0 P63 POR A209 is | a ie |e leer en 48 1:12 [> tha. Setncntevvescaghecessssardll Delo O°) EMO) | S28) SB 1 QAR 2a teste (ase | 07. ples] meet Oe) alae 50 1:16 Dilvedivsestscdgstvcdvsieeeee eh | eee ees] ED ak? 83 2b | O°) Sil A 2d ts Ae 37 0:86 Gai ihisine oagtieasaevscesel Oh AO Be Ghee O6 Ce eel re shaadi Aa ema Ae 58 1:35 rQ 68 ON SOME WATER-WORN AND PEBBLE-WORN STONES. Discussion. Rey. Epwin Hitz asked if the Author had any means of telling how far chemical action had operated. Mr. Huxxe wished to know if there was any record of oka positions of the six stones given in the table. Prof. HueHrs was desirous of learning whether the stones taken into account were an average sample of the stones. Mr. Wuiraxer commented upon the rapid waste. Mr. Carrutuers asked whether the rock was homogeneous. Rev. H. H. Wirxwoop inquired as to the nature of the pebbles. The PresipENt wished to know whether from the percentages taken some datum could be given for estimating the average loss from the whole surface of the apron. The AurHor believed the action was principally abrasive, as there was only a small proportion of lime in the stone which would be the subject of chemical action. The weir was placed diagonally across ~ the river, and the stones referred to, which were average samples, were taken from the apron at the upper end of the diagonal, where the abrasive effect appeared to be greatest. The pebbles were prin- cipally of quartzose description. The rock from which the stones were taken was of a homogeneous character. In the case of stone No. 1, supposing the action to have been uniform, the abrasion would represent a loss of nearly one foot three inches from the surface of the stone as originally placed in the apron, and the others in pro- portion. [In accordance with a request made by the Council, the Author has added the following notes :— The stones referred to, which were average samples of the pebble- worn stones which had ‘been removed, were taken from the apron of the weir within a distance of from 40 to 100 feet from the Island end of the weir, where the abrasive effect appeared to be greatest. The stones, however, forming the central portion of the apron, be- tween 100 and 200 feet from the Island end of the weir, were almost similarly abraded and perforated, and would also shortly require to be removed. Taking the whole surface of the apron, the stones which were not affected by the action of the pebbles were worn down by the action of the water and ice passing over them for an average depth of about 2 inches from their top faces. That might be taken as the average effect of the water and ice alone passing over the weir in the 43 years during which time the apron had been subjected to the action of those forces.—January 20, 1891.] DR. E. HULL ON THE PHYSICAL GEOLOGY OF TENNESSEE, FIC. 69 7. On the Puystcat Grotocy of TENNESSEE aid ApsoInINe Disrricrs in the Unirep Srares of America. By Epwarp Hutt, M.A,, LL.D., F.R.S., F.G.8., late Director of the Geological Survey of Ireland. (Read December 10, 1890.) ConTENTs. Part I. § 1. Introduction. § 2. Physical Features. 1. The Valley of East Tennessee. 2. Cumberland Plateau; Walden’s Ridge. 3. The Sequachee Valley. 4. Rocks of the Cumberland Table-land. Part II. Development of the chief Physical Features. 1. The Cumberland Plateau. 1. The Stratification. 2. Epoch of Greatest Terrestrial Movements. 3. Direction of Greatest Vertical Movement and Erosion. 4. Formation of the Cumberland Plateau. 11. The Gorge of the Tennessee through the Cumberland Plateau. Part I.—§ 1. Inrropvcrion. A RECENT visit to the Southern States of North America induces me . to lay before the Society some observations on the physical aspect of a peculiarly interesting region traversed by the Tennessee River in the State of the same name and the bordering districts. The geological structure of this district has been ably described by Professor James M. Safford, the State Geologist *. The region is now in process of being re-surveyed topographically and geologically under the direc- tion of Major Powell, U.S. Geological Survey, to whom I am much indebted for kind assistance in procuring maps and informa- tion t. In the present communication I do not propose to enter at any length into the geological structure of the district here described, but only to single out the most striking features connected with its physical structure, and to endeavour to show how they can be accounted for upon those principles of interpretation which, after many years of discussion and research, are generally adopted amongst geologists. Amongst others we shall have to explain the formation of table-lands, and of the erosion of the gorge by which a great river, the Tennessee, traverses a mountain-plateau in pursuing its way towards the ocean, instead of taking a much more direct course. * «Report on the Geology of Tennessee’ (1869). + The sheets, prepared in the Geological Survey Office, are on a scale of , and are contoured at intervals of 100 feet vertical. A very fine minera- 1 125,000 logical map of Tennessee, on a large scale, constructed by Major Kelly, is placed in the Town Hall of Chattanooga. 70 DR. E. HULL ON THE PHYSICAL GEOLOGY OF § 2. Puystcat Features. 1. The Valley of East Tennessee.—The physical features of East Tennessee are, when viewed on a large scale, extremely simple, and are a faithful index to the geological structure. Along its eastern margin, where Tennessee joins North Carolina, the State follows the crest of the Unaka Range, which may be regarded as one of the parallel ridges of the Alleghanies, and is nearly con- tinuous with the Blue Ridge of Virginia. This ridge is composed chiefly of granite, gneiss, and crystalline schists, presumably of Archean age, and forming a prolongation of Professor J. D. Dana’s “‘ Archean Protaxis.” It attains an elevation of 6760 feet in Black Mountain in North Carolina *, and ranges in a general south- westerly direction. From its base stretches the great plain known as ‘‘the Valley of East Tennessee,” which extends south-west into: Georgia and Alabama, and in an opposite direction is continued — into the Valley of Virginia or Shenandoah. This rich and fertile plain has an average breadth of about forty miles, and along its course winds the Tennessee River, a noble stream of about 450 yards in average width. The plain itself is closely furrowed by parallel valleys and ridges, all trending in north-east and south-west direc- tions, parallel to the strike of the beds. The ridges and furrows are in fact the outcrops of the harder and softer strata. The whole valley is underlain by Cambrian and Silurian formations, often highly inclined or thrown into numerous flexures. This series is surmounted by the Devonian beds, here very thin, and consisting chiefly of black shale, which lie close to the base of the northern margin formed by the Cumberland Table-land, which I now proceed to describe. 2. Cumberland Plateau; Walden’s Ridge-—The north-western margin of the Valley of East Tennessee is formed by the escarp- ment of the Cumberland Plateau, which rises abruptly above the plain to a height of 1300 to 1600 feet, or 2000 to 2200 feet above the ocean. The crest of the escarpment, formed of massive grit and conglomerate of Carboniferous age, breaks off into mural precipices, often perfectly vertical. As the Tennessee River hugs the base of this escarpment for many miles, the full height of the cliff is thus obtained at one sweep; and as the slopes as well as the summit of the ridge are covered with primeval forest, except where the naked cliff offers no footing for vegetation, the view of this grand escarpment is as striking as it is beautiful. The escarpment above described forms the south-eastern margin of the Cumberland Table-land, the surface of which is slightly undulating, formed of Carboniferous beds. and which, below Chat- tanooga, immediately on the west, is traversed by the Tennessee River through a deep and winding gorge about twenty miles in length, where the States of Georgia and Alabama on the south join * The granitoid rocks of North Carolina are remarkable for the number, beauty, and size of the minerals they have yielded; specimens are exhibited in the museum of the Smithsonian Institution, Washington. TENNESSEE AND ADJOINING DISTRICTS. 71 on to that of Tennessee on the north (see Map, fig. 1, facing p. 74). We shall have to discuss the mode of formation of this remark- able gorge later on. The Cumberland Plateau has a breadth of about forty miles north of Chattanooga (lat. 35° 15’ N.), and it breaks off along the north-western margin in a precipitous and lofty escarpment, as along the valley of East Tennessee, but much indented by valleys and coves; while the south-eastern escarpment is seldom broken, but sweeps along the banks of the river in a nearly direct or gracefully-curving line, the indentations of the streams being hardly noticeable *. The Cumberland Table-land is the southerly prolongation of the Appalachian Mountains ; and, though deeply indented by the Cum- berland River and its branches in the North-west, is nowhere absolutely cut through by these streams; so that it is only in the gorge of the Tennessee, close to Chattanooga, that the complete intersection of the range is effected. To the south of this gorge the table-land continues into Northern Alabama, till the Carboni- ferous strata sink down and disappear beneath those of Cretaceous and Tertiary age which border the shores of the Gulf of Mexico. Several terraced and nearly isolated hills, portions of a once con- tinuous plateau, occur along the Tennessee near Chattanooga, of which “ Lookout Point,” rising abruptly from the river-bank to a height of 2126 feet above the sea. or 1450 feet above the stream, is the most conspicuous example (see Map, fig. 1). The average elevation of the Cumberland Plateau may be taken at, 2000 feet above the surface of the ocean, and 1350 feet above the Tennessee River at Chattanooga; but towards Pennsylvania on the north, at Cross Mountain, it rises to about 2800 feet T, where its structure becomes more complicated. Confining our attention, however, to the region of Tennessee and the borders of Kentucky, we observe that this table-land has the character of a well- defined plateau, formed of massive grit and conglomerate, or other strata, of Upper-Carboniferous age, and intersected by deep ravines, which open out to the south and west, and form the channels of streams draining into the Tennessee, the Cumberland, and the Ohio. (See Sections, figs. 2 and 3.) Over its whole surface and its flanks this table-land is enveloped in almost continuous virgin forest, consisting of trees of great variety and often of noble stature, with an undergrowth of smaller plants. Nearly fifty varieties of forest-trees may here be counted, including, amongst others, cedars, pines, maples, chestnuts, satin- wood, poplars, and oak of several varieties. These forests give cover to many wild animals, including pumas, bears, deer, hogs, and smaller game. Rattlesnakes and other venomous reptiles le con- cealed under the fallen logs, and at night the groves and low-lying woods at the foot of the plateau are lighted up by myriads of fire-flies, while the air is resonant with the croaking of the * According to Professor Safford. t In Walden’s Ridge, east of the Sequachee Valley, there are tracts reaching the 2300 or 2400 feet level. te DR. E. HULL ON THE PHYSICAL GEOLOGY OF tree-frogs. From the crest of the escarpment at various points beautiful and extensive prospects may be obtained of this region of wooded plateaux and wide valleys, where the white man has as yet done little to alter the natural landscape, or to diminish the extent of the primeval forest *. 3. The Sequachee Valley—tThe table-land thus described is inter- sected longitudinally by a remarkable valley, that of the Sequachee River, for a distance of sixty miles, in a nearly straight line north- eastward from the banks of the Tennessee near Jasper, with an average breadth of four miles. The narrow plateau thus formed be- tween the valley of East Tennessee and the Sequachee is known as “‘Walden’s Ridge” (see Map, fig. 1). he direction of the Sequachee Valley is therefore parallel to that of the eastern boundary-scarp of the table-land itself, where it overlooks the Valley of Kast Tennessee. On either side it is bounded by steep and densely-wooded slopes, generally crowned by cliffs of grit or conglomerate ; and at its upper end the Sequachee River has its origin in copious springs issuing forth at the foot of the sandstone cliffs. I was unable to visit the source of this stream, but, from the accounts I had from observers in the district, it must be most remarkable. From the foot of the cliff the waters flow down the steep slopes into a natural caldron, formed in the soft shales and grits overlying the Carboniferous Limestone. The latter here forms a barrier, holding back the waters which have hollowed out a tunnel through the rock, and on issuing forth they descend into the valley in a series of cascades. The flanks of the Sequachee Valley are composed of Carboniferous erits and shales resting on limestone, from below which the Devonian and Silurian strata emerge with a dip in the direction of tbe sides of the valley (see figs. 2 and 3). The valley is therefore clearly in the line of an anticlinal axis; and to this it probably owes its origin, though it is possible that there may be a fault here running in a | parallel direction, along which river-erosion has aeted through a lengthened period. It is a striking example of valleys of this kind, The Little Sequachee, a smaller valley further to the west, is pail ably due to a similar anticlina] flexure. 4. Rocks of the Cumberland Table-land.—The geological structure of the Cumberland Table-land is extremely simple. The strata of which it is formed consist of grits (sometimes pebbly), sandstones, and shale, with beds of coal, all of Carboniferous age, resting on Mountain-Limestone, which crops out in two beds, separated by soft red sandstone, all along the base of the escarpment; the two series constitute in part the ‘ Carboniferous” and ‘ Sub-Carboni- ferous”’ groups of American geologists (see figs. 2 and 3). The Carboniferous series is succeeded, in descending order, by dark Devonian shales, which, owing to their friable nature, have * This region was the abode of Cherokee Indians, who some years ago were transplanted to the Indian Reserves in the Western States. Shell-mound Station is the site of the terrible slaughter of this tribe by its white and more civilized brethren in 1816, under Major Bond. TENNESSEE AND ADJOINING DISTRICTS. 73 doubtless facilitated the work of erosion; and these again by the members of the Upper and Lower Silurian groups, occupying the plains and the central portions of the valleys. The Silurian strata, which are thrown into numerous flexures along the valley of East Tennessee, ultimately give place to others of Cambrian age as we approach the Archean Protaxis of the Unaka range, forming the south-eastern margin of the plain. Parr I].—DerEvrELorpMENT OF THE CHIEF PuysicaL FEATURES. 1. The Cwnberland Plateau. The physical features, the origin of which I here propose to discuss, are (1) the Cumberland Plateau, and (2) the Gorge of the Tennessee River where it traverses this plateau below Chattanooga. The discussion of the origin of these two leading physical features necessarily involves some reference to the mode of formation of the adjoining areas, that of the Valley of Eastern Tennessee on the east, and that of the Silurian plain of the Cumberland River, or of Nashville, on the west. An inspection of the longer diagrammatic section, from the Archzan Protaxis of the Unaka Range to the plain of the Cumberland River at Nashville, shows in order of suc- cession from east to west—(1) the Unaka Range; (2) the Valley of Kast Tennessee ; (3) the Cumberland Plateau; (4) the Silurian dome or uprise of Nashville; together with the generalized stratifi- cation of this tract. (See Sections, figs. 2 and 3.) 1. The Stratification—In dealing with this subject I have to observe that from the base of the Cambrian beds, where they rest discordantly upon those of the Archean Protaxis, the whole series of Lower- and Upper-Paleozoic formations succeed each other in apparently conformable sequence, except at the junction of the Lower- and Upper-Silurian series, where a probable discordance occurs *, Throughout the prolonged period during which these formations were being deposited, there was continuous subsidence, with occasional pauses, over the region lying to the west of the Archean Continental area, and successive formations of marine strata were laid down in vast sheets over the bed of the ocean, never probably very deep. In later Carboniferous times the marine deposits gave place to those of lacustrine or estuarine origin, but still without any apparent discordance in the stratification ; so that the Upper and Lower Carboniferous beds are apparently conform- able to each other, and these again to the Devonian and Upper Silurian 7. * According to Professor J. D. Dana, this discordance is very marked in the New-England States, where the Lower-Silurian beds have been metamorphosed and elevated with the Archean rocks, In their southern prolongation this is not so evident, but highly probable. See J. D. Dana, “ Areas of Continental Progress,” Bull. Geol. Soc. America, vol. i. 1889. t I use the expression “apparently conformable,” because, though there may be discordances of stratification, they are so small as not to have been recog- nized, Ee 74 DR. E. HULL ON THE PHYSICAL GEOLOGY OF 2. Epoch of Greatest Terrestrial Movements.—The prolonged period of subsidence and deposition above described at length gave place to an epoch of elevation and contraction of the crust, acting with greatest effect and intensity along the line of the Alleghanies, and parallel with the Atlantic sea-board, where the Paleozoic strata are folded, flexured, and even reversed, along parallel axes, as so admirably illustrated by the late Professor H. P. Rogers*. The foldings of the strata, it is well known, generally subside in a westerly direction towards the Valley of the Ohio, and ultimately pass Into widely extended dome-shaped centres of elevation with intervening areas of depression. Amongst the former are the ‘Cincinnati uplift” and the anticline of the Nashville Silurians ; amongst the latter is the region of the Cumberland Plateau, which hes along the centre of a broad syncline. 3. Direction of Greatest Vertical Movement and Erosion.—From what has been said, it clearly follows that the greatest amount of vertical movement, consequent on powerful lateral thrust, was along the Archean Protaxis of the Alleghanies. All along this line the Paleozoic strata were elevated thousands of feet above the ocean, and subjected in consequence to great denudation ; this process was doubtless facilitated by the flexures and fissures accompanying the movement. Away from this axis of disturbance, the strata (as has been already observed) were but slightly moved, with the result that they remained under water and undenuded, or but slightly emergent, long after those on the border of the Archean Protaxis were being subjected to extensive erosion. Under these conditions denudation proceeded more rapidly along the tract bordering the Protaxis, and especially along the arches or anticlinal flexures. The synclines, or trough-shaped areas, were protected from erosion to a greater or less degree. In the region with which we are specially concerned, the line of the Unaka Range and Blue Mountains, which was perhaps never altogether submerged, was upraised gradually into high land. The Cambrian and Silurian strata were subjected to erosion; and streams carrying the materials flowed down the flanks of the emergent land into the sea or estuary to the westward. This process was going on all through the Meso- zoic period. As time went on, these western tracts, wherever in the line of anticlines, were themselves elevated and eroded, and ultimately the synclines themselves ; but the necessary result of this unequal process of erosion would be to leave the synclinal tracts relatively higher than the anticlines. At a later epoch the Cumber- land Plateau began to be formed by the cutting back of the strata in the direction of their dip ; the massive Carboniferous grits, resting on softer strata largely formed of shales, presenting the necessary conditions for the development of a crested ridge. 4. Formation of the Cumberland Plateau.—We are now in a position to understand the primary conditions under which this plateau was developed. First there are the required stratigraphical * “Geology of Pennsylvania.’ [To face p. 74. ’ Q. J. GS. vol. xlvii.] s [Lo face p. 74, , Fig. 1.— Map of a part of the State of Tennessee, showing the Gorge of the Tennessee River near Chattanooga. MAP AND SECTIONS 0 H 2 MILES. TO ILLUSTRATE : INMAN J FF DR. E. HULL'S PAPER ON THE PHYSICAL GEOLOGY _ OF THE TENNESSEE AND PARTS OF ADJOINING DISTRICTS IN THE UNITED STATES OF AMERICA. Fig. 2.—Section across the Cumberland Plateau to the Hast Tennessee Valley. Cumberland Table-land. Valley of Hast Tennessee, r ao Little Tennessee River W. Sequatchee Sequatchee Valley, Walden’s North of i. Ridge. Chattanooga. 1 ELE RIVER, ‘ q i (> | Bee <> Z = LSS — YY ff CON cLOMEGATE . oon, SHALES & SANDSTONES CARBONIFEROUS, WH COAL-SEAMS - LIMESTONE W/TH BAW OF SANDSTONE, Vertiradl Sele "54,9 190020005000, WG DEVONIAN. = UPPER SILURIAN. 5 5 ° ( 2 3 4 5 Eaplamation of Big, 2, “7m sles Fig. 3.— Generalized Section across the State of Tennessee. (Length about 200 miles.) The Unaka Monnaie Blue Ridge. Reshwolle Sequatchee Tennesseo or blu nag Anticline. Valley. River. N.W. ' THE CUMBERLAND) PLATEAU. a Cc { - SSS SS = —— LTD z 2 SS : ST SILURIAN LOWER SILURIAN. ¢ GNEISS & SCHISTS. G SILURIAN STRATA. 4) UPPER xs ata EISS. GRANITE, &¢ 2(r Yes ies s | a) UPPER & LOWER SILURIAN, CARBONIFEROUS Ewplanation of Fig, 3. se. uwesrove, § ; ETM) cananran, ie eas GRANITE, VOWIAN, a a aenaan {cies AND WOLa aA] HORNGLEWOK SCHIST, TENNESSEE AND ADJOINING DISTRICTS. 19> conditions, namely hard grits or sandstones resting on soft strata, and these occupying the line of a low synclinal axis, ranging in a N.E. and §.W. direction. The strata in this position being the latest which were upraised, were preserved almost intact; while those continuous with them, and forming the flanks of the parallel anticline, were denuded away. The simple conditions here stated are somewhat modified by the two secondary anticlines along the Sequachee Valleys; but these do not affect the general position, and are themselves examples of lesser valleys eroded along anticlinal axes. It should also be observed that the Tennessee River, continued into the Clinch River, keeps close to the base of the escarpment of the Cumberland Plateau (Walden’s Ridge); and we may suppose that, as this escarpment was cut back in the direction of the dip, the river itself gradually moved westward, or in the same direction*. Thus the Cumberland Plateau was developed by the erosion of the Valley of East Tennessee on the one hand, and by a somewhat similar series of physical operations along the Valley of the Cumber- land River on the other or western side. un. The Gorge of the Tennessee through the Cumberland Plateau. The course of this stream, the fourth in size in the United States, is most remarkable, and requires to be explained on geological prin- ciples. Descending (under the name of the Little Tennessee) from the Blue Ridge (or Archean Protaxis), it crosses the Unaka ridge in a north-westerly direction to Kingston ; here it joins the Clinch River, coming down the Shenandoah Valley in a south-westerly direction ; and this course it retains, flowing along the foot of the Cumberland Plateau to Chattanooga, when it changes its course, and traverses the plateau by the gorge already described. Ultimately the Tennessee, instead of continuing its course in a southerly direction into the Gulf of Mexico, makes a great sweep to the northward and joins the Ohio at a distance of about forty miles above the junction of that river with the Mississippi, thus adding to its course a length of about 800 miles! The east and west saddle or water-parting, from which the streams drain into the Tennessee on the one side and into the Gulf of Mexico on the other, descends to a level of about 920 feet above the waters of the Gulf a few miles south of Chattanooga. The level of the saddle is only 270-280 feet above the river at Chattanooga; so that (to put the case in popular language) we may say that the Tennessee, rather than take a direct course towards the Gulf by crossing a saddle which is only 270-280 feet above its bed, has preferred a channel through a table-land rising 1400-1500 feet above its bed—a course * Tf we regard the direction, the Tennessee River is the real continuation of the Clinch downwards, and the Little Tennessee is a lateral tributary. ‘The Tennessee at a former period probably ran in a channel further east and at a higher level. ' 76 DR, E. HULL ON THE PHYSICAL GEOLOGY OF which shows that the original relative levels of the saddle and the plateau have been absolutely reversed. In brief, therefore, we infer that when the river began to erode its channel in the region of the Cumberland Plateau, this tract was relatively lower than that to the south of its present course. By the process of denudation these relations have been reversed ; but the river, having once begun to wear down its channel, continued to deepen it as the land rose; so that, having once selected its course, itnever afterwards left it. If it be permitted to compare small things with great, we may say that the process of valley-erosion as applicable to the Tennessee is somewhat analogous to that which took place in the South-east of England during later Tertiary times, in consequence of which streams, such as the Medway and the Ouse, pass into the sea by channels traversing the escarpments of the Chalk and Lower Greensand. The high grounds forming the sources of these streams in the centre of the Wealden area represent the ridge of the Unaka and Blue Moun- tains ; the plain of the Weald Clay represents the Valley of Hast Tennessee, and the escarpments of the Greensand and Chalk the Cumberland Plateau. How these channels were formed, together with the adjoining escarpments, has been ably explained by Messrs: Foster and Topley in their joint paper on the ‘* Denudation of the Weald” * and further illustrated by Sir Andrew Ramsay. ‘The principles of interpretation which have been adopted in the one case are applicable in the other, though on a larger scale, and need not be repeated ¢. The effects of denudation here described were doubtless accelerated during the “‘ Pluvial” or “‘ Champlain”’ Period, corresponding to the later stages of the Glacial Period. ‘This region was, it is true, far to the south of the limits of the great ice-sheet of North America, as shown by Mr. T. C. Chamberlain §; but the evidences of extra- ordinarily copious rainfall and of the former erosive and transporting action of the rivers over the regions lying along the margin of the great ice-sheet are abundantly evident, and are fully recognized by American geologists. Along the eastern side of the Alleghanies the representative of this epoch is the ‘“‘ Columbia Formation” de- scribed by Mr. W. J. M‘Gee || ; and to a similar stage is probably referable the remarkable deposit of red loam by which the surface of the country in the valleys of the Tennessee and Sequachee is over- spread to a depth of many feet or even yards. The effects of extensive aqueous erosion, and the consequent deposition of sediment in the valleys beyond the reach of existing streams, are everywhere mani- fest in this part of America. * Quart. Journ. Geol. Soc. vol. xxi. t ‘Phys. Geol. & Geogr. of Great Britain.’ {t It is right to observe that Professor Safford and Mr. J. P. Leslie account for the preservation of the Cumberland Plateau by faulting, which has relatively lowered the Carboniferous strata ; but the well-defined escarpment with which the strata crop out along the Valley of East Tennessee near Chattanooga seems to me to show that such a cause is insufficient. § ‘Seventh Annual Report U.S. Geol. Survey,’ p. 155. || -Zbid., ‘* Taxonomy of the ‘Columbia Formation,’” p. 611, &e. TENNESSEE AND ADJOINING DISTRICTS. ‘i DIscussion. Mr. Torrey thought the parallel drawn by the Author with the Wealden area was in part justified, but there were differences connected with minor points. The structure had been worked out in detail for the Wealden area, but a similar state of things existed in other parts of England. The watershed between the Hast Ten- nessee valley and the Gulf of Mexico must have been greatly lowered. Prof. Hueues asked if the gravels of the high terraces were com- posed of Silurian or Carboniferous detritus, as he wished to know whether the Carboniferous beds of the plateau had been continued over the East Tennessee valley at the time of the formation of the gorge. The northerly direction of the river after leaving the plateau suggested change of level. Mr. Witts compared the area described with the gorge of the Avon at Bristol. Dr. Hytanp had not been led to any definite conclusion during his short stay in the region. The Prestpent found difficulties here, as elsewhere, in realizing the form of the ground when the rivers began to flow, and in dis- covering whether there were subterranean movements which affected the denudation. He felt that the explanation of the topography might not be so simple as Prof, Hull made out, and would like to have more details as to the structure of the ground. The AvrnHor, in reply, concurred with the remarks of the Pre- sident as to the complex character of the subject. He thought the fault drawn by Prof. Safford on the east side of the plateau had little to do with the formation of the escarpment. He had no evidence to adduce in answer to Prof. Hughes’s question. There was no more reason why the river should have flowed south on the west than on the east of the plateau. hs a , Ob! aa at eae FA uh Pag ot re ow Joa Yat ibe ee , ; ‘% y dog tay * Ks “4 oy ee a} Sey Fe oo . “Oi (tuba ut er tt, ial ¥ ry ; A ‘eo we oN y Chive ae i> at eS ene rk oy Oe at, Se i oc OM EPRIT th 2s Ps ee _ Pely bait te A Ae thane? TENNESSEE AND ADJOINING DISTRICTS. Td DIscussiIon. Mr. Tortry thought the parallel drawn by the Author with the Wealden area was in part justified, but there were differences connected with minor points. The structure had been worked out in detail for the Wealden area, but a similar state of things existed in other parts of England. The watershed between the East Ten- nessee valley and the Gulf of Mexico must have been greatly lowered. Prof. Hucues asked if the gravels of the high terraces were com- posed of Silurian or Carboniferous detritus, as he wished to know whether the Carboniferous beds of the plateau had been continued over the Kast Tennessee valley at the time of the formation of the gorge. The northerly direction of the river after leaving the plateau suggested change of level. Mr. Wits compared the area described with the gorge of the Avon at Bristol. Dr. Hyzanp had not been led to any definite conclusion during his short stay in the region. The Presiprnt found difficulties here, as elsewhere, in realizing the form of the ground when the rivers began to flow, and in dis- covering whether there were subterranean movements which affected the denudation. He felt that the explanation of the topography might not be so simple as Prof. Hull made out, and would like tc have more details as to the structure of the ground. The Avruor, in reply, concurred with the remarks of the Pre- sident as to the complex character of the subject. He thought the fault drawn by Prof. Safford on the east side of the plateau had little to do with the formation of the escarpment. He had no evidence to adduce in answer to Prof. Hughes’s question. There was no more reason why the river should have flowed south on the west than on the east of the plateau. Q.J.G.8.. No. 186 G 78 REV. EDWIN HILL AND PROF. T. G. BONNEY ON THE 8. On the Nortu-west Reeion of Caarnwoop Forssr, with other Notes. By the Rev. Epwin Hitt*, M.A., F.G.S., and Pro- fessor T, G:. Bonney, D.Sc., LL.D., F.R.S., V.P.G.8. (Read January 7, 1891.) ConTENTS. I. Introduction. II. The North-west Region and Bardon Hill. . General Description of the North-west Region. 2. The Porphyroid of Peldar Tor. 3. The Porphyroid of Sharpley. 4, Field Relations of the Peldar and Sharpley Rocks. 5. Bardon Quarry. III. Additional Notes. G. Stable Quarry, Bradgate Park. 7. ‘Uhe Igneous Junctions. 8. Brazil Wood. 9. The Blackbrook Group. 10. Fragments and Pebbles. 11. Glacial Phenomena. 12. Age of the Clastic Charnwood Rocks. 15. Age of the Igneous Rocks, 14. Corrigenda, — I. Inrropvuction. Ir is now more than ten years since the last of our papers on the pre-Carboniferous rocks of Charnwood Forest was laid before this Society 7. At that time, as we stated, we had no expectation of writing further upon the district. But since then, though little has been changed in the Forest, beyond the enlargement of some quarries, the general progress of knowledge has affected our inter- pretation of some of the facts which we had ascertained, and much has been learnt in regard to the whole subject of metamorphism, especially as to the effects of pressure, due to movements of the earth’s crust, in modifying rock-structures and initiating, if not producing, mineral changes. We have, we hope, more knowledge and a wider experience, so that our interpretation might be altered, though the facts might be unchanged ¢. Now there was one district in the Forest, that of Peldar Tor, * Mr. Hill desires to state that throughout this paper all references to microscopic evidence are due to Professor Bonney. He himself has taken part only in the field work. Tt Quart. Journ. Geol. Soc. vol. xxxvi. (1880) p. 337. + Not much has been published since the date of our last paper. There is a good account of a visit of the Geologists’ Association published in their ‘ Pro- ceedings, vol. x. (1888) p. 472, by Mr. J. D. Paul, to which is appended a useful note on the microscopic structure of some of the rocks, by Major-Gen. McMahon, who expresses the opinion that the rocks of Sharpley and Peldar Tor are lavas. Mr. W. J. Harrison refers, in some papers on the pre-Carbo- | niferous floor of the Midlands, to the rock of Charnwood (‘ Midland Naturalist,’ vol. viil.), and describes the syenites of S. Leicestershire (2b¢d. vol. vii.), but adds nothing material to our notice, to which he does not refer, NORTH-WEST REGION OF CHARNWOOD FOREST, 79 with Sharpley and Bardon Hill, in regard to which, as stated at the time *, we had found great difficulty in deciding whether the rocks had been originally ashes or lavas; that is to say, whether the fragmental structure, which could be dimly traced, had been present from the very first, or had been superinduced. We dis- cussed the question at length, giving, to our best ability, the arguments on each side, and coming to the conclusion that the rocks of Peldar Tor, of Sharpley, and of Bardon Hill were all of pyro- clastic origin. Still, as expressed at the time, this view was not without considerable difficulties, especially in regard to the first and second, and during the following three or four years new factors were introduced into the problem, of which account had to be taken. The researches of Lehman, Heim, and others, coupled with our own work in other fields, indicated that earth-movements were far more effective than had hitherto been supposed in producing breccia- tion and clastic structures in large masses of rock, which originally had been homogeneous or crystalline. Again, increased experience showed us that from fiow-brecciation and other causes a fragmental structure was of commoner occurrence in a true lava than we had expected, and that the general uniformity of character which was presented by the rocks of Peldar and Sharpley over such large areas was very difficult to parallel in the case of tuffs. One thing, how- ever, more than any other made a reconsideration of the question absolutely necessary: we referred in our discussion to the porphy- roids of the Ardennes as presenting very close resemblances to the Peldar and Sharpley recks, especially to the latter. These had been examined by eminent geologists, who had discussed their origin, and denied it to be igneoust. In 1882, however, one of us was able to visit the Ardennes, and came without any hesitation to the con- clusion that these porphyroids were simply igneous rocks modified by subsequent pressure t. This of course struck away one of our chief supports, and led us to examine the Charnwood district anew. We were the more hopeful of some result, because the publication of the six-inch map made it possible to record our observations with a detail which was impossible on the ordinary one-inch map, and thus to obtain a clearer idea of the form of the areas occupied by these rocks, and of their relation to others in the neighbourhood. Accord- ingly, during visits in the spring of 1887, 1889, and 1890, we carefully re-examined not only all the north-western part of the Forest, but also a few other localities about which we felt some difficulty or were hopeful of additional evidence. Il. Tae Norru-west Recion anp Barpon Hit. 1. General Description of the North-west Region.—This district is entirely included within lines drawn through Abbot’s Oak (or Greenhill) and the Whitwick-village Quarry (Pinfold Quarry) on * Op. cit. pp. 541-348. t ‘Les Roches Plutoniennes de la Belgique,’ &e. p, 246. + Bonney, Proc. Geol. Assoc. vol. ix. (1885) p. 247. GZ 80 REV. EDWIN HILL AND PROF. T. G. BONNEY ON THE the one side, through the plantations of Strawberry Hill and Cat Hill on the other, and from Abbot’s Oak, skirting Timberwood Hill, across these two. These boundaries are probably natural ones, for they follow the features of the ground, and do not leave any out- crops outside, though many come close to or are upon them. The first line is parallel to the well-marked horizon of the Black- brook Group, and the second (about a mile from the first) only makes an angle of 7° with that horizon. The third, if produced, also skirts the slopes of Bardon Hill and suggests a fault. The boundary of visible rock in the remaining direction (N.W.) is irregular. The rocks within these limits have few representatives in the rest of the Forest. The outcrops from the $.E. boundary up to Peldar Tor and Sharpley are, in general, confused heaps of agglo- merate, and ashes without large fragments are rarely seen. Then come the areas occupied by the porphyroids of Peldar Tor and Sharpley which are in contact along a very short line (from Abbey | Grange to Spring Hill Farm, about a quarter of a mile). The remainder of the region also contains agglomerates and ashes, the latter, without conspicuous fragments, being more frequent than in the southern part; yet even here the majority of outcrops are agglomerates. Even the very flinty slate at the Car-Hill Quarry, Whitwick, and at that near the Forest-Rock Hotel do not indicate thick masses, and are seen to pass into beds of ashy materials. 2. The Porphyroid of Peldar Tor.—We have little to add to or correct in our published description of the general macroscopic characters of this rock. The dull green colour of the matrix, the rough external surface, the rugged and almost lumpy weathering are features markedly characteristic. We have observed, however, that some of the outcrops on the western margin of the mass approach the Sharpley type in that they display a smoother surface and a tendency to bleach in weathering. Included fragments are not numerous, and are generally small, though occasionally a fragment several inches in diameter may be found*. Some resemble the porphyritic felstone which is common in the neighbouring volcanic breccias; others (the more numerous) are a rather fine- grained reddish-grey rock. ‘The Peldar porphyroid, here and there, * The largest which we observed was 18’ in diameter—an exceptional size. This specimen under the microscope exhibits a groundmass consisting of nume- rous small felspar erystals—often about ‘02'’ long—some with Carlsbad twinning and resembling orthoclase, some plagioclase. The intervals are blackened with opacite or occupied by viridite. We find also grains of epidote and of iron oxide (?ilmenite more or less altered). In this groundmass are one or two small grains of quartz and several larger crystals of felspar, some showing plagioclastic twinning. All exhibit a rather rounded central part speckled with gpacite, surrounded by a clearer margin, which has a more rectilinear boundary. A slide from a compact-looking fragment, collected trom the 8. side, much resembles in general character the rock of the “‘ purple porphyritic” fragments at Ratchet Hill, &., though the quartzes and felspars are smaller, and its ground- mass puts on occasionally the slightly “granular” aspect characteristic of the urdinary Peldar rock. Now and then these “granules” give indications of spherulitic structure. There is a nest of small felspar crystals and epidote (?), an inclusion from a yet older rock. ae [Yo face p. 80. MAP OF THE AREAS OF SHARPLEY ano PELDAR TOR IN CHARNWOOD FOREST SHARPLEY noc (ea MISIBEe Lal 1s KS) Inrerrco VISIBLE WZ \wreRREO Ash on AGGLOMERATE VISIBLE PELOAR nocn| A. .AGGLOMERATE of.. PURPLE PORPHYRITIC A..COMPACT PURPLE Y.. PORCELLANOUS ¢.. SvYEniToID FRAGMENTS INcLUDED VS fy 4 42 £,2 42 + bts pease EES £ set “Hich Cape maN's 9494049 DryeRoox 9 4: x AB Wooo ; yee Hotty Haves Wooo 8 FURLONGS 1 MILE. Q, J. G. 8. vol. xlvii.] EXPLANATION OF THE MAP. Turis Map, traced from the Ordnance Survey on the six-inch scale, shows the areas in which the porphyroids of Sharpley and of Peldar Tor are exposed, and the regions beneath which they probably occur. The latter are obtained approximately by connecting together the outermost outcrops of each type of rock with straight lines, so as to define an area within which no other rock is visible. The Map also marks the outcrops of other rocks surrounding the above-named areas, and indicates their nature in each case. No attempt has been made to give an accurate representation of the outlines of the visible rock-masses. ‘The place has been determined by the indica- tions on the Map, and the outline sketched in from memory; more- over, in some cases a number of separated outcrops, which are in proximity one to another, are represented as if they formed, as no doubt they do, a continuous mass. [Vo face p. 80. MAP OF THE AREAS OF SHARPLEY ano PELDAR TOR IN CHARNWOOD FOREST SHARPLEY ROCK = Mi INFEAREO (BD Visieve PELOAR ROCK la een ASH on AGGLOMERATE Pip] VISIBLE A AaG.omeRATe Ri of. PURPLE PoRPHyRITIC A..Coupacr PURPLE Y., PORCELLANOUS @.. SYENITOID ae 9 Pienration eres 2 8B fe Pree mvencex G4 (i.sar Wooo a . [One Basen ie "ARI FRAGMENTS: IkcLuoED ) 2 FURLONGS 1 MILE. NORTH-WEST REGION OF CHARNWOOD FOREST. 81 has an aspect which suggests a “ fragmental” origin, and its mode of weathering certainly more resembles that of an indurated homo- geneous volcanic ash than of a lava. A rough cleavage can generally be discovered on closer examination. Its strike is approximately W.8.W.—E.N.E. (sometimes more nearly approaching W.-E.), with a very high dip on the northern side. Though much attention has been paid to the microscopic structure of the rock, we have little to add to our former remarks. The larger felspars are sometimes to a great extent replaced by epidote, which, by the mode of its occurrence, appears to have formed direct from the other mineral, almost as a paramorph, instead of resulting from a general exchange of constituents with the neighbouring minerals. This change would most readily occur in the case of an andesine- or labradorite-felspar, provided some of the soda and of the silica were removed by solution. But an addition of some iron would seem requisite. This, however, might have been obtained from enclosures of iron oxide or ferriferous glass. Epidote and viridite are sometimes associated: the latter also occurs alone, and varies from an isotropic mass to an aggregate of flaky minerals, which have only a feeble action on polarized light; probably these belong to the chlorite group, although in some cases they may be nearer to serpentine. ‘These may indicate the former presence of a member of the pyroxenic group, but neither the external form nor the structure of the grain helps us to a conclusion. We think it probable that the grains composed of viridite and abundant opacite, for the origin of which garnet or even olivine was vaguely suggested in our former papers, are an iron oxide, in some cases at least ilmenite, where the grain has been partly converted into a mineral allied to chloropal. The quartzes are cracked, but, as a rule, do not exhibit strain-shadows. The cracks sometimes are occupied only by viridite or an allied secondary mineral, and appear connected with a linear structure indicative of crushing of the adjacent rock, in which case we refer them to subsequent pressure. But occasionally they are partly occupied by material corresponding with the matrix, and suggest that they are due to strains set up before it solidified. For instance, in one slide, small angular bits of quartz are excep- tionally numerous, and occupy positions in relation to an inlet in a large quartz grain (like a wedge cut out of a round cake); and this seems as if they came from the gap. Sometimes the quartzes have a fairly-defined crystalline outline, but generally they are more or less rounded, and occasionally seem to have been invaded by the matrix. Some difficulties as to the structure of the last have been cleared up by additional study. We called attention formerly to a peculiar spotted character produced by thin lines of a green mineral which seemed to traverse the whole in a kind of network. This, it is now clear, is a secondary product, replacing a black iron oxide, and so it is most likely a variety of chloropal. As the rock cooled, pro- bably the iron, as usual, separated out of the glass, and then a coneretionary action was set up in the latter, expelling the opacite, 82 REV. EDWIN HILL AND PROF, T. G. BONNEY. ON THE and causing it to aggregate as a kind of network between their walls. Subsequent change, perhaps devitrification, and certainly the action of water, brought the rock to its present condition. The rock which we mentioned as occurring on the north side of the Bardon-Hill Pit, and apparently identical with that of Peldar Tor, has been much more fully exposed, owing to the enlargement of the quarry. It is indubitably, macroscopically and microscopically, identical with that of Peldar Tor. It contains rock-fragments, similar to that described above. In one place a slightly more com- pact variety occurs. The matrix of this does not exhibit the usual spotted structure, but gives a very faint indication of a fluidal struc- ture. This Peldar rock overlics, with a rather irregular base, a porphyroid which bears, as will be shown in our account of the pit, a general resemblance to the rock of Sharpley, and, in this, fragments and possibly lenticular streaks of the characteristic Peldar rock are sometimes abundant. We have thus been led to abandon our idea that the Peldar porphyroid had a pyroclastic origin, and now regard it as a lava, somewhat modified by various secondary changes. The appended analyses *, kindly made in duplicate for us by Miss E. Aston, B.Se., in Professor Ramsay’s laboratory at University College (London), indicate that, as we had heen led by microscopic examination to expect, the rock is rather intermediate in its position; but on the whole it is more nearly allied to the dacites than to the rhyolites. Hence it should be named either an altered dacite or a porphyrite. 3. The Porphyroid of Sharpley.—In our last paper we called attention to the very close resemblance which this rock presented to a lava, but, for reasons there given, preferred to regard it as the result of the alteration of a tuff which had a rather uniform character. Further study, however, has increased the difficulties which existed in the latter view, and diminished greatly those in the former. The fragmental character of the normal rock is due, we now believe, to the pressure which has produced the schistosity, and led to the development of films of sericite. To the same cause the eracking of the quartzes, at any rate in most cases, must be attri- buted. We are, however, still of opinion that a small mass of pyro- clastic rock occurs near the west end of the southern ridge, and a yet smaller one some distance east of it, as already described. These appear to pass imperceptibly into the normal rock, and thus to support our former view; but we can see that this difficulty in find- ing a division might arise if the lava had originally a rather smooth, slagey surface, and the ash consisted of fragments and powder of an identical rock. That this was really the case there is, as will presently be shown, good reason to believe. The differences in * SiO, Al,O, Fe,0, FeO CaO MgO Na,O K,O Total. ios. 22s. 71-44 10°54 381 223 533 2:95 193 0:84 99-07 No. Il. ...... 71:68 1039 409 2293 545 250 1:93 084 99-11 Only one analysis was made of the amount of Na,O and K,O; the loss by drying and ignition was not estimated. NORTH-WEST REGION OF CHARNWOOD FOREST. 3 microscopic structure between the Sharpley and Peldar rocks are not conspicuous. In the former, the dusty opacite is, as a rule, only very rarely replaced by the green silicate already mentioned, and is less regularly aggregated; so that the rock, under the microscope, has a more streaky or blotchy aspect instead of the peculiar spotted or speckled look of the other. Still this may be seen occasionally. Our old specimens have been repeatedly studied, and some new ones examined, with the result that we feel more confidence in attributing the occasional indications of a fragmental structure, which the rock exhibits, to mechanical movements subsequent to its consolidation—that is, to the cause which has produced its schis- tosity ; and we find, in its general structure, increased resemblances toa glassy lava, which has been subsequently devitrified. ‘Traces of pyroxenic minerals are rarer in the porphyroid of Sharpley than in that of Peldar Tor. But in some slides, brown, almost opaque, rather roughly shaped belonites, about :005" long, are fairly common. ‘These occasionally become almost translucent and eolour- less, perhaps owing to the conversion into chalybite of the colouring limonite. They are feebly anisotropic, and may possibly indicate the former presence of a pyroxene rich in iron. Larger grains of an iron oxide, somewhat decomposed, occur occasionally, and the eranular blackened mineral, described in the Peldar-Tor rock, is here much more rarely seen. The order of the phenomena in both rocks seems to have been the same, and may be summarized as follows :-— (1). Formation of quartzes, felspars (with a slight tendency to occur in groups of two or three), iren oxides, and other minerals (é. g-5 pyroxenic). (2). Partial corrosion of the quartzes and felspars, with occasional fracture of the former (at least). (3). Consolidation of the matrix, segregation of opacite, forma- tion of minor structures, (4). Production of a rude cleavage ; principal cracking of the larger included minerals. (5). Devitrification, &«.—Order of (4) and (5) uncertain; the latter doubtless a very prolonged process. An analysis of the Sharpley rock was made by Mr. Berry in 1882%*, and a partial one is given in our last papert. The large amount of Na,O compared with K,O in the former analysis appeared strange, for microscopic examination had indicated that a fair amount of the porphyritic felspar was orthoclase. At our request, Miss Aston kindly undertook to determine the amount of these constituents in another specimen; this gave Na,O=2:43 and K,0=2:18, which we believe better expresses t the normal composition. The percentage of S10, was high, about 77-3 +; Mr. Berry obtaining 67-6, the other observer 68-05. Probably the fragment recently analysed contained * Quart. Journ. Geol. Soc. vol. xxxviii. (1882) p. 199. t Ibid. vol. xxxvi. (1880) p. 342. ¢ One analysis gave SiO,=77°70 ; another, 77'8 84 REY. EDWIN HILL AND PROF. T. G. BONNEY ON THE a greater number than usual of the included quartz-grains. Unless a very large amount of the rock be pounded up, discrepancies of this kind are inevitable. About 70 per cent. of SiO, would probably be near the general average. Thus the rock represents an ancient dacite rather than a rhyolite, but, as is often the case in the Forest, is somewhat intermediate in character. The Sharpley rock, when fresh, allowing for the presence of the quartzes,; might have had a general resemblance to the older andesite of Krakatoa *. 4, Field Relations of the Peldar and Sharpley Rockst.—It is re- markable that the areas occupied by these two porphyroids are so constantly environed by agglomerates. With only one exception i, fragments will be found in the outcrops, nearest to any point of their boundaries, and the agglomerates in the spinneys east of the Peldar moorland are among the coarsest in the Forest. The agreement between two boundaries of this region and the usual directions of strike suggest that they may be in some way connected with surfaces of deposition. The Peldar rock also occu- ples a position which would agree with its being part of a stratified series, regularly overlying the Sharpley. But this view is nega- tived by the abrupt termination of the former against the agglome- rates and ashes of Ratchet Hill, which are exactly in the usual direction of strike ; it is limited as completely in the opposite direc- tion. If we define the two areas of Sharpley and Peldar rocks by straight lines drawn joining their outermost outcrops, we find the Sharpley rock covering a rude parallelogram §, about three times as long as it is broad, while the Peldar occupies a much ruder parallelogram of only one-third the size, touching along a small portion of its longer side the Sharpley area (see Map, facing p. 80). It is scarcely possible to account for the restriction of these rocks to such strictly limited and peculiarly-shaped areas by faulting. But these difficulties disappear if the rocks are regarded as two lava-flows. A rock which we found in a spinney (rudely trilobate in form) not quite one-third of a mile W.N.W. of Swanymote Rock, between the northern end of Cademan Wood and the road, may be mentioned here. It occurs near the outside of the southern end of the spinney, and can be traced fairly continuously over an area extending about 30 yards in a northerly direction, and perhaps 4 or © yards wide. Macroscopically, it is rather intermediate in cha- racter between the normal Sharpley and the “ purple porphyritic ” rock which is common in the agglomerates; that is, it is like the former, but the individual crystals are not quite so large. _ Under the microscope the felspars are similar, suggesting in their outline fracture or corrosion. They contain sometimes frequent enclosures, * Report of Krakatoa Committee of the Royal Society, plate iii. fig. 1. . t See Map, facing p. 80. : t Drybrook Wood (S. end), where Sharpley rock is seen within a few feet of ashy rocks. § It is also noteworthy that, corresponding to an indentation in the Sharpley boundary at Ratchet Hill, there is a similar indentation in the opposite bound- ary by Gun Hill, as if the parallelogram had been broken across and the parts displaced. NORTH-WEST REGION OF CHARNWOOD FOREST. 85 now yiridite, but suggestive of having been a glass. The quartzes, though fewer and smuller, are like those of Sharpley and Peldar: there are the occasional small clusters of epidote and the composite grains, darkened with opacite. The slide includes at one edge a small portion of a felted mass of small plagioclastic felspars with a little epidote and black iron oxide, which recalls the structure of the fragments included in the Peldar rock, and the matrix now resembles that of the latter, now that of the Sharpley rock—the two being rather “streaked” together, as if a slight fluidal structure were present. ‘The mode of occurrence and aspect of this rock in the field suggests that it is a dyke, and this view accords fairly with the microscopic structure. If so, it confirms our present interpre- tation of the typical Sharpley and Peldar rock. It is, we may add, the only case in which field evidence strongly suggested the presence of a dyke in this region, and this is certainly remarkable, seeing that agglomerates and ashes are so abundant. The outcrop called the Swanymote Rock*, at the north-western extremity of the Sharpley massz/, presents considerable difficulties. The mass may be roughly divided by a line running from rather N. of N.W. to rather S. of 8.E. The portion on.the eastern side con- tains fragments of a dull-purple porpbyritic rock, like that often seen in the neighbouring agglomerates. In that on the western side it is doubtful whether any of these fragments are present, but pieces of slate, sometimes quite 2 feet long, occur, generally green in colour, but in a few instances purple. Once or twice they are distinctly banded, and they have been bent, but the bending was apparently anterior to the production of a cleavage in the matrix, which has hardly produced any effect on them. The matrix of the massif is rather irregular in structure—quartzes and felspars abound. The rock sometimes appears identical with the normal porphyroid of Sharpley, but occasionally is more suggestive of a pyroclastic origin. An adjoining knoll, to the W., in parts resembles the purple por- phyritic rock just described as occurring in fragments, bvt in others contains numerous and large quartzes, and exhibits in its purple groundmass a curious mottling of a light grey colour, which is suggestive of a flow-brecciation. Possibly we may be here just on the edge of the: lava-flow, and the mass may be more or less a true pyroclastic rock. The outcrops between the Swanymote Rock and Cademan Wood are fairly compact ashy grits, one of which contains occasional quartzes and felspars. One feature of these volcanic materials, whether lavas, agglome- rates, or tuffs, is rather remarkable. ‘This is the absence in the larger masses of any very determinate characters. In those now considered to be lavas, the lath-like microliths, so common in ordi- nary trachytes, are never more than very imperfectly seen, and they are often wholly wanting. Fluidal, perlitic, and spherulitic structures have not yet been found. In the fragments in the agglo- * Referred to in former papers as being “ near the last letter of the word Swanymote” on the one-inch map. 86 REV. EDWIN HILL AND PROF. T. G. BONNEY ON THE merate, the first is not often suggested, and it is never conspicuous. A spherulitic structure, not indeed very distinct, has once occurred. and this is in a pebble, obtained only in 1890, from one of the bands of conglomerate at Hanging Rocks. The exterior of the fragments is seldom at all scoriaceous; often. especially in the larger, it is quite smooth. In the examination of some ten dozen slides of pyroclastic rock from Charnwood, Professor Bonney has never come across a fragment which was indubitably vesicular. This seems indicative of a rather general absence of water from the volcanic foci which supplied the materials *. , do. Bardon Quarry.—Durin¢g the past ten years the great quarry at Bardon Hill has been much enlarged. The owners have afforded us every facility on each of our visits; but in so busy a place minute examination is rather difficult, and we have never found ourselves able to be there when the workmen were absent. At the present time (1890) the quarry is divided into three stages, the lowest and smallest of which has been opened since 1880. It lies rather on the northern side of the common axis of the quarry. The points which were chiefly studied during our visits were— (i.) the order of occurrence of the rock masses, (ii.) their nature, whether indurated pyroclastic or somewhat altered igneous rocks. As regards (1.), we are now convinced that formerly we mistook the significance of the *“‘ shaly bands,” in regarding them as indi- cative of bedding. Further examination in the light of new know- ledge has satisfied us that these schistose beds are only ‘ crush- bands,” where the rock has yielded to exceptional pressure. This has produced a rude cleavage, on the surfaces of which a filmy micaceous mineral has been rather largely developed, probably at a subsequent period, by the percolation of watery. But, fortunately, this error does not very seriously affect our description of the pit =. We then regarded these bands as indicating a general dip of the beds to a point a little N. of N.N.E., but now, so far as we can trust the indications of succession, we consider the dip to be very nearly north. Gi.) At each visit we carefully recorded our impressions as to the nature of the rocks and their apparent succession. ‘To describe these in detail would be tedious, and perhaps needless; therefore we content ourselves with a summary of the results, requesting future visitors to remember that not a few of the data on which our con- clusions are founded have disappeared concurrently with the exten- sion of the excavations. To the south of the middle pit rises the knoll of breccia described in our papers for 1877-78. The matrix is a volcanic ash, containing fragments of slate, mostly purple, but sometimes greenish in colour, which vary in size, but are occasionally over 1 foot long, and in one * See Professor Judd’s suggestive remarks on the lavas of Krakatoa, Geol. Mag. dee. iii. vol. v. (1888) p. 684. t On the southern side, in 1889, we were able to examine a part where the rock had been less severely crushed, and found that it was really identical with the brecciated rock of that part of the pit. + Quart. Journ. Geol. Soc. vol. xxxii. (1877) p. 789. NORTH-WEST REGION OF CHARNWOOD FOREST. 87 case more than 2 feet. A rude cleavage affects the fragments as well as the matrix, dipping approximately at about 60° to a point a little N. of N.N.E. This mass, as indicated in our section’, underlies the rocks of the quarry. A little distance south of the edge of the same pit, and about 70 yards from this knoll, are (at present) outcrops of another rock more or less brecciated. In one reef just close to the edge of the pit-wall an agglomeratic character is very distinct, the fragments, by weathering, standing out from the matrix. Freshly broken surfaces exhibit a mottled structure, which somewhat reminded us of the Kite-Hill rock +. To this suc- ceeds a brecciated rock, now well exposed in the south wall and adjacent floor of the quarry ; fragments, sometimes 4 or 5 inches in diameter, of a speckled rock occurring in a dull greenish matrix, from which they are distinguished by the pinker, redder, or some- times slightly yellowish tint of their ground-colour. This rock seems to pass up—no hard-and-fast boundary being determinable— into the compact, green felstone-like rock, which was chiefly quarried in the older workings. Over this comes a brecciated rock, in which the apparent fragments are parted by a compact streaky matrix of a purple-red colour. To this succeeds another brecciated rock in which a green or yellowish-green colour predominates, which is followed by another breccia, pinker in hue. ‘This last seems to pass into a dull purplish rock, which is rather fissile, and resembles generally the Sharpley porphyroid, especially that variety which is exposed in the knoll on the moor near Spring Hill Farm, except that the quartzes and felspars are rather smaller in size. This porphyroid, in the upper part, has a very ashy look, and contains fragments which sometimes are numerous. Most of these are typical Peldar rock, but a few are a purple porphyritic felstone, and over this porphyroid comes the main mass of rock, identical with that of Peldar Tor. The nature of the last-named has been already discussed. If, then, it be a volcanic rock, we have to investigate the underlying porphy roid. ‘This rock appears to be rather variable in thickness, and to pass almost imperceptibly into the more or less brecciated greenish rock in which the greater part of the excavation is made. Was it originally a lava or an ash? If the former, the presence of fragments of Peldar porphyroid in its upper part must be explained by supposing it to have broken through a mass of that rock; but on that hypothesis it is a little difficult to account for the occasional association of the purple porphyritic felstone. The distribution also of the Peldar fragments in all the masses which we have examined, while not absolutely incompatible with the above explanation, cer- tainly accords better with the idea of a pyroclastic origin {. Micro- scopic examination of the matrix does not help us very much. The rock evidently has been subjected to considerable pressure. ‘The * Op. cit. p. 780. t The microscopic structure also is rather similar. t They may have been ejected from the vent which supplied the lava, and have fallen in advance of it—the Sharpley lava and tuff coming from a neigh- bouring vent. 88 REY. EDWIN HILL AND PROF. T. G. BONNEY ON THE usual filmy ‘“‘ sericite”’ has been produced. It contains crystals of felspar, sometimes well developed, sometimes rather rounded in outline or cracked. We find also the usual greenish, black-spotted minerals *, Of the two specimens examined microscopically, one closely resembles a specimen of Sharpley rock, except that quartz grains are practically absent; the other exhibits a more granular structure, rather suggestive of a fragmental origin, and in it aggre- gated patches of a granular mineral, giving rather rich tints of pink and green with the two nicols, are notuncommon ; these, in some cases at least, appear to replace felspar +. Subangular patches also of a brownish mineral are not rare, which has rather weak depolarizing action, and is speckled and bordered with dots of brown iron oxide. This in places seems to be composed of aggregated folia, resem- bling an altered mica, or more probably a chlorite. Their outline is not sufficiently definite to give any real clue to the mineral which has been replaced. Almost certainly it was a ferro-magnesian sili- cate, possibly biotite, but more probably a member of the pyroxene group. In the microscopic structure of the rock there is nothing incompatible with its having been a tuff, but there is nothing to prove it—no confused association of constituents, no bits of indu- bitable scoria ; in short, no definite structure can be detected. But taking all circumstances into consideration, especially the field evidence, a pyroclastic origin seems the more probable one; that, in any case, the rock is closely related to the porphyroid of Sharpley we think may be safely assumed. We pass on to the brecciated rocks, in which the quarries are chiefly opened. Here we seem to find almost insensible gradations from the compact felstone-like green rock to rather coarse breccias, which sometimes are also green (fragments and matrix differing slightly in tint), sometimes, as has been said, contain fragments of a colour more or less buff or pinkish. That the structure of these rocks has been to some extent modi- fied by subsequent pressure seems incontestable, but the effects of this do not generally appear to be very conspicuous. The purple- streaked breccia most resembles a crushed rock, but microscopic examination certainly does not negative the hypothesis of a pyro- clastic origin. This is favoured, by both microscopic and field evi- dence, in the case of the outcrops already mentioned to the south of the middle pit, and by some formerly seen east of the upper pit (at present,'we believe, quarried away), while the rocks near the summit of the hill, which now seem to be nearly on the same line of strike, are certainly agglomerates. Microscopic examination of the breccia from the pit itself has not helped us much. That the materials have an igneous origin is beyond doubt; that the whole, if this be a clastic deposit, is practically from the same source; that in it we have as yet failed to find bits of indubitable ash, which are common * See page 81. t I have little doubt that they belong to the zeolite group, but think it safer not to attempt to name them. ‘To do this would involve a long investigation, which, for my purpose, would be a waste of time.—'T. G. B. ‘a NORTH-WEST REGION OF CHARNWOOD FOREST, 89 in many of the volcanic breccias of Charnwood Forest, is cer- tain. The structure might be explained by flow-brecciation, but there is nothing to suggest this, and the absence of any fluidal structure in the matrix is opposed to this hypothesis. Though . indications of some mechanical disturbance can be perceived almost everywhere, and brecciation may be occasionally due to it, those usually significant of the crushing in situ of a large mass of rock are certainly not common. The “shaly bands,” of course, prove mechanical action ; but our examination of the less crushed portion (on the southern side of the pit) convinced us that the breccia had practically arrived at its present condition prior to the crushing. This appeared to be due to the same cause as that which produced the rough cleavage more or less perceptible throughout the mass. On the whole, after examining several carefully selected specimens, we still incline to the view of a pyroclastic origin for the main mass of Bardon Hill, while we would not exclude the possibility of some portions being small flows of true lava. The fact is, as we stated long ago, that the micro-mineralogical changes—the devitri- fication, and subsequent decomposition, the formation of viridite, epidote, &c.—have so “ blurred” the structures as to leave us always in a state of uncertainty as to the right interpretation *. We are not aware that any analysis of the Bardon rock has been published ; therefore the following, kindly made for us by Mr. Lord in the laboratory of University College (London), may be interesting. The specimen selected represents the compact green rock, without brecciation, which, as mentioned above, has the closest resemblance toafelstone. Hvidently, like the rocks of the Peldar-Sharpley region, it is rather intermediate in character, and nearer in composition to an andesite than to a sanidine-trachyte. It is rather more acid than the Markfield “syenite,” and less so than that from Croft Hill. l | | | | SiO, | Al,0,|Fe,0,; FeO | CaO | MgO} K,O | Na,O | Total. paths ssid gut: No. 1....[5986 {1600 | 447 | s64| 806} 390 | 120} 260 |oo73 No. IL....| 59:00 |1600 | 450 | 3-70} 8:00} 420] 1:36 | 1-84 | 98-60 Loss by drying and ignition not estimated. There have been shown to us some small portions of copper-ore found in quarrying, which contained apparently, with malachite and cuprite, a little native copper. en eae 118 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN therefore more susceptible of alteration, No. 543 * being within one yard from the Uriconian, and No. 544 about six yards distant. The former is a mudstone, the latter a grit; and the materials of both are largely volcanic. Neither Prof. Bonney nor myself could detect any signs of alteration or any traces of chiastolite. Indeed, Prof. Blake himself has omitted to give us any microscopic evidence on these points, though it is obvious that they are of critical importance. As this crucial case breaks down on examination, I thought it needless to re-examine the other masses of Uriconian which appear on the line of the great Pontesford-Linley fault. 2. Tur ALLEGED UNCONFORMITIES. Roughly speaking, the Longmynd rocks may be divided into am upper member, coloured purple, and a lower member, coloured green; though purple rocks occur in the lower part, and green rocks are occasionally found in the upper. By Sir R. Murchison and the Geological Survey the two members were united into an un- broken series, and this view has been generally accepted. In Haughmond Hill there is a marked break between the purple and the green beds, as I pointed out to Prof. Blake; but it can be demonstrated that this is due to faulting, and, from a general acquaintance with the Longmyndian rocks, I have not been led to question the received opinion. Prof. Blake is the first to dissent from the old view. He con- siders that the Longmynd Series is divisible into two groups sepa- rated by a marked unconformity. The upper member he correlates with some part of the Cambrian, the lower he places in his so-called “¢ Monian ” system. I contend, however, that Prof. Blake has not proved his case, and that, if he had done so, he would hardly have refuted my evidence from included fragments, since well-rounded pebbles of volcanic rock are found in the lower series as well as in the so-called “¢ Cambrian.” (a) The supposed Break i the Longmyndian.—The general sec- tion + offered by Prof. Blake appears on the face of it improbable. He draws the upper series as dipping at about the same angle and in the same direction as his so-called ‘‘ Monian”; but, just at the contact, the basement beds of the “Cambrian” are represented as creeping up over the edges of the ‘‘ Monian,” and lying to the east of the junction in outliers, horizontal ‘“‘ Cambrian” resting on nearly vertical ‘“‘Monian.” So extraordinary a section surely requires some explanation. In pursuance of my plan, I have carefully examined one of Prof. Blake’s critical sections, the “outlier” at Narnell’s Rock. The actual junction is figured by Prof. Blake on page 395 +, and the * So numbered in my cabinet. t Quart. Journ. Geol. Soc. vol. xlvi. (1890) p. 392, t Op. cit. THE BASAL ROCK-GROUPS OF SHROPSHIRE Lis grit (‘*‘ Cambrian”) is shown as squeezed into the slates (‘‘ Monian”) so as to be nearly conformable. ‘There is in truth no unconformity whatsoever. ‘The slate, it is true, is somewhat abruptly succeeded by the grit, which is very massive and shows little clear bedding ; but a conformable passage between the two is apparent without any very critical observation. The reading of the section is represented in fig. 1, below, and my note-book supplies the following par- ticulars :— Approaching the section from the east, we pass alternations of purple slates and grits. At Prof. Blake’s ‘‘ junction,” we have, on the east side, purple slaty beds interlaminated with seams of purple grit, both slates and grits being rather micaceous. On the west side is a bed of the grit, which appears to have been a little squeezed into the slate, so that here and there a trivial unconformity between Fig. 1.—Section at Narnell’s Rock. £.S8.E. WNLW. “agee8 \ . Junction of Prof. Blake’s “ Monian” with his ‘‘ Cambrian.” e, Conglomerate, with flakes of shale. s. Shaly and slaty beds. g. Grit. the two beds is apparent. A seam of slaty rock overlies the first grit-band. A few yards farther on is a thin bed of conglomerate. The included pebbles are largely of quartz, but there also occur numerous thin flakes of purple shale, and these lie flat in the plane of the seam. A little farther west there appears a band of thin- bedded grit, with abundant mica on the lamination-surfaces. These three beds—slate, conglomerate, and laminated grit—have the same dip (N.W. by W. at a high angle) as the underlying ‘“ Monian.” The alleged unconformity is thus absolutely disproved, so far as this section is concerned. All the grit in the crags to the west is apparently massive, and as there are no exposures in the slopes below the section, there is an appearance of what Prof. Blake calls a “‘ horizontal crag.” That this grit is not an outlier, but an integral part of a regular series, is further evident from the fact that it can be followed on the strike into Callow Hollow on the south, and into Lightspout Hollow on the north, to say nothing of more distant localities. | _ It seems to have been supposed * that the flakes of purple shale found in the upper series~are derived fragments from an earlier formation. This argument, if it were taken into consideration, * Op. cit. pp. 392, 393. Lll4 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN would prove too much; for I have found similar flakes low down in the great grits of the so-called ‘‘ Monian” on the southern slopes of the Ashes Hollow. I need hardly state that they are devoid of significance. | In pointing out the insufficiency of the above attempt to refute the received views, I do not wish to prejudge issues outside the present enquiry. I have not critically studied the whole of the Longmynd area, and must leave some points for future research. (b) Presence of Rolled Fragments of Volcanic Rock in the so-called ** Monian” Sertes.—These are not very common; but they exist. One good specimen I found in the green grit of Ashes Hollow. It is a well-rounded pebble of about one inch in diameter. The material is certainly volcanic. In the field it looked like felsite, and under the microscope it appeared to me to show a micro- crystalline structure. Prof. Bonney leans to the opinion that it is allied to obsidian, though he does not absolutely exclude the alter- native of a highly altered mudstone. As there are no lava-flows in the Longmynd Series, and, so far as I know, no mudstones resembling felsite, it would seem that such pebbles must be derived from an earlier formation. 3. Tue ConGLOMERATES AND GRIts. Prof. Blake truly remarks (op. ct. p. 408) that the age of these conglomerates and grits is “‘ of supreme importance in connection with the general interpretation of the district.” With equal accuracy, he adds that they have been taken by the present writer to be ‘“‘ part of the volcanic series.” He is not, however, correct in saying that I regard them as proving “the clastic origin of the latter.” It is true that a grit contains the evidence of its own ‘‘clastic origin”; but it certainly does not carry with it the genesis of any lava-flows that may be associated with it. Prof. Blake contends that these clastic rocks do not belong to the Uriconian, but are outlying patches of the ‘‘ Cambrian” strata which form the western half of the Longmynd Series. He mentions examples occurring on the Cardington massif, on Caer Caradoc, and on Charlton Hill. I will deal with these seriatem. (a) The Cardington Massif.—Certain patches of grit at or near the Gaer Stone are said to be “superficial,” but no evidence is offered in support of that opinion. One of the numerous patches that ie near Willstone Hill is noticed; but again its ‘“ Cambrian ” age is assumed. I have examined a large number of exposures of grit in the western part of the Cardington mass, but could obtain no evidence of unconformity. Farther east, however, there is clear proof in at least two localities that this grit is intercalated in the voleanic series. One mass occurs a little more than half a mile east of the Gaer Stone in the ravine which opens on to the Hope- Bowdler road, opposite ‘‘ The Yells.” This rock is of the usual type, as described by Prof. Bonney in an appendix to my 1879 paper *, * Quart. Journ. Geol. Soe. vol. xxxv. p. 607. THE BASAL ROCK-GROUPS OF SHROPSHIRE. L15 consisting mainly of quartz and red felspar. It varies in the degree of coarseness between microscopic grains and bits of 3 inch in diameter, and this character gives us the meaus of deter- mining the strike. At two spots on the western side of the glen, a clear strike may be made out by following the coarse seams, In one case the direction is EK, and W., in the other N.W. by W. The dip is vertical or very high. At the southern end of the section the erit is almost black, and this peculiar variety crops out on the opposite slope about 50 yards to E. by a littleS. The strikes agree therefore in a general way with the usual strikes of the volcanic rocks in the Cardington mass. See the accompanying map, facing « 120, : The other locality is in the quarries at Woodgate, where is exposed the clear series of rhyolites and grits described in my 1879 paper (op. cit. p. 658). These rocks are admitted by Prof. Blake to belong to the true volcanic group. I have recently detected at the back of this quarry, at the western extremity, and almost on the strike of the green grits, a reddish grit, composed of quartz and red felspar, with some bits of rhyolite. In hand-specimens it is seen to be quite of the ordinary Uriconian type. The annexed section (fig. 2) shows ~ the relation of the grits of these two localities to the associated volcanic rocks. Fig. 2.—Seetion across the Hope-Bowdler mass. N. 8. 7 OER RRS TH aD Vi ua ------—~~—-—- Quarry. -----—~—~—~— Woodgate. —— “ Wi, Ey Up: app. a TY g. Grit. r. Rhyolite. (b) Caer Caradoc.—The mass of grit near the south-western end of the ridge is regarded by Prof. Blake as a mere surface-patch. There is, however, very clear proof that it is intercalated in the Uriconian. In one place it is distinctly bedded. The reddish variety alternates in regular seams with a dark grit, with a shaly band, and with a fine-grained compact rock, like hilleflinta, but whether igneous or aqueous is uncertain. Prof. Bonney inclines to the former belief. These beds dip very clearly to the N.E. Both above and below numerous exposures of the ordinary hiilleflinta of the district are seen to crop out, with the same dip, or, where the dip is not apparent, with a strike to N.W. or NNW. (c) Charlton Hill.—Clearer evidence for the pre-Cambrian age of the grits and conglomerates could hardly be desired *. After Prof. * Quart. Journ. Geol. Soc. vol. xxxv. (1879) p. 655; vol. xl. (1886) p. 483. 116 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN Blake’s communication was read, I went several times over the ground and examined the critical sections foot by foot. Weathering and cart-wheels had exposed new outcrops, and the evidence I obtained strongly confirmed my published conclusions. The annexed section (fig. 3) illustrates the main points. I will take the section from N. to 8. Fig. 3.—Section across Charlton Hill. c. Conglomerate. | Z. Andesitic lava. 9. Grit. h. Halleflinta. p. Pebbly grit. g. Cambrian quartzite. pg. Grit and pebbly grit. The volcanic grits forming the northern part of the hill do not show any clear dip. The first rock that attracts attention, as we keep along the western side, is an andesitic lava, showing a flow-structure, which under the microscope becomes very distinct. The strike is about E. and W., and the dip nearly vertical. We next come to the well-known conglomerate on the top of the hill. It displays no clear dip or strike, but the majority of the pebbles have their longer axes trending N.W. by W. A few yards south of the strike of this conglomerate, in the surface of the road, a very instructive section has been exposed. Grit-bands alternate with finer compact material. One of the grits is of the ordinary Wrekin type, the fragments. being chiefly felsite. Separated from this by a compact seam is a coarser grit of which the fragments are mainly quartz, red felspar, and red granite, some of the larger pieces being well rounded. Small pebbles of mica-schist also occur in this band. ‘Thus, within 2 feet in the same solid mass, we have one bed which is distinctly Uriconian, and another of the type which Prof. Blake would probably call ‘‘Cambrian.” The strike of these beds is clearly N.W. by W., and therefore parallel to the conglomerate. ’ We next come to the well-known section seen in the northern bank of the hollow road to the S.E. of the last spot. Prof. Blake’s drawing * shows a mass of structureless hilleflinta with the surface hollowed into two depressions in which lhe small synclnes of ‘Cambrian grit.” He says that the grit, in two masses, dips “in certain parts towards the W., but becomes almost horizontal through disturbance.”” At only one spot could I find a westerly dip; but the rock was dolerite, and the sloping slabs are produced by joint-planes. Nor could I detect horizontal bedding ; but, since. the strike of the beds is nearly parallel to the section, the outcrops, viewed from a distance, appear nearly horizontal. Caution is there- fore necessary in working out the stratification. As we descend the * Quart. Journ. Geol. Soc. vol. xlvi. (1890) p. 409. THE BASAL ROCK-GROUPS OF SHROPSHIRE, 419 road, and examine the re-entering angles in the vertical rock-surface, we can make out the true dip, bed after bed, and measure a thick- ness of from 20 to 30 feet. The upper strata are strictly typical of the voleanic series, as it is exposed in Lawrence-Hill Quarry, at the foot of the Wrekin. Associated witb these grits and hornstones are bands which are even more distinctly sedimentary, consisting of a sort of claystone, with thin seams of small pebbles composed of materials substantially identical with the included fragments of the main conglomerate. At a recent visit, | was accompanied by my friend Dr. Geo. Deane, F.G.S., who, with compass and clinometer, ascertained that these beds dipped to the north at 70°—75°, a result which differs very slightly from my observation already recorded in the Quarterly Journal *. At nearly the bottom of the section, just before we reach the Cambrian quartzite, we come to another conglomerate, with large well-rounded pebbles of quartz, quartzite, granite, and various schists. In hand-specimens, the matrix looks like a slightly por- phyritic felsite. Prof. Bonney finds this matrix very difficult of determination; but he is disposed to regard most of it as clastic volcanic material. I confess I am not quite satisfied with this opinion, but the point is not very important. Further proof of the Uriconian age of the conglomerate is found in the small elevation to the 8. of Charlton Hill. Typical Uriconian ashes and grits, with an E. and W. strike, are exposed on the plateau. ‘Towards its southern margin, forming a small crag, there is an outcrop of the ash containing an irregular band of conglomerate, Sor 9 feet from E. to W., by 3 or 4 feet in breadth. Some of the pebbles are of unusual size, reaching a diameter of 5 or 6 inches.’ Hand-specimens are procurable, in which one or more rounded fragments of granite and other rocks are embedded in as typical an ash as any that can be found in the Wrekin area. I may add that in the Lawrence-Hill Quarry, and on the Wrekin itself, grits and conglomerates occur which differ only from those just described in the larger proportion of felsite-fragments. South of the Wrekin and Wrockwardine masses, the signs of purely vol- canic action decrease. They are less conspicuous even at Charlton Hill; but, in the Church-Stretton area, rhyolites are much less pro- minent and felsite fragments decrease in number, while distinctively sedimentary material increases in proportion. It is almost superfluous to point out that the proof of the Uriconian age of the conglomerates involves the pre-Uriconian age of the granite and metamorphic land-masses from which so many of the fragments were derived. Some of the granite-fragments, as Prof. Bonney and myself have shown’, are similar to the rock which occurs at the two opposite ends of the Wrekin chain. This state- ment has an obvious bearing upon Prof. Blake’s opinion that the granite of the Wrekin is intrusive in the volcanic series. * Vol. xlii. (1886) p. 483. t Quart. Journ. Geol. Soc. vol. xxxv. (1879) p. 654; vol. xlii. (1886) pp. 485-485. 118 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN 4, Turn Granitic Rocks. Prof. Blake appears to have discovered some new outcrops of the granite. He says “there are two small patches of it in the midst of the rhyolites near the south-eastern end of the Wrekin.” But, as the Wrekin is a narrow ridge trending N.E. and 8.W., it cannot have a “ south-eastern end,” and we are therefore left in the dark as to the situation of the new masses. Wherever they are, they must be very small. At Primrose Hill, the granite is thought by Prof. Blake to “suggest a neck.” He states that it ‘“‘almost seems to pass into . rhyolite.” I can find, however, no rhyolite near the granite. The rock into which the granite really passes is fine-grained and com- pact; but, under the microscope, it is seen to be merely a crushed state of the granite, as first suggested to me by Prof. Bonney. A . series of microscopic slides shows the actual stages of the crushing process. There remains the difficult and obscure section at Ercal Hill. Very little is to be learned in the field. At one spot I once saw a mass of felsite included in the granite; but all the rock was too much decomposed for satisfactory study. On the other hand, I have never found a vein of the granite in the felsite. The junction is nearly a straight line, and is marked by some minute crushing. There is some reason to suspect a fault. Since writing the above, I have received a note on the subject by Prof. Bonney. He has examined a new junction-specimen of mine, and re-examined several slides of his own. He says, “‘ Certainly there is nothing to show that the granite is intrusive in the felsite,” and he thinks there are signs of fault-brecciation. He cannot find any distinct proof of the intrusion of the felsite in the granite. We may then fairly conclude that Prof. Blake’s opinion receives no support from the study of the rocks, either in the field or under the microscope, and the evidence from included fragments, given in the last section, may be allowed its due weight. The pebbles of granite and schist in the Charlton-Hill conglome- rates clearly prove the unconformity between the Uriconian and an earlier rock-system. The nature of the break between the Uri- conian and the Longmyndian will be more fully explained in Part III. Ill. Tur Rerarion BETWEEN THE URICONIAN AND THE LoNGMYNDIAN. The age of the Longmynd rocks must be considered as sub gudice. The antique facies of the fauna of the Hollybush Sandstone and the gap between that formation and the Upper Cambrian led me many ~ years ago to doubt the Upper-Cambrian age of the former, and I have provisionally regarded it as Menevian. The obvious hiatus between the quartzite underlying this sandstone and the Longmynd Series rendered the Cambrian age of the latter highly improbable, THE BASAL ROCK-GROUPS OF SHROPSHIRE. 119 and I thought it convenient in 1887 * to give to this great group the local designation of ‘“ Longmyndian.” Prof. Lapworth has since announced the discovery of fossil evidence for the Lower- Cambrian age of the Hollybush Sandstone; but, pending the pub- lication of his details, it would hardly be wise to positively assign a pre-Cambrian age to the Longmyndian system. Our present object is to ascertain if there is a time-break between the Uriconian and Longmyndian, and, if so, whether it is a small or a great one. That there is a break will, I think, appear from the following considerations. 1. Tue DiscorDANCE OF STRIKE BETWEEN THE TWO GROUPS. It is well known that the normal strike of the Longmynd rocks is N.N.E. Ihave made observations in hundreds of localities, from Haughmond Hill on the north to a newly discovered inlier on the borders of Herefordshire on the south; and I have found that, though there are occasionally slight deviations and abrupt twists, the normal strike, even in faulted masses, is maintained with re- markable uniformity. The strikes in the map which faces the next page have all been personally verified. On the other hand, as I pointed out in 1879, the normal strike of the Uriconian is more or less transverse to the above. Ina volcanic series, we should not expect the strikes to be uniformly parallel or persistent for great distances; nevertheless, it will be found that the exceptions to the above rule are not very numerous. I have recently reviewed the original evidence with great care, and have discovered new strikes in all the chief masses. The enlarged evi- dence is summarized in the map, which represents the minimum number of strikes observed, but where several occurred in the same vicinity only one is usually drawn. I have been careful to supply nothing from the imagination. Even when a curve in the strike appeared to be fairly deducible from several disjointed exposures, I have not connected the broken lines, unless they were so near together that it would be mere pedantry to refuse to join them. I will now take the chief areas in order. (a) Lilleshall Hill.—Strikes are clearly seen almost from end to end of the ridge. Some of them are in rhyolite, others in slaty ash-beds. They vary between E. and W., and N.E. by E. (b) Lhe Wrekin Chain.—A strike, originally mapped as occurring in the so-called “ granitoidite” of the Ercal, must be erased. The rock is now regarded as a true granite, and its band of fragments is probably a crush-breccia. ‘The transversal strikes in the rhyolite at the southern end of the Ercal, as also those in the hornstones, ashes, and agglomerates of Lawrence Hill and the northern end of the Wrekin, are confirmed. Ihave moreover made out a number of new strikes in rhyolite along the N.W. side of the Wrekin nearly as far south as the summit, and I find that most of them trend E.N.E. ; * Trans. Shropsh. Archzxol. Soc. for 1887. 120 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN that is, parallel to the overlying grits. The strikes in the Wrekin thus agree with those of Lilleshall Hill. (c) Zhe Wrockwardine Mass.—The rock is mainly rhyolite. There is great variation in the strikes, and they are frequently curved, as we should expect. There is, however, a predominance of what we may call “ Longmyndian strikes,” such as N. and §., or N.N.E. and S.S.W. Lava-flows are of course less reliable as indications than clastic beds. At the edge of the area, near Leaton, is a small section of hornstone and volcanic grit, with a high westerly dip. ‘This is the only case in the Wrekin area, so far as I know, in witch clastic rocks have a Longmyndian strike. (d) Charlton Hill—The numerous clear strikes visible in this area are all, with one exception, in clastic beds. The andesitic lava, the conglomerates, grits, hornstones, and argillites have a steady dip to the north, or a little east of north. The strike in the lava as well as several very distinct strikes in grit and conglomerate — have been recently discovered, and they agree with my old results. (e) Fhe Lawley.—I have found here but one exposure of sedi- mentary rock with a distinct strike, and that is about E. and W. Prot. Blake remarks that some of the rhyolitic-looking rocks in the Church-Stretton district, when microscopically examined, turn out to be slates. I can confirm this statement, though I would rather describe some of these fine-grained compact rocks as argillite, or even hornstone. In the field it is sometimes impossible to dis- tinguish between a sedimentary deposit and a true lava-flow. Prof. Bonney has examined for me several varieties of compact rock from Caer Caradoc and the hills to the south. He identifies the majority of them as a sort of volcanic mud, one or two as probably igneous, and one or two as doubtful. True lava-flows occur in the middle of Caer Caradoc, in the Hope-Bowdler district, and perhaps elsewhere ; but recent investigations demonstrate that sedimentary material forms a larger proportion of the Uriconian rocks of the Church- Stretton area than we had supposed. This will make the strikes of the region more reliable for our present purpose. Where IJ have not been able to identify a hard, compact, flinty rock as either igneous or sedimentary, I have found it convenient to employ the vague and, I think, useful term “hilleflinta.’’ The strikes in these rocks are usually indicated by a coloured banding or by a platy structure. (f) Caer Caradoc.—The strikes are numerous andclear. Many of them are in halleflinta, some are in rhyolite, and at the 8.W. end there are very good strikes in grit and shaly mudstone. It will be seen from the map that the prevailing strikes are almost at a right angle to the strike of the Longmyndian, which here comes tlose up to the Uriconian ; so that on the western side of the hill the discordance of strike may be seen within a couple of hundred yards, the green Longmyndian slates striking N.N.E., while the beds of halleflinta and grit on the slopes above trend to the N.W. Some of the Uri- conian strikes are seen to form curves, and one of them comes out very clearly. A band of hilleflinta strikes across the ridge in a south-easterly direction, then it bends round to the E., then turns’ [Zo face p. 120. LILLESHALL ae Prot OUTLINE MAP OF THE ONIAN ano LONGMYNDIAN DCKS oF SHROPSHIRE. (ES IN THE URICONIAN. HYOLITE & MUDSTONE. Rica, « GCONGLOMERAT ES .------+<------" ES IN THE LONGMYNDIAN.------ LINES OF FAULT ---------- e SCALE OF MILES. 5 7. Ce = eure v4 i f Q. J. GS. vol. xlvii.} [Lo face p. 120. HAUGHMOND ait tee LILLESHALL 77 HILL eae 2 42 7 | ro. - a - ee _ 7 ennoy & : N. es Hj Y . - PONTESFORD ve 5 HILL acs rirenronoffinuien Win? y 7 7 7 co \ wa (ed wy : vy D LITTLE 2 éhRaboc OUTLINE MAP (CAER CARADOC OF THE URICONIAN ann LONGMYNDIAN ROCKS oF SHROPSHIRE. STRIKES IN THE URICONIAN. RHYOLITE & MUDSTONE. GRITS & CONGLOMERATES.-- STRIKES IN THE LONGMYNDIAN.----- - LINES OF FAULT ------~-- - SCALE OF MILES. — ° 5 10 eee ESS) THE BASAL ROCK-GROUPS OF SHROPSHIRE. 121 and keeps to the N. for some distance; finally it curves abruptly, and runs straight to the N.W., nearly (or quite) reaching the fault. Thus is formed an oblique parallelogram, whose north-western side is wanting and whose eastern angles are rounded off. All the dips slope inward, and the centre of the figure is occupied largely by dolerite. South of this figure the dips are to the N.E. and N. of it; while in the centre of the hill the dips are very variable. (g) Helmeth Hill—There are no good exposures, except along the summit-ridge, which is mostly occupied by grit and hiilleflinta. A red band in hiilleflinta displays a north-westerly strike, but I do not like to rely upon it and have not inserted it in the map. (h) Hazler Hill.—At the northern end, a banded argillite strikes N. and 8. This Longmyndian strike occurs within 70 yards of the true Longmynd slate with the normal trend. On the eastern side of the hill, a small quarry exposes a good section of volcanic beds, dipping to the N.W. In ascending order, the strata are purple fine-grained ash, pale green porcellanite, coarse grey ash, and fine-grained grey ash. ‘The beds are very irregular, and thin out like wedges. This is one of the most Pebidian-like sections that I have seen in Shrop- shire. (i) Ragleth Hill—The rocks forming the mass of this hill are very sedimentary, consisting mainly of argillite and grit, with some hiilleflinta. I have not found in it any igneous rock except intrusives. At both the north and south ends the strikes are about N.W. by W.., but in both cases the beds at their eastern ends curve round and keep along the eastern side of the ridge with a N.N.E. strike. We have thus another figure similar to the imperfect parallelogram in Caer Caradoc, the north-western side, as in that case, being absent,. and the dips being inward. (k) The Cardington and Hope-Bowdler Mass.—The materials of this group of hills are like those of Caer Caradoc, with certain differ- ences. Coarse grits are abundant towards the N.W. part; felspar- porphyry and rhyolite predominate towards the 8. and8.W. Strikes are recorded on the map from nine principal localities. Near the Gaer Stone, one strike is E. and W., another is to the N.W. North- east of Hope Bowdler a conspicuous band of felspar-porphyry trends east and west. Just at the back of the village this rock is well exposed in a farm-road, and the strike is quite clear, bands with numerous porphyritic crystals alternating with more compact seams. The strikes in grit in the ravine opposite ‘The Yells” and in Woodgate Quarry have been already noticed. In the glen above Woodgate, a thick band of rhyolite strikes parallel to the grits (E. and W.), and the lavas and ashes in the quarry run in the same direction. South-west of ‘* Middle-Hill” Farm, hilletlintas at several spots have a strike to W.N.W., in others to N.W. Towards the eastern end of the massif, at Stone Acton, there have been noted in hiilleflinta one strike to E.N.E. and one with an E. and W. trend. Some of these strikes are, for volcanic rocks, of considerable length. ‘Thus, the green grit of Woodgate Quarry reappears on the 122 DR. CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN strike to the east at a distance of nearly a quarter of a mile. Also the felspar-porphyry of Hope Bowdler runs E. and W. for about half a mile. Besides calling attention to the occurrence of structural strikes, it is of importance to point out the trend of large masses. Thus, the rhyolite in the Woodgate glen appears to strike almost E. and W. across the two ravines tio the west, and to be produced less distinctly in the opposite direction to Stone Acton, so that it is probably con- tinuous from end to end of the Cardington massif. Also the strike of the grit in the glen opposite “‘ The Yells ” is in a line connecting masses ‘of grit at ‘Gaer Stone with others in the Woodgate ravine, and this line is parallel to the strike of the adjacent rhyolite. On the other hand, I have not observed that any bedded or banded rock in this district trends in a direction transverse to the prevailing strikes. Summarizing the details of all the districts, we notice that the strikes in the Wrekin area are usually either E. and W. or E.N.E. and W.S.W.; in the southern area, they are E. and W. or some point between W. and N.W. The only strikes in the Uriconian which lie parallel to the normal Longmyndian strike are those near Wrock- wardine, at the northern end of Hazler Hill, and on Little Caradoc and the last-named is really a little to the west of north. The bands in Caer Caradoc and Ragleth which trend to the north can hardly be reckoned, for they curve round at each end, and abut on the fault at nearly a right angle to the Longmyndian strike. It will be observed that all the Lonemyndian strikes in Uriconian rocks are near the fault. Also that, in such important masses as Charlton Hill and Caer Caradoc, the transverse Uriconian strike comes close up to the fault. It is hardly rash to assume that the strikes at a distance from the great fault are least likely to have been disturbed, and are therefore most reliable as indications of the original lie of the rocks. 2. Tue Favrrep Juncrion. The two formations come into visible contact along a line of only about 23 miles. The fault drawn on the Survey map runs in a straight “line from the western margin of Caer Caradoc to Little Stretton, throwing down Silurian (Wenlock) rocks against Ordo- vician and older systems. The area east of the fault is marked “b*” (Caradoc) on the map, but it is correctly recognized by Prof. Blake as Longmyndian. It is this mass which abuts upon the voleanic series. The contact is certainly a fault at the northern end, near Caradoc Coppice. The dislocation can be taken up again about a mile to the south, near the cluster of cottages called Hazler. An interesting section of the junction-rocks is seen on the old road up from Church Stretton to this locality. The ordinary green slates of the Longmyndian crop out at intervals up the road tor nearly half a mile, with the normal strike to N.N.E. As we approach the hamlet, the slate loses its lamination, and, near a well, it comes to an abruptend. At this point it has a burnt appearance, THE BASAL ROCK-GROUPS OF SHROPSHIRE. 193 and the cracks are injected with red felspar. No igneous rocks are seen near, but the alteration-effects just described are similar to those observed at the contact of slate’ with dolerite east of Ragleth Hill (p. 111). About 70 yards farther on, we come to slaty and gritty rocks of a Uriconian appearance, which obviously could not have produced the alteration. The most probable explanation of the facts is a fault, with an intrusion of the dolerite which occurs in force on the same line of fault farther south. There is very little discordance of strike between the two systems at this locality. This fault does not, of course, prove an unconformity ; but it helps to invalidate the hypothesis of conformity. 3. DIFFERENCES IN THE ConpriTIons oF Deposit. The Uriconian is essentially a volcanic formation. In the Wrekin area, the rhyolites and felsite-grits form the chief mass ; beds with rounded fragments being very inconspicuous. In the Church-Stretton district, sedimentary matter, chiefly in the form of volcanic mud, is more prominent. But, throughout the Uriconian area, the comparative rarity of clear stratification is a marked feature. The ashes and hornstone of Lawrence-Hill Quarry, taken as the typical Uriconian section, were described by Sir R. Murchison and the Survey as intrusive “greenstone.” ven the clear sedi- mentation of Charlton Hill was overlooked, and the mass appears on the map as “greenstone.” Large parts of the Wrekin have been subjected to close and repeated examinations without yielding any evidence of bedding. The conglomerates of the Wrekin area can hardly be called beds at all; they are mere irregular patches, some- what elongated in the direction of the strike of the ash-beds with - which they are associated, and in a greater or less degree they shade off into the finer materials. In the Uriconian hills of Church Stretton, although I have observed grits in thirty or forty different localities, in only four or five of them has a strike been detected, and in only two of these cases could it be proved to extend for many yards. The strikes are more per- sistent in the hilleflinta, but a glance at the map reveals how rarely they can be followed for great distances. ‘he irregularity in the shape of the masses of grit in this area is a marked feature. One example will serve as a type. It occurs at Hazler, and forms part of the Uriconian rock lying east of the fault (p. 122). When I first approached this spot, I saw what looked like a granitic rock intruding into slates. The seeming granite was without trace of bedding or lamination. It cropped up in the road in irregular masses amidst laminated slaty rocks; on the northern bank it rose up into the slate in a somewhat dome-like form, and on the southern side it also appeared irregularly. A slight examination with a lens dispelled the illusion, and proved that the rock was a quartz-felspar grit. These irregular lumps of grit are explicable only as the result of direct volcanic action. Contrast the irregular arrangement of the volcanic ejectamenta of 124. DR, CH. CALLAWAY ON THE UNCONFORMITIES BETWEEN the Uriconian area with the even sedimentation of the Longmynd Series. The uniformity of the strikes and the regularity of the bedding in the latter are too well known to need comment. Even conglomerates, which hardly form beds at all in the volcanic series, can be traced on the strike mile after mile almost from end to end of the Longmynd chain. This change from almost pure vulcanism — to pure sedimentation must surely indicate a break in time. The unconformity between the Uriconian and the Longmyndian is not necessarily a very great one. In so early a period of the earth’s history, when the crust was more easily bent and broken, and in a region of energetic vulcanism, new axes of upheaval might be formed with comparative rapidity. Then, too, it must be admitted that the grits and slates of the Longmyndian do not materially * differ in their composition from the grits and mud- stones of the older series. Nevertheless, the unconformity appears to be real, and I submit that the terms ‘“ Uriconian” and ‘ Long- myndian” are worthy of a place in our nomenclature. | Discussion. Prof. Buaxxe had not regarded the volcanic rocks as zntrusive, but as having burst out from amongst the slates. -Any evidence de- rivable from Lyd’s Hole he had shown to be immaterial. The main point in his (the speaker’s) paper was the subdivision of the Longmynd rocks into two great groups by an unconformable over- lap betwixt the Upper and Lower groups—the similarity of strike being due to subsequent pressure. No satisfactory conclusion could be derived from the examination of a single section such as that at Narnell’s Rock; but his assertion of an unconformable overlap had been arrived at after tracing the junction across the district. He had seen fragments in the Lower group hke the specimen exhi- bited, and admitted that they were puzzling, as he had mentioned in his paper. It was not of primary importance whether the conglo- merates on Charlton Hill were superficial or not, for he regarded the whole as lying atthe very base of the Cambrian. He had no doubt that there was plenty of crushed granite amongst the rocks on Primrose Hill. He was not much concerned in the relations of the Ercal-Hill Red Rock, but would like to know where the rhyolite was entirely surrounded by it. He would ask, if the whole of the Longmynd rocks were to be called Longmyndian, where was the Cambrian of the western part of the district ? Prof. Bonney had seen the Charlton-Hill section, and thought that the evidence was in favour of Dr. Callaway’s views, and he might say the same of the Ercal-Hill section. The so-called “ chiastolite rock ” had been examined by him without finding any sign or pro- bability of the existence of chiastolite. Dr. Hicks, so far as his examination of the district had gone, was” inclined to follow the Author rather than Prof. Blake. He main- * The chief difference is that mica is abundant in the newer series and rare in the older: THE BASAL ROCK-GROUPS OF SHROPSHIRE, 125 tained that the source of supply of the Longmynd Series was the Caer Caradoc voleanic group. ‘There was no indication of a volcanic group in the Longmynd rocks, and it was much more likely that the volcanic rocks existed previously to their deposition. He as yet saw no reason for separating the Longmynd Series from the Cambrians. The Avrnor had difficulty in understanding Prof. Blake’s theory of evirusion as distinguished from intrusion, The rocks at Lyd’s Hole were important, because it was the only locality where Prof. Blake aftirmed the existence of contact-alterations. With regard to the apparent conformity produced by squeezing, he would ask why the outliers were not also conformable. He could not follow Prof. Blake’s arguments concerning the conglomerates on Charlton Hill; because the outcrops were circumscribed, it did not follow that the patches were Cambrian, and, as a matter of fact, the rocks have a dip and strike conformable with that of the associated rhyolites. The Cambrian occurs on the west side of the Longmynd, under the Stiper Stones; for the Shineton shales are seen there, and there is also a fault, cutting out the Lower-Cambrian rocks. Q.J.G.S. No. 186. K | 126 PROF, PRESTWICH ON THE AGE, FORMATION, AND 11. On the Ace, Formation, and Svuccrsstve Drirr-Sraces of the Vauiny of the Darent; with Remarks on the PALmOLITHIC Imritements of the Disrrict, and on the Oriein of tts CoaLk Escarpment. By Josep Presrwicu, D.C.L., F.R.S., F.G.8., &e. (Read January 21, 1891.) [Puatzs, Vi., Vil. & VILL] Contents. Page § 1. General Character and Age of the Darent Valley ...........0...++. 126 2. The Chalk-Plateau Drifts and the associated Flint Implements. 128 3. The Initial Stages of the Darent Valley’...............cssse-r«canseun 135 4. The High-Level or Limpsfield-Gravel Stage ...........sscsceeeeeees 137 5. Contemporaneous Drift in the Cray Valley .........c0..ssseeneees 144 G. The Brick-earths of the Darent Valley .......0500-s000sssovesswansen 145 7. Other Drifts of the Darent Valley : the Chevening and Dunton- | Green Gravel: n,n. cccss sccccrecasneo negates sne-scasteee ee ae 147 8.-The Low-Level WValley-Gravels. 0 ..0..:50. 52 .2020.4 scctaeesccoan enema 151 9. The Rubble on the Sides and in the Bed of the Valley............ 154 10. The Alluvium and the associated Neolithic Implements ......... 156 11. On the Chalk Escarpment within the Darent District ............ 156 § 1. Generat CHaracTER AND AcE or THE Darent VALLEY *, In former papers f I have touched incidentally upon the drift phe- nomena of this district, and on the occurrence of a peculiar group of flint implements found on the adjacent Chalk plateau. I now purpose to limit my observations to the circumscribed valley of the Darent, which I have bad more special opportunities of studying since my residence at Shoreham. This valley, including the district surrounding it, is of peculiar interest, from the circumstance that its geological history, beginning with pre-Glacial times, may, with few breaks, be traced to Neo- lithic times; as also from the light it throws upon the age of some of the Thames-Valley drifts, and from its distinctive groups of . Paleolithic implements. It is moreover free from the complication produced in the valleys north of the Thames by the presence of foreign-drift elements, for here the drift is restricted to débris derived from its own drainage-area. The Darent Valley is one of the few which run through the Chalk escarpment into the so-called Wealden area §, though it does * A general account of the drift-beds and denudation of this valley is given by Mr. Topley in his ‘Geology of the Weald,’ pp. 188-194, and 270, in Mem. Geol. Survey (1875), to which I shall often have occasion to refer. See also Messrs. Le Neve Foster and W. Topley’s ‘ Superficial Deposits of the Valley of the Medway, etc.,’ Quart. Journ. Geol. Soc. vol. xxi. (1865) pp. 443-474, and the Maps of the Geological Survey. t Quart. Journ. Geol. Soc. vols. xlv. (1889) p. 270, and xlvi. (1890) p. 155. t Another of these valleys, that of the Wey, was described in 1851 by the late R. A. C. Godwin-Austen in Quart. Journ. Geol. Soc. vol. vii. p. 278. § Taking the Wealden area to mean physiographically the whole of the area encircled by the escarpment of the Chalk. DRIFI-STAGES OF THE DARENT VALLEY, Mie 9 not pass beyond the first outworks, being shut out by the range of the Lower Greensand from the central Weald. To the south of the Chalk range, the valley branches westward in the line of tho main stream to near Limpsfield, and eastward to near Ightham in directions parallel with the ranges of the Lower Greensand and the Chalk, and is terminated by watersheds which separate it in the one case from the Oxted stream, and in the other from the Ightham stream (the Shode), both of which run from the foot of the Chalk hills, and flow into the central or Medway drainage-area of the Weald. ‘The valley is thus isolated, and its basin is of very limited extent, though at one time it would appear to have been larger, in consequence of the greater importance of the affluents from the Tertiary area (see Map, Pl. VII.). The first indent of the Darent Valley was, for the reasons given in two papers referred to in the last page, clearly subsequent to the deposition of the Lenham Sands, which are of Pliocene age, of the Red Clay-with-flints, and of the Southern Drift, while it com- menced with the general great denudation of the Weald. It is there- fore of late pre-Glacial or very early Glacial date *. At the former of these periods, the great valley separating the Chalk and Lower- Greensand ranges of hills was still bridged over by the Chalk and overlying strata, and it is to the denudation of these latter that both valley and escarpment are due (PI. VI., figs. 1, 2, 3). As the Lenham Sands are only of local occurrence, our object will be best answered by taking the Red Clay of the Chalk Plateau, with its sprinkling of Southern Drift, as our base-line. Without at present going into the question of the origin of the Red Clay- with-flints +, beyond mentioning that it is of local derivation, [ may ~ state that it is newer than the Tertiary strata, the outliers of which it encircles, while it seems to be older than the Southern Drift, with which it is closely associated. Besides the main valleys of the Medway and Darent, the Chalk Plateau, with its “‘ Red Clay,” is intersected by a system of lesser valleys, which, starting near the crest of the escarpment, run north- wards into the main valley of the Thames. These valleys commence on the Red Clay in very slight deflections on the surface, which rapidly increase in depth, and enlarge into the deeper valleys above which the Red Clay is left high on the adjacent plateau. These valleys, therefore, like the larger ones before named, are posterior in time to the Red Clay, as well as to the implement-bearing old drift with which the latter is associated. ‘The difference of level between this older drift and the drifts of these other valleys, though * In further proof of the sub-glacial action before noticed at the time of the Southern Drift (Quart. Journ. Geol. Soe. vol. xlvi. (1890) pp. 157, 174), I should mention that Mr. B. Harrison has since found on the summit of the Chalk escarpment at Terry’s Hill above Wrotham, and at a height of 770 feet, some small angular boulders (‘as large as quart measures’) of the Oldbury chert, mate ate smaller blocks of Iron-sandstone from the Lower-Greensand range to the south. + For an account of this Red Clay, see Mr. W. Whitaker’s ‘Geology of London,’ vol. i. chap, xviii. 1889. K 2 128 PROF. PRESTWICH ON THE AGE, FORMATION, AND necessarily small at the head of those valleys, equally indicates, as when the difference becomes greater, a marked difference of age. It will be our object to show the relation of both these valley- systems, with their drift-beds, to the Chalk-Plateau drift, and also to the Glacial and post-Glacial drifts of the Thames Valley *. § 2. Tue Cuarx-Piateav DRIFIS AND THE ASSOCIATED Fuint IvpLements. Since the publication of my Ightham paper, Mr. Harrison has traced the rude Palzolithic implements of the Chalk Plateau to West Yoke, 1 mile N.W. of Ash, and very near to the line where the Red Clay with flints ends abruptly on the brow of the hill over- looking the lower plain of bare chalk, which, except where the Swanscombe Hills intervene, extends to the Thames (Pl. VL, fig. 2). The relation that the Red Clay with the associated Palzolithic im- plements here bears to the adjacent valleys is shown in the following section (fig. 1):— Fig. 1.— Transverse section of the West-Yoke plateau. wm. E. Speedgate Fawkham West Longfield Hill. Valley. Yoke. Valley. 490. 270. 460. 280. es a. Red Clay-with-flints. e. Low-level valley-gravel. 2. Chalk. v- Site of Palzolithic implements of the Plateau-type. (The vertical scale of all the general sections in the text, unless mentioned otherwise, is jy inch = 100 feet. The base-line represents the sea-level, and the figures which give the height above it are taken from the six- inch Ordnance maps, which are indispensable in work of this sort.) The valleys on either side of the plateau at West Yoke are about 180 feet deep, while the small central depression (*), which eventually joins and belongs to the same valley-system, is here in an incipient state (10 to 20 feet in depth), showing that the mere question of level is not always conclusive in determining the relative antiquity of these drifts. The great antiquity of the plateau-drifts can, however, be better realised by the N. and S. section (Pl. VL., fig. 2), which extends from the Lower-Greensand hills to the Thames, and shows the _* Tuse these terms for convenience, meaning to embrace the whole of the. cold period from the earliest pre-Glacial to the latest post-Glacial times. The pre-Glacial, Glacial, and post-Glacial cycles pass one into another in a continuous series marked only by different and fluctuating degrees of intensity: of cold. The term ‘ post-Glacial’ conveys an incorrect meaning. ‘ Later- Glacial’ would be a better term. DRIFT-STAGES OF THE DARENT VALLEY. 129 relation that these drifts bear to the river-drifts of the Thames Valley. This section passes through the summit-level of the Swanscombe Hills, which are there capped by Tertiary strata and an outlier of the older drift. Though the height of this hill does not much exceed 300 feet, it corresponds with the level that the gradient of the plateau at West Yoke and Ash should have, if extended thus far. North of this hill, at Milton Street, near the village of Swanscombe, and at a level here 200 feet lower than the plateau-drift, the high- level river-drift of the Thames Valley is met with. It contains flint implements of a distinct and more advanced type than those of Ash and West Yoke, while at a lower level still are brick-earths and gravel with Mammalian remains and implements of a yet later period. This is, I conceive, conclusive of the great antiquity of the Chalk-Plateau drift and implements, and if we are to assume, as there is every reason to suppose, that the great denudation of the valleys has been the work of Glacial times, then these implements may prebably be assigned, as I have before suggested, to a pre- Glacial or early Glacial period. The plateau which constitutes the table-land west of the lower Darent Valley presents features precisely similar to those at Ash, Bower Lane, and other places on the plateau east of the Darent Valley. There is the same spread of Red Clay-with-flints over all the Chalk Plateau, and the same slight sprinkling in places of a drift of much worn brown-stained flints, with a few subangular fragments of chert and ragstone from the Lower Greensand *. I have found this drifé on the hills just above Shoreham. Chert and ragstone are particularly abundant in the field over the railway tunnel opposite Colegates Farm. They oceur less abundantly around Halstead, and have been found by Mr. Harrison en the very summit of the escarpment, at a height of 700 feet, on Morant’s Court Hill (see Pl. VL., fig. 3). Farther west, Mr. De B. Crawshay has found the brown-stained worn flints on Betsom Hill (790 feet) above Westerham, and on Titsey Hill (864 feet) above Limpsfield, both being on the crest, and forming the highest summit-levels of the Chalk escarpment. The inter- mediate ground between Morant’s Court Hill and Betsom Hill has at present yielded no specimens, though the Red Clay-with-flints is continuous throughout. On the hill above Stonehouse, north of Halstead, I have found a considerable proportion of the brown- stained flints with numerous Tertiary flint-pebbles, some Tertiary sandstones, and a little Lower-Greensand débris. At the time my Ightham paper was read, the only Paleolithic fliunt-implement known on this western plateau was the one at Currie Farm, south of Halstead +, found 20 years ago and described by Dr. John Evans. Its surroundings and position were such as to lead me to group it with the Ash specimens as of early Glacial or pre-Glacial age. My friend Dr. Evans, however, considered that * T include any Lower-Greensand débris, such as grit and ironstone. t+ The Rev. R. Ashington Bullen, the Vicar of Shoreham, has recently found a very similar specimen in the same field. 130 PROF. PRESTWICH ON THE AGE, FORMATION, AND although it was found at the high elevation of nearly 600 feet, the position of the site above the extreme head of the valley of the Cray was so slight that this specimen might belong to the later or post-Glacial * drift of that valley, and not to the older level t> which I would assign it. The lines of drainage of the Cray Valley being also from south to north confirmed him in this opiniont. If such a view could be sustained, it might invalidate the antiquity of the Currie-Farm specimen, and by inference the antiquity of those of the Ash district. But though it is true that both drifts are due to currents from the south, the one system of drainage which extended from the central Weald was in existence before the exca- vation of the Holmesdale Valley, whilst the other (the present Cray and its tributaries) dates from a period subsequent to the severance of the Chalk Plateau from the Lower-Greensand hills. We now, however, have more decisive corroborative evidence of the age of the Currie-Farm specimen. Some time elapsed before any new locality was discovered in this district, but within the last two years Mr. De B. Crawshay has found similar implements at other localities on the N. of Halstead. The interest of these finds is that they occur on the northward prolongation of the Red-Clay plateau at a point where, owing to the valley gradient being more rapid than that of the plateau gradient, the difference of level between them—which near Currie Farm does not exceed a few feet —amounts to more than 100 feet. At one spot, 13 mile distant from Currie Farm, 480 feet above O.D., and a little north of Stone- house (fig. 2), Mr. Crawshay has found seven flint implements, two of which are of the rudest Ash type and of the usual dark- brown colour, whilst five are of a light yellow colour and more closely allied to those found at Snag Lane (see p. 145). These latter were found on one of the Broke-Farm fields and may be of later date. North of the valley and beyond the Halstead station, Mr. Craw- shay found a large rude flake on Hewit’s Farm at the level of about 470 feet, and more to the east, on the edge of Shacklands Wood (525 feet), two stained flakes. In another direction, on a hill where Tertiary flint-pebbles abound, west of Northstead Farm, he records four implements of the Ash type. All these places are on the Red-Clay plateau, here intersected by the dry upper Cray Valley and its tributaries, between which the Halstead and Northstead hills project as promontories, as shown by the plan and section on the next page, figs. 2 and 3. East of Well Hill, at the level of about 430 feet, Miss H. Waring found a pointed specimen of the Amiens type on Cockerhurst Farm, near Shoreham. Farther westward, Mr. Crawshay has discovered on the highest summits of the Chalk escarpment three other implement-bearing localities. The first of these is at Betsom Hill, near Westerham, at the height of 750 to 790 feet; the second is a little off the summit * * Ancient Stone Implements of Great Britain,’ p. 531. t+ Quart. Journ. Geol. Soc. vol. xlv. (1889) p. 295. DRIFT-STAGES OF THE DARENT VALLEY. 131 at Ivy Cottage, near Tatsfield, at the level of 780 feet, and the third at Titsey Hill, where the escarpment attains its greatest height of 864 feet (see Pl. VI, fig. 3). Fig. 2.—Plan of the Halstead and Northstead promon‘ories, 4 | : WW. BE: Stoneh. & Hewit’s Pratt’s Br.-F. fields. Farm. Northstead. Bottom. 480. 470 v : gQ es Oo fev) Q — me = “ee > 1 pea =) fa a | es) 5 Qu i} Sit oS HI 3 fh 5 i a =F =] GQ ° = t+ SS oO TR o i=] H — H Ee @o 5 =} iw) iG. = e. Unstratified Flint-gravel lying in the dry bed of the upper valley of the Cray covered. . Site of flint implements of the Plateau or Ash type. Site of implements of the high-level River-valley type. Chalk. 90 4 Still more recently, Mr. Harrison has found an outlier of the brown- stained-flint dritt, together with some scarce rudely-fashioned flints, on the top of Morant’s Court Hill, forming the very summit of the escarpment south of, and $ mile distant from, Currie Farm. They are scattered on the surface of two fields within the 700-feet contour- line. This ground cannot possibly be connected with any existing line of drainage, as it forms the summit-level of that part of the escarpment. Its position relatively to Currie Farm and Halstead, as 132 PROF, PRESTWICH ON THE AGE, FORMATION, AND also to the escarpment, is given in Pl. VL., figs. 1 and 3, the former showing the gradient of the plateau from Morant’s Court Hil to the Cray Valley, and the latter the relation of the hill to the adjacent part of the escarpment. These make it evident that the older drift is clearly independent of both the valley-systems, and that the drift on this Chalk plateau, like that on the Ash plateau, is of greater antiquity and distinct from that of the valley of the Cray and its tributaries. The following lists, brought up to date, of the approximate num- ber of specimens of the Chalk-Plateau or Ash type, found at the several localities above mentioned, will show how important this plateau-group has become. For those on the west plateau, with the exception of Morant’s Court Hill and Currie Farm, I am in- debted to Mr. De B. Crawshay, and for those on the east plateau to Mr. B. Harrison. Mr. Crawshay has likewise a considerable number of specimens from Ash, West Yoke, and Bower Lane. . Chalk Plateau, West. (See Map, Plate VII.) Height above sea-level toe "Flakes. in feet. eee 1. Stonehouse and Broke Farm, Halstead* ... 480 5 2 2. Shacklands Wood, field west of ............... 5385 0 2 oa ewsts Harm, Chelsfield ..2...0ccseeeeeeeee ee 470 0 1 a oNorthstead, hill west, Of. ices «ss sseus sce cence 485 3 ify 5. Betsom Hill, near Westerham ................0. 790 15 6. Titsey Hill, near Limpsfield ...2..5....cscec-0006 864 3 ieeivy Cottage, Tatsfield ......05.c.esteeee enone 780 8° SOurrie Harm, Halstead «:-.cecacteteeceee eee 590 2 9. * Morant’s Court Hill, 8. of Halstead ......... 700 12 Single specimens have also been found farther north at Park Gate, Lullingstone, and at Cockerhurst, both above the 400-feet level. Chalk Plateau, East. Height above Numbeces senitea I Implements f. TPAC drestavan one odds ancuaxbennte eee cicuane 490 80: inva’ EE OOUME AS... IMPLEMENTS. Fic.1. SECTION FROM THE UPPER PART OF THE CRAY VALLEY TO SEVENOAKS. SHOWING THE RELATIVE LEVELS OF THE PLATEAU — AND VALLEY — Drirts AND OF THE THREE GROUPS OF PALEOLITHIC - SECTIONS DIAGRAM a. hed Cay with Flints EXPLANATION. 1. Tertiary Strata. 2. Chalk. Sevenoaks 500 as te Se = SE Hill Green 270 Broughton Dunion. 357 Bs ries g)se8 reall S$: E33 ~s|S] 5s} DOO. 3. Upper Greensand and Gault. d. Cheverang and Durtton-breen Gravel. 4. Lower Greensand. @ Stained, Flints. @ chert & Ragstone Fragments. 0 Implements of the High- + Localities. Yaldham 430 Fawkham FIG.2.SECTION FROM THE THAMES To OLDBURY HILL NEAR IGHTHAM. Fic.3.SECTION ALONG THE CREST OF THE CHALK ESCARPMENT FROM LIMPSFIELD To IGHTHAM. EEE 7 inch =7mile. Milton Dinch= 100 feet. Seale, —— Quart. Journ ( A I ee 10! OOC. VO; TWIT LaVii Quart Journ Geol Soc Vol XLVI! Pi Vit 0°10’ 0°15" MAP oF tHE DARENT BASIN, ABOVE FARNINGHAM : ( IN PART FROM THE GEOLOGICAL SURVEY. ) t 1 2 3 Ns aioe ae See ee eee Scale of Miles. | i Allarum. © Implements of the Upper Villey Gravely i. a Vee i Si Angular Dritt . Vv a , » Hateau hype Lower Valley Gravels. ~~ Watershed of the Daren Cherening ad " ! Danton Green Grards. ! Gta lines 100, £00,900. 000, WO frat abave Lullingston Upper taller Grarels Meane-Sea level 2 > ' -—---- “Lines of Sectaons. KAS Soe 400 Ca Va 51°20' 15°20 a eal heat} Ee Ae hell @ 2 oe xSNCErs Al tr os fourt. ~, ce sChevening >= io Dey x ZB = + Chipstead eye Q Sundridge 3 + ‘ 9 @b / 3i i sy 3 AN . 2 W ee eee. | tN ha Sem eee s Meo 4 > & > Ss ™~s 51°15’ fo} WEALDEN The Dangertield Printny compary Lonwon % Quart.dourn. Geol. Soe. Vol. XLVI. PL. Vu. . 2. (406). WS Tomkir del Mintern Pros. lith. FLINT IMPLEMENTS OF THE CHALK PLATEAU. Type Specumens Ya natural size. DRIFT-STAGES OF THE DARENT VALLEY. 163 for the search being limited to the surface*. These implements were, however, always found in close association with those drift-beds, and confined to the area over which they spread. These are never met . with in the intervening valleys, except an occasional derived speci- men in the newer drifts. Besides this, the colour and incrustation of the specimens show that they have been embedded in a surface- drift, which, with others, has suffered denudation, and it is probable that a large number of the specimens have been brought to the surface on which they are now found in course of working the land. With respect to the Limpsfield gravel, it certainly wanted some of the characters of a river-drift, but ice and snow may have had a good deal to do with its lodgment. It is evident also that it has been derived from the Tertiary outliers on the adjacent Chalk escarpment 300 to 400 feet above Limpsfield Common, whence the fall would be exceptionally rapid. At the same time, the Limps- field gravel assumes much more the character of a river-gravel as it descends the valley, in consequence of receiving tributary streams and acquiring greater water-power and deeper waters. Respecting the brick-earth and the Chevening gravel, the Author pointed out that the disturbed condition of the former could be best explained by floating ice, and of the latter by a covering of ice and snow. That there should be an absence of striated surfaces and scratched stones was no more than might be expected, considering the want of hard rocks. Other evidence would, however, be found in the paper, which, from its length, he had found it necessary to omit in reading. There was certainly an appearance of striz on some of the implements and older flints, but whether that arose from ice-action or from the rubbing and knocking about they. received in the old drift-streams he would not at present like to pronounce. The formation of the Chalk escarpment presented great difficulties. In the North of England, where the great ice-sheet passed over high hills, the escarpments would no doubt suffer defacement, but here the character of the ice-action would be different. The Author did not suppose that the great northern ice-sheet extended over this area. A southern central ice-area may then have existed in the Wealden highlands, and the ice and snow in these valleys have been local, The height of the glacial period preceded the Limpsfield gravel, and the W. and E. and the 8. and N. directions of the flows were the result of different physiographical conditions at different periods. The Author then expressed his obligations to the three gentlemen who had so greatly assisted him by their researches in the field, the results of which were to be seen in the large collection of Flint Implements exhibited. * T have now seen the fine specimen mentioned on p. 133. It is 6 inches long by 3} in. wide, very flat and round-pointed, and shows no wear. It more resembles one of the large St. Acheul types. It was found on the top of the soil last thrown out of the hole. 164 PROF. H. G. SEELEY ON AGROSAURUS MACGILLIVRAYI. - 12. On Acrosavrus Macettiivrayt (Seeley), a SavriscHran Reprise * from the N.E. Coast of Austratia. By Professor H. G. Srxtzy, F.R.S., F.G.8. (Read January 21, 1891.) In July 1879 the Geological Department of the British Museum obtained by purchase from Mr. E. Charlesworth some fossil bones, which were dispersed at the sale of the collection of Mr. 8. L. Waring, F.G.8., of Norwood, then recently deceased. They are labelled, in a small, delicate handwriting, “‘ Fly, 1844. Jn. Mac- gillivray, from the N.E. coast of Australia.” I believe this to indicate that the specimens were collected by Mr. Macgillivray during the voyage of the ‘ Fly,’ from some locality which was then unnamed. The bones were placed in the Mammal Gallery, where they have since remained. They comprise a complete left tibia, a less perfect proximal end of the corresponding right tibia, a fragment which I regard as a portion of a fibula, attached to matrix, which besides other fragments of bone contains two laterally compressed claw-phalanges. ‘There is necessarily no direct evidence of their geological age. But as they indicate a new Saurischian reptile, which has its nearest known allies in the lower Secondary rocks of Europe and the Trias of South Africa, it is not improbable that the animal belongs to the Lower Oolites or Trias. I have not noticed any reference to the specimens by Professor Jukes in the ** Voyage of the ‘ Fly’” or in his other books and papers, or in the writings of Mr. Macgillivray. The left tibia (figs. 1, 2) is about 20 cm. in extreme length, with the usual sub-triangular flattened proximal articular surface: That surface is nearly 5 cm. deep by 4 cm. wide posteriorly, and is flattened on the hinder and fibular borders, which are inclined at a right angle, and convex on the antero-internal contour, so that a distinct thick anterior crest is defined without obliterating the right- angled triangle form. The articular surface is slightly inclined towards the posterior and fibular borders, partly because there is a small patelloid convexity above the pre-cnemial crest, then a trans- verse concavity, behind which is the larger part of the articulation, consisting of two areas which correspond to condyles. These surfaces are divided by a shallow groove, and the larger condyle was on the external border. The posterior angles of these condylar surfaces are rounded and prolonged backward beyond the shaft, and there is a slight concavity between them. The proximal end of the bone is expanded as compared with the unusually slender sub-cylindrical shaft, which is 12 millim. in diameter. The anterior internal surface, though flattened, is gently convex, and it rounds into the posterior surface of the bone, as well as on to the strong anterior convex ridge of the pre-cnemial crest, which is prolonged down the shaft for about 6 cm. with a gently convex © * See Proce. Roy. Soc. vol, xliii. p. 165, and Quart. Journ. Geol. Soc. vol. xliv. (1888) p. 79. —— Q. J. G8, vol. -xlvii.] | To face p. 164, EXPLANATION OF FIGURES. Agrosaurus Macgillivrayi (Seeley). % natural size. Left tibia—fibular aspect, showing expansion of the ends of the bones. Left tibia—anterior aspect, with lateral notch for the astragalus, Fig. 3. Proximal articulation, showing impress of condyles of femur. Fig. 4. Distal articulation, showing distinctive quadrate form and attachment for astragalus. Fig. 5. Section of shaft, showing its cylindrical form, Fig. 6. Claw phalange, PROF, H. G. SEELEY ON AGROSAURUS MACGILLIVRAYI. 165 contour. The width proximally of the antero-internal surface is about 5 cm., but it narrows rapidly. The posterior contour of the shaft is concave. The surface of the bone at the proximal end is moderately concave, with a slight vertical channel for the fibula towards the external crest. ‘This is augmented by crushing, in the left tibia, but the imperfect proximal end of the right tibia shows no post-mortem compression. The slender portion of the shaft includes more than its middle third, and is more attenuated than in any Saurischian hitherto figured. Its section is ovate, and the bone appears to be thin and hollow. The distal end expands gradually, so that there is no point in which it is sharply defined from the shaft ; it is about half as deep as the proximalend. The articular surface is more quadrate than the distal end of any Saurischian tibia figured, with the exception of Dimodosaurus poligniensis, in which the proportions are similar. It is 3 em. wide, and measures rather more in depth, with two pairs of sides nearly parallel, though the lateral borders converge a little forward. A deep notch indents the anterior margin in the middle, and marks a division of the basal surface which descends on the fibular side like a broad talon and indicates a corresponding modifica- tion of the astragalus. I infer from the quadrate shape of the articulation that the distal end of the fibula was carried external to the tibia, and not in front of it as in Ornithischia, which have the distal end of the tibia expanded transversely. The somewhat saddle- shaped articular surface is clean, and gives no indication of close union with the astragalus. Another fragment may be the distal end of the fibula; it is less than 5 cm. long, with a slender shaft 1 cm.in diameter as preserved. The articular surface is oblique, flattened, and measures 2°5 cm. in width by 1:8 cm. in depth; it is convex on the tibial side, and less convex externally. A fracture in the matrix displays two claws, which are flattened and appear to be compressed from side to side. They are of the type which is usual in carnivorous reptiles. The larger of them is 2-5 em. long and 1°8 cm. deep at the slightly concave articular border: the contour of the bone is convex above and concave below. The smaller claw is more slender; it is 2 em. long, and 1°3 cm. deep at the posterior articulation. The distinctive character which determines the affinities of the fossil is the distal end of the tibia. It shows an ordinal resemblance with Pekilopleuron and Cetiosaurus, but with Dimodosaurus from the top of the Keuper its resemblance is so close that the two must be regarded as nearly allied. I consider the fossil now described as generically distinct from Massospondylus and all known types, and defined by its slender shaft, by the enlarged proximal end which curves backward, by the slight development of the cnemial crest, by the uniform increase in size of the distal end, and finally by the moderate excavation of the distal articulation on the inner side. The remains indicate an animal about as large as a sheep. 166 PROF. H. G, SEELEY ON SAURODESMUS ROBERTSONI. 13. On SavropEsmus Roperrtsonr (Seeley), a Crocopitran ReEpTILE from the Ruzttic of Linxsrietp, vn Exern. By Professor H. G. SEELEY, F.R.S., F.G.S. (Read January 21, 1891.) LiInksFIecpD is north of Elgin and west of the River Lossie. The locality was fully described by Professor Judd, F.R.S., in 1873 *, and the strata, previously regarded as Lias and Wealden, were interpreted as a large boulder of Rheetic beds in Boulder-clay Tf. A number of freshwater shells occur in these beds, associated with land plants, marine invertebrata, fishes, and reptiles; but very few of the species are identical with those found in other Kuropean localities. And, although there is not much room for doubt as to the age, I could have wished the evidence stronger in determining the horizon of the interesting specimen now described, which was found by Mr. A. Robertson, of Inverugie. It appears to be an isolated bone, first noticed nearly fifty years ago by Sir R. Owen in his second report to the British Association on British Fossil Reptiles. ‘“ I have been favoured by Mr. Robertson of Elgin with the examination of a Chelonian femur 43 inches in length from a stratum at Linksfield in which remains of Plesiosaurus and Hybodus occur; and this femur, though not identical in form with that of any Trionya with which I could compare it, yet resembles the modifications of the bone in that genus more closely than in Tortoises, Emydians, or Turtles.” This guarded deter- mination has sometimes been read as referring the specimen to Trionyx, but it amounts to no more than a statement that the bone is a femur, and probably Chelonian. It was figured in 1842 by Mr. Patrick Duff in his ‘ Geology of Moray,’ pl. v. fig. 10. At that time only the dorsal aspect was seen. It was acquired by the British Museum in 1854, by purchase, from a sale at Stevens’s, and registered by the late W. Davies as “? humerus of a? Chelonian” . In 1889 it was removed from the shelly matrix, so as to expose the ventral aspect of the bone. The latest reference to it is in the B. M. Catalogue of Fossil Reptiles and Amphibia, pt. in. (1889), p- 223, where Mr. R. Lydekker places it after the remains of Chelytherium with the following description :—‘‘The imperfect right humerus or femur of a Chelonian, perhaps referable to this or an allied form .... The head is wanting, and there is no distal groove or foramen. The specimen differs very markedly from the corresponding bone in any existing type.” I am not aware of evidence which would associate the specimen with Chelytherium, and the association is not discussed. Mr. Lydekker’s cautious determination amounts to no more than a belief that, while the * Quart. Journ. Geol. Soc. vol. xxix. pp. 1385-138. t The Rey. Dr. Gordon informs me that he has seen glacial markings on the rock beneath this boulder. t Inan article on the Dinosauria in the ‘Popular Science Review’ for Oct. 1879, p. 46, I referred it with doubt to the Dinosauria. -_ PROF, H. G. SEELEY ON SAURODESMUS ROBERTSONI, 167 bone is Chelonian, it cannot be referred to an existing type, and may be either a humerus or a femur. There is in the mutilated proximal end of the bone some resem- blance to a mammalian femur, such as that which I described from Stonesfield *, but the distal end of the bone forbids a comparison. And it is as certainly not the femur of any reptile. I am not familiar with any Chelonian in which the femur closely resembles the humerus. The bone is, I submit, a right humerus, but the characters have not hitherto been enunciated which would refer the bone to the Chelonia. The characters which suggest Chelonian affinities have little value in classification. They are limited to the general form of the bone. The proximal end is expanded, with a saddle-shaped ventral surface, but this condition occurs in Ornithosaurs, in Hyperodapedon, and many extinct genera of reptiles. The distal end and the articular head of the bone appear both to be in the same plane, so that there is no twist in the shaft; but since the head is not preserved, it is not impossible that it may have formed an angle with the distal articu- lation. , I am indebted to Mr. Boulenger for the opportunity of examining bones of recent Chelonia in the British Museum, but I have found no evidence to sustain the Chelonian hypothesis, nor any closer resemblance than is seen in the right humerus of the genus Hardella. Among the characters which I believe to constitute differences from Chelonians are (1) the cellular medullary cavity in the shaft of the bone, which is filled with calcite; for I am not aware that the humerus is hollow in any Chelonian; (2) the straight shaft is unparalleled in any Chelonians in which an approximate comparison could be made, for they have the superior contour convex, and the infero-posterior contour concave ; (3) the remarkably open concave curve between the radial and ulnar crests is not paralleled in Chelonians, which have the more or less lamellar, sub-triangular, radial crest close to the head and reflected downward ; (4) the radial crest is distinct from the head of the bone, does not extend so far proximally, and is relatively small, as in no Chelonian; (5) the slight development of the ulnar crest, the rough muscular attachment along its extent, and its convex contour in length are unknown among Chelonians; (6) the thickness of the ridge by which the shaft appears to have united with the articular head in the fossil is greater than among Chelo- nians, and it is possible that the shaft had an appreciable extension proximally beyond the fracture which limits its preservation ; (7) the transverse curved ridges on the ventral side of the head from the radial process to the ulnar crest are distinct in their divergence from the not dissimilar ridges sometimes seen on the Chelonian humerus ; (8) the distal end of the shaft is unlike Chelonians in the great transverse extent, in the flattening of the ulnar side of the bone, which is concave in length, in the compression of the * Quart. Journ. Geol. Soc. vol. xxxv. (1879) p. 456. 168 PROF. H. G. SEELEY ON SAURODESMUS ROBERTSONI. radial side of the bone above the articulation to a sharp edge, which makes the longitudinal contour convex on that side at the distal end, and in the dissimilar form of the distal articular surface with its well ossified detail and definition. Even disregarding the absence from the fossil of a radial groove or perforation, the characters enumerated seem to me to outweigh the slight resemblances to Chelonians which it seems to present, and it would follow that there is practically no evidence of value in favour of the Chelonian nature of the fossil. At first sight the bone (figs. 1-6) is not very like the humerus of a Crocodile, yet its affinities with the Crocodilia are more important. Careful consideration of the radial process in the fossil shows that its inner prolongation is a sharp almost knife-edge, as in Crocodiles, and that the proximal articular part of the bone prolonged the shaft beyond the radial process in a way only paralleled among Crocodiles. ' On the other hand, the radial crest is never reflexed forward so much in existing Crocodiles, and there is no Crocodile in which the ulnar border is compressed to a sharp muscular edge, though there is, perhaps, a faint suggestion of such a ridge in a proximal angle of the ulnar tuberosity ; and in Gavials, some recent species of Croco- dilus, and some fossil Crocodiles like Crocodilus Hastingie, the character is slightly marked. The bone as a whole is much more expanded at both ends than in Crocodiles, and is even straighter, but the distal articulation is essentially the same in plan, with like details of condylar structure and a like compression of the bone on the radial side, though the ridge in the living types is very slight compared with its development in this fossil. These are, however, homologous characters, and Crocodiles have the limb-bones hollow; so that, as the indi- cations from the proximal and distal ends and from the internal structure all point to the same result, it may be concluded that the Linksfield fossil indicated a primitive Crocodilian stock; and that the intensified characters which it shows are feebly preserved in its surviving representatives. The chief differences from Crocodiles are that the radial crest is directed more forward and less downward ; that the ulnar side is sharply compressed, ends in a muscular ridge, and has a convex curve; that the shaft is straighter ; and that the distal end is relatively wider, with its radial border much more compressed. In Lizards there are some approximations in these points which are worthy of remark. The extremities of the bone are more ex- panded transversely than among Crocodiles, but then the shaft is twisted. Both the radial border at the distal end and the ulnar border towards the proximal end are compressed in Lizards, but then the distal end has enormous articular condyles equally unlike those of the Linksfield fossil and Crocodiles. The concavity below the head of the humerus is more open transversely in Lizards, but then a strong rounded ridge connects the radial crest with the articular head. In no respect, however, either in characters of the proximal or distal end, can Lizards be said to approach so near to the fossil as do Crocodiles. Nevertheless, there may be a tendency Q. J. G.S. vol. xlvii.] [To face p. 168. EXPLANATION OF FIGURES, Saurodesmus Robertsoni (Seeley). # natural size. Fig. 1. Dorsal aspect of right humerus, showing fracture of head of the bone. Fig. 2. Outline of the proximal end of the bone, showing the relation of the radial crest to the shaft. Fig. 3. Transverse section of the middle of the shaft, showing cancellous tissue. Fig. 4. Outline of the distal articular end of the bone, showing form of condyles. Fig. 5. Ventral aspect of the bone, with an approximate restoration of the proximal end. Fig. 6. Posterior lateral aspect of the bone, showing the straight shaft, the curvature of its extremities, and the condition of the ulnar margin. PROF. H. G. SEELEY ON SAURODESMUS ROBERTSONI. 169 towards a generalized Lacertilian type, in so far as the characters are not Crocodilian, which is especially shown in the compressed distal radial margin. The Trias of Elgin has already yielded Telerpeton, Hyperodapedon, and Stagonolepis, and the fossil is well distinguished from these. It is somewhat smaller than Hyperodapedon, which has the proximal end of the bone greatly expanded and concave, though less expanded than in Stagonolepis, but in neither genus is there the same resem- blance to a Crocodilian type which is seen in the fossil under review. Among the extinct Orders it is in some ally of the Ornithosaurs that an approximation to the Linksfield type might be expected, for it is only in the humerus of the Pterodactylia that a close general resemblance to the fossil is found in those distal characters in which it varies from Crocodiles. As preserved, the bone is 8*2 cm. long, and when perfect may have been from 1 to 2 em. longer. At the fracture it is 3:5 cm. wide, and the bone is 1 em. thick in the middle, where a muscular impres- sion marks the large angle made with the ulnar and radial sides of the head. The middle of the shaft, which is nearly cylindrical, is 1-1 cm. wide. The distal end is 3:2 cm. wide. The articulation ascends the superior surface a little on two moderate ridges with a concavity between them. It is about 1°5 cm. thick on the ulnar side, and thinner on the radial side, the two parts being defined by the anterior and posterior concavities. These differ from the corre- sponding constrictions in the humerus of Crocodiles in that the depression in front is much narrower, while the inferior concavity is much wider. The form of the distal articulation indicates, I think, that the bones of the forearm were placed as in Crocodiles, and not as in Lizards or Anomodonts. The compressed ulnar margin (supposing it to be unbroken), with a muscular attachment at its edge, would constitute an ordinal difference from existing reptiles. The fossil, if grouped with the Crocodilia, belongs to a suborder hitherto unknown, and defined by a combination of Crocodilian and Lacertilian characters which is not Saurischian. DiscussioN ON THE ABOVE TWO PAPERS. Mr. Lyprexxer agreed with the Author in regarding the Australian tibia as that of a Dinosaur, but asked how it was generically distin- guished from Dimodosaurus or Massospondylus. He was glad that the Author termed the bone from Elgin a somewhat unsatisfactory specimen ; in the speaker’s opinion it was not worthy of being made the type of a genus. He differed from the Author in regarding the bone as being solid, and expressed his belief that although it might belong to a Rhynchocephalian or an extremely generalized Chelonian, it was certainly not Crocodilian, in any accepted sense of that term. He further enquired the Author’s meaning in using the expression ‘* Lacertilian affinities ” in an apparently loose way. He concluded by protesting against the use of the term “Saurischia” for the typical Dinosauria. It was perfectly permissible to divide the Dinosauria 170 PROF. H. G. SEELEY ON SAURODESMUS ROBERTSONI. into two orders, but if this was done the original name must be retained for the typical forms. An analogous instance occurs in the separation by some writers of the Lemuroidea from the Primates, the latter being retained for the typical members of the order. Any other course would be unjustifiable. The Avrnor thought that if Mr. Lydekker visited Paris and sought the aid of Prof. Gaudry in making comparisons, he might learn the nature of Dimodosaurus and the relation of the Australian fossil now described to that type and its allies. He used the term ‘“ Saurischia ” rather than “ Dinosauria” in defining the position of this animal, because new ideas in classification needed new names for their adequate expression. It might be that the groups Ornithischia and Saurischia were provisional, for there were indications of a third — group which could not be defined as yet. He thought there could be no more justification for the proposal to restrict the name ‘ Dino- sauria”’ to one of these groups than there would be to restrict the term ‘“‘ Mammalia” to the Monotremata or Marsupialia. With regard to the Linksfield fossil, he had carefully compared it with every available specimen in the British Museum without finding evidence of near affinity with the Chelonia, though without doubt as to its osteological identification. This was the first necessity in making a determination of the bone. As Mr. Lydekker had been unable to determine whether the bone was a humerus or a femur, he did not know how it was possible for him to have arrived at any reference of it to the Chelonia or any other group. But when the form of the distal end was appreciated as fixing its place in the skeleton, it followed that only in Crocodiles and Ornithosaurs could any parallel be found to the characters of the proximal end, so as to bring it into harmony with the distal end of the bone. He fully admitted the difficulty in restoring the head of the bone in a new type of animal. ~ MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, 171 14. Nores on Rocx-Srrcruens collected by W. Gowtann, Esq., A.R.S.M., F.1L.C., F.C.S., ox Korna*. By THos. H. Hotranp, Esq., F.G.S., A.N.S.S., of the Geological Survey of India, late Berkeley Fellow of the Owens College. (Read November 12, 1890.) (Communicated by Prof. Jonn W. Jupp, F.R.S.) ConTENTs. I. Literature. | IV. Metamorphic Rocks. II. Introduction. V. Sedimentary Rocks. III. Igneous Rocks. | VI. Summary. (1) Acid. VII. Appendix. (2) Intermediate. | (3) Basic. I, LiveRaturRe. 1704. A Collection of Voyages and Travels. London. Vol. iv. pp. 623-632. 1818. Account of a Voyage of Discovery to the West Coast of Corea and the Great Loo-Choo Island. Captain Basil Hall. London. Pp. 1-57. Appendix, p. exxi et seq. 1819. Voyage of H.M.S. ‘ Alceste. John McLeod, M.D. Third Edition. London. 1834. Journal of Three Voyages along the Coast of China. Rev. Charles Gutzlaff. Third Edition. London. Pp. 227-249. 1873-74. China-Sea Directory. 1881. ‘* Notes on the Geology of the Corean Archipelago,” Nature, vol, xxiii. p. 417. | 1886. “ Geologische Skizze von Korea.” Dr. C. Gottsche. Sitzungs- berichte k. Akad. Wissensch. Berlin, vol. xxxvi. p. 857. 1886. ‘‘ Beitrage zur Petrographie von Korea.” Prof. J. Roth. Ibid. p. 875. Il. Lyrropucrion 7. Although his visit in 1884 was primarily intended for purposes of archeological research, Mr. Gowland has succeeded, under considerable difficulties from the ill-concealed hostility of the natives, * Mr. Gowland has kindly supplied me with the spelling of the names as adopted in the system of transliteration employed by Satow, Aston, and Chamberlain. t In studying Mr. Gowland’s collection of rock-specimens I have been greatly assisted by the maps and information he has so readily supplied. I would like to take this opportunity also of acknowledging my great indebted- ness to Professor J. W. Judd, F.R.S., for facilities of study and kindly direction in the Geological-Research Laboratory of the Normal School (now Royal College) of Science, and to Professor W. Boyd Dawkins, F.R.S., for kindly poche at my disposal the use of the Geological Laboratories of the Owens. ollege. 2 172 MR, T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. in collecting numerous specimens, and in making important observa- tions on the geological structure of the interior of Korea. Mr. Gowland traversed the country from Soul, the capital, to Fusan, on the south-east coast—the route thus roughly following up the basin of the Han River to the watershed (which forms a natural line of boundary between the provinces of Chhung-chhéng and Kyéng-sang), and then from the pass near Mungyong, following the course of the Naktong River to its estuary on the south-east coast. The topography of the part of Korea thus crossed may be roughly ascertained from the list of aneroid-readings made by Mr. Gowland and appended to this paper. Although the country may be looked upon as distinctly hilly, there are no great elevations in the southern provinces. Occasional peaks may attain altitudes of 3000 feet ; but there is nothing comparable to the heights observed in the Pepi-shan mountains of the north. The highest point reached by Mr. Gowland was attained in crossing, at a point about 20 h * south- east of Brambe, the range of mountains which runs the whole length of the peninsula. Before the treaties of 1883, by which Europeans were admitted to the country, nothing whatever was known of the geological structure of the interior of Korea; the only observations made being on the numerous islands in the adjacent seas and on parts of the coast-line. The earliest geological observations recorded were made by Capt. Basil Hall, R.N., in 1818, and are set forth in an appendix to his “ Account of a Voyage of Discovery to the West Coast of Corea and the Great Loo-Choo Island.” Beyond the mention, in 1834, by the Rev. Chas. Gutzlafft, of the occurrence of columnar ‘“‘ bay-salt ” on the west coast, nothing more was done with regard to the geology of the country until Dr. Gottsche, taking advantage of the treaty with Germany in 1883, undertook a tour through the country in the latter part of that year and the beginning of 1884. The observations then made were published by Dr. Gottsche in 1886+, together with a paper on the petrographical results by Prof. J. Roth §. Whilst a careful study of the specimens in Mr. Gowland’s collec- tion has Jed to a confirmation, in general, of the results obtained by the above-mentioned authors, I consider that, from the fact of their being obtained for the most part in localities not visited by Dr. Gottsche, a description of them might prove of some interest as an addition to our knowledge of the geological structure of Korea ||. * A Korean /i is about one third of an English mile. Tt Op. cit. p. 233. t Sitzungsber. k, Akad. Wissensch. Berlin, vol. xxxvi. (1886) p. 857. § Lbid. p. 875. || An interesting account of the general characters of the country and people will be found in a paper by Mr. W. R. Carles, H.M. Vice-Consul in Korea, in the ‘ Proceedings of the Royal Geographical Society,’ vol. viii. (1886) pp. 289 to 312. he MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, 173 III. Ienrovs Rocks. Dr. Gottsche, in classifying the specimens he obtained from Korea, divided the igneous rocks into an older eruptive series (dltere Hruptivgesteine) and a younger eruptive series (jungere Eruptivgesteine). In the former group he includes granite, granite- porphyry, felsite-porphyry,’ diorite, hornblende-porphyrite, gabbro, and diabase; whilst basalts and dolerites are the only Korean igneous rocks which are not, according to Dr. Gottsche, of pre- Tertiary age. As it is impossible, in the central and southern provinces, to do more than prove that the eruptive rocks are of a later date than the crystalline schists, I shall classify the rocks only with reference to their structure and composition. There is abundant evidence of former igneous action, both plutonic and volcanic, in this country as well as in the neighbouring parts of China, described by Von Richthofen; but there is no evidence of present volcanic activity, nor is there any record ot earthquake-action within the memory of man. The only manifestations of the activity of internal forces consist in a few hot springs in different parts of the country. Dr. Gottsche mentions the occurrence of such springs, having a temperature of 76° C., near Tongnai, Kydng-sang Do; and in the north at Tamni, north of Unsan, in Phyéng-an Do, having a temperature of 45° C.* Mr. Gowland found a hot spring at Brambe near the base, on the north- west side, of the Mungyong pass. The water of this spring had a temperature of 105° F. (45°5 C.) in the bath; it was perfectly clear and transparent; without action on litmus, quite devoid of taste or smell, and leaving no deposit. (1) Acrip Ervurrive Rocks. a. Plutonic. Of the eruptive rocks, granite seems to be by far the most abundant in the southern provinces, forming most of the principal hills and exhibiting its characteristic weathering on the craggy summits. In the mountains to the north and west of Soul, curiously weathered crags of a coarse-grained granite rise to a height of 2000 feet. A specimen of this rock in Mr. Gowland’s collection has a specific gravity of 2-613. Under the microscope the large grains of quartz are seen to contain the usual bands of secondary inclusions. In places the quartz itself is secondary, and is seen to encroach on a decomposing orthoclase, sometimes retaining the kaolinized pro- ducts, which show the old line of demarcation between the crystals, Vermicular chlorite is occasionally included. Orthoclase occurs in large flesh-coloured crystals, frequently kaolinized in the centres. Decomposition has given rise to the secondary formation of minute nests of a colourless, micaceous mineral, possessing a high double * Op. cit. p. 862. Q.J.G.S. No, 186. N 174 MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. refraction. Occasional flakes of these are of sufficiently large dimensions to determine the biaxial character of the crystals. Similar brightly polarizing crystals have been frequently observed, Dr. Hatch, for example, refers them to muscovite or kaolin *. Plagioclase is represented in the Soul granite by smaller crystals exhibiting extinction-angles agreeing with those of oligoclase. Albite and muscovite are found associated as secondary products, infilling cracks in the rock. Most of the biotite originally existing in the rock has been converted into chlorite, wbich exhibits the pleochroism : E=yellowish-green, O=dark green, and which, when treated with sulphuric acid, decomposes with the formation of gelatinous silica. Of accessory minerals, magnetite occurs, and less abundantly apatite. In the coarse-grained granite of Soul there occur veins of a much finer-grained dark grey biotite-granite. Many of the crystals in this rock seem to exhibit signs of growth after the formation of the original crystal-outline (fig. 1). Much of the quartz, by intergrowth Crystal of orthoclase showing secondary extension of the felspathic material, irregularly intergrowing with the neighbouring crystals. In biotite-gramite from the mountains north-west of Soul. with the other crystals in the rock, produces a distinct pegmatitic structure. Colourless acicular inclusions are common in the quartz. Although the felspars, both orthoclastic and plagioclastic, are dis- tinctly kaolinized, the biotite has undergone little or no apparent alteration; chlorite is present in very small quantities. Near the boundary between the provinces of Kyong-kwi Do and Chhung-chhéng Do, between Yukei and Eumsong, and for some miles around the latter town, there is an extensive development of * ‘The Spheroid-bearing Granite of Mullaghderg, Co. Donegal,’ Quart. Journ. Geol. Soc, vol. xliv. (1885) p. 550. ——— MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, 175 granitic rocks, breaking through the crystalline schists in that area, and forming the low rounded hills of the district. A specimen taken from Castle Hill, Eumsong, is a medium-grained rock, with a specific gravity of 2°61. Under the microscope, we recognize quartz, in granular crystals, biotite, changing into chlorite, orthoclase, con- siderably kaolinized, and plagioclase, exhibiting examples of crystals with a progressive zonal development from a more basic plagioclase in the centre to the more acid types at the periphery. From its microscopic characters this rock would come under the definition of the granitite of Gustave Rose. A closely related rock, but more decomposed, and having a specific gravity of 2°58, occurs near Fusan, Kyéng-sang Do. A specimen of pegmatite, found by Mr. W. G. Aston near Soul, is almost devoid of mica, and consists of a decomposed aggregate of flesh-coloured felspar and quartz. The rock might be described as an aplite. Aplite has been described by Prof. Roth* from Tsu- shima, an island off the south-east coast of Korea; and Capt. Basil Hall describes some specimens from two small islands on the south- west coast, lat. 34° 23!’ N., long. 126° E., as “a decomposing, fine- grained rock with flesh-coloured orthoclase, white quartz, and porcelain clay” f. Besides the above-selected localities, granitic rocks are found in yarious parts of the country breaking through the crystalline schists, and forming, in the southern provinces, the principal factor in the formation of the mountain-peaks and -chains; one of the highest in the south being Mount Kimonsangsan, reaching an altitude of about 3000 feet, situated W.S.W. of Sonsan, in west Kyéng-sang Do. Mr. Gowland has found rocks of similar type cropping out also at the following places along the line of route :—In Kyong-kwi Do, at Yong-in, about 70 i south of Séul, at Pekkemmi, north-west of Chuksan, and at Chuksan; in Chhung-chhong Do, at Chhung-ju, at Brambe, and at various points over the Mungyong pass to the province of Kyéng-sang, where granite occurs at Yuko, 40 1: S.S.E. of Mungyong, and thence at intervals to Hamchhang, Sangju, Sénsan, and Haiphyong. At Yangsan it is again exposed, and occurs plentifully along the route from that town to the coast at Fusan. The granite is found in all stages of decomposition in the same district. Near Soul, for example, there is to be found a beautiful example of the decomposed rock in which the felspathic constituents have been almost wholly reduced to a powdery mass of kaolin, leaving the unaltered quartz-crystals, and, here and there, a clear plagioclase, whilst cubes of pyrites, of from 2 to 5 millim. edge, have been developed. The rock gives no effervescence with dilute acids. Eurites.—Acid eruptive rocks are found varying in crystalline characters from the true granitic type to those presenting the * Op. cit. p. 876. t Op. cit. Appendix, p. cxxviii. 176 MR, T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. characters of rocks which are known as felsites and eurites *. They occur in various parts of the country breaking through the crystal- line schists and granites. Between Chhungju and Brambe, Mr. Gowland obtained specimens of a mass of eurite breaking through granite. The rock has a specific gravity of 2°53. Amongst the porphyritic constituents the most conspicuous are quartz, in large transparent crystals, and Carlsbad-twins of orthoclase. The rock is stained with ferruginous decomposition-products, and contains irregularly-scattered minute cubes of hydrated oxides of iron occurring as pseudomorphs after pyrites. Small flakes of biotite are not uncommon. Prof. Roth has described a somewhat similar rock as a granite-porphyry from a locality between Paikchi and Ikujang, in the western province of Hwang-hai f. In the Mungyong pass, the larger quartz-crystals of the eurites exist in well-developed bi-pyramidal forms, the faces being generally considerably etched. Orthoclase, plagioclase, and hexagonal plates of biotite occur. in association with the quartz. Iron pyrites and magnetite are present in smaller quantities. In a pass a little to the west of Chhungju Mr. Gowland secured specimens of a similar rock, with a specific gravity of 2°56. Eurites occur also S.E. of Undon, and, like the granites, are frequently traversed by veins of quartz. b. Voleanie. Corresponding in chemical composition with the foregoing there occur in Korea rocks which, from the structures they exhibit, have had an undoubted voleanic origin, and which have, since their eruption, suffered from a devitrification of their originally glassy magmas, with a production of a secondary felsitic structure, similar to the structure exhibited by many ancient British lavas with which we are familiar from the researches of Rutley, Allport, and Bonney. . In the province of Chhung-chhong, there occurs to the S.E. of Chhungju an example of this nature associated with granitic rocks. The compact, greenish or greyish hand-specimen shows distinct banding ; and this is confirmed under the microscope by the fluidal arrangement of the microliths around the porphyritic con- stituents. With crossed nicols, the field becomes broken up into doubly-refracting patches with no apparent relation to the irregularly distributed masses of nebulous, green, and brown material. Quartz occurs in irregular grains and as bi-pyramidal crystals; and, in greater abundance, orthoclase in flesh-coloured crystals. The relics of partially decomposed felspars show that plagioclase was by ne * The name ‘ eurite’ has been shown by Messrs. Cole and Jennings (Quart. Journ. Geol. Soe. vol. xlv. (1889) p. 435) to have been employed with scientific precision by d’Aubuisson before such names as ‘ felsite,’ ‘ quartz-porphyry,’ &e. were proposed. I have, therefore, thought its use preferable to the terms more commonly employed in this country. t Op. cit. p. 876. MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. 177 meaus an unimportant constituent of the original rock. During the changes which the rock has undergone, epidote has been formed, biotite has been changed into chlorite, and a considerable develop- ment of secondary quartz occurs, infilling cavities produced by the removal of decomposition-products from the felspars. Magnetite occurs in cubic crystals. The rock has a specific gravity of 2°64. Prof. Roth records the occurrence of a somewhat similar rock from Puphyéng, Kyéng-kwi Do, and another trom Deer Island, off the coast near Fusan *, Specimens of an acid volcanic rock were obtained by Mr. Gowland south-east of Milyang, in the province of Kyoéng-sang. Porphyritic crystals of clear quartz and decomposed felspar are seen embedded in a grey fluxion-structured matrix. Occasional fragments of green material, apparently caught up in the mass, are found, on microscopic examination of the section, to be inciuded masses of altered rocks of the andesitic type. Many of the quartz- erystals exhibit the geometrical outlines of idiomorphic crystals ; the majority, however, are so far corroded by the magma as to be devoid of any trace of original form. The greatly kaolinized felspars are seen in some cases to exhibit plagioclastic twinning. An interesting example of the micrographic intergrowth of quartz Micrographic intergrowth of quartz and felspar, some of the latter being plagio- clastic, and the whole considerably decomposed. The quartz has extended its borders with an attempt at the formation of its normal, crystalline outline; and has, by the development of alternate forms, produced a rude ‘babel’ quartz. In felsitic lava from Milyang, Kyéng-sang Do. and felspar is represented in fig. 2, in which it will be seen that the quartz has, by secondary growth, increased at the borders with an * Op. cit. p 876. 178 MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, attempt at the assumption of its normal crystallographic outiine, and producing, by alternate development of forms, an imperfect example of “ babel” quartz. Hydrated ferruginous products, titanoferrite, green chloritic material, and secondary quartz, infilling cracks and cavities, have accompanied the distinct devitrification of the magma. The con- tortions and twistings (during the original flow of the material) of the patches of unlike chemical composition have given rise to a structure similar to that to which Mr. Rutley has given the name ‘“‘ damascened ” *, (2) InrermeEDIATE ErvuPriveE Rocks. The rocks to which Prof. Judd in 1876 gave the name “ inter- mediate ” are represented in Mr. Gowland’s collection of Korean specimens only by members of the series in which plagioclase is the predominating felspathic mineral. These comprise examples of plutonic origin as well as those formed as lavas. a. Plutonic. Diorite—An exposure of a fine-grained variety of this rock occurs in a gorge near Yukei, in the province of Chhung-chhoéng. Scattered through a fine-grained, almost aphanitic groundmass are patches of coarser-grained aggregates of hornblende and fel- spar, forming the “ glomero-porphyritic ” structure of Prof. Judd. Under the microscope, hornblende, in green and brown crystals, small flakes of biotite, and plagioclase-felspar are seen to be the principal constituents. Occasional granules of quartz, with nume- rous acicular inclusions of apatite, lumps of magnetite, and rarely crystals of zircon, occur as accessories. The hornblende in this rock, besides occurring in the form of the numerous small green erystals with an extinction-angle (cA ¢) of 15°, is found more rarely in large crystals of a deep brown colour, with a narrow zone of the green variety around, and in crystallo- graphic continuity with the brown mass in the centre. The cleavage is well marked in the brown crystals, which exhibit a very strong pleochroism from a=straw-yellow to y=deep brown. The strong absorption (y > >a) increased the difficulty of making an accurate determination of the position of extinction ; but 5° was obtained as an average of numerous measurements made from the vertical axis. The properties of the brown variety are thus identical with basaltic hornblende, whilst the green zone exhibits the optical characters of common hornblende. The major part of the plagioclase-crystals are zoned by a gradual passage into succeeding layers of different chemical composition from the centres of the crystals to their borders. Distinct, narrow, and irregular fringes with a different extinction-angle sometimes surround the crystals, and were apparently formed after the con- * «The Study of Rocks,’ 1879, p. 181. MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. 179 solidation of the rock, in the sume manner as was shown by Prof. Judd to be the case with the porphyritic plagioclase in the labra- dorite-andesite of Dun da Ghaoithe, in the Isle of Mull *. Near Sokul, a small town to the north-east of Soul, occurs a medium-grained, friable rock, with a specific gravity of 2°82, in which quartz, felspar, black mica, hornblende, and sphene are recognizable. In addition to these, the microscope reveals the presence of such accessories as apatite, in bacillar crystals, zircon, and magnetite. The brown “uniaxial”? mica and hornblende are found intimately intergrown; the latter sometimes exhibiting its idiomorphic outlines. Cross-sections of such hornblende crystals sometimes show, in addition to the more usual prismatic and clino- pinacoidal faces, the traces of the orthopinacoidal plane. The pleo- chroism is a=straw-colour, }=grass-green, y=bluish green, and the extinction-angle varies from 13° to 15°. Twinning on the plane 100 (co Po) commonly occurs. The biotite seems to be more generally included in the hornblende, the latter being some- times quite subordinate in quantity, and occasionally existing merely as irregular and discontinuous fringes to the mass, but which may be recognized, from the same physical orientation over large areas, as fragments of one crystal. Of the felspars, plagioclase predominates. The crystals are fre- quently composed of zones having different extinction-angles, and in the centres are frequently kaolinized over a definitely marked area, the surroundirg zones being apparently untouched by weathering agents. These centres are so well marked, and so deeply decom- posed, that one might well suspect their allothigenous origin, whilst the remainder of the crystal might have been produced during the general and final consolidation of the rock. Quartz in fair quantity exists as crystals, which are allotrio- morphic to the other constituents of the rock. The presence of sphene is confirmed by the characteristic reactions for titanium obtained during a chemical examination of the rock. Evidence of the pressure to which this rock has been subjected is afforded in the bent twin-planes of the plagioclastic felspars, and the “ undulose” extinctions of the quartz crystals. Mr. Gowland obtained, in a stream 8.E. of Yuko, pebbles of a rock somewhat similar in mineral composition, but coarser in crystalline character. Owing to the depth of the soil, exposures were rare in this district ; and consequently their absence prevented the determination of the geological characters of the area. Mr. Gow- land was compelled at this part of the journey to contend with somewhat exceptional difficulties from the hostility of the natives, who several times attempted to stone him and his party, until, at Taiku, he was rescued by the guards sent by the Governor, and thus accompanied throughout the route to the coast. * Quart. Journ. Geol, Soc. vol. xlv. (1889) p. 178. 180 MR. T, H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. b. Volcanie. An interesting series of rocks of the andesitic type oceurs in Korea. The specimens I have examined invariably exhibit signs of secondary changes, resulting, in some eases, in a deyitrification, more or less complete, of an originally glassy matrix, and passing gra- dually into the structures characteristic of the rocks to which Von Richthofen gave the name of propylites. This change is sometimes accompanied by others due more precisely to subaérial agencies, whereby we find the original alkaline and ferro-magnesian silicates replaced by chlorite, epidote, and calcite; whilst magnetite has been oxidized and hydrated to the usual ferruginous products. Between Indong and Tawon, Mr. Gowland found boulders of a black, glassy-looking, porphyritic andesite in the river-bed. The specimens were associated with stratified sandstones, grits, conglo- merates, and dark slaty rocks of doubtful stratigraphical position *. The andesite, under the microscope, exhibits the most beautiful Huidal structure. Patches of brown, black, and green microliths have been formed in the moving mass. Tbe brown and black microliths, which are seen under the higher powers to be merely minute specks and streaks of ferruginous material, have aggregated into wisp-like radial groups with irregularly curved outlines. These minute bodies closely resemble the minute black rods and granules of magnetite observed by Prof. Judd in a magma-basalt from Gribun, Isle of Mull f. Although the matrix appears so beautifully glassy, it is seen under polarized light to be completely devitrified in granular areas, which are apparently formed irrespective of the spherulitie aggre- gates of microliths. The most prominent porphyritic constituent is plagioclase, in large crystals very slightly affected by kaolinization. Twinnings, both on the pericline- and the albite-type, are exhibited, and the erystals u1e occasionally zoned from change of chemical composition during growth. The grouping together of the felspars occurs in such a manner as to suggest the “ glomero-porphyritic” structure of Prof. Judd, thus pointing to their probable allothigenous origin. Occasionally the felspars have been decomposed in the centre, and the cavities infilled with quartz (or, perhaps, albite). On stirring the erushed and sifted rock in a heavy solution of boro- tungstate of cadmium, the majority of the white felspathic grains were just held in suspension when the liquid possessed a density of 2°66, thus showing them to be of a composition bordering on andesine and labradorite. This is confirmed by the results obtained in an examination by Szabd’s method of flame-reactions. It was found that nine grains agreed closely with the characters given by Prof. Szab6 for andesine, whilst six grains showed the characters of labradorite. A silica determination of the very small quantity: * Dr. Gottsche marks this district in his map as being probably of Carboni- ferous age. _ + ‘On the Gabhbros, Dolerites, and Basalts of Tertiary Age in Scotland and Ireland,’ Quart. Journ. Geol. See, vol. xlii. (1886) p. 69, and pl. vi. fig. 7. MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. 181 available gave 56°45 per cent. of silicic acid, the mineral being thus slightly more acid than the compound Ab, An, *. Secondary quartz with chlorite and acicular apatite infill cavities which present the characteristic shapes of the pyroxenes originally in the rock. ‘Together with the quartz, occasional lumps of mag- netite and, more rarely, granules of yellowish-green epidote occur in these cavities. The rock has a specific gravity of 2-65, A porphyritic andesite occurs about 15 i south-east of Milyang, in Kyéng-sang Do. ‘The specimens have an average specific gravity of 2°62. Under the microscope, the fluidal structure is distinctly shown by the bands of dusty microliths in the brown matrix. The black and brown microliths are so closely felted (Mikrolithenfilz), that it is impossible to determine the interstitial glass. Throughout the sections, and following the direction of flow, there are grano- phyric bands, which, in places, distinctly exhibit the structure to which Becke, in 1881, gave the name of “centric.” From the way in which these bands are drawn out in the direction of flow, they suggest the existence, in the molten material, of bands of different chemical composition; but there is not, in the present case, any evidence to prove that the structure has been formed in the manner suggested by Prof. Lagorio+, or whether it is the result simply of secondary changes. “Tt is worthy of note that the felspathic material thus intergrown with the quartz is invariably decomposed and kaolinized ; and the same structure has, in other cases, been proved to be of secondary origin by Prof. R. D. Irvingt, Prof. J. W. Judd §, and Miss Raisin ||. The plagioclastic felspars are represented in this rock by clear, glassy crystals, which, when examined by Prof. Szabo’s method of flame-reactions, agreed in character with oligoclase; and, on sepa- ration from the crushed matrix by means of a solution of boro- tungstate of cadmium, were found to possess a specific gravity of 2-65. In some cases it is easy to prove that the felspars were formed as such before the rock in which they are now found—that, indeed, some of them previously existed as porphyritic constituents of a plutonic rock. In fig. 3, I have represented a case in which a plagioclastic crystal is seen to possess a centre probably much older than the surrounding zones. The central core has been schillerized in four distinct planes, the inclusions being arranged in lines crossing one another at the angles shown in the diagram (fig. 4). There is evidence of corrosion of this central core, and, afterwards, * Owing to an accident, part of the bulk analy of the rock was lost. ‘The silica, however, was found to amount to 68°42 per cent. ; in this respect the result of the analysis closely resembles that obtained from the dacite- glass of Krakatoa, described by Prof. Judd (‘Report of the Krakatoa Committee of the Royal Society,’ 1888, pp. 31, 32, and 34). + ‘Ueber die Natur der Glasbasis, sowie der Krystallisationsvorginge in eruptiven Magmen,’ T’schermak’s Min. u. Petrogr. Mittheil. vol. viii. p. 421. t U.S. Geol. Survey, Monograph V.—The Copper-bearing Rocks of Lake Superior (1883), p. 113, pl. xiv. figs. 1, 2, 3, & 4; and pl. xv. fig. 4. § Quart. Journ. Geol. Soc. vol. xlii. (1886) p. 73, pl. vii. fig. 8. | Ibid. vol. xlv. (1889) pp. 252 and 253. 182 MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. of accretion in zones of a decidedly more acid character. Some of the twin-planes of the central, old core are common to the sur- rounding zones, and run the whole length of the compound crystal ; Fig. 3. Crystal of porphyritic plagioclase with a core of older, felspathic material which has been schillerized in four directions and corroded before the formation of the clear zones of aJess basic character around. In addition to the series of twin-planes running the whole length of the crystal, another set, perhaps older, are limited to the central schillerized core. In andesite - from Milyang, Kyong-sang Do. Diagram showing the directions of the planes of schillerization in the above (fig. 3). but there are a large number which extend out only to the borders of the schillerized patch. These latter, perhaps, existed before the formation of the new felspathic material around them. MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. 183 Magnetite, in cubic crystals, is scattered throughout the rock. The original ferro-magnesian silicate seems, from the shapes of the masses of secondary products, to have been a pyroxene. It is now replaced by viridite, magnetite, granular epidote, and quartz. Fragments of other andesites are caught up in the rock. They exhibit a greater development of lath-shaped crystals, and, from the amount of ferruginous material present in them, are apparently derived from rocks more basic in character. Specimens of a vesicular andesite were found in the same district, a little nearer Milyang. The plagioclase-felspars, frequently almost wholly converted into epidote, are arranged in a direction parallel to the flow of the rock. The vesicles are frequently lined with opaline silica, and sometimes filled with brightly-polarizing radial aggregates of zeolites. The opaline silica is seen under high powers to be composed of minute, globiform bodies, similar to the structures figured and described by Vogelsang from the so-called quartz- trachyte of Hliniker Thal, Schemnitz*. The opal is stained green, and frequently drawn out into streaks. In opal similarly occurring in a hornblende-dacite from Santorin, MM. Fouqué and Michel- Lévy suspected the green staining to be due to the presence of silicates of iron and magnesia 7. Small plates giving dull polarization-colours suggest the presence of tridymite; but although many of these are hexagonal in outline, I have found no definite cases attributable to the characteristic fan- like twins described by Vom Rath. Silica has crystallized in this rock also assecondary quartz. ‘The relics of the porphyritic felspars give extinction-angles which indicate a composition more basic than that of the small, lath-shaped crystals exhibiting binary twinning. Assuming the latter crystals to be developed along the edge oP: 0 Pm (001: 010), as pointed out by Michel-Lévy +, then the microliths in the present case have a chemical composition closely agreeing with that of oligoclase. They are more acid, and probably belong to a later period of consolidation than the larger crystals, which are thus por- phyritic in the sense in which that term is employed by Rosenbusch §. The porphyritic felspars are greatly kaolinized, and have frequently been partially converted into yellowish granular epidote, which occurs in some cases as isolated granules, and at other times is spread out in patches, extending to the borders of the felspar- crystal, but never exceeding such limits. ‘These facts agree with Mr. Rutley’s suggestion as to the formation of epidote from kaolin, by the secondary action of carbonates of lime and iron in solution ||. In a rock like this, vesicular in structure and abounding in zeolites and hydrated ferruginous products, the conditions could not have been * «Die Krystalliten,’ 1875, p. 139, pl. xv. fig. 1. + *Minéralogie micrographique,’ 1879, p. 179, pl. xviii. + Annales des Mines, sér. 7, vol. xii. p- 451. ‘ ‘Ueber das Wesen der kérnigen und oA eae Structur bei Massen- gesteinen,’ Neues Jahrbuch f. Min. &c. (1882) p. || ‘On the possible origin of some Bpidcrjtes, Quart. Journ. Geol. Soc. vol. xliv. (1888) p. 740. 184 MR, T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. unfavourable to such a change being brought about in a kaolinized felspar. The ferro-magnesian silicates which originally existed in the rock have been destroyed ; but the casts of secondary products bear the form of hornblende crystals. Around each crystal there is a zone of opacite (magnetite?) granules, so commonly occurring in horn- blende-andesites—for example, in that of Altsohl, Hungary. Some- times, however, the whole crystal has been destroyed. If these borders be due to decomposition subsequent to consolidation, one might reasonably expect that the changes would extend to the centres of the crystals in a case where alteration has been so thorough. It seems more likely that the dark zones have been produced by the caustic action of the magma before consolidation, and after the relief of pressure in the rock, as in the manner suggested by Zirkel*, and confirmed by the experiments of Becker f. (3) Basto lenEous Rocks. The basic rocks are especially well represented in the southern provinces, where they break through the crystalline schists, paleozoic rocks, and granites. The same succession has been observed by Dr. Gottsche in several parts of the country, and Capt. Basil Hall mentions a “ whin” dyke cutting through the micaceons schists on Hutton’s Island (lat. 36° 10’ N., and long. 126° 13’ K.%). The basic rocks which have been collected by Mr. Gowland are all fine-grained in texture, and have, in nearly all instances, under- gone considerable alteration; they may be classed as altered varieties of dolerites and basalts. There is a gradual passage represented from the intermediate to the basic group; and in some cases it becomes a matter of considerable difficulty to distinguish augite-bearing andesites from the basalts. Although the examples of basic rocks from the southern provinces are nearly all altered to a considerable extent, there occur, to the north-east of Soul, plateaux of basaltic lavas of comparatively recent date. These are mentioned by Dr. Gottsche and by Mr. W. R. Carles §. Besides these inland occurrences, Charles Gutzlaff, in 1834, recorded the occurrence of columnar “ bay-salt” on the west coast near Changsan ||. * Ueber d. Kryst. Gesteine lings d. 40-Breitgrade in N. A,’ Bericht k. Sachs. Gesellsch. Wissensch. (1877) p. 181. t “Ueber die dunklen Umrandungen der Hornblende und Biotite in den massigen Gesteinen.” Prof. Judd, in his paper on the ‘ basic’ rocks of Scotland and Ireland (Q. J. G. S. vol. xlii. (1886) p. 79), refers to the crystallization, in the effusive basic rocks, of large quantities of magnetite, which must, in the deep-seated types, have crystallized out in the ferriferous enstatites. In these andesites, therefore, there might be, on eruption and relief of pressure, a corrosion of the ferriferous silicates, previously formed under greater pressure. t Op. cit. Appendix, Dp: EXev. § Proc. Roy. Geogr. Soc. vol. viii. (1886) p. 305 e¢ wh || Op. cit. p. 233. MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, 185 Amongst the specimens collected by Mr. Gowland, one of the freshest types of basaltic rock obtained was from a river-gorge near Chhéngdo, in the province of Kyéng-sang. In the hand-specimen the rock is a dark-green, compact variety, with an abundance of long felspar-crystals arranged in approximately parallel directions. The specific gravity of the rock is 2°81. The porphyritic felspars, judging from the angles of extinction measured, have a composition approximating to that of bytownite. They are well twinned, principally on the albite type, and contain frequent inclusions of glass, especially in marginal zones. These erystals are porphyritic also in the sense in which this term is employed by Rosenbusch. The early consolidation of these large erystals of felspar is further indicated by the occasional occurrence of specimens which have been split asunder along the Gleitfldchen and cleavage-planes, with intrusions of the matrix, which has either erystallized out, like the general matrix of the rock, or consists solely of augite, which has been partly changed into green fibrous hornblende. The original ferro-magnesian silicate has been converted into the various green decomposition-products. In some cases, this is dis- tinctly serpentine, either alone or mixed with chlorite. Patches of green fibrous hornblende, with its characteristic pleochroism, occur with these green products. The constituents of the matrix have consolidated in the order which they follow in the description: Magnetite occurs in well- marked, cubic forms, seldom accompanied by titanoferrite. Small, Jath-shaped crystals of plagioclase are abundant, both in binary and in repeated twins, which, from their angle of extinction, are closely allied to labradorite or andesine. Angite fills in the intervening spaces, with the formation of the ophitic structure of M. Michel-Lévy on a small scale—the micro-ophitic structure described and figured by Prof. Judd in certain basaltic lavas of Mull*. The augite in the Chhéngdo rock has partly changed into uralitic hornblende. The occurrence of ophitic structure indicates, according to Prof. Judd, consolidation of the rock under condi- tions of comparative quiescence ; hence in this rock the porphyritic felspars must have acquired their approximately parallel directions before complete consolidation, or before the rock assumed the quiet conditions preparatory to solidification. As might be expected from this, the microlithic felspars of the groundmass exhibit, with regard to direction, no such regularity of arrangement. A compact dark-green porphyritic basalt occurs S.E. of Milyang, in Kyéng-sang Do. The specific gravity of the rock is 2-80. Under the microscope, plagioclase, augite, and magnetite are porphyritically developed in a finely granulitic matrix. ‘“ Pro- gressive’ zones, due to changes in the chemical composition of the erystallizing magma, not unfrequently characterize the plagioclastic crystals. Between the extinction-angles of the outer zones of the * Quart. Journ. Geol. Soc. vol. xlii. (1886) p. 68, and pl. vy. figs. 3, 5, & 7. 186 MR. T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA. porphyritic crystals and those of the smallest felspathic microliths there is a curious correspondence, the larger crystals of the matrix being found to correspond to the inner zones, which exhibit grada- tions to anorthite. Zonal inclusions of the vitreous magma are common; and decomposition, with the formation of brightly polar- izing flakes of a hydrous mica, muscovite, zeolite, or kaolin, has frequently destroyed the centres, or extended along fracture-cracks in these crystals. Augite occurs in very large and almost colourless crystals, which have suffered from incipient uralitization and separation of dusty magnetite, together with the formation of chlorite, occasionally vermicular, and secondary quartz in the cavities. In the matrix of the rock there are small fibrous hornblende- crystals interspersed with cubic granules of magnetite. Epidote occasionally accompanies other secondary minerals in the cavities. It is impossible to form any reliable estimate as to the amount of glass in the rock after its primary consolidation. A further stage in the decomposition of the Korean basic rocks is exemplified in a specimen from the hills behind the Fusan settle- ment, in the south-east of the peninsula. The porphyritic plagio- clase-felspars, which are exceedingly numerous, have been almost completely kaolinized, and a considerable amount of epidote formed in granular crystals and radiating branches, together with chlorite and the usual green products of an altered basic magma. There are irregular straw-coloured patches of material, in many respects serpentinous in appearance. Mr. Rutley mentions the occurrence of a somewhat similar substance in the augite-andesites of St. Minver, Cornwall *. The original ferro-magnesian silicates have been utterly destroyed. Magnetite is plentifully scattered in small crystals. The specific gravity of the specimens in Mr. Gowland’s collection averages 2:78. A compact dark green rock from the Mungyong pass contains a large amount of calcite, which effervesces with cold acids. The specific gravity of the rock is 2-815. The felspars are seen, under the microscope, to be converted completely into zeolitic products, which give brilliant aggregate polarization-colours. Small granular crystals of augite, showing the usual twin-phenomena and frequent zoning, occur in great abundance scattered through the section. In one section, two large quartzes, with the usual bands of inclusions, have evidently been entrapped in the rock and partially fused, Around each of the crystals, which are situated very close to one another, there is an inner zone of brown, felsitic material not unlike the matrix in many of the acid felsites, and it seems in this case to have been produced by the fusion of the quartz with the more basic matrix in which it occurred as a foreign inclusion, Outside the felsitic zone there occurs a zone of granular crystals, * «On some Ernptive Rocks from the neighbourhood of St. Minver, Cornwall,’ Quart. Journ. Geol. Soe. vol. xhi. (1886) p. 393. Mr. Rutley (footnote, bid. p- 393) quotes Prof. Bonrey’s opinion as to the palagonitic nature of some of the yellow patches. MR, T. H. HOLLAND ON ROCK-SPECIMENS FROM KOREA, 187 each with a high refractive index, strong double refraction, and wide extinction-angle. ‘These granules, so far as it is possible to determine, present all the characters of the granular augites of the matrix. Quartz crystals zoned around in this manner in the magma were mentioned by Prof. Zirkel as occurring in some augite- andesites of Arran *. More recently Dr. Hatch has observed similar sporadic quartz-crystals in some specimens collected by the Rev. Mr. Baron in Madagascar f. Secondary quartz and chlorite occur in this rock. Magnetite exists in abundance. ‘The high powers of the microscope reveal the presence in the matrix of a series of yellow-brown pleochroic crystals of (probably) hornblende. . A rock having a specific gravity of 2°73, from a river-gorge near Chhungdo, presents, in the hand-specimen, characters similar to the foregoing example. It is dark-green and compact, with an abun- dance of stout prismatic crystals of plagioclase, which, with augite, form the principal porphyritic constituents. As shown by their extinction-angles, the felspars vary in chemical composition from labradorite to bytownite. The crystals are frequently split along the gliding-planes with intrusions of the magma. The augites are partly converted into hornblende, with bye-products of epidote and chlorite. Iron oxides occur as granules of magnetite, and as smaller flakes of red hydrated products, scattered through a grey matrix of closely felted plagioclase-microliths, exhibiting a fluidal arrange- ment. The structure presented in the matrix of this rock is similar to that to which Rosenbusch gave the name “ pilotaxitic ” 7, and might have been at one time hyalopilitic. This rock presents many of the appearances of the true andesites or propylites ; in fact, the majority of the specimens obtained from the south-east of Korea exhibit characters bordering on the augite-andesites, propylites, and basalts. In an andesitic rock from Pumasa, we have examples of the com- plete uralitization of augite. In addition to these, some of the erystals, whilst retaining the cleavage, form, and twin-structures characteristic of augite, exhibit the peculiar pleochroism of horn- lende. A separation in lines of dusty magnetite frequently characterizes the uralitic crystals. The large plagioclase-felspars of rather a basic type are rendered cloudy in appearance by the inclusion of a very fine, black dust, the particles of which are scarcely capable of distinct individualization even under high powers of the microscope; and, although frequently occurring arranged along irregular, curved lines, they sometimes occur in rectilinear bands parallel to the twin-planes. From the way in which they follow in intensity the zoning of the crystals, they present the appearance of being original. Dr. Hatch mentions the occurrence of similar dusty inclusions in the felspars of a Malagasy olivine- * 30°S. of E. 25° W. of N. <— zz a) 3 . ; = a Ss ras = | S zz eh ~ Q . = DQ 3 3 Gs] s ‘ ~ — be 2, a= Se 950 <:'0° N. of W. i FQ i MD wm —»25° E. of S. ‘ ’ 1 H 1 t H wt : ' ' \ 1 i S - 7 ; ‘ ' ‘ \ = i : ' { SN 1 ‘ ae ' C ‘ ‘ Sy ' { ; = 1 A =. —_s— ss : ; ne ' ' Rs A 1 2 3 9 SCALE {MILE c. Carboniferous Conglomerate. 6. Upper Coniston Limestone. } 2. Rhyolites, rhyolitic ashes, and Part of D. Dyke. 5. Calcareous Breccia. | Coniston breccias. Borrowdale 9. Brathay Flags—Part of Coniston Flags. 4. Rhyolite. f Limestone 1, Andesites and andesitic ashes. series. ; 3. Lower Coniston (Stile End) series. Limestone. ) 274 MESSRS. A. HARKER AND J. E, MARR ON Shales, and the higher portions of the Upper Limestone. The lower part of the latter at the Spa Well consists of 15 feet of nodular limestone, and banded limestones in calcareous shales, all somewhat hardened, but not greatly metamorphosed. It passes down into the breccia, which is very calcareous in its upper six feet, and crowded with small fragments of rhyolite, whilst the lower thirty feet consists of calcareous shales and grits with few rhyolite- fragments. ‘The lower part of the breccia contains casts of Lind- stremia and fragments of large trilobites in places, whilst a thin grit associated with it yields numerous Z'entaculites. After a short interval, 10 feet of flaky rhyolitic ash is seen dipping down stream at a low angle, and then a break of many yards is occupied by the red conglomerate. At the top of the Hotel Plantation is a claret- coloured felsite with quartz crystals and tolerably large porphyritic felspars. Although this is in the position of the rhyolite between the two limestones in Stockdale and elsewhere, we have satisfied ourselves that it is intrusive. There is considerable disturbance at the junction between it and the underlying limestone, which is about sixty feet thick, and consists of calcareous bands and nodules interstratified with more shaly beds. The top of the Rhyolitic Group is a nodular rhyolite, and, after another interval occupied by Carboniferous conglomerate, we meet with an alternation of more or less altered rhyolitic ashes, breccias, and lava flows, and one thin andesite, which extend up Blea Beck to the W. side of the high road, where they are faulted against the vesicular andesites and banded ashes of the south side of Tewsett Pike. Hitherto we have merely considered the general succession of the beds without reference to their changes of strike and the faults by which they are affected, and it remains to say a few words concerning these. The normal E.N.E.-W.S8.W. strike of the Silurian strata and of the Coniston Limestone Group is slightly changed in the vicinity of the granite, curving round its southern margin. The Rhyolitic Group has its strike more strongly deflected, for the rocks turn somewhat sharply towards the 8.E. on the western side of the granite, and appear to curve round with a mean N.E.-S.W. strike on the eastern side. The andesites dip at first in a southerly direction on the 8.W. margin of the granite, though farther north they appear to turn over, and maintain a general northerly dip all along the northern margin of the granite; so that on the N.E. margin, the Andesitic Group is seen dipping away from the junction of the Andesitic and Rhyolitic Groups, as represented on the map. The absence of the Stockdale Shales indicates the existence of a strike-fault running between the Coniston Flag and Coniston Limestone series. This fault has been recently described by one of us, in a joint paper with Prof. Nicholson, as running across the district *. * See Quart. Journ. Geol. Soc. vol. xliv. (1888) pp. 662 e segg. rc THE SHAP GRANITE AND ASSOCIATED ROCKS, 270 A second fault occurs at the base of the Stile End Limestone, as shown by its attenuation in Stockdale and its disappearance else- where ; also by the change of strike in the Rhyolitic Group below the Stile End Limestone of Blea Beck. iG Another fault separates the Rhyolitie Group from the Andesitic rocks to the north of it. he proof of this is furnished by the difference of dip of the two groups on the north-east. margin of the granite, where the south limb of the anticline in the Andesitic Group is entirely cut out. Besides these faults, cthers of minor importance cut through the Rhyolitic Group to the east of the granite. Their general position is shown on the map, though, owing to the quantity of drift which here covers the ground, we have not been able, in all cases, to indicate their exact position. [As the existence of these minor faults does not directly bear upon the subject of this paper, we have not fully discussed the evidence for their general trend and hades. We hope to recur to this, however, in a future communication.—March 11th, 1891.] $ IL. Descrrerion or tHe GRANITE, The well-known “Shap Granite” is familiar, not only as an ornamental building-stone, but also as the material of the famous boulders which have so often been made use of in tracing lines of glaciation in the north of England. The most striking feature of the rock is the occurrence of flesh-coloured crystals of felspar, one or two inches in length, in a matrix of moderately coarse texture and usually of greyish hue. In this matrix orthoclase, quartz, biotite, and a striated felspar may be detected by the unassisted eye or with a lens. In strict accuracy the name “granite” is not quite applicable to the rock, which is differentiated from typical granitic rocks by its porphyritic character; but the name “ granite-porphyry ” has not been applied in this country except to rocks with a matrix of very fine texture, and it will be sufficient to designate the Shap Fell rock by its popular title, “ porphyritic granite.” We shall see, however, that its micro-structure also presents a departure from the rules which hold in most granitic rocks, the quartz being of anterior consolidation to the orthoclase. The preponderance of felspars in the rock is partly explained by its chemical composition, the silica-percentage being rather low. Mr. E. J. Garwood, who has kindly made several analyses for us, finds 69°78, and Dr. J. B. Cohen, in another specimen, 68°55 per cent. of SiO,. The rock is thus less acid than the Skiddaw and Eskdale granites, which yield 75-223 and 73-573 per cent. of silica respectively. Dr. Cohen’s analysis is given on the next page (mean of two) :— 276 MESSRS. A, HARKER AND J. E. MARR ON L i; sa, ask ie. 68°55 68°54 ae Oethio.s DR eeky 16°21 15°82 POM. oscggly OM 2-26 2°26 Heth £0; ). seats not estimated — APS), At ae ee 0°45 0-52 RGD tacts eaer oa. 1-04 0°93 eas ste. darkts letes | 2°40 1-99 CE ee eae 4:08 4°23 |S Eee ee 4°14 4°45 EPA SPER. Vike not estimated not estimated 99°13 98°74 I. Shap-Fell granite, bulk-analysis from an average specimen weighing 10 lb.; anal. J. B. Cohen. II. Caleulated composition obtained from the analyses of the porphyritic felspars and the groundmass given below (p. 278), on the supposition of one part of porphyritic crystals to nine of groundmass, The specific gravity of a specimen of the “light ” Shap granite was found to be 27687. The microscope reveals several minerals in addition to those enumerated above *. Apatite is present in all the slides examined. It occurs in little prisms with hexagonal cross-section, and sometimes in very slender needles. Zircon, in small quantity, is also a constant constituent, forming small prisms terminated by an obtuse pyramid. These two minerals are always the earliest products of consolidation, and contain no inclusions. Magnetite is usually present, in little octahedra, in clusters of partly-developed crystals, or in less regular patches. The bulk of the mineral has separated at an early stage, but sometimes a portion is seen to mould the mica and later minerals. Sphene is always present and often abundant. It commonly shows good crystal forms; namely, n (123), ¢ (001), and y (101), in Miller’s notation ; but twinning is not met with. The cleavage- traces are often apparent, making acute angles with the bounding lines of the section. The colour in thin sections is light to moderately deep brown, with well-marked pleochroism. Longitu- dinal sections show the three forms mentioned, n being the best-deve- loped ; one of the axes of elasticity (a) is nearly parallel to the length of such a section, and vibrations in this direction give a pale straw colour, in the perpendicular direction a light reddish brown. Trans- verse sections are parallelograms bounded by n-faces only, and show a deeper colour than the others, with slightly less dichroism. The absorption formula is— y>B >a. * The specimens illustrating this paper, with about 150 slides, are in the Woodwardian Museum at Cambridge. The figures given in square brackets refer to the numbers of the slides. — THE SHAP GRANITE AND ASSOCIATED ROCKS, 277 Looking at Lane’s* two types of rock-forming sphene, it appears that this corresponds to the type which he associates with rocks poor in alkalies and rich in magnesia and iron-oxides. The sphene in our rock is free from inclusions, excepting occasional crystals of zircon and magnetite. The mineral sometimes occurs in granular patches, but there is no reason to doubt that these also are of original formation. Dark mica is the only ferro-magnesian silicate proper to the rock. It forms moderately small flakes, which almost always, when well bounded, show the pseudo-hexagonal appearance with large basal plane. In connexion with certain marginal modifica- tions of the rock, however, there occur larger plates of mica with a different habit. These have the shape of long narrow blades, bounded apparently by the forms ¢ (001) and 6 (010), with irregular terminations. They are often as much as an inch long, with a breadth of less than =}, inch. When fresh, the mica is of a deep brown colour with intense pleochroism, vibrations perpendicular to the a-axis (7. ¢. nearly parallel to the cleavage-traces in a section) being absorbed almost to opaqueness. The bisectrix is not quite perpendicular to the basal plane, as may be verified by a slightly oblique extinction in sections. This also enables us to detect in some crystals a lamellar twinning parallel to the base. The mica encloses occasionally any of the previously-named con- stituents, besides its own secondary products. Its most usual mode of alteration, exhibited in almost all the slides, results in a partial decoloration, or more frequently a green colour in place of the brown, and a considerable diminution in the absorption and pleochroism. The process is effected along the cleavage-planes of the mica, and often gives rise to irregular lamelle of green colour alternating with the browny. A separation of granular magnetite invariably accompanies this mode of decomposition. Side by side with flakes so affected there are often others converted in their interior into a reddish-brown substance free from magnetite. This shows less intense pleochroism than the fresh mica, the absorption being rather less parallel to the 6 and y axes, and greater parallel to a. The optical properties are retained so far as to show the lamellar twinning between crossed nicols, but the cleavage is obliterated. Probably this represents a further stage of change than the green mineral, the secondary magnetite having been reabsorbed in the form of ferric oxide. The marginal parts of the flakes so affected are usually green, and still show cleavage-traces in the sections. An examination of some basal sections of mica in the slides, or, better, of thin films carefully flaked off from the mineral, frequently reveals numerous minute needles of rutile disposed in three directions parallel to the boundaries of the hexagon. ‘These we have met with only in the decomposing mica, and they may possibly be secondary * Tscherm. Min. u. Petr. Mitth. (N.S.) vol. ix. (1888) p. 207. t See Teall’s ‘ British Petrography’ (1888), pl. xxxv. fig. 1. 278 MESSRS. A, MARKER AND J, E. MARR ON products rather than original inclusions *. They are well seen in the blade-like micas already referred to. Mr. J. A. Phillips mentioned hornblende as a constituent of the Shap granite, but it is not found in any of our numerous specimens and slices. The felspars of our rock fall under three heads: the earlier ortho- clase, the plagioclase, and the later orthoclase. The first builds the large red crystals which give to the rock its porphyritic appearance. The crystals show the common habit, and are usually twinned on the Carlsbad law. The mineral is clearly monoclinic. Mr. Phillips speaks of it in one place as ‘“ microcline,” but as the specimens alluded to formed part of the facade of a building, it is clear that they could not have been subjected to any decisive test. These large felspars enclose crystals of apatite and sphene, besides occasional flakes of mica and prisms of striated plagioclase. More rarely they contain little patches of quartz, or even a well-bounded crystal of that mineral [395]. Sometimes, however, we find numerous round grains of quartz enclosed in the marginal portion of the felspar crystals [796]. The inference is that these porphyritic crystals were formed at a time when the accessory constituents of the rock had already separated out, and the mica and plagioclase had begun to form, and that their growth only occasionally continued into the stage at which free silica began to separate. For the following analysis (I.) of the porphyritic felspars we are indebted to the kindness of Dr. J. B. Cohen. The figures are the mean of two determinations :— ie aT III. SIE ek eel 64:48 68°89 AUQUE ony 19-04 15-48 Fe Oe. on: tk “i 2°46 Bett ge.i. 0-51 te traces MnO .... _ traces ap 0°58 i 0 ee 0-0] 1:02 1:04 (OF (6 Nae 0°73 we 2:13 po Fr 6 epee 2°15 2°64 4:69 i Oo Bea tawdl tec: 10°74 3°70 H,O .... not estim. 0-78 (ign.) not estim. 99°01 98°70 98°97 I. Porphyritic pink felspar of the Shap Fell granite: anal. J. B. Cohen. II. Dominant felspar of the granite of Glenmalure, Co. Wicklow : anal. Galbraith ; cit. Haughton, Quart. Journ. Geol. Soc. vol, xii. (1856) p. 178. IIT. Groundmass of the Shap Fell granite: anal. J. B. Cohen. The percentage of soda is worthy of notice; a small part of it is * Rosenbusch, ‘ Mikr. -Physiogr. d. petr. wichtig. Miner.’ pe ed. (1885) p. 303. See also W. Maynard Hutchi ings, Geol. Mag. (1890) p. 2 THE SHAP GRANITE AND ASSOCIATED ROCKS. 279 probably due to enclosed crystals of plagioclase ; but, making liberal allowance for this, the potash- and soda-felspar molecules must be combined in the mineral in some such ratio as 4:1. The felspar resembles the dominant one in the granites of Leinster, investigated by Dr. S. Haughton, and the analysis of one of these is here quoted for comparison (II.). » Dr. Cohen has also analysed the groundmass of the granite: that is, the rock excluding the porphyritic felspars. From his figures (IIT.), and remembering that part of the potash must be contained in the mica, we see that among the smaller felspars of the rock plagioclase is the dominant variety. Comparing the figures in columns I. and III. with the bulk-analysis of the rock given above (p. 276), it is seen that the porphyritic crystals constitute about one-tenth of the whole mass of the rock (see column II. on p. 276). The plagioclase felspar occurs in idiomorphic crystals, often en- closing zircon, dark mica, &c., but moulded by the quartz and later orthoclase ; these facts sufficiently fix the time of formation of the mineral. Alhite-twinning is always seen, the lamella being rather narrow. Carlsbad-twinning sometimes occurs in addition [395 a], and more rarely a lamellation answering to the pericline law [395, 876]. The optical properties point to oligoclase. The crystals are frequently turbid, being filled with a fine dust doubtless due to de- composition, and calcite is also to be detected, besides minute fan- like groups of fibres; probably of some soda-zeolite. The quartz and later orthoclase call for no special remarks. The last-named mineral, being commonly the latest product of consolida- tion, is for the most part without crystal boundaries, and moulds the irregularly shaped or rounded grains of quartz (see Pl. XI. fig. 1). Micropegmatitic intergrowth of the two is not found in the normal type of granite. The structure of the Shap Fell granite seems to warrant some inferences as to the conditions under which it was injected into its present position. The intrusion must have occurred soon after the close of the Silurian period. Taking the thickness of the Silurian strata as 14,000 feet, we obtain an approximation to the depth at which consolidation took place. By the consolidation of an igneous rock we must understand the consolidation of such of its constituents as erystallized in situ, and in particular of the one last formed, which in this case is the later generation of orthoclase. Earler-formed minerals may have separated out from the magma at greater depths and been carried up to their present position. From a study of the fluid-cavities enclosed in the quartz of this rock, Mr. Clifton Ward deduced that it was formed under a pressure equivalent to the weight of 46,000 feet of strata, instead of the 14,000 which formed its actual cover; but this conclusion, as has been said, must be applied to the mineral, not the rock. It is improbable that the overlying strata would be able thus to withstand an upward pressure equal to more than three times their weight. Mr. Ward shows that, as regards this wide discrepancy, the Shap Fell rock is exceptional among the Lake District granites: but he fails to notice that it is 280 MESSRS. A. HARKER AND J. E. MARR ON also exceptional in that its quartz erystallized prior to the final consolidation of the rock. If we suppose this mineral to have been brought up by the magma from its place of consolidation at a greater depth, the difficulty vanishes. One of the most striking characters of the Shap Fell granite s the occurrence of distinct patches of darker colour and somewhat finer texture than the surrounding rock. These patches are abundant in the quarries, and may be well studied in the polished slabs and pillars used in building *. They are of rounded outline, though not usually spheroidal, and have a sharply defined boundary. Most of them are only a few inches to a foot or two feet in diameter, but there is one large enough to be separately quarried for setts. They contain, though rather more sparsely, porphyritic felspars like those of the normal granite ; and Mr. Phillips gives instances of felspars lying partly in the dark patches and partly in the surrounding rock, The large felspars within the inclusions frequently have, however, a rather rounded outline, and present other peculiarities which will pe described below. It is evident that these phenomena cannot be explained by sup- posing the liquid granite to have caught up fragments of rocks broken through in its irruption and metamorphosed them to a crystalline condition. There are, indeed, some inclusions in the granite which represent highly altered fragments, but they are much Jess common than the type under consideration. They show acloser texture, and never enclose porphyritic felspars. Further, their form is quite irregular, and usually angular, and one large specimen in the Woodwardian Museum has evidently been a shaly or slaty rock, which has been partly split and penetrated by tongues of granite in the direction of its laminz. Leaving out of account evident included fragments, we have a type of inclusion possessing very definite characters, and agreeing with what is observed in many other granites, granophyres, and syenites, The inclusions are constantly of finer texture, greater density, darker colour, and more basic composition than their matrix. In the following table of silica-percentages the figures for the Shap Fell rock are obtained from Mr. Garwood, those for the other granites being quoted from Mr. J. A. Phillips’s paper :-— Matrix. Inclusion. Gready, Cornwall ........ 69°64 65°01 Baterleadie 52 ios owen a ae 73°70 64:39 Ardshiel (Fort William) .... — 52°43 Shap Re ee een ee end: (69 TS 56°95 It will be seen that the difference between matrix and inclusion is greater in the Shap Fell granite than in those of Gready and Peterhead. An average specimen of an inclusion from the Shap quarries was found to have a specific gravity of 2°769. This is considerably * E.qg. at the Midland Grand Hotel, St. Pancras Railway Terminus. THE SHAP GRANITE AND ASSOCIATED ROCKS, 281 higher than the sp. gr. of the normal granite, and the difference is greater than in Mr. Phillips’s rocks * Matrix. Inclusion. BONG Te 24/50. Haske e aude 2°72 2°73 PaterWenae ody say ins 2°69 O78 Ardshiel (Fort ila fer 2-93 Pra ely). Seabee Shs ORF 2°769 It is interesting to compare the dark patches in the Shap Fell rock with those described by Dr. Ch. Barrois + in the intrusions near Rostrenen in Britanny. This, too, is a biotite-granite with large porphyritic crystals of orthoclase, which, however, are not red but white. In it occur patches of darker colour than the normal type, containing less orthoclase in the groundmass, more plagioclase, and that of a more basic variety, and more apatite. In these patches, however, the porphyritic crystals of orthoclase are wanting. This seems to be explained by the fact that these crystals, unlike those of the Shap rock, are of rather late consolidation, being posterior to the mica. In the Rostrenen rock, too, the patches are described as graduating into the normal rock, which would seem to indicate a greater degree of fluidity at the time of injection than in the case of the Shap granite. Microscopic examination shows that these dark patches differ in some respects from the normal granite of the quarries, in both the relative proportions and the arrangement of the constituent minerals (see Pl. XT. fig. 2). Apatite occurs rather plentifully, though locally, in small clear needles. _ Zircon is less abundant, but a few crystals occur, chiefly in the mica. Magnetite is present rather sparingly in little crystals and grains, as in the normal granite. Sphene and dark mica occur in much greater abundance than in the typical Shap granite, and it is the latter mineral which gives the prevailing dark colour to the patches in question. It is mostly in rather small flakes, and shows much of the green decomposition- product noticed above. The sphene is sometimes almost as plentiful as the mica; it forms acute-angled crystals, as already described, or rounded grains, with deep brown colour and strong pleochroism. The felspars here are almost constantly idiomorphic, and _ besides the porphyritic orthoclase, to be separately noticed, occur in larger and smaller crystals. Among these the triclinic felspar predominates over the monoclinic, and is more abundant than in the normal * Among foreign rocks, the biotite-granite of the Barr-Andlau district in the Vosges compares very closely with that of Shap Fells. The figures given by Rosenbusch (‘ Steiger Schiefer,’ pp. 147, 154, ed. 1877) are :— Matrix. Inclusion. Silica-percentage ......... 68-967 57894 ' Specific gravity ......... 2°680 2779 + Ann. Soc. Géol. du Nord, vol. xii. (1885) p. 6. Q.J.G.S. No. 187. x 982 MESSRS. A. HARKER AND J. E. MARR ON granite ; it shows Carlsbad twinning, as well as fine lamellation on the albite-law. The orthoclase is in Carlsbad twins or simple erystals. Both felspars in the larger crystals exhibit zones of growth with slight variations of optical characters, The quartz, which in the normal granite is of anterior consolidation to the orthoclase, has here been in general the latest mineral to form, and occurs interstitially in wedges, or often in granular patches. More rarely there is a micropegmatitic intergrowth of this mineral with part of the orthoclase [1046]. Again it is not uncommon to find isolated round grains of quartz, ;/, to + inch in diameter, with no inclusions except an occasional grain of sphene ; these must belong to a rather early stage of the consolidation [984, 1068-1070]. The porphyritic crystals of flesh-coloured orthoclase which occur within the dark patches are essentially identical with those in the normal granite, and must belong to a rather early stage of con- solidation ; but they present certain peculiarities which suggest that they have been subjected to chemical corrosion by the surrounding magma. They show very generally a somewhat rounded outline, and frequently have a well-marked narrow border distinguished by a white colour. Under the microscope it is seen that this border does not consist of orthoclase, but for the most part of plagioclase and quartz. The former is partly in lath-shaped forms, partly more irregular, and is moulded by the quartz. All the plagioclase crystals around any one orthoclase have a common orientation, presenting the usual crystallographic relation towards the monoclinic felspar, even when they have no point of contact withit. Thisfact, together | with the rounded outlines of the central crystal and of the whole aggregate, point to the effects of corrosive alteration rather than an original intergrowth. It may also be observed that the border contains no inclusions other than those found in the orthoclase itself, as it would probably do if it were an actual addition of later date. A singular modification of the granite is seen in a large loose block to the south-east of the intrusion and just below the footpath that runs along the north side of Wasdale Beck. Unfortunately we have not found this type in situ. Here, on a cursory examination, we seem to have something very like a gradual passage from the granite to a metamorphosed rock, or at least a contact of a very intimate character, the two rocks dovetailing into one another in a manner which makes it difficult to draw any definite line of demar- cation between them. It seems as if little parallel veins of a pink felspathic rock proceeded from the granite penetrating the darker metamorphosed rock. In, or on the line of, these veins are large flesh-coloured felspars identical with those of the normal Shap granite; but the veins are sometimes too narrow to completely enclose these crystals, and the felspars also occur in the line of the veins beyond the point where these can be traced. The whole presents a striking resemblance to a section given by Dr. Ch. Barrois* * Ann. Soc. Géol. du Nord, vol. xii. (1885) p. 15. THE SHAP GRANITE AND ASSOCIATED ROCKS. 283 to show the dying-out of apophyses of the porphyritic granite of Rostrenen, which seems to have many analogies with the Shap Fell rock. Any ideas based on the general appearance of this rock are, how- ever, dispelled by a closer scrutiny, which proves that the whole is granite, and the semblance of a contact quite illusory. The granite differs somewhat from the normal type, especially in possessing a general ‘‘ parallel structure.” This parallelism is shown not only in the banding of the rock and the imitation of intrusive tongues, but also in the orientation of the large felspars and their restriction, for the most part, to particular lines. The pinkish colour of the rock along these lines must be referred to subsequent chemical action, as in the case of the pink granite in the quarries; and it here follows fine cracks which have probably served as channels for infiltration. Slides of the rock [1071, 1280, 1281] show some curious cha- racters. There is a distinct banded structure on a small scale. In some bands quartz is abundant, and then tends to be idiomorphic towards the felspar as in the normal granite: in other bands felspar is far in excess of quartz, and is then moulded by it, as in the dark patches in the quarries described above. Another link with these dark patches is the abundance of magnetite and apatite, but we have not identified any sphene. The rock, moreover, has peculiarities not found in either the normal granite or the dark patches. The magnetite, mainly occurring in streaks parallel to the banding, shows some crystal forms, but in some cases moulds the felspar. The brown mica is partly of early consolidation, but partly posterior to the felspars. Much of this mineral shows green coloration or bleaching, and finally conversion into a yellowish-brown substance with complete loss of the original structure. ; The most striking feature, however, is the abundant occurrence of andalusite in idiomorphic, though rather rounded, prismatic crystals, usually coated with little flakes of yellowish or greenish-brown mica (see Pl. XI. fig. 3). The andalusite is usually clear and colourless, only occasionally showing the characteristic pleochroism : a(c), pale rose-pink; 6 and y, colourless or very faint green. The inclusions are of magnetite, zircon, and mica, and around some of these, especially the zircon, the well-known pleochroic halo* is well seen, the colours being: a(c), bright yellow; 6 and y, colourless. Andalusite as a regular constituent of granite has been recorded by Mr. Teall tT and Dr. E. Cohen ¢ of Greifswald. As an accessory in granitic dykes it is also recorded in Spain §, Cornwall, and * Rosenbusch, ‘ Mikr. Physiogr. d. petr. wichtig. Miner.’ 2nd ed. (1885) p- 380. t Min. Mag. vol. vii. (1887) p. 161. t Neues Jahrb. (1887) vol. ii. p. 178. * § Macpherson, Ann. Soc. Esp. Hist. Nat. vol. viii. (1879) p. 229. E fe 2384 MESSRS. A. HAKKER AND J, E. MARR ON Alsace*, while Von GiimbelT mentions it as occurring in pegmatite- veins in Bavaria. In connexion with the parallel structure in this rock, it may be observed that at one place in the quarries the granite has a banded appearance not unlike some gneisses, a phenomenon doubtless due to a certain fluxional movement of the mass. In the general bulk of the granite the only indication of flow is an occasional rude parallelism of the long axes of the porphyritic felspars. It remains to allude to some other special mineralogical and tex- tural modifications exhibited in certain parts of the granite mass. It may be noticed that two varieties of the rock are recognized for building purposes, the difference being one of colour only. In both the large porphyritic felspars are of a flesh-red tint, but in the most common type the other felspars of the rock are white, while in the “dark” variety they too are red. The relations of the two rocks as seen in the quarry suggest that the latter is a modification of the former, produced by secondary actions, and connected with infiltration along fissures. This is certainly the case with some other granites, which are red in the neighbourhood of joint-surfaces, but grey in the interior. In one place in the Shap Fell quarry, extensive weathering along a main divisional plane, assisted perhaps by some degree of sliding, has converted the granite for some distance into a soft, greenish, earthy material. The ordinary granite is in some places distinctly cut by small veins of a lighter coloured and somewhat finer-grained granite without porphyritic crystals. Although thus clearly posterior to the main intrusion, these may reasonably be referred to the same general source. The texture of the normal granite itself seems to be very constant throughout the mass of the intrusion. It does not become finer in the marginal parts, nor, usually, in the nearest offshoots connected with it, so far as our observation goes; but the large porphyritic crystals are wanting in the small ramifying veins on the border of the mass, as if the narrowness of the fissures, though these are wide- enough to contain the felspars, had offered some impediment to their floating in. . On the other hand there are, in one or two places at least, mar- ginal modifications of the granite, which present a coarser texture than the normal type, as well as some mineralogical differences. This is seen on the hillside above Wasdale Head, about 350 yards N.W. of the farm. Here, at the contact with the metamorphosed rocks, the granite consists almost entirely of large crystals of pink felspar, with very little quartz, and the flakes of dark mica have the long blade-like habit already mentioned. Mica also occurs at the same place in the form of thin films adherent upon the crystal-faces of the felspar, which is partly idiomorphic. Again, the junction of the granite with metamorphosed ashes (altered to the appearance of mica-schist) is exposed in the tramway-cutting at the north-east * Rosenbusch, ‘ Mikr. Physiogr. d. massig. Gest.’ (1887) p. 31. + Geogn. Beschr. Konigr. Bayern, vol. ii. p. 317. THE SHAP GRANITE AND ASSOCIATED ROCKS. 285 corner of the granite mass. Here a narrow band in the granite consists mainly of pink felspar, but has some quartz intergrown with it as a rude pegmatite [794-796]. There is also mica with the blade-like habit, as in the other case. The pegmatite band does not border the granite, but runs horizontally at right angles to the vertical face of junction. A remarkable section is seen on the west side of Sherry Gill, Here the granite is seen underlying the altered rocks with a low angle of dip, and is probably a large sill rather than the main mass of the intrusion. Along the junction runs what at first sight appears to be a quartz-vein ; but on examination it is found that the vein must have been a rather coarse-grained aggregate of felspar and quartz (‘‘ pegmatite” of some writers), in which the felspar has been largely replaced by quartz. The former mineral had often erystal outlines, and the process of replacement, which began in the interior of the crystal, is seen in various stages. A similar vein ents through this, as well as through the granite and the meta- morphosed rock, proving that veins of this kind were not all pro- duced simultaneously. As a somewhat analogous phenomenon may be mentioned a large cavity seen in the heart of the granite-quarries. It. occurs in con- nexion with a joint, the surface of which is laid bare, and it has a width of about six inches from the joint-surface. This is lined with large felspars and quartz showing crystal faces, while around it is a narrow margin of pegmatite with graphic structure. The geodes frequently contain well terminated crystals, and, in addition to the minerals mentioned, we have noticed im these and the joints tale, calcite, fluorite, malachite, iron pyrites, copper pyrites, molybdenite, and mispickel (?). The replacement of the felspars by quartz at Sherry Gill, pre- sumably an operation involving the agency of water, must belong to a late stage in the history of the intrusion. Perhaps we may assign to the same period the production of white mica along joint- faces in the metamorphosed rocks, accompanied by modifications ’ extending to a very short distance from those planes. They have been observed in the Andesitic Group near Wasdale Pike, in the limestones of Wasdale Head Farm, and in the Coniston Flags of Wasdale Beck. Mr. E. H. Acton has kindly examined spectro- scopically the mica from the last locality, and finds in it no trace of lithia ; it is apparently an ordinary potash-mica. § III. Tue Dyxes anp SILLs AND THEIR RELATIONS TO THE GRANITE. An interesting group of intrusions is well exhibited in a valley about a mile south of Shap Wells Hotel at Stakeley Folds and Gill Farm. Stakeley Folds is two thirds of a mile from the nearest granite outcrop. Here, and within three or four hundred yards to the south-east, four distinct sills are seen, injected one above the other at slightly different horizons in the Coniston Grits. 286 - MESSRS, A. HARKER AND J. E. MARR ON i The highest sill, which, owing to the dip, is the lowest down the — valley, shows a grey compact ground, studded with little quartz-_ i grains and flakes of dark mica, and enclosing porphyritic felspars, some of which are one or two inches long. ‘The quartz grains are mostly rounded, but occasionally a bipyramidal crystal is seen. The most conspicuous feature of the rock is the occurrence of ‘i ‘felspars of flesh-red colour, some with Carlsbad twinning, which at. il once recall those of the Shap Fell granite, but have the rounded outlines and often the well-marked borders associated especially f with the dark inclusions in that rock. Besides the above minerals, | the thin slices cut from this sill [1157, 1158] contain apatite | prisms, occasional zircons, and abundant acute-angled crystals of | brown pleochroic sphene, like those so characteristic of our graniti¢e inclusions. When the zircon is enclosed by mica, it is surrounded 1 by an intensely absorbent pleochroic halo—a character which we have noted in the Shap Fell granite itself, and which is well known i in many others. The mica is of the usual brown colour, Its mode {\ _ of alteration is sometimes like that of the mica in the granite; ‘| while sometimes it gives rise to the interposition of lenticles and i streaks of calcite along the cleavage-lamelle in the fashion usually i seen in lamprophyric rocks. The rounded grains of clear quartz have inlets and enclosures of the groundmass, which is that of an ordinary quartz-porphyry, in which, however, part of the felspar has separated out in little prisms. The porphyritic felspars enclose a few mica-flakes, as well as the earlier accessories. Both ortho- clase and oligoclase are represented. The latter sometimes occurs in clusters of small crystals, with irregular junction with one another, but presenting crystal forms to the surrounding groundmass, Mr. Teall *, in describing similar clusters of felspar crystals in the Tynemouth dyke, has pointed out that this accords with the view that such crystals were formed under plutonic conditions, and floated up in the magma into their present position. The next sill has, to the eye, a similar grey ground, with perhaps rather more mica, and encloses little plagioclase crystals and scattered grains of quartz about 1, inch in diameter, but apparently ~ none of the large red felspars. ‘The microscopic characters accord with those of the former rock, except that there is very little sphene present [1159]. The preceding rocks may be called micaceous quartz-porphyries. The next sill has in the field a dull brown ground crowded with flakes of brown mica, and would naturally be mapped as amica-irap. It contains, however, large red felspars with rounded outline, and a few scattered blebs of quartz. The microscope shows that these felspars are of a striated variety, probably near oligoclase, with a narrow border of orthoclase [1160]. The interior of each erystal is twinned according to the albite and Carlsbad laws, and the Carlsbad twinning is continued into the border of orthoclase. The slide contains abundant brown mica, which has suffered altera- tion chiefly of the kind producing calcite: there is but little mag- * Geol. Mag. (1889) p. 481. THE SHAP GRANITE AND ASSOCIATED ROCKS. 287 netite, either original or secondary. The groundmass of felspar and quartz is too far decomposed for minute examination, but it is clear that the rock has been of a much more acid type than such mica- traps as those found, for instance, in the Sedbergh district. The lowest and thickest sill, at Stakeley Folds itself, consists of a quartz-porphyry in which the porphyritic elements are much more crowded than in the foregoing, forming a considerable proportion of the mass. The quartz grains are rounded, but with occasional idiomorphic faces, and the felspars comprise both orthoclase and oligoclase: The slice shows plenty of brown mica, altered into the green mineral along cleavage-planes [1161]. Two sills are seen near Gill Farm, farther down the same valley. The lower of these two is a red quartz-porphyry with little blebs of quartz. These average about ;/, inch in diameter, and have the usual “‘ corroded ” appearance, with enclosures of the groundmass, which is almost cryptocrystalline [1156]; there are, moreover, clusters of small porphyritic felspars like those noticed above. The upper sill has to the eye a much more lamprophyric appearance. Not far east of Gill Farm is a large dyke which has all the appearance of an ordinary minette. No quartz is evident, but there are small porphyritic felspars, usually not more than 3} inch long; some light red, others, with rounded edges, colourless and glassy. These latter are found under the microscope to consist of striated plagioclase with a narrow border of orthoclase, like those noted in the third of the sills at Stakeley Folds. As before, the two felspars have Carlsbad twinning in common [1155]. The brown mica has the usual hexagonal habit, but its extinction in transverse sections is oblique enough to show vaguely the repeated lamellar twinning already remarked in the granite. The flakes are frequently bleached in the interior, in the fashion familiar in the mica-traps of various districts. The inclusions of zircon, apatite, &c. are sometimes ranged parallel to the basal plane. Magnetite occurs in rather large patches through the rock, as well as in numerous minute octahedra. The general ground consists largely of little felspar prisms, with a few more shapeless crystals of concentrically zoned felspar, and subordinate quartz. Except for the quartz in the groundmass, which seems to be at least in part an original con- stituent, this dyke compares closely with mica-traps such as those described by Prof. Bonney and Mr. Houghton * in the Kendal and Sedbergh districts, and by Dr. Hatch t near the latter locality, or with similar rocks to be seen near Ingleton and in the district west of the Cross Fell range. Viewed as a whole, the set of neighbouring intrusions briefly described above, while presenting a considerable range of differences, have at the same time some curious points in common. Further, while they have characters which seem to connect them on the one hand with the Shap Fell granite, and particularly with its darker patches, they, are unmistakably linked on the other hand with the normal * Quart. Journ. Geol. Soc. vol. xxxv. (1879) p. 165. + Brit. Assoc. Rep. 1890 (Leeds Meeting), pp. 813, 814. 288 MESSRS, A, HARKER AND J. E. MARR ON type of “ mica-traps ” found at greater distances from the Shap Fell intrusion. For instance, the rounded quartz-blebs, which are found in all the Stakeley Folds rocks, occur occasionally in mica-traps as far away as Swindale, near Knock, 14 miles from the granite, although there the groundmass contains no original quartz. The Swindale intrusions, too, have here and there a crystal of felspar, either of the red or of the colourless glassy-looking kind, the edges showing the rounding already noted in our rocks. The ‘ glomero- porphyritic ” clusters of small felspars have been noticed by Mr. Late * in one of the Ingleton dykes. At the same time, the special characters of the Stakeley Folds rocks are met with more rarely at greater distances from the Shap Fell granite. From Prof. Bonney’s descriptions we gather that of the seventeen dykes examined by him (at distances of 5 to 14 miles from the granite), only one had original quartz-grains, and he adds that “ their appearance suggests the possibility of their having been caught up by the molten rock.” We do not find in published descriptions anything to compare at all closely with the above group of intrusions as a whole. Jt may be worth noting that the well-known “ porphyroide” of Mairus in the Ardennes, described by MM. de la Vallée Poussin and Renard 7, is a biotite-quartz-porphyry in which the porphyritic felspars show phenomena of rounding and bordering in some respects similar to those noticed above. The largest dyke in this part of the district is one exposed on the moorland some four or five hundred yards south of Wasdale Old Bridge. It strikes in a nearly N.W.-S.E. direction, and is remark- able for containing porphyritic crystals of monoclinic felspar (in the form of partially-interpenetrating Carlsbad twins nearly two inches long), It has also porphyritic quartz in good crystals up to 1 inch, showing prism- as well as pyramid-faces; and these occur in great numbers enclosed in the large felspars, as well as in the general mass of the rock. The felspars have a very pronounced tabular habit, parallel to the clinopinacoid, the thickness of a crystal being less than one-fifth of its length. The forms present are the usual clinopinacoid, prism, basal, and hemidome, with another form (Ak1) not determinable on the specimens. The mineral has a strong glassy lustre, and the third cleavage (parallel to the orthopinacoid) is well developed. These characters, with the tabular habit, are the chief mineralogical grounds on which sanidine is usually separated from orthoclase, and there seems to be no reason why these crystals should not be named “ sanidine.” A number of other dykes, showing in some degree a radial arrangement about the granite, are marked on the Geological Survey map, and we have examined several of these between Shap Fell and Tebay, and farther west and south. In many cases the rocks are deeply weathered, and detailed descriptions would not be very profitable. It is sufficient to note that some are ordinary quartz-porphyries ; others are normal mica-traps with no original * Brit. Assoc. Rep. 1890 (Leeds Meeting), p. 814. t Mém. couronn, Acad. Roy. Brux. (1876). THE SHAP GRANITE AND ASSOCIATED ROCKS. 289 quartz—e.g. the dyke between Crookdale and Borrowdale [1163] ; while others, again, belong to intermediate types—e. g. the dyke or series of dykes on Potter Fell [792]. In this latter rock, at a distance of six miles from Shap Fell, are found long flakes of dark mica with the blade-like habit of that noticed in some parts of the granite margin. } None of the various intrusions we have alluded to can be traced as continuous with the granite at the present surface. If we are right in regarding them as apophyses, they are in connexion, not with the visible granite mass, but with a deep-seated extension of it. The small dykes found in close proximity to the margin of the granite outcrop, such as the two seen not far west of Wasdale Head Farm, are ordinary quartz-porphyries with no special peculiarities, except that they sometimes contain little crystals of brown sphene [757], an uncommon mineral in such rocks, though it might almost be expected in any offshoot of the Shap Fell granite. The same constituent occurs sparingly in the brown-coloured rock with por- phyritic quartz and felspars which is prominent in the Blea Beck section, apparently forming an irregular sill at the summit of the Lower Limestone [882]. There is no doubt that many dykes which we have failed to observe exist to the north of Shap Fell. Such dykes would be less noticeable in the Volcanic Series than in the Silurians to the south, and the quarter-sheet of the Geological Survey map for the northern part of this district is not yet published *. Sedgwick found a quartz-porphyry not very different from that of Blea Beck in Wet Sleddale [803]. This may be a dyke or sill belonging to the inner group, connected at or near the surface with the Shap Fell granite. A large dyke belonging to the outer group of apophyses cuts the Skiddaw Slates at Goodcroft Farm, near Rossgill. It is about four miles north of the granite outcrop, towards which it bears directly. It is twenty to forty yards wide, and encloses entangled masses of indurated slate. The rock is a normal quartz- porphyry [1164}. When these apophyses are considered in conjunction with the patches of darker rock caught up in the mass of the granite, they appear to throw light upon the origin of several of the dykes which penetrate the Lower-Paleozoic rocks of the district, and which are thickly clustered in some areas, whilst they are much rarer in others. They abound within a radius of fifteen miles of the Shap granite (see Map, fig. 4, p. 290), whilst others are found in great numbers around the other granite areas. These latter are usually felsitic, whilst those more immediately in the neighbourhood of the Shap granite are both felsites and mica-traps, and, so far as we are aware, the latter are chiefly confined to the east end of the Lake District, * [Since this paper was read we have, by kind permission of the Director- General, compared our map with the six-inch MS. map in the Survey Office, and inserted several additional dykes in the country north of the granite.— March 11th, 1891.] GRANITE i: LOWER | | POST - i SILURIAN = x) OOK) x as aa ee = oe oe ae = 2s eee Sses = ~~ <-———--— ‘ound the Shap Granite. eA a a= i me ROS 6 S82 es ye Fig. 4.—Sketch-Map showing the Distribution of Dykes and Sills rs al v4 i 4 , ( HAY 4 MA . a? ‘ ’ ‘aban | , ¥ tN a 1 ' FY ae: Se = a, aS ee ee So ee WINDERMERE a il x Jatt THE SHAP GRANITE AND ASSOCIATED ROCKS, 291 and to the Cross Fell, Sedbergh, and Ingleton areas, and do not occur around the Eskdale granite, though the minette of Sale Fell and the mica-traps of Dodd are near the granite of Skiddaw. Both the felsitic and micaceous rocks have abundant porphyritic felspars in the neighbourhood of the Shap granite, as we have shown in our description of the rocks from Stakeley Folds and elsewhere, and these felspars are in every respect so similar to those of the granite itself that it seems impossible to disconnect them from that mass, especially as we find that the felspars abound near the granite contact, and become rarer as we recede from this, whilst at the same time the more distant dykes show other indications of having consolidated at a greater distance from the then deep-seated magma, as evidenced by the occurrence of vesicles in the dyke at Castle How, near Tebay, which contains few porphyritic crystals. We have already pointed out the resemblances between these dykes and sills and the dark patches included in the granite. It would seem that a magma occurred beneath the Shap granite of a more basic character than the granite itself, and that from this the micaceous dykes were sent out, whilst the more acid portion of the magma was injected into rocks at a higher level than the main source of supply to form the Shap granite and the felsite apophyses, portions of the more basic part being carried up as “clots” in the granite, consti- tuting the dark patches contained therein. In favour of this view, we may note that the evidence points to the Shap granite and the mica- traps having been formed at the same geological period, as shown by the intrusion of undoubted apophyses of the granite along with mica-traps in rocks of late Silurian age, whilst the pre-Carboni- ferous age of both the granite and the mica-traps is generally recognized. From the great abundance of mica-trap dykes it is evident that a magma like that which we suppose to have existed at a lower level than the Shap granite must have been situated at some point below this region, and the strong resemblance of the micaceous dykes of Stakeley Folds and the Gill to the patches in the granite almost certainly demonstrates that they have had a _ common source. _ The possible presence of this more extensive magma underneath the Shap granite, and at one time connected with it, is of great importance in discussing the origin of the actual granite mass. In connexion with the foregoing observations, a few remarks concerning the nature of the intrusive mass of Shap seem to be necessary, though, in the absence of any certain knowledge, our - comments must be brief. We have attempted to show that the Shap granite is merely a subsidiary offshoot of a deep-seated igneous mass of much greater extent. It is interesting to observe that just as the Shap granite occurs at the point of contact of two sets of disturbances, viz. that which has produced the normal strike of the Lake District rocks, and that which gives the beds associated with the Skiddaw Slates to the west of the village of Shap a general N.W.-S.E. strike, so the principal dykes which we have attempted to connect with the 292 MESSRS, A, HARKER AND J. E. MARR ON deeper-seated magma occur where the above-mentioned normal strike is complicated by the great fractures ranging down the Lune Valley in the neighbourhood of Tebay and Sedbergh. We are now confronted with the question, what is the nature of the Shap intrusion itself? It does not appear to be a simple — laccolite, and, on the other hand, we have seen that several sills ure protruded from it around the margin. ‘The strike of the rocks may have been bent by the granite, or, on the other hand, the intrusion may have taken place in a region where a weak place was caused, owing to the bending of the rocks. The usual greater density of the altered rocks seems to indicate some compression, though whether this and the preceding cause would be sufficient to leave space for the granitic intrusion is doubtful. On the other hand, the abnormal alteration of the rocks around a mass with so small a diameter would suggest the passage of molten matter for a considerable period through the channel which is now tilled with granite, though whether this channel ever communicated with the surface, giving rise to volcanic outbursts similar to those which occurred in regions farther north during Old Red Sandstone times, we have no evidence to show. On the whole, the phenomena presented here seem to us to be most easily explicable upon the supposition that molten matter was for a long period forced from the underlying magma through a channel which may have been “ punched out” in the way suggested by Dr. Ch. Barrois in the case of the Rostrenen granite *, and that it finally consolidated therein in the form of a “ cedar-tree” lacco- lite, 7. e¢. in a form corresponding with that of the gabbro shown in the theoretical representation drawn by Dr. A. Geikie f. Lastly, in connexion with the deep-seated magma, one naturally recalls the magnetic observations of Profs. Thorpe and Riicker = on the occurrence of a “ ridge ” in the neighbourhood of Appleby ; but when we consider the subsequent injections of more basic rock in the neighbourhood, such as the Great Whin Sill, which was also doubtless connected with deeper masses of a somewhat similar nature, we are not inclined to lay much stress upon the coincidence, § LV. Muramorpuism oF THE SuRROUNDING Rocks. A. THE ANDESITIC GROUP. The lowest rocks affected by the granite intrusion are those which may be classed as andesitic. Owing to the anticlinal folding © described above, this group occupies a considerable area of ground to the north and west of the granite, being in contact with the intrusion around one-half of its circumference. The total thickness of the andesitic lavas is probably made up of a succession of com- paratively thin flows; this is inferred from the fact that the rocks * Ann. Soe. Géol. du Nord, vol. xii. (1885) p. 105. T Trans. Roy. Soe. Edin. vol. xxxv. (1888) p. 142. t ‘Nature,’ vol. xli. p. 598 (April 24th, 1890). THE SHAP GRANITE AND ASSOCIATED ROCKS. 293 are vesicular throughout, and it is verified in one or two places where ash-bands are interbedded with the upper part of the lavas. The heterogeneous nature of the materials which constitute pyro- clastic rocks makes it, however, difficult to classify them strictly into families. The ashes included here have probably a composition not yery different from that of the andesitic lavas, and give rise by metamorphism to almost identical products, but they are not very sharply marked off from the overlying rocks which we place in the Rhyolitic Group. Examined at a distance from the granite, as, for example, in Stockdale, the andesitic lavas are found to have suffered considerable changes by the ordinary processes of weathering. These changes— the destruction of the augite, the filling of the vesicles with secondary products, and the formation of little veins of calcite and quartz— date from a time anterior to the intrusion of the granite. This fact is abundantly proved by an examination of the metamorphosed rocks, and is essential to the interpretation of their phenomena. It must be remembered that this part of the Volcanic Series was perhaps subaérial, and rocks of this character were certainly exposed to denudation while the Coniston Limestone was being accumulated. A section of the weathered andesite of Stockdale shows crowds of little felspar-prisms embedded in a pale green decomposition-product which appears to represent ophitic augite. The felspar may be referred by its extinction-angles to a variety near andesine in composition. A few larger felspar-crystals are scattered through the rock, and all are more or less turbid owing to secondary changes. The pale-green mineral polarizes in deep indigo tints, and shows the properties which seem to belong to delessite. There are none of the little bastite pseudomorphs so characteristic of decomposed rhombic pyroxenes in the hypersthene-andesites. There are apatite prisms and a few magnetite crystals of rude form, besides a certain amount of secondary magnetite-dust contained in the delessite. A little dust of calcite is also present, and a considerable amount of silica has been set free in the form of quartz. The ovoid vesicles vary in size from one-twentieth of an inch to one or even two inches. The smallest ones are filled sometimes with quartz, sometimes with the pale-green product referred to delessite, which has a regular radiate arrangement. Others are lined with this substance, and have their interior occupied by confusedly crystallized quartz. The larger ones may have detached radiate aggregates of delessite in their centre, or more irregular patches due to the breaking away of the lining from the wall of the cavity before the silica was deposited. Other vesicles, again, are partly or wholly occupied by calcite, usually in a single crystal, and this kind of amygdule may often be seen side by side with the others here mentioned. In a typical specimen of andesite taken between Wasdale Pike and Great Yarlside, where the metamorphism has been but slightly felt, Mr. Garwood finds 59-95 per cent. of silica. This confirms the 294 MESSRS. A. HARKER AND J. E. MARR ON character of the rock, and compares closely with the figures found for other Lake District andesites. Two rocks from Mr. Clifton Ward’s Falcon Crag section (near Keswick) gave 60°718 and 59°511 per cent., and one from Lingmell Beck, about two miles — north-west of Scawfell Pike, 59°151 per cent. To these we may add a large boulder at Manfield, near Darlington, doubtless from the — Lake District, in which Mr. W. F. K. Stock found 59°87 per cent. of silica *. The andesitic lavas afford some of the most beautiful and instruc- tive examples of thermo-metamorphism in the district, and as they can be followed along their line of strike from localities free from alteration into the ‘‘ contact-aureole” and up to their Junction with the granite itself, the process of transformation can be traced in all its stages. j At places at considerable distances from the junction, such as Little Saddle Crag (1350 yards), we notice that the vesicles contain a quantity of epidote in addition to delessite, quartz, and calcite [1277, 1278]. In some districts epidote has been recorded as a product of thermal metamorphism, but we cannot satisfy ourselves that in these andesites it is other than an ordinary result of weathering action. Setting it aside, the rocks at the locality in © question give no marked indication of metamorphism by the granite. Proceeding eastward, however, we find substantial alteration setting in, being first skown in the weathering-products of the original rock. A specimen taken some distance west of Wasdale Pike, and nearly 800 yards from the granite-boundary, is crowded with minute flakes of brown mica, apparently developed at the expense of the decomposition-product (delessite ?) disseminated through the weathered andesite [1205]. In the vesicles part of the delessite is altered into green hornblende, while part remains unchanged (see Pl. XI. fig. 4). A little farther east (at 750 yards) the flakes of brown mica are rather larger and more collected [1204]. The pale green product in the vesicles is still only partially altered, and a little epidote is still present, but this mineral is not found nearer to the granite. | The formation of brown mica, which, it will be seen, is the most characteristic mineral in the altered andesites, is thus the first clear result of the metamorphic action, while, almost concurrently with it, green hornblende begins to appear among the contents of the vesicles. The greatest distance at which we have verified meta- morphic action is 1150 or 1200 yards, on Low Fell, where the mica-fiakes are chiefly collected about little grains of magnetite, a frequent occurrence in such rocks [1279]. In the field the early stages of metamorphism are indicated chiefly by the vesicles, in which lustrous, greenish-black aggregates of hornblende are to be detected by the eye. Proceeding towards the contact, the hornblende becomes more distinctive, showing good cleavage-planes, while the quartz filling other vesicles takes on a * ‘Naturalist’ (1889), p. 304. Yr HE SHAP GRANITE AND ASSOCIATED ROCKS. 295 whiter and more evidently crystalline appearance, and stands out like pebbles on a weathered surface, often enclosing a kernel of hornblende. At the same time the dull grey ground of the rock becomes blacker and more compact, and often contains greenish crystalline streaks of hornblende, or more rarely a pyroxenic mineral in its place. Pyrites is of common occurrence in these streaks and in the interior of some of the vesicles. The dark colour of the rock is due to the development of mica, and nearer to the granite this mineral imparts a brown or purplish-brown sheen to the rock, and eventually becomes apparent to the eye. Near the contact the vesicles lose something of the distinctness of their external boundaries, their contents being to some extent merged in the general recrystallization of the rock. The microscope brings out more clearly the nature of the transformations undergone by these rocks. The chloritoid substance which we identify with delessite is found to have disappeared completely in the thoroughly metamorphosed specimens. It is replaced most frequently by a deep brown, intensely dichroic mica, which is disseminated through the rock in very minute flakes, with occasionally a few larger ones in clusters and patches. This mineral occurs almost universally, but is not uniformly distributed. Instead of it in some parts of the sections we find a green horn- blende, in crystalline grains of varying size, showing the prismatic cleavage-traces, and giving the usual absorption-formula : y, grass-green ; (,aslightly less deep green; a, pale yellow-brown ; v¥5B> >a. With the hornblende, or locally replacing it, is seen occasionally a green fibrous actinolite in characteristic sheaf-like bundles. These amphibole-minerals are very generally confined to streaks varying from a very narrow width to half an inch or an inch, so that a slide may show hornblende as the characteristic mineral in one half of the field and brown mica in the other. Less common is a pyroxenic mineral, colourless in thin slices and having the general characters of monoclinic augite. It occurs in well-cleaved crystal- line grains with the hornblende, and less frequently a vein is seen consisting entirely of a mosaic of crystalline augite [759]. _ Magnetite is a very common mineral, usually building minute but rather perfect octahedra. It is assuciated more frequently with the hornblende than with the brown mica, though the latter mineral sometimes encloses a few grains of magnetite also. Pyrites * is another mineral having the same association. It forms little cubical crystals or irregular patches. More remarkable is the very frequent occurrence of sphene, almost always in those parts of the altered rocks which contain hornblende. ‘The sphene is in little rounded grains or clusters of minute granules, less commonly in rather imperfect crystal forms. It exhibits unusually strong * Judging from the colours in reflected light, both pyrites and pyrrhotite occur, but it would be very difficult to isolate the little granules for examina- tion. = 296 MESSRS. A. HARKER AND J. E. MARR ON pleochroism, between purplish-brown and colourless. Apatite — occurs very rarely in little veins of quartz and mica, in a fashion which seems to indicate a metamorphic origin [798]. The above-mentioned minerals collectively make up a consider- able part of the metamorphosed andesite. ‘The remainder of the rock consists of a finely granular groundmass, the precise nature of which is less easily studied. A portion of it is quartz, but careful scrutiny detects here and there in the grains the evidence of twinning and even of twin-lamellation. It is doubtful in some cases how much of the original felspar of the andesites is preserved as such in the less metamorphosed examples. The process of recon- struction is seen, however, in some of the occasional porphyritic — felspars. One of these will be found to be studded with little flakes of brown mica and partly transformed into a granular ageregate, while enough of the original felspar-substance remains to vaguely indicate the twinning between crossed nicols {799}. In the vicinity of the granite, the whole substance of the rock is certainly transformed, and the granular aggregate in which the — coloured minerals are embedded assumes the perfectly clear ap- pearance so well seen in many “ crystalline schists.” The twinning — of the granules can be verified only occasionally, although it is evident from chemical considerations that a considerable proportion of the aggregate must consist of felspar. These highly altered rocks share with many of the products of dynamo-mctamorphism their singular immunity from subsequent secondary changes, a property which seems to require some physical explanation. It is curious to compare in a junction-slice of altered andesite and granite the fresh minerals of the former with the turbid felspars and discoloured micas of the latter. The contents of the vesicles and of certain narrow cracks posterior to the filling of the vesicles have undergone instructive transforma- tions. Itis here that the first effects of the metamorphic agent were manifested. ‘The silica is found to have recrystallized in a mosaic of clear quartz-grains, free from fluid-cavities. In the less meta- morphosed examples this process is incomplete, the central portion retaining its confused, almost cryptocrystalline, structure. In the highly altered rocks the change toa rather coarsely-granular mosaic is universal, and this at its outer boundary is not very sharply separated from the surrounding rock. The delessite is here almost constantly replaced by the green — hornblende described above, the brown mica occurring but rarely either in the vesicles or in the occasional narrow veins which represent cracks in the rock [798]. The vesicles contain horn- blende even when the surrounding rock is densely charged with flakes of mica (see Pl. XI. fig. 5). The hornblende is well- cleaved, and sometimes the whole or a large part of the mineral within one vesicle is in crystalline continuity. One slide [897] shows one of the narrow veins alluded to running straight through the section. It contains clear quartz or, in some parts of its length, hornblende. The vein traverses two vesicles, and in each case the THE SHAP GRANITE AND ASSOCIATED ROCKS, 297 portion of it within the vesicle is occupied by hornblende in crystal- line continuity with the adjacent hornblende. Little magnetite-crystals are not infrequently found in the altered vesicles, and sometimes pyrites. Sphene occurs in both the vesicles and the narrow veins, usually in round granules, occasionally in a characteristic acute-angled crystal. The metamorphic origin of all these minerals is abundantly proved by their manner of occurrence. In one instance only was a minute garnet found embedded in a quartz within the vesicle [759 ]. Felspar does not appear to have been commonly formed in the metamorphism of the contents of the vesicles. One slide only shows good crystals of that mineral, often twinned, occupying a considerable portion of some of the cavities, and accompanied by brown mica instead of the customary hornblende [1203]. This specimen was taken north of Wasdale Pike, about 400 yards from the nearest outcrop of granite. The felspar is here moulded by the mica, and occupies the marginal part of the vesicle (see Pl. XI. fig. 6). ‘ The weathered andesites before metamorphism appear to have been traversed in places by little veins of chalcedony. One slide [1205] shows such a vein, now transformed into quartz, but retain- ing the mamillated form of deposit so characteristic of chalcedonic infiltrations. Other specimens, nearer to the granite, show veins of quartz-mosaic, which may or may not represent altered chalcedony, but are evidently recrystallized during the metamorphism of the rocks. Among other inclusions, this quartz contains minute patches of brown mica with rounded outlines [1201}. In the field the altered andesites sometimes show silvery mica on planes which seem to have been joints in the rock prior to the metamorphism,—a feature observed in some other rocks near the Shap Fell intrusion. The distribution and association of the various minerals met with in the metamorphosed andesites seem to admit of a certain amount of explanation on chemical grounds, supposing that the substances formed at any point within the mass depended on the chemical composition at that point of the weathered andesite prior to the metamorphism. The phenomena described above, and especially those connected with the altered vesicles, sufficiently prove that con- siderable weathering had already taken place. It appears that the augite had been completely, and the felspars partially, destroyed, with the formation of quartz, calcite, and a chloritoid mineral as the chief secondary products. In accordance with their usual behaviour, the chloritoid substance was mainly in pseudomorphs occupying the place of the augite, but probably disseminated also through some of the larger felspars; the secondary quartz was confined mainly to the felspars; and the calcite formed granular patches or collected in veins and streaks. The vesicles were filled with quartz, or with calcite and the green product, or with the usual associations of these minerals, as already mentioned. Taking the chloritoid mineral to be delessite and the brown mica biotite, the addition of some silica and the loss of most of the Q. J. G. 8. No. 187, | ¥ 298 MESSRS, A. HARKER AND J, E. MARR ON water would be almost the only changes involved in the conversion of the former into the latter mineral, ‘and this appears to have been the usual mode of alteration in the rocks in question. To produce hornblende, however, would require the taking up of lime, as well as silica, and the distribution of this mineral in patches and streaks © in the metamorphosed rocks, and particularly within the vesicles, seems to show that its formation, instead of biotite, depended upon ~ the presence of calcite in the immediate neighbourhood of the delessite. Augite contains much more lime than hornblende, and — its mode of occurrence as a metamorphic mineral accords well with | our suggestion. The veins of pure augite [759] may be taken as © representing veins of crystalline calcite traversing the weathered andesite before its metamorphism. The mineral is identical with that to be described below as one of the most abundant silicates in metamorphosed rocks of the calcareous group, a is presumably a variety rich in lime. With respect to the sphene, it is not easy to say in what form the titanic acid existed before the metamorphism. Imenite does not appear to be a common constituent of the original andesites, though it occurs rather abundantly, with secondary translucent sphene, in some of the ashy beds [766]. The sphene, a lime- bearing mineral, naturally occurs in association with the hornblende and colourless augite rather than with the mica, but the titanic acid may have been distributed uniformly through the weathered andesite and be now partly contained in the last-named mineral. The biotite of Miask is known to have 4°73 per cent. of titanic acid *, and this substance is beginning to be recognized as a wide- spread constituent of the brown rock-forming micas. The magnetite, again, is mostly found in association with horn- blende, but it is possible that the other parts of the rock contain as much iron, which is there incorporated as part of the brown mica. The flakes of the latter mineral are too minute to allow of any precise study which might determine whether they should be referred to biotite, haughtonite, or lepidomelane. The strongly pleochroic brown mica of thermo- metamorphic rocks (Hornfels, &e.) is usually stated to be biotite, but we shall allude to this point again below. The specific gravity of a Hisliy an ceeeRphOAat vesicular andesite from near the northern border of the granite was found to be 2-800, the figures for the non-metamorphosed rock in Stockdale being 2°736. We shall see that the metamorphism of the rocks around the Shap granite is in general accompanied, as in this case, by a condensation of bulk. In this place we may most conveniently notice the metamorphism of certain ashes, fine agglomerates, &c., which are closely associated with the andesitic lavas. The fragmental volcanic rocks, having usually a heterogeneous constitution, do not admit of any very strict * R. Schlipfer, ‘Rech. sur la compos. des micas et des chlorites,’ Schaffhausen (1889). See also Koch, Zeitschr. deutsch. Bool, Gesellsch, vol. xli. (1889) p. 165 ; W. M. Hutchings, Geol. Mag. (1890) pp. 272, 273. J THE SHAP GRANITE AND ASSOCIATED ROCKS, 299 classification into rhyolitic, andesitic, &c. ; but the rocks in question seem to be made up largely of pyroclastic materials of the same general nature as the associated andesites, and give rise when meta- morphosed to very similar results. They are probably on the whole somewhat more “ acid ” in composition, since they frequently enclose rhyolite-fragments broken up by the explosive outbursts by which the accumulations were produced, and mingled with the fine dust, fractured crystals, and andesitic material. The non-metamorphosed rocks of this type may be studied in Stockdale, Wet Sleddale, &c., where the influence of the intrusion has not been perceptibly felt. The recognizable fragments are partly crystals, partly pieces of lava. The crystals are similar to those which occur porphyritically in the andesites themselves. They very frequently lie with their length nearly at right angles to the lamination of the finer matrix, indicating that they have been dropped into their place [895]. This appears to be a characteristic feature of pyroclastic rocks, especially those accumulated on land, and affords a useful criterion in other districts where ashes and lavas, chiefly of acid type, have been altered (not by thermo-meta- morphism) almost beyond recognition. Some of the lava-fragments are of andesite, showing the usual densely-packed felspar-prisms, and occasionally enclosing small vesi- cles [875]. Others are of rhyolite, as already remarked. The matrix of the mass is usually a finely-divided clastic material. Its lamination is emphasized by the development along it of a pale- yellowish or colourless sericitic substance which winds past the enclosed fragments, and imparts a “schist ”-like appearance to the sections. Crystals, fragments, and matrix have undergone the ordinary weathering processes, with the production of secondary quartz, the usual pale-green product, a little magnetite dust, and some calcite, which is more uniformly distributed than in the weathered andesite lavas. Metamorphosed representatives of these rocks, which we may term andesitic agglomerates and ashes, are met with intercalated among the lavas at various horizons. The minerals produced are in general those already described in the metamorphosed andesites. Mica is the commonest of the coloured constituents. It is usually of the highly pleochroic variety already noticed, giving a very deep, rather greenish-brown colour for vibrations parallel to the cleavage- traces. Sometimes it has less intense absorption, and is apparently partly bleached [875]; or again, it is partially decomposed, giving a green colour with secondary dust of magnetite. There is, however, a rather different type of mica seen in some of the slides, having a more ruddy brown colour and giving : 6 and y, chestnut-brown ; a, nearly colourless. This mica, when partially decomposed, loses its cleavage and some of its pleochroism. Its characters would seem to indicate a variety having a different chemical composition from the former, but although the two types usually occur separately, they are in some » a” 300 MESSRS. A. HARKER AND J. E. MARR ON cases associated in the same flake. Hornblende and actinolite both occur with the same characters as in the metamorphosed an- desites [902 and 875], associated with one another, and in one instance with colourless augite [902]; but here these minerals are much less abundant than the mica, which is commonly the only ferro-magnesian mineral present in the slides. Hornblende, how- ever, occurs as usual in the metamorphosed vesicles of enclosed andesite-fragments [875]. Magnetite is found in octahedra and less perfect forms [796], but it is less abundant than in the metamorphosed andesites, and is often wanting in those specimens most rich in mica [896, ete.]. Sphene has not been observed. These facts accord with the sug- gestions offered above ; the titanic acid and most of the iron oxides contained in the rocks have been incorporated in the brown micas. The remainder of the rock is a granular aggregate resembling that seen in the metamorphosed andesitic lavas, though not quite so fine-grained. Twinning and twin-lamellation are to be observed between crossed nicols, and it is evident that a large part of the rock consists of reconstituted felspar. This is brought out also by a certain amount of turbidity in the felspars, distinguishing them from the clear quartz, which they sometimes mould. It is noticeable in these metamorphosed andesitic rocks that the originally fragmental examples show signs of subsequent weathering which are not found in the associated lavas, The embedded felspar-crystals have been replaced by an aggregate of new felspar and quartz, with more or less brown mica, and ex- ceptionally a considerable quantity of yellow epidote [900]. In the less metamorphosed examples the original twinning can be vaguely discerned; in specimens taken close to the granite-junction the structure is totally destroyed, and the pseudomorphs are recognized merely as areas poorer in mica than the surrounding rock. The character of the metamorphosed andesitic ashes and agelo- merates is sufficiently indicated by the foregoing remarks. It only remains to be said that there is often a marked laminated structure well indicated by the parallel disposition of the flakes of mica [797, 896], and increasing the general resemblance of these highly altered rocks to true crystalline schists. In the thermo-metamorphism of rocks in the vicinity of an igneous mass, it is an important question how far the total chemical com- position has been modified by the changes produced. To give a satisfactory solution of this question would demand a detailed — chemical investigation. With respect to the andesitic rocks of Shap Fell, Mr. Garwood has examined for us specimens of highly altered andesite and ash from near the northern margin of the granite, and finds that they contain only 50°75 and 50-90 per cent. of silica respectively ; 2. ¢. 9 per cent. less than the non-metamor- phosed andesite. This apparent loss of silica is a fact for which we are unable to offer any explanation. There is no strati- graphical reason to suppose that the specimens analysed differed in their original composition from normal augite-andesites such as those THE SHAP GRANITE AND ASSOCIATED ROCKS. 301 of Stockdale, but they probably do not occupy quite the same horizon. Thermo-metamorphism in andesitic rocks has hitherto received but little attention. Prof. Judd* has adverted very briefly to some changes of this kind in the andesites or “‘propylites” of the Western Isles of Scotland. He alludes to the formation in the contact-zones of colourless secondary pyroxene, magnetite, and deep brown biotite, with possibly melilite and felspar; but we do not gather that these phenomena are exhibited on any extensive scale. As regards the metamorphism by heat of augitic rocks in general, the first important record is that of Mr. Allport +, who showed that in the neighbourhood of the Cornish granites the augite of the ““ oreenstones ” has been replaced by hornblende and actinolite. It appears from his description, and from the only Cornish examples we have examined, that there, as in the Shap Fell andesites, much at least of the augite must have been converted into secondary minerals before the metamorphism [1129]. Prof. Lossent, however, describes appearances in the metamorphosed diabases in the Harz, which leave no doubt as to the direct “ uralitization ” of augite under the influence of a granitic intrusion ; and a series of slides from specimens taken near Rosstrappe, Thale, one of his typical localities, show that hornblende has been formed both directly from augite and also from its decomposition-products [469-473]. The officers of the Geological Survey of Saxony § have described the conversion of diabases into actinolite- and anthophyllite-schists around the syenite of Meissen. B. THE RHAYOLITIC Rocks. The rhyolites of the district, whether associated with or underlying the Coniston Limestone, have characters familiar to geologists who are acquainted with Ordovician volcanic rocks in other parts of. Britain, and we do not propose to euter into many details with respect to their general features. Moreover, owing to their com- paratively simple chémical and mineralogical constitution, they do not present such diversities in their modes of metamorphism as have been described in the case of the andesites. In the field, indeed, the rhyolites seem to show little or no change, as they are traced along their strike into the aureole of metamorphism; but this idea is dispelled by a closer study of the specimens. At a distance from the granite, the rhyolites may be studied in Stockdale and Long Sleddale. They often have a grey colour with a rather flinty appearance ; when this is wanting, they are pink or cream-coloured, but always of compact texture. One type is lami- nated parallel to its flow-lines, and often has a fissile structure in * Quart. Journ. Geol. Soc. vol. xlvi. (1890) p. 370. t Ibid. vol. xxxii. (1876) p. 418. t ‘Erlaut. zur geol. Specialk. Preuss.,’ Blatt Harzgerode (1882), pp. 79, &e. § ‘Erlaut. zur Specialk. d. Konigr. Sachsen’ (1889), K. Dalmar, Section Tanneberg, Blatt 64; A. Sauer, Section Meissen, Blatt 48. 302 MESSRS. A, HARKER AND J. BE. MARR ON the same direction. Another is coarsely nodular, the spheroidal nodules varying from an inch to a foot in diameter. Such nodular rhyolites are well known in other districts, and have been discussed by one of us in the case of the Ordovician lavas of Caernarvonshire *. The alterations there described in what appear to have been giant spherulites, and in particular their partial and total replacement by eryptocrystalline silica or quartz, are exhibited on a magnificent scale on Great Yarlside and at other localities in our district. The peculiarity is not confined to true lava-flows; for an apparently intrusive rock in Blea Beck plantation, near Shap Wells, contains good silicified spheroids. The rhyolites are never notably porphyritic, resembling in this and other respects the corresponding rocks in North Wales. Indeed the microscope shows that much of the material of the rocks was but very imperfectly individualized into felspar and quartz, pre- senting rather the features which are referred by many English petrologists to devitrification. Vesicles are found in these rocks only rarely, and they are usually of microscopic size. An idea of the chemical composition of the rhyolites may be gathered from Mr. Garwood’s analyses given below ; the figures for two Caernarvonshire rocks are quoted for comparison. It will also be seen from columns I. and II. that, whatever metamorphism has operated in the rhyolite near the granite-contact, it has not materially affected the bulk-analysis of the rock. The specific gravity of the specimen (I.) analysed is 2°608, which agrees exactly with that of similar rocks from North Wales. One of the most metamorphosed rhyolites from Wasdale Head gave 2-623, showing no great difference. £ E. EET. Ly. oS AM epee 75°95 76:95 74°88 avo CS aaa 137% 15°50 12°00 9°7 ee 3°48 2-C0 3°50 61 a ee not estim. not estim. 0:20 not estim. 1 6 ee trace. wee 1-28 A on aoe 0:25 1:05 0:34 si Na,O ut 2 r 2-49 0-3 K,O oO § 6°55 4°50 477 58 (Ignition) } 2s 1-20 0-4 100-00 100-00 100°66 99-8 I. Spherulitic Rhyolite, Stockdale ; anal. E. J. Garwood. II. Nodular Rhyolite, close to granite, near Wasdale Head Farm; anal. E. J. Garwood. III. Rhyolite, Pitt’s Head, 2} miles S.W. of Snowdon; anal. J. Hughes, Trans. Roy. Ir. Acad. vol. xxiii. (1859) p. 615. IV. Rhyolite, Cwm-silyn, above Nantlle Valley ; anal. E. Hamilton Acton and J. T. Hewitt; ‘ Bala Vole. Ser. of Caern.’ (1889) p. 138. A very characteristic rock is that which forms the lower part of the Coniston Limestone rhyolite at Stockdale and Long Sleddale. It is of the laminated, fissile variety, the overlying rock being * «The Bala Volcanic Series of Caernarvonshire ’ (1889), chap. iii. THE SHAP GRANITE AND ASSOCIATED ROCKs. 303 nodular,—an arrangement noted in some other localities also. ‘This laminated rock has a typical microspherulitic structure, being almost entirely built of densely-packed minute spherulites, each of which gives a distinct black cross when a section is examined between Beonsed nicols [861]. In some places the growth, instead of being centric, is linear, and then follows the lines of flow. Slighter dif- ferences in structure in different parts of the slide also follow the fluxion-lines. A beautiful figure of this rock, showing the micro- spherulitic and perlitic structures, has been given by Mr. Teall *, and the same rock has been described and figured by Mr. Rutley T The latter author has expressed the opinion that the ise structure is here an effect of devitrification subsequent to the perlitic cracking; but we are unable to see that he has given any reasons for this view. The practice of assigning a secondary origin to special structures in the older acid lavas has perhaps been pushed to excess in some quarters. In the Westmorland rhyolites there are traces of perlitic fissures traversing rocks which are now microcrystalline, and other appearances pointing to the alteration of an originally glassy mass; but we find nothing to suggest that the spherulitic and allied structures are of formation posterior to the consolidation of the lava; and the breaking up of the vitreous material of the rocks examined seems to have “been in many cases a chemical, not merely a molecular change. In addition to occasional small crystals of quartz and felspar— mostly plagioclase—the only original minerals found in these rhyo- litic lavas are scattered magnetite-crystals, and very rarely prisms of apatite [802]. Probably a little angite or biotite formed part of the original rocks, but a few scraps of the usual pale-green decom- position-product are the only thing to indicate the former presence of these minerals. Another secondary constituent is a yellowish- brown filmy mineral, like sericitic mica, which usually occupies perlitic cracks. A frequent type of alteration in the rhyolites, shown in many of © our specimens, is what we may conveniently term ‘ silicification.” The groundmass of rocks so affected presents a finely crystalline appearance, and consists mainly of quartz in a fine-grained mosaic, passing in irregularly disposed patches into a rather coarser grain. Included crystals of felspar are frequently pseudomorphed by a similar quartz-mosaic, and the process is sometimes made very evi- dent by portions of the twinned crystals remaining still unaltered [802]. The occurrence of these silicified rhyolites shows no relation to the proximity of the granite, and we do not ascribe it to meta- morphism by the agency of the intrusion. Identical phenomena are observed in various localities in Caernarvonshire remote from any igneous intrusion, and Miss Raisin + has suggested a “ percola- * «British Petrography’ (1888), pl. xxxvili. t+ Quart. Journ. Geol. Soc. vol. xl. (1884) p. 345, pl. xviii. fig. 6; Mem. Geol. Surv. (1885), ‘The Felsitic Lavas of England and W ales,’ p. 12, pl. il. fig. 1. “t Quart. Journ. Geol. Soc.-yol. xlv. (1889) p. 267 304 MESSRS, A. HARKER AND J. E. MARR ON tion of heated waters carrying silica in solution” during a “ Solfatara- stage ” which may have marked the decline of Ordovician vulcanicity in the area. Whether this explanation hold good or not, it is difficult to believe that the alteration observed in some of these rocks could be effected except by the introduction of silica in some manner ; and this addition of silica from without probably explains the high percentages of that substance found in some published analyses of rhyolites. Passing on to the thermo-metamorphism of the rhyolites, we find a few points worth recording. Specimens taken north and east of the spot marked “ Tunnel,” at distances of about 600 or 700 yards from the margin of the granite, have suffered some alteration of the groundmass, which is in places of a microcrystalline texture, showing felspar as well as quartz. This is apparently quite reconstituted, but curving perlitic cracks are still clearly evident throughout the mass, marked out by micaceous films. The rock here encloses small por- phyritic felspars, which are either quite unaltered or partly silicified, as mentioned above. One specimen has numerous vesicles, which are filled by crystallized quartz, partly idiomorphic; and there is no evidence that this quartz has recrystallized under metamorphic action [801]. Near Wasdale Head Farm the rhyolite may be examined close to its junction with the granite, and here more distinct evidences of metamorphism are obtained. Some specimens show a microcrys- talline aggregate of recognizable clear felspar and quartz, similar to that noticed in the metamorphosed andesites, and leaving no doubt that the whole has been reconstituted by metamorphic agency [907]. Other examples seem to have been silicified prior to the intrusion of the granite, and the quartz which forms most of their bulk cannot be stated with certainty to have recrystallized during the metamorphism [880, 881]. The same is true of the quartz- veins which traverse some of the slides, the quartz in them often showing partial crystal contours. Besides quartz and felspar, these metamorphosed rhyolites have minute flakes of pleochroic brown mica and some colourless mica giving brilliant interference-colours. Larger flakes of brown mica occur, grouped in a fashion which suggests their derivation from the pale-green decomposition-product seen in some of the non-metamorphosed rhyolites. e) agree with anatase, and there is but little doubt of the identity of the mineral. Further, it seems to be formed at the expense of rutile, for Mr. Hutchings points out that the ‘“ clayslate-needles,” which he finds in the less metamorphosed flags lower down the beck, are here almost absent. The locality of the specimen is apparently about 800 or 900 yards from the granite-outcrop.—March 11th, 1891.] Still approaching the granite, we come on to the spotted or “knotted” rocks. A specimen taken at 500 yards from the contact does not materially differ, except as to the spots, from the last, the re- crystallized mosaic, in which some rectangular sections clearly point to felspar, being visible only in some portions of the slide, while the THE SHAP GRANITE AND ASSOCIATED ROCKS. 319 rest is very obscure. The rather irregular spots, 1, to 5\,; inch in diameter, are differentiated by their comparative freedom from mica 1219]. 7 At 60 yards from the granite the spots are more regularly ovoid and their boundary more sharply defined, the brown mica in the interspaces forming distinct small flakes arranged tangentially to the outlines, The central part of each spot contains smaller flakes, often rather rounded, but the marginal zone is free from mica [1218]. (See Pl. XII. fig.5). In some specimens from this neigh- bourhood the mica in the general body of the rock has a marked parallel arrangement, which corresponds to the lamination of the original flags [864]. An example from Packhouse Hill has less mica, and that of a pale colour, but here pyrrhotite is exceptionally plentiful, and has presumably used up most of the iron which has elsewhere gone into the usual brown pleochroic mica [1222]. This is at 600 yards from the granite. A specimen from Collyrag Quarry, a hundred yards nearer, shows similar characters [1079]. These rocks are on a slightly higher horizon than the preceding. They show little or no indication of “ spots,” have rather abundant clastic quartz, and present a considerable resemblance to the Upper Coldwell beds exposed farther south. The normal brown mica of the metamorphosed flags resembles in general characters that which has been produced in the andesitic and other rocks described above. Such mica has a special quality as seen in reflected light, which gives a peculiar purplish-brown sheen to the rocks in which it is abundant. With this goes a very intense pleo- chroism in thin sections, the absorption being almost complete for vibrations parallel to the cleavage-traces, while, if the nicol be turned a very little away from this position, a distinctly greenish-brown colour is seen. Similar characters have been described by various writers in the mica of *‘ contact ” rocks in other districts, and it would be interesting to ascertain whether the mineral is chemically different from the brown micas of igneous rocks. The only investigation we can find on this point is in Lang’s brilliant paper on the Christiania district, already referred to above. He and Jannasch separated and carefully analysed the brown mica of a Ghimmerhornfels in that district. They found it to contain 7-98 per cent. of magnesia and 21:94 of ferrous oxide, ferric oxide being entirely absent: also titanic acid occurs to the extent: of 3:40 percent *. Except for the absence of ferric oxide the figures differ but little from Schlapfer’s analysis of the biotite of Miask. The last-named author has shown that the earlier analyses of micas leave much to be desired in point of accuracy. The nature of the “spots” in such rocks as these is not an easy question, and it seems clear from the literature of the subject that the phenomena of spotted and knotted slate-rocks arise in several different ways. We find nothing of the local accumulation of the “‘ pigment” of the rock into spots, which characterizes the * Op. cit. p. 318. 9? 320 MESSRS. A. HARKER AND J. E. MARR ON outer ring of metamorphism (Knotenthonschiefer) in some districts of argillaceous strata, such as Rosenbusch’s * Steiger Schiefer. The spots in our rocks are comparatively free from coloured constituents, and do not make their appearance until after considerable develop- ment of secondary minerals. The original pigment of organic matter is dissipated as the first result of metamorphism. More- over, the spots in the Brathay Flags, when best developed, show a distinctly crystalline structure between crossed nicols, being evi- dently imperfect crystals charged with a large quantity of foreign inclusions. The ovoid form seems to be that of imperfectly formed crystals, for the general ground of each spot extinguishes parallel to the long axis of the irregular oval, which has no universal direction, but lies quite at random. Possibly the mineral may be andalusite. Apart from the spots, it will be noticed that andalusite is absent, as well as other characteristic aluminous “ contact-minerals.” It would appear that the rock contained sufficient alkalies to build up a large part of the alumina present into secondary felspars. The veins of white mica have already been mentioned. The rock adjacent to these shows some curious modifications, being built in great measure of a clear colourless mica similar to that occupying the veins [949, 1080]. This mica is partly in minutely- matted aggregates, but mostly in well-defined flakes with rough parallelism, moulded by a clear crystalline mass of grains, some of which show felspar-twinning. The brilliantly polarizing flakes are also moulded by another micaceous-looking mineral with a very pale greenish-grey colour and feeble dichroism, but not sensibly bi- refringent. This may be one of the ripidolite group, but we have not established its identity with any described variety. A little magnetite and granules of the supposed pyrrhotite occur. Brown mica is only sparingly associated with the white in this marginal modification of the rock. At about an inch from the actual vein, however, the spotted character of the rock is apparent, and brown mica occurs as usual in the interspaces between the spots. There is still a considerable amount of white mica, mostly in exceedingly minute scales ‘within the spots, but partly in more conspicuous flakes near their margin. One slide [949] shows a crack running at right angles to the main vein, its course marked by a slightly coarser aggregate of colourless mica and quartz, with some clear | felspar, a little of the ripidolite-like mineral, and occasional grains of yellow-brown tourmaline, touched here and there with blue. Of the remaining members of the Silurian formation we have made no systematic examination, but the few specimens studied offer some points worth recording. The Upper Coldwell beds and the Iess caleareous portion of the Middle Coldwell, viz. the lowest strata of that division exposed at Packhouse Hill, bear a general resemblance to the uppermost beds of the Brathay Flags at the ‘same locality, and the resemblance is borne out by the microscope. Numerous minute shreds of a mineral like tremolite disseminated through these metamorphosed flags perhaps point to a certain * “Abh. zur geol. Specialk. v. Elsass-Lothr.’ vol. i. (1877) part 2. THE SHAP GRANITE AND ASSOCIATED ROCKS. 321 amount of carbonates of lime and magnesia in the original rock. The metamorphism is evidently incomplete, and the clastic grains of quartz show no change. The Coniston Grit and the Lower Coldwell beds (or grits in the Coniston Flags) resemble one another very closely. They are ordinary grauwacke grits. A specimen of the former, taken near Stakeley Folds, shows in a section subangular grains of quartz and felspar with some interstitial dusty matter like kaolin, and little patches of finely granular calcite. The felspar has minute twin- lamellation and is rather abundant, though subordinate to the quartz. ‘There is no other clastic element except very rarely a flake of white mica. The rock is freely veined with quartz [1165]. For comparison we take a specimen of the Lower Coldwell grit at Packhouse Hill, about 600 yards from the granite-contact. The contrast is evident in hand-specimens, the “metamorphosed rock showing the vitreous appearance of a quartzite, in which the granular structure is only faintly discerned. Under the microscope [1223] we see a mosaic of quartz and felspar, the irregular grains of which show the ‘“sutural” junction characteristic of crystal- lization in situ. It is not easy to judge of the proportion of felspar present, since the grains are all perfectly clear, and twinning is rarely seen. The twinning is never compound, and the grains showing it give rather lower polarization-colours than the average, which seems to point to orthoclase. Besides these minerals there are numerous little rounded brightly-polarizing granules, colourless or very faint yellow, and precisely similar to those so commonly seen in the metamorphosed Coniston Limestones. We regard these asa lime-augite. The graaules are aggregated together, especially in irregular vein-like streaks. The slide shows also some small irregu- larly-shaped granular patches, so densely packed as to be opaque, and appearing yellow in reflected light. These are, at least in part, of the same pyroxenic mineral, which corresponds closely in its distribution with the calcareous decomposition-product in the non- metamorphosed grit. It is noteworthy that neither mica nor garnet has been found. The kaolin seems to have gone with the carbonates to form pyroxene. (See Pl. XII. fig. 6.) The quartz in this rock encloses many irregularly-grouped minute cavities, round or more frequently shapeless, with bubbles of various relative size. Judging by the apparent relief of cavities and bubbles, both glass- and fluid-pores may be represented, but no movement was verified in any of the bubbles. The calcareous Middle Coldwell beds, as. seen on the top of Pack- house Hill, exhibit a high degree of metamorphism. As in the Coniston Limestone, this is shown especially by the development of lime-bearing silicates, and, although we have no analyses of these Silurian strata, a comparison with the general character of the unaltered beds makes it appear that a very moderate proportion of calcareous matter, which would not cause a field-geologist to describe the rocks as limestones, is sufficient to make the meta- — morphism follow this line. 322 MESSRS, A. HARKER AND J, E. MARR ON Probably more than one lime-silicate is present. The dominant one gives the interference-colours of a pyroxene, and has marked cleavage-traces, parallel to which it extinguishes. This may be referred with some doubt to wollastonite. It is partly collected in crystalline patches and streaks, but smaller granules of the same or a similar mineral make up a large part of the rock, in conjunction with a clear substance polarizing in grey tints and occasionally showing the twinning of felspar. In the pyroxenic patches occur grains of a yellow opaque mineral, probably pyrrhotite. Here and there among the pyroxene is seen a little grain of calcite, showing that here, at 600 yards from the granite, the elimination of the carbonic acid is not quite complete [1225]. In hand-specimens this rock has a compact homogeneous appear- ance, with a pinkish-grey or pale violet colour, and a hardness rather less than that of orthoclase. The specific gravity of an average specimen is 2°874, which agrees with the identification of the chief constituent as wollastonite. The pale violet colour figures frequently in descriptions of foreign lime-silicate rocks. Specimens of the Middle Coldwells taken at a point 8.S.W. of Wasdale Old Bridge show a similar compact porcellanous appear- ance, but with a light grey colour, They resemble very closely the Upper Coniston Limestone of Wasdale Head, but have a rather higher density, 2°899, owing, as the microscope shows, to a larger proportion of pyroxene. ‘The dominant mineral here is the colour- less lime-augite, which is largely developed, in crystal-plates en- closing the felspar, &c. in ophitic fashion [1306, 1307]. At this locality, about 460 yards from the probable outcrop of the granite, there is no longer any trace of calcite remaining. It would appear that, in these impure calcareous rocks, the particular lime-silicates produced vary from point to point, as determined, perhaps, by com- paratively slight differences in the chemical composition of the mass. Some light is thrown on the conditions governing the formation of augite, wollastonite, &c., by Vogt’s* interesting researches on slags. An interesting feature in the Packhouse Hill section is a meta- morphosed fault-breccia, which intervenes between the Lower and Middle Coldwell beds. The lowest beds seen here in the Middle division are ordinary flags, but there appears to have been a lower calcareous band similar to that described above, for fragments of the characteristic pale-violet rock occur in the breccia, mingled with pieces of the dark flags and vitrified-looking fragments of the under- lying grit. The whole is united by a greenish finely-crystalline cement of pyroxene. The fragments of grit appear in sections as a mosaic of clear crystal-grains of quartz and felspar, evidently of metamorphic formation. It is impossible to estimate the proportions of the two minerals, but a fair number of the grains show twinning and seem from their properties to be orthoclase. No repeated twinning is observed [1286, 1287]. Among the grains of the mosaic, and * Arch. f. Math. og Naturvidensk. vol. xiii. (1890) pp. 34-71, Christiania. THE SHAP GRANITE AND ASSOCIATED ROCKS. o2p enclosed by them, are patches of rounded granules, highly refringent and birefringent, which must be referred to a pyroxene, probably the lime-augite already frequently alluded to. ‘These granules are mainly collected at the margin of the fragments or in the neigh- bourhood of little vein-like cracks. The angular pieces of flag show a marked lamination defined by streaks of opaque dust. Their metamorphism is similar to that of the corresponding rocks in situ, except where the fragments are traversed by cracks and veinlets, which evidently represent a per- meation by carbonate of lime and other substances. In these places a number of special minerals may be detected, lime-silicates pre- dominating. ‘The usual colourless augite is abundant in irregular crystalline patches, often accompanied by clear felspar and probably quartz. There are also streaks composed entirely of a minutely matted aggregate of rather fibrous tremolite [1285]. Near these there is frequently a pale yellow-brown pleochroic mica, in clusters of small flakes. A pyrites mineral occurs among the tremolite and felspar, and by its colour would be assigned to pyrrhotite. The metamorphosed fragments of the more calcareous flags in the breccia generally show a finely granular mass, mostly polarizing in bright tints, but too minute to be precisely determined. The general character of the mass may, however, be inferred from those constituents which are here and there developed in larger crystalline patches. Of these the most usual is colourless augite, readily identified by its cleavage, extinction-angles, and interference-colours. Another conspicuous mineral is light brown, pleochroic sphene, which occurs plentifully in grains and good crystals (habit, n, c, y) scattered through the fragments. — The cementing material of the breccia is almost exclusively colourless augite, building a relatively coarse-grained crystalline ageregate, and enclosing plenty of little sphene crystals [1286]. This cement makes up on the whole a small part of the mass, and it, . with the smailer veins traversing the fragments, clearly represents a calcareous infiltration filling the interstices of the original fault- breccia. No calcite now remains. In conclusion we may note one or two points with reference to the metamorphism of the Shap Fell rocks as a whole. The pro- duction of new minerals is confined to distances of not much more than 1200 or 1300 yards from the granite-contact, or about equal to the mean semidiameter of the intrusive mass itself as exposed at the surface. The width of the metamorphic aureole, as thus defined, seems to be tolerably uniform in different directions from the granite. Moreover, this extreme limit of metamorphic action is very nearly the same, whether we consider the andesitic rocks, the rhyolitic ashes, the various calcareous strata, or the Brathay Flags. _ Within the metamorphic aureole the changes increase in degree 324 MESSRS, A. HARKER AND J. E. MARR ON as we approach the granite, and, with few exceptions, the rocks in the vicinity of the contact have been completely reconstituted. Our results, however, lead to the conclusion that any division of the aureole into distinct rings or zones would be arbitrary and arti- ficial, and certainly could not be made to apply alike to the various kinds of rocks metamorphosed. In the andesites, for example, the transition from the least altered to the most altered types is so gradual that no lines of division can be drawn either in the field or by minute examination. In the rhyolitic ashes our descriptions show two different types, but the distinction of these two would probably resolve itself into one of degree rather than of kind, if it were possible to examine the rocks ‘between 300 and 600 ‘yards from the granite, between which limits we have found no exposures, As to the calcareous beds, these have been described in other districts as showing a very complete alteration to points even beyond the limit of the aureole in the associated slates, though with no grada- tions in metamorphism within those limits. But, although we find in our calcareous rocks a high degree of metamorphism extending to a considerable distance from the actual contact, this seems, so far as we can judge from the rocks exposed, to die away gradually to the boundary of the aureole. The flags in the Shap district are not well enough exposed to warrant any sweeping conclusions, but it would be difficult to draw any divisional line in those seen within the metamorphic region. Zones of metamorphism may perhaps be usefully laid down in certain cases, as, for instance, when a mineral like chiastolite is developed in the outer part of the aureole and disappears in the inner; but such divisions do not appear practicable in the Shap Fell district. It is noticeable that the chemical effects of the metamorphism were first produced in those constituents of the rock which owed their origin to weathering, decomposition, &c., such as ‘delessite, calcite, and carbonaceous matter. In other words, the substances which had been formed under normal atmospheric conditions were the least stable when subjected to the high temperature which accompanied the intrusion of the granite. The minerals of direct igneous origin in the volcanic rocks were less susceptible to thermal metamorphism, and the original quartz-sand in the flags proved especially refractory. The several minerals detected in the various metamorphosed rocks as products of the metamorphism are summarized in the table given below. The absence or rarity of some characteristic ‘‘ con- tact-minerals” of other districts is rather striking. Some of these are products which probably require special ‘“ mineralizing agents ”’ to co-operate in their manufacture; such as fluorite, tourmaline, lithionite, and axinite; but the almost complete absence of anda- lusite, staurolite, and garnets (other than lime-garnets) is more remarkable. In the table the occurrence of the minerals in the different rocks studied is marked by an asterisk (*). Parentheses ( ) indicate rarity or occurrence only under special conditions, ¢.g. in the vicinity of veins. mineral. THE SHAP GRANITE AND ASSOCIATED ROCKS. 325 A query (?) indicates some doubt as to the identity of the The more doubtful ones, such as the possible andalusite in the spotted flags, are omitted altogether ; as are also minerals, like epidote, of which the metamorphic origin is not satisfactorily established. Table showing the distribution of Minerals of Metamorphic n o p=) : iy 2] Minerals. Wit.) R 9 o = ~~ ~ —_ — mn Nn o 5) Gs MG! ee: q =< ROMMMED hoc ae to Bec cae cud % % Mrtnoclane’. 6.0 Boo ee. x * Plagioclase (various) ...| * x Origin in the chief rocks exanined. Colourless Mica (Musco- vite) Brown near Ripidolite ? et 8 Green Hornblende ...... ae eee ee ee ee ey Mica (probably Fiotite) Fo. 80.2... sant 196 ee ry Ltt a * PEPEXHOMUG 5c caste es cio hs Colourless Augite (pro- bably Wollas rich in lime)...... ¥ MOWEDS 0. eee atti PGDERAOD, Fibs Bonen ost lek Grossularite (ene 1 pte REE Sei EEE (*) Essonite (isotropic) ...... Pe Common Garnet ......... (?) Mourmialine: ...8..9. 000242. dis RI ETIG fae chic: ac Ben ec Seenic? “ Ee ee (Saeaeeeen WAALS oac0n vec occecnazss UL a Cee (*) Cpa So ics Betas passer Fo. wet Pe ae | * * BIMGONHC... si oBeduwcd.sacs Wigeat "es Pyrites and Pyrrhotite...; * ... rapiite! BU owsesa ses. awe a aa | Rhyolites. hes) Oe ea ee & | Rhyolitie Ashes. * OK KS | Lower Limestone. one. + c Limes Upper * (x) ; Brathay Flags. - *K ow OK * * OK | | Mid. Coldwell beds and Fault-Brececia. %* OK XK 326 MESSRS. A. HARKER AND J. E, MARR UN EXPLANATION OF PLATES X., XI., & XII. PuaTE X. Map illustrating the relations of the Shap Granite and associated rocks. (The figures are all drawn in natural light, and, except P1. XII. fig. 5, are magnified 20 diameters. The numbers in brackets [ | refer to the slides.) Prats XI. Fig. 1 [902]. Shap Fell granite, normal type; showing clear quartz, turbid felspar, and flakes of brown mica. A flake near the lower right-hand aoe contains a small zircon surrounded by a strongly plsdeatete order. Fig. 2 [399]. Dark patch in Shap Fell granite; showing quartz, felspar, and mica, as before, but the last more plentiful; also grains of sphene, octahedra of magnetite, and little needles of apatite. See p. 281. Fig. 3 [1281]. Special modification of Shap Fell granite, containing andalusite ; not found in place. The portion of the slide figured is rich in anda- lusite, which forms imperfect prismatic crystals, coated with brown mica and enclosing magnetite, zircon, mica, &c. Around some of the inclusions, especially zircons, are pleochroic halos, changing from bright yellow to colourless. The bulk of the rock is a mosaic of felspar and quartz with abundant crystals of magnetite and occasional apatite. See p. 283. Fig. 4 [1205]. Metamorphosed vesicular andesite, near Wasdale Pike, about 800 yards from the granite. The upper half shows a vein of chal- cedony converted into quartz. The lower half shows a vesicle in which the delessite (represented dark for distinctness) has been partly replaced by green hornblende. The clear mineral in the lower part of the vesicle is quartz. This rock represents an early stage of meta- morphism. See p. 294. ; Fig. 5 [897]. Metamorphosed vesicular andesite, Wasdale Pike, about 500 yards from the granite ; showing the groundmass of the rock converted into a fine-grained aggregate of brown mica, felspar, quartz, and magnetite. Within the vesicle is green hornblende instead of mica. A patch of granular sphene is seen at the lower edge of the figure, on the line of a small crack. See p. 296. Fig.%6 [1203]. Metamorphosed vesicular andesite, north of Wasdale Pike, about 400 yards from the granite ; showing an unusual type of altera- tion, brown mica and felspar (in relatively large crystals) being formed in the interior of the vesicles, as well as in the groundmass, See p. 297. Prats XIT. Fig. 1 [1169]. Idocrase-garnet-rock in the metamorphosed Lower Coniston Limestone, Wasdale Head, about 100 yards from the granite; show- ing dodecahedra of grossularite garnet embedded in ophitic erystals of idocrase. Both minerals contain granular pyroxene and other matter, and the idocrase encloses groups of small needle-like crystals. See POLL Fig. 2 [909], Ovoid nest of colourless lime-augite, bordered by a zone of felspar erystals, in the metamorphosed Calcareous Breccia of the Upper Coniston Limestone, Wasdale Head, about 250 yards from the granite. Two quartz grains, of clastic origin, are seen in the lower part of the figure. Flakes of brown mica cluster round these and round the augite-felspar nest. See p. 314. Fig. 3 [1215]. Tremolite-rock in the metamorphosed Caleareous Breccia at the same locality. See p. 314. Quart. Journ.Geol.Soc.Vol. XLVI1 .PI X. RXBONIFEROUS. stom Grits Zz Coldwell (flags) s Le Colawell (calc. ) 5 or Colwell (grits) ai ay Flags. n a ‘ Pa Quart. Journ.Geol.Soc.Vol. XLVI .P1 X. ’ : 8 $ 8 AS K Ss ae : i NVIDIAOGHYO NVISN TIS aria Granite. -) Lower Gonistow Lonestone Upper Coniston Limestone witty Caleareous Brecciu. Upper Coldwell (flags) Middle Coldwell (c ti) Lower Glowell (grits) | Brathay Flags. c ee CARBONIFEROUS. 4 2%| Suis and yr Dykes. MAP ittustratinc THE RELATIONS of tHESHAP GRAN ITE anpb THE yee] Se — le) i op) < O (@) {tp) Seed o 9 Ie oa < Yr = 2 § ae = Z 2 & Hn SD =O) oes Z 6 = az wl eK < O re) a x : ox Wasdale Hendi o 100 200 300 400 500 yards One Mice oy oa Ay y RE Ret ie \y en ve. aS = wo +, Ey EL FEE WS a tee en Wy, \ on : 2 , ML \ Ul Mintern Bros. hth. Bie rkcer. del. : SECTIONS OF SHAP GRANITE AND METAMORPHOSED ANDESITES Mintern Bros. lth. F METAMORPHOSED ROCKS NEAR THE SHAP GRANITE mmc lLLONS O a THE SHAP GRANITE AND ASSOCIATED ROCKS, 32 Fig. 4 [873]. Lime-augite-rock in the metamorphosed Upper Coniston Lime- stone at the same locality ; showing colourless augite, both in crystal- line aggregates with good cleavage and in granules and granular patches. See pp. 314-315. Fig. 5 [1218]. Spotted schist in the metamorphosed Brathay Flags, north of Wasdale Beck, about 360 yards from the granite; showing little spots comparatively free from the secondary brown mica. This figure is magnified 10U diameters. See p. 319, Fig. 6 [1223]. Quartzite, with colourless lime-augite, in the metamorphosed Lower Coldwell beds (grits) just nerth of Packhouse Hill, about 580 yards from the granite. The pyroxene occurs in distinct rounded granules and in finely granular patches. See p.321. Discussion. Prof. Bonney said that it was almost impossible to discuss a paper of such wide bearings, but it appeared to him to be one of great value. The most important points were the clear demonstration of the occurrence of felspar as a product of contact-metamorphism, and the effects of the intrusion of an igneous mass on pyroclastic rocks. He mentioned some cases which illustrated the uncertainty as to what mineral might be produced by contact-metamorphism. Prof. Le Neve Foster said that the point which struck him as a miner, with reference to the intrusive boss of granite, was the absence of tin-ore. The Authors had remarked that no axinite or fluorspar had been found, and that tourmaline was very rare. It was interesting to note that where there was a lack of minerals containing boron and fluorine there was a complete absence of cassiterite. Mr. Srrawan asked for an explanation of the connexion referred to between the intrusion of the granite and the Pennine movements, by which were usually meant post-Carboniferous movements. The map exhibited of a necessity showed the dykes in a diagrammatic form. He enquired if this radial arrangement with reference to the granite would appear if they were shown on a true scale. In’ the Cautley neighbourhood micro-granites of the same age as the Shap granite occurred as sills in Coniston Limestone, and were cut across by mica-trap dykes, which seemed to show that the more basic rock was later than the more acid, and not earlier as argued from inclusions in the Shap granite. Mr. Rurtey thought that the red felstone-like dykes (often more or less micaceous) were probably apophyses of the Shap granite, as indicated by the Authors. With regard to the dark micaceous dykes being in any way related to the highly micaceous inclusions met with in the granite, he felt considerable doubt. Judging from the alterations produced artificially in rhyolitic rocks by heat, he was inclined to believe that the temperature under which the alterations in the rhyolites had been effected in the Shap area was a compara- tively low one. The occurrence of the peculiar polysynthetic structure in the garnets which the Authors described was, he believed, the first notice of such a structure in British garnets, since, hitherto, it had only been observed in ouwarowite and in the 3238 THE SHAP GRANITE AND ASSOCIATED ROCKS. garnets of one or two localities in Saxony. The paper appeared to be one of exceptional interest and value. Mr. Barrow was much interested in the Authors’ list of minerals developed by contact-metamorphism. The light they had thrown on the origin of cyanite was particularly valuable to geologists working in the Central Highlands, where cyanite schist occurs on a large scale. In one instance a broad belt of this schist follows the outcrop of an igneous gneiss for some miles in such a manner as to suggest contact-metamorphism. The crystals of cyanite show little or no signs of deformation, and if developed by contact-meta. morphism seem to point to the conclusion that the igneous rock originally consolidated as a gneiss. Mr. Marr, in reply, recapitulated the reasons which had caused the Authors to connect granite, felsites, and mica-traps alike with the existence of a deep-seated magma, without asserting which por- tions of this were first consolidated. The movements in the Pennine Chain to which they had referred were those pre-Carboniferous ones which affected only the Lower Paleozoic rocks. Though the map of dykes exhibited was necessarily diagrammatic, the directions of those dykes which they had not themselves examined were taken from the published maps of the Geological Survey. He believed that the metamorphism produced by the granite might throw some light upon the changes which had occurred in the rocks of a“ regionally metamorphosed” area. The Authors had attempted to show that the Shap-granite intrusion was connected with earth-movements.. If such movements had taken place to a greater extent, dynamic metamorphism would doubtless have altered the granite, the dykes, and the various sedimentary and vol- canic rocks, but the pre-existing contact-metamorphism might still remain as a factor in the process of regional metamorphism. Mr. Harker remarked that although new-formed felspar occurs in the most metamorphosed types of all the rocks studied, the minuteness of its grains and their pellucid appearance render it in many cases difficult to distinguish from quartz. Cyanite as a ** contact-mineral”’ had been recorded by Lossen in the Harz. Mr. Tratt and Dr. Harcn also spoke. THE CAMBRIAN SERLES IN N.W. CAERNARVONSHIRE. 829 18. On the Tower Limit of the CamBrian Sertes in N.W. CaeRNARVON- soire. By Carnertne A. Raisty, B.Sc. (Read February 25, 1891.) (Communicated by Prof. T. G. Bonnry, D.Sc., F.R.S.) ConTENTSs. Page Rea NE PMUATDIOT! 5, So BAe occ phi asubya cad enlivevase rete 329 (a) Lithological Character of the Strata ................sceeeees 329 (6) Unconformity below the Cambrian Conglomerate ...... 330 WG) AaeOniie OL OULCTODS: <.21.. Bl xi. fig. 3, Q.J.G.8. No. 187. 28 346 MR. R. LYDEKKER ON A LABYRINTHODONT SKULL that the two forms are generically identical. In describing the ver- tebral skeleton of Jchthyerpetum, Prof. Huxley remarks that, in comparison with the other Jarrow forms, ‘it is more pisciform”; and in noticing the impressions that appear to have been formed by the ventral scutes, he observes that ‘‘the ventral surface of the trunk presents numerous minute, more or less parallel ridges, pointed at each end, and taking a general course obliquely down- wards and forwards to the middle line.” In describing Pholido- gaster pisciformis, the Professor first of all states that he proposes the name on account of its “ tish-like form”; and in treating of the ventral scutes or scales, observes that ‘‘ they are seen to be some- what oat-shaped,” and are “so arranged as to form oblique series, directed inwards and forwards, and meeting in the middle line.” When we recollect that the term ‘‘inwards” is precisely equivalent to ‘downwards ” in these cases, it will be apparent that the de- scription of Jchthyerpetum might be transferred to Pholidogaster, or vice versé; and | am thus strongly inclined to believe that the Jatter is founded upon a larger species of the former. Additional evidence in favour of this opinion is afforded by Prof. Huxley’s description of the skull of Pholidogaster. Were he observes that “in front of the symphysis of the mandible, the under surface of the premavxilla is visible, bearing the stumps of two teeth. These teeth are situated at some distance (about 0-7 of an inch) from the middle line, and pass outside the ramus of the mandible. They are conical, and round in transverse section..... The bases of the teeth are marked by strong longitudinal grooves.” Remembering that we have three in place of two premaxillary teeth remaining in the skull of Jchthyerpetum hibernicum, this description will apply word for word. Further evidence is, however, afforded by the skull of Dendrer- petum, where, according to Dr. Fritsch’s figure*, the premaxillary teeth are enlarged, separated by a diastema in the middle line, deeply fluted at their base, and, judging from the outward inclina- tion, apparently biting outside the mandible. In this connexion I find that Sir J. W. DawsonT long ago pointed out that Pholidogaster was closely allied to Dendrerpetum ; and it is to my own mind not at all improbable that they may prove to be identical. All these lines of evidence point, therefore, very clearly to the conclusion that Jchthyerpetum and Pholidogaster are identical, and also suggest that they may be inseparable from Dendrerpetum, which is the earliest of the three names. I refrain only with hesitation from adopting the name Pholidogaster in place of the later Ichthyer- petum ; but in view of the possibility that both these may prove to be synonyms of Dendrerpetum, I have taken the course of referring the specimen to the genus to which I absolutely know that it belongs. I may avail myself of the opportunity of observing that any objec- * ‘Fauna der Gaskohle,’ vol. ii. pl. xlix. fig. 1. } ‘ Air-Breathers of the Coal Period’ (1863), pp. 22, 23. FROM THE KILKENNY COAL-MEASURES, 347 tion that might be taken against the identification of Jchthyerpetum and Pholidogaster on the ground that, while the type species of the former occurs in the Upper, that of the latter was obtained from the Lower Carboniferous, is nullified by the circumstance that both Loxomma and Anthracosaurus (and apparently the same species of each genus) range through the whole of the Carboniferous system. Further, if these two Labyrinthodonts be really generically iden- tical, it will be obvious that the vague suggestion of the rhachito- mous nature of the vertebral column of Pholidogaster, made in the British-Museum Catalogue, at once falls to the ground. Finally, taking it as proved that Pholidogaster and Ichthyerpetum are closely allied to the so-called ‘‘ Brachyopina,” we now have evidence that a type of Labyrinthodonts common throughout the European Carboniferous (and unknown there after the base of the Permian) was represented in the Lower Gondwanas (? Upper Per- mian of India) by the genus Brachyops, while, as we go farther east- wards, we find it surviving in the Hawkesbury beds of Australia (which are of somewhat later age), where it is represented by Bothriceps ; a member of the latter genus, together with the allied Micropholis, also occurring in the great Karoo system of South Africa, some portion of which is probably the equivalent of the Hawkesbury beds. This seems, therefore, to be another instance of the persistence of types in the Indian, Australian, and Ethiopian regions duriug jong ages after their total disappearance from the Palvearctic area. 348 MR. J. W. GREGORY ON THE 20. The Tupor Specimen of Eozoon. By J. W. Grecory, Esq., F.G.8., F.Z.S., of the British Museum (Nat. Hist.). (Read March 11, 1891.) Tue unanimity with which the view of the organic origin of Hozoon canadense was received on its announcement in 1865 by Sir William Logan, Dr.(afterwards Sir J. W.) Dawson, and Dr. W. B. Carpenter *, in the Quarterly Journal of this Society, was first seriously broken by the publication, in the succeeding volume, of the memoir by Professors King and Rowney “On the so-called Eozoonal Rock” 7. In the following year the Quarterly Journal contained a series of '“ Notes on Fossils recently obtained from the Laurentian Rocks of Canada, and on objections to the organic nature of Eozoon,” by Dr. (afterwards Sir J. W.) Dawson #, the most valuable contribution in which was the description of a specimen found by Mr. H. G,. Vennor in a limestone belonging to the Hastings series at Tudor, Hastings county, Ontario. This was identified as Hozoon canadense, though as possibly a new variety, by Sir J. W. Dawson, who seemed to consider that this discovery relieved him of the necessity of making any detailed reply to the arguments of his critics, as ‘‘ furnishing a conclusive answer to all those objections to the organic nature of EHozoon which have been founded on comparisons of its structures with the forms of fibrous, dendritic, or concretionary minerals—objections which, however plausible in the case of highly crystalline rocks, in which organic remains may be simulated by merely mineral appearances readily confounded with them, are wholly inapplicable to the present specimen ” §. The importance of the new discovery depended on the fact that all the previously known specimens of Hozoon consisted of aggrega- tions of calcareous with serpentinous minerals, a fact upon which ereat stress had been laid by the objectors to its organic origin. But it was claimed by Sir J. W. Dawson that the Tudor specimen was a true Hozoon preserved in limestone alone. The claim was not altogether a new one, as Sir J. W. Dawson had previously discovered Hozoon in the Madoc limestone, and had emphasized the value of this point in a letter which was published by Dr. Car- penter among his ‘ Supplemental Notes’ ||. But, as was admitted 4 in the memoir on the Tudor specimen, Sir J. W. Dawson “did not then venture to describe as a fossil” this very imperfect fragment, * W. EK. Logan, ‘On the Occurrence of Organic Remains in the Laurentian Rocks of Canada,’ Quart. Journ. Geol. Soe. vol. xxi. (1865) pp. 45-50; J. W. Dawson, ‘On the Structure of certain Organic Remains in the Laurentian Limestones of Canada,’ op, cit. pp. 51-59, pls. vi. & vii.; W. B. Carpenter, ‘ Addi- tional Note on the Structure and Affinities of Zozoén canadense,’ op. cit. pp. 59- 66, pls. viii. & ix. t Quart. Journ. Geol. Soe. vol. xxii. (1866) pp. 185-218, pls, xiv. & xv. t Dhid. vol. xxiii. (1867) pp. 257-265, pls. x, & xii. § Tbid. pp. 257-258. || ‘Supplemental Notes on the Structure and Affinities of Zozoon canadense,’ Quart. Journ. Geol. Soe. vol. xxii. (1866) p. 228, “| Zbid. vol. xxiii. (1867) p. 261. TUDOR SPECIMEN OF EOZOON. 349 and but for the discovery of better material probably no value would have been attached to it by other writers. But as the disbelievers in Hozoon were as ready then as now to admit that the production of such a specimen would at once settle the whole controversy and conclusively establish its organic origin, the claims based by Sir J. W. Dawson on the Tudor specimen had an enormous influence in con- firming geologists in their acceptance of the supposed Laurentian fossil. The value attached to the preservation of Hozoon in a lime- stone may be illustrated by the following quotation from Sir War- ington Smyth’s Presidential Address to the Geological Society in the year of the publication of the Tudor memoir :—*“ The elaborate argu- ments of Messrs. King and Rowney in favour of the mineral origin of ‘EKozoonal’ structure had at one time a strong show of support in the fact that these appearances were always observed in serpen- tinous limestones (ophicalcites) only... .. But the announcement made by Dr. Carpenter, in the Quart. Journ. Geol. Soc. for August last, of Dr. Dawson’s discovery of Hozoon preserved in carbonate of lime pure and simple would appear to close the discussion ’’*. Sir J. W. Dawson’s view apparently is that the specimen consists of a slab of Hozoon 6 inches long by +4 inches wide and 2 lines in thickness, broken off at right angles to the septa. Profs. King and Rowney f subsequently pointed out the improbability of so large and thin a slab being thus formed transverse to the lamine, and suggested that the calcite veins were merely produced by infiltration into a series of fissures or cracks. But as this opinion was only based on second-hand information no great weight seems ever to have been attached to it, and from that time onwards the Tudor specimen has always remained the great obstacle to the acceptance of the mineralogical explanation of the structure of Hozoon. Thus Prof. Moebius, after an examination of the specimen, told Mr. C. D. Sherborn and myself that it alone had ever suggested to him doubts as to the truth of his conclusions. Prof. Nicholson, moreover, allows me to say that he has always felt that if the Tudor specimen should exhibit the characteristic canal-system of Hozoon, it would afford a strong presumption in favour of the view that Hozoon is organic; and that after an exami- nation of the specimen he saw nothing that would justify the assertion that its nature was organic, or even that the specimen was one of Hozoon at all. The specimen having recently Pe aii sent to England for examina- tion by a committee “which had arranged to work through the enormous mass of Evzoonal material collected by the late Dr. W. B. Carpenter, I have had the opportunity for a careful study of it. For this | am indebted to Dr. P. H. Carpenter, F.R.S.; I must also express my best thanks to Dr. R. A. C. Selwyn, C.M.G., for his kindness in allowing a further section to be prepared, which has been skilfully cut by Mr. Ryley. Dr. Selwyn has moreover allowed * Op. cit. Proc. p lxiv. ' t ‘On Lozoon cunadense,’ Proc. Roy. Irish Acad. vol. x. (1870) p. 511. 350 MR. J. W. GREGORY ON THE the specimen to remain for some time in England, and thus several specialists, including Profs. Moebius and Nicholson, have been enabled to examine it. Considering, therefore, the importance that has been attached to this specimen throughout the whole controversy, it has been thought that a redescription may be of value. The note upon it is sent to this Society, as the plates illustrating the specimen were first published in its Journal. (a) General Form.—As reference to the original figure * will show, the specimen consists of a series of narrow white bands of calcite separated, though often imperfectly, by bands of a darker-coloured limestone, which are often continuous with the matrix on either side. The former, which are neither as numerous or regular in the specimen as in the figure given in the Journal, are wider at one end than the other, and thus mark off a somewhat clavate-shaped area of the slab. The view taken by Sir J. W. Dawson is that the white bands of calcite form the original ‘“ intermediate skeleton ” of an Hozoon colony, while the darker layers between them represent the ** body-cavities ” filled up by the material that forms the mass of the limestone. (b) The Microscopic Structure of the Rock.—The rock itself is a calc-mica-schist in which the remains of the bedding-planes are recognizable though obscure+. When examined under the microscope the cleavage is seen to be due to the development of crystals of a white mica, while the dark colour arises from minute particles of graphite scattered irregularly throughout. A few small quartz- fragments, which may represent original sand-grains and patches of a grey calcareous mineral (probably dolomite), are also to be noted. Mr, Teall has very kindly examined the slide; he observes that it reminds him much of some of the Assynt limestone which has been altered by contact with granite, and, as he accepts the micas as authigenous, there can be no doubt that the rock is a true schist. The abundance of graphite gives it a resemblance to the grey ** cipolinos ” of the St. Gothard, but it is less altered than these. (c) The * Eozoonal Bands.”-—The ‘* Hozoon” is preserved on the surface of a slab of the calc-mica-schist, and an examination with the naked eye shows that three sets of structure traverse the rock. The cleavage is parallel to the face of the slab, and this it was that Sir J. W. Dawson regarded as “the plane of stratification,” a view which an examination of a transverse section clearly shows to be untenable. The true bedding-planes cross those of cleavage at a fairly high angle and run parallel to the obliquely-truncated upper margin ; their traces on the sides are intensely crumpled and con- torted. They can be clearly seen on the upper surface, but are not shown in the original figure. They are, however, well marked in a * Quart. Journ. Geol. Soc. vol. xxiii. (1867) pl. xi.; reprinted in Amer. Journ, Sci. ser. 2, vol. xlvi. (1868) pl. i. ; Dawson, ‘The Dawn of Life’ (1875), pp. 111, &e. s) The rock is a limestone, as Sir he W. Dawson has stated; but on pe examination it is seen to belong to the schistose-micaceous variety termed by petrologists a cale-mica-schist, ———————— TUDOR SPECIMEN OF EOZOON. 351 photograph, a copy of which has kindly been presented by Mr. H. B. Woodward to the British Museum (Nat. Hist.). The third set of structures are the white calcite-bands for which an organic origin is claimed. These consist of veins of crystalline calcite, which rarely extend to a depth of more than 54, inch. ‘heir relations to the rest of the rock are very irregular. They may end off abruptly or break up into slender ramifications, which are sometimes connected by other calcite-bands developed along the cleavages, so that a reti- culate series results (see fig. 4, facing p. 354). The boundaries between the “‘ Eozoonal’”’ layers and the normal calc-mica-schist are excessively irregular, as is seen in fig. 3; there is no ‘ proper wall” to be seen at the junction—at least, I have been unable to recognize even such traces as might have been expected had the pores been obliterated by the infiltration of calcite, as has been suggested *. Nor does the evidence for the canals seem more satisfactory. Sir J. W. Dawson figured f as such a series of carbonaceous inclusions, and though by the kindness of Dr. Selwyn I have been enabled to study the original slide, I fail to see any reason for regarding them as the infillings of organic canals. There seems no essential difference between the graphitic bodies in the matrix and those in the calcite, though the latter are as a rule more minute in size. In their irregularity of form and arrangement they seem to be very different from the canals in any known foraminifer. Sir J. W. Dawson’s own figure, magnified though itis 120 diameters, fails to carry conviction of the origin which he assigns to them. After a careful examination of all the slides and figures, and con- sideration of Sir J. W. Dawson’s interpretation, 1 must contess myself absolutely unable to recognize in the specimen any trace of the “* proper wall,” *‘ canals,” or ‘‘ stolon passages ” which are claimed to occur in Hozoon t, or any reasons for regarding the calcite bands as the “‘ intermediate skeleton ” of a foraminifer. There are points in Sir J. W. Dawson’s figure which might pass as “ stolon passages,” but these appear very different in the photograph, and the specimen agrees with the latter. But the case against the organic origin of the Tudor specimen does not rest on negative evidence alone: in addition there seems plenty of positive proof against this view. The circumstance that while the rock has been intensely cleaved and crumpled the twin lamiuz and the planes of crystalline cleavage in the calcite-bands are not bent, suggests that the bands are of secondary origin; and this seems to be conclusively established by the fact that the bedding- planes can be often traced right across the specimen, traversing the limestone in the supposed body-cavities, and broken only by the calcite-layers (see fig. 1). This is faintly indicated in the photograph * Quart. Journ. Geol. Soc. vol. xxiii. (1867) p. 259. t Ibid. pl. xii. fig. 1. t But it should be noted that Dr. Carpenter abandoned the view of the organic origin of the ‘ proper wall’: see Whitney & Wadsworth, ‘The Azoic System and its proposed subdivisions,’ Bull. Mus. Comp. Zool. (Harvard) vol. vii. (1884) pp. 535-536, and J. W. Gregory, ‘Science Gossip,’ vol. xxiii, (1887) p. 103, oon MR. J. W. GREGORY ON THE already referred to, though as the bedding-planes do not come out clearly till the surface be wetted, they are not well shown. The explanation of the relations of the bedding-planes to the calcite- bands on the organic hypothesis is beset by three difficulties: Ist, — it requires that this thin strip of Kozoon should have been buried vertically, which, considering its proportions, is not very probable ; 2nd, had it been covered by the deposition of a caleareous mud we might have expected this to have been piled up around it, so that the stratification should not meet it with such regularity; and 3rd, had the body-chambers been filled by the washing in of the lime- stone material, any planes of bedding would have accommodated themselves to the irregularities of the cavities, and the bedding would hardly have been continued so exactly in the same straight line across all the layers traversed by any plane. It seems to the writer quite impossible to account for the continuity of the bedding- planes across the specimen, except on the view that the calcite-bands have been formed later than the dark layers of amorphous carbonate of lime between them. The fact that the cleavage-planes are not continued across the calcite and that no mica has been developed in this would alone be sufficient in the minds of many geologists to settle the relative ages of the crystalline and amorphous parts of the limestone, But it cannot be expected that those who regard the foliation of the pre- Cambrian schists as an original structure in the rocks will attach much weight to this argument. (d) The Origin of the Calcite-bands.—It now remains to be con-- sidered what explanation of the origin of the crystalline calcite can be offered without the assistance of any organic agency. Profs, King and Rowney suggested that they were formed by the infil- tration of calcite into a series of cracks, but this is not an adequate explanation. In some cases a small patch of the cale-mica-schist can be seen completely surrounded by the calcite, and this alone is sufficient to overthrow the fissure hypothesis ; moreover, the remark- able irregularity of the junction of the crystalline and amorphous carbonate of lime, the distribution of the graphite particles in the former, and the absence in it of any banding, are all difficult of explanation on this view. It seems more probable that the calcite-bands were formed by the solution of the limestone and its redeposition along the lines on which the water percolated through the rock. To explain why these curved bands were formed is probably impossible without a knowledge of the position which the slab occupied when i situ. The whole surface of the specimen has been slightly altered to a depth of half an inch. The irregularity and apparent capriciousness of the action by which the bands were formed are too well known for any serious objection to be raised to this explanation while the field relations of the specimen are unknown. In most cases the solvent has acted along the lines of weakness and started from the weathered surface. As a rule, the secondary calcite has been deve- loped along the bedding-planes, but at times some has also been TUDOR SPECIMEN OF EOZOON, 353 formed along the lines of cleavage. In one case a thin vein extends across the whole thickness of the slab along a line that was possibly a true crack. In one or two places the solvent acted along a kind of minute pipe, so that the calcite appears as a circular patch entirely surrounded by the calc-mica-schist. Some of the bands have the whole thickness formed of one crystal, but in other cases they are occupied by a mosaic, the separate constituents of which have the characteristic irregular polygonal outlines. There is a further convincing proof of the later origin of the erystalline calcite where projections of the cale-mica-schist extend into the calcite ; this has been deposited in such intimate connexion with these patches that a certain amount of crystalline continuity has been established, and the cleavage-planes pass uninterruptedly from the one to the other (fig. 3). Further, the regular distribution of the graphite seems to show that the transparent calcite was formed by the solution and recrystallization of the schist, as the inclusions appear to have been primary impurities rather than to have originated as infiltrations. In conclusion, it should be pointed out that no opinion is here expressed as to the nature and origin of the other types of Hozoon. It is only maintained that as the Tudor specimen lacks all the structures (except the mere alternation of irregular layers of different composition) which caused the typical Hozoon to be regarded as organic, the argument based upon it to the effect that all these structures have been preserved in calcite alone is not sustained by a further examination of the specimen. But if it still be contended that Sir J. W. Dawson has rightly identified this Tudor specimen, then Hozoon is not organic, as in this case it is due to secondary alterations produced long after the consolidation of the limestone, and eveu after the metamorphic action which converted it into a ealc-mica-schist. (e) Stratigraphical Position of the “ Hastings serves.”—In regard — to the stratigraphical position of the specimen it should be remarked that the Hastings series, including the Tudor limestone, cannot now be regarded as Lower Laurentian, to which horizon it was assigned by Sir Wm. Logan in 1867 *. The detailed mapping of Mr. H. G. Vennor f has conclusively disproved this opinion and established the correlation of these beds with the Grenville series, and thus led to the abandonment of the term ‘“‘ Hastings series” as applied to a separate group >. It is quite possible, as Dr. Lawson § has : ‘On new Specimens of Kozoon, Quart. Journ. Geol. Soe. vol. xxiii. (1867) ». 254. ' t ‘Report of Mr. H. G. Vennor on Hastings County,’ Rep. Progress Geol. Surv. Canada, 1866-69 (Montreal, 1870), pp. 143-171 ; H. G. Vennor, ‘Reports of Surveys in the counties of Renfrew, Pontiac, and Ottawa, &c.,’ op. cit. 1876- 77 (1878), pp. 244-320. { Op. cit. 1876~-77, p. 256. ; § A. C. Lawson, ‘The Archean Geology of the Region North-west of Lake Superior,’ p. 86 of the ‘Etudes sur les Schistes cristallins,’ of which separate copies were issued at the London Session of the Congrés géologique inter- national, but have not yet been published. Ss ee 304 MR. J. W. GREGORY ON THE suggested, that the limestones may be inclusions in the intrusive gneisses. But, at any rate, the rock itself is post-Laurentian and is now included by Messrs. Selwyn* and Vennor in the Huronian, while the bands of crystalline calcite to which it owes its fame may have been formed at any time between the cleavage of the rock and the discovery of the specimen by the officers of the Canadian Geological Survey. EXPLANATION OF FIGURES. Fig. 1, The lower part of the Tudor specimen ; nat. size. The part in the upper portion of the left-hand side of the specimen shows the unweathered surface of the slab with the lines of structure (regarded as bedding- planes). ‘To the right of this is the weathered surface containing the white ‘ Eozoonal’ bands (); the latter are seen to interrupt the bedding- planes (as at the lower 0), to be developed along them (as at d), or to terminate abruptly against them (asate), This can best be seen when the specimen is moistened. Fig. 2. Section transverse to the ‘ Eozoonal’ bands along the right-hand margin of fig. 1; x2diam. In the lower part the schistosity is recognizable, but in the layer containing the ‘ Eozoonal’ bands it has been obli- terated ; the general form of the ‘ Kozoonal’ bands in cross-section is shown (but see fig. 3); one of the bands is continued across the specimen along a crack. Fig. 3. One of the ‘ Eozoonal’ bands of the same slide as fig. 2 (x30 diam., reduced 2), showing their relations to the weathered cale-mica-schist : b is one of the ‘ Eozoonal’ bands, and 0’ b' are parts of the two next bands; they show the irregularity of the junction of the erystalline calcite with the rock, parts of the latter projecting into the former; in places the cleavage of the calcite is continued across the inclusions of the rock (as at a). In addition to the particles of graphite (g), the matrix contains crystals of mica (mm), but the foliation has been obliterated, Fig. 4. Another part of the same slide, showing above the weathered surface with the ‘ Eozoonal’ bands, and below the normal calc-mica-schist. One of the ‘ Eozoonal’ bands (0) is continued down across the slide (0'), and in places (as 0?) branches along the planes of sehistosity. Discussion. Mr. Surrporn remarked that the Society must be congratulated on having at last the Tudor specimen upon the table. It was for- tunate, too, that the specimen had been examined by Mr. Gregory. He agreed with Professor Moebius that the Tudor specimen had nothing whatever to do with the Kozoonal structure, beyond the fact that both had somewhat parallel layers of calcite. He had no hesitation in saying that, had Dr. W. B. Carpenter scen the Tudor specimen at the time of the publication of the first account of it, he could not have put his name to the statements, The Presipent regretted that no paleontologist was present to break a lance in support of the organic origin of Hozoon. He had * R. A.C. Selwyn, ‘Report of Observations on the Stratigraphy of the Quebec Group,’ Rep. Progress Geol. Surv. Canada, 1877-78 (Montreal, 1879), p. 144. ai) (To face p. 354. x 30 diam. (red. 2). Natural size. x 30 diam. (red. #). TUDOR SPECIMEN OF EOZOON. 355 examined the Tudor specimen, and could not recognize in it the Kozoonal structure with which he was familiar in the ordinary spe- cimens. At the same time the specimen, though by no means so distinct as the published plate of it represented, had a curiously organic-looking aspect; and he thought it probable that the palseon- tologists would not surrender it to the mineralogists without a more vigorous struggle than had been attempted that evening. Prof. Buakn, having satisfied himself of the inorganic nature of Kozoon at St. Pierre, had been to Tudor without obtaining any light on the specimen. The crystallization of the material and its passing into the cracks might be due to subsequent miueralization, and the evidence of bedding passing across the specimen was doubtful, but the specimen itself was perfectly convincing. He was sure that Sir J. W. Dawson never expected the so-called “ fossil” would die out like cracks within 3 inch of the flat surface, which was quite incompatible with its being organic. The AvrHor, in reply to Prof. Blake, recapitulated the evidence in favour of the bedding-planes, and stated that several objections to Sir J. W. Dawson’s explanation of the thinness of the specimen were stated in the paper. ‘The supposed definite margin of the specimen was due to the fact that the calcite-bands had only been developed on the weathered surface of the slab. Every English paleontologist who had examined the specimen and slides agreed that it was inorganic, | | ' | 356 MR. A, STRAHAN ON A PHOSPHATIC 21. On a Puospwatic Cuatk uth BrLEMNITELLA QUADRATA at Tartow. By A. Srrawan, Esq., M.A.,F.G.S. (Read March 25, 1891.) [Communicated by permission of the Director-General of the Geological Survey. ] WuiLE engaged in the course of my duties in arranging the specimens collected for the Rock Collection in the Museum of Practical Geology, my attenticn was attracted. by a chalk of unusual character. The specimen was collected by John Rhodes, when the second edition of the ‘‘ Geology of London” (Memoirs of the Geological Survey) was in preparation, from an old pit at the Lodge of Taplow Court, the seat of W. H. Grenfell, Esq. By perniission of the Director-General, I examined the pit, and made the following descending section :— : Lodge Pit, Tuplow Court. 5) ae Sort wihiterenalk, Op TNOb SEAN vivakscsc=s orp esas -k-nnmeniess Oadaner gence 12 0 passing down into Brown chalk with Ostrea acutirostris, Nilss., Belemnitella quadrata, Defrance (both abundant), Echinocorys vulgaris, Breyn., and Cidaris sceptrifera, Detrance, abouts. ..++:s.ss»2s suche satel eee 8 0 White chalk traversed by numerous tubes and cavities filled with | brown chalk ; a hard and blocky top, forming a marked floor to ihe'bcown chalk hove: sfc... -:sascc-0scde-ccdoraicoss ot ae eam 3 0 White chalk, mostly inaccessible, and not examined in detail ...... 14 0 White chalk with scattered brown grains. .......2.0ces0.0-cecausnsouuee 2 6 passing down into Brewin Ghalk, BOG ti. ncs sasesncecsinnscad'atukonseaneeace pede ose Ee See 4 0 Hard crystalline chalk with nodular structure and greenish markings (like Chalk Rock), about: . 2... sassack son.cesneatateene neon 1 0 White chalk, piped with brown chalk as above ....................006 L SPC DEO WN Laver j..o4.; <5. nq 0ne sedecanerun cet augach seeeth eae enone 0 Gg White chalk, piped with brown chalk ..........ccccccc0ccc0ccee ceeeeeees 2 6 White chalk to the bottom of the pit, the first flints occurring 12 feet below the lower band of brown chalk .................sse0005 15 O+ At the top of the pit there occurs a thin streak of intermingled red clay and green sand, with a few green-coated flints, which has clearly been washed but a few feet down from the Tertiary outlier on which Taplow is situated. ‘he true base of the Tertiary strata lies probably not more than 10 feet above the highest chalk seen in the pit. The dip in the pit and along the hillside northwards is about E. 10°8. at 4° or 5°. In the memoir referred to (vol. 1. 77) a correlation of this chalk with the well-known Margate Chalk, or the zone of Marsupites, is suggested on the evidence of the occurrence of Belemnitella quadrata and the scarcity of flints. The total thickness of Upper Chalk existing in the neighbourhood is believed to be between 250 and 300 feet. [A shaft, which has been sunk in the hillside above the pit since the reading of this paper, gives the exact depth of the phosphatic CHALK WITH BELEMNITELLA QUADRATA, 357 bands below the baso of the Tertiary strata. The following mea- surements were made by Mr. Lodge, and kindly supplied to me by Mr. Grenfell.—May, 1891.] ft. in. Reading tee. ainpes/iks wa dan CAR MAUR UAAG Dee eRA MRA How eis ngekien 7 10 Beds, | Green sand with green-coated flints................4. 2 6 8 EO a A a 18 0 P pewaisn White CHALE ccctseisare-stqrcnevesanusessys cs 2 0 Upper j} Brown chalk [upper phosphatic band] ... ........ IO Chalk. ita White Ghee. cisarsss Ser ciaieds vexesbacdeddacasaeg 12 0 | Soft brown chalk [lower phosphatic band]......... 4 0 Peer Bard White Ghali: is ides Pascua ng datas tgsvacsay The specimens from which the following notes have been drawn up were collected from the brown bands of the above section. Both bands owe their colour to a multitude of brown grains, and pass up gradually into a rock less rich in brown grains, and of a pale brown or greyish tinge. The chalk is irregularly bedded, perhaps current- bedded, and traversed by irregular joints, so as to break readily up into blocks, which crumble into a rounded form and finally into powder. Its friability forms a marked feature, and is due to the softness of a white chalky paste in which the brown grains are embedded. Under the microscope these grains are seen to be almost entirely of organic origin, foraminifera, the prisms of Jnoceramus- shells, and small oval pellets forming the bulk. The following list of foraminifera has been kindly prepared for me by Prof. T. Rupert Jones, F.R.S. :— List of Foraminifera in Chalk from Taplow, Textularia ylobosa, Ehrenberg (common). Bs sp., angular edges (1 specimen). Sp., square edges (a fragment). Spiroplecta biformis, P. & J. (1 specimen). Bolivina textularoides, Reuss (1 specimen). Nodosaria radicula, Linn. sp. (1 specimen). Cristellaria rotulata, Lamarck (common). os cultrata, Montfort (1 specimen). Planorbulina ammonoides, Reuss (common). Rotalia Beccarii, Linn. (1 specimen. Very rare in the Chalk), Globigerina cretacea, D’Orb. (common). ft linneana, D’Orb. (a fragment), és bulloides, (1 specimen), Belew both bands of brown chalk there occur some few feet of white chalk, traversed by cavities and tubes of all shapes and sizes up to an inch in diameter, and filled with the brown chalk. The top of this piped chalk is in both cases hard, crystalline, and of a nodular structure, so as to form a clearly marked floor to the soft brown chalk, Finding that 97-7 per cent. of the brown chalk dissolved in cold 358 MR. A. STRAHAN ON A PHOSPHATIC dilute hydrochloric acid, I submitted a specimen to Mr. J. Hort Player, who was kind enough to make a complete analysis of it, the result being to show that the rock consists chietly of carbonate of lime, but contains from 18 to 35 per cent. of phosphate of lime with a. little fluoride of calcium. ‘The analysis is given in full on p. 364. The brown colour was attributed by Mr. Player to the presence in the rock of a substance which he believed to be humic acid, and of which he extracted 1 gramme from 1000 grammes of the chalk. His identification of this acid was confirmed by the following analyses, made (by the kind permission of Prof. Thorpe, F.R.S.) by Dr. Tingle, under the direction of Prof. Japp. “* Analyses of Humic Acid from Taplow Chalk. “The substance, dried at 130°, gave the following results on analysis :— Analysis I. Analysis IT. Analysis III. Carbon 3. y0ns 65°37 per cent. 63°00 per cent. Hivdrogen’.. ..° D0 5°60 INDEEO@CI 6 cart all eek vais 3°20 ORV ORs npn 20°65 22°63 aN Salers 5°08 5°52 96°80 96°80 “ Urrer Karoo. J [Olive shales of the Karoo Formation. | [Campbell- Randt or Kaap Rocks, and the old schists of the Vaal. | Jones, Uprrr Karoo, | Lower Karoo. J PERMIAN ro TRIASSIC. Penning. High-Veldt Beds. Kimberley Beds, ) Klip-River Series, Witwatersrand Series. De-Kaap-Valley Beds. Megaliesberg Formation. OOLITIC, DEVONIAN. SILURIAN, \ q ‘ 4 - ¥ a ‘ the waste of the underlying Silurian beds.” The conglomerates, and in a less degree the sandstones, are auriferous. Subsequent obser- vations convinced me that the inference was correct, and that the Lydenburg rocks referred to later on are of the same age as those, 200 miles distant, now to be briefly described. As in the De-Kaap Valley and Northern Swaziland, so between Pretoria and Johannesburg, the rocks have been partly tilted by, and partly faulted against, a mass of granite, which, by its more rapid disintegration, here also forms the lower ground, surrounded by the tilted and broken stratified deposits. This upheaval was more recent than that of the De Kaap, as in the latter region the Silurian (?) rocks only were disturbed, whilst the Devonian (?) still occupied very nearly their original horizontal position. (a) The Witwatersrand Series.—These rocks have been described as *‘ chiefly sandstones, sometimes micaceous, with occasional shales, cherts, and quartzites, especially towards the base of the formation, which abuts against the mass of granite already mentioned. The intrusion of this granite has tilted the rocks to a high angle... . further south they gradually assume a less inclination, and eventu- ally are seen in a nearly horizontal position” *. Here they pass in under a very large and quite conformable sheet of dolerite, the out- crop of which (from beneath an upper series of strata) has recently been traced south-eastward to beyond Heidelberg, the beds rising from beneath it, in due order, all round its eastern extremity, and again along its southern margin. ‘This establishes as a fact, what was previously only a surmise, that they form a practically continuous basin, the southern extension of which passes in under the Coal- formation to an indefinite distance in the Orange Free State. From several sections across this basin, and many observed dips all over the area, I have worked out the curve which these beds must follow in their downward course ; and, as the results fairly tally at dis- tances of some miles apart, they may be taken as appreximating to the truth. I find that the surface of the granite along the central line of the basin, if assuming as regular a curve as the outcrops indicate, would lie at a depth of about 18,000 feet below the ground ; and, as this has an average elevation of about 5000 feet, the granite along that line is no less than 13,000 feet below the level of the sea. Of this fact the continuity of the beds all round the eastern end of the depression, with mere local fractures and without much change in thickness, affords satisfactory evidence. The regularity of the lessening dip implies general absence of folding; and, even if the beds be folded, there must still be room for their great thick- ness, unless (as is unlikely) they have thinned out towards what is — now the middle of the depression. ‘The width of outcrop of the rocks of this series near Johannesburg is five miles, and their dip decreases from almost 90° through 60°, 45°, and 20° to 10° (where they pass under the dolerite), with an average of about 40°, which 454 MR. W. H. PENNING ON THE * Journ. Soc. Arts, vol. xxxvi. (1888) p. 435. GEOLOGY OF THE SOUTHERN TRANSVAAL. 455 would give them a thickness of about 17,000 feet, thus corroborating the estimate of 18,000 feet obtained from the assumed regular curve in the present lines of their stratification. ‘There may be nothing unusual in such hollows in the granite, whether above or even far below sea-level ; but a consideration of this basin (now so far inland and so high at its edges) suggests to me that the Megaliesberg formation may possibly have had a marine origin, and not necessarily have been, as hitherto generally supposed, formed at the bottom of a lake at a considerable elevation. The great basin may certainly have been formed by subsidence of its centre, and not by upheaval around its margin; or there may have been both upheaval and depression. It is geologically and practically in- teresting to determine whether the granite ridge completely circum- scribes this Rand basin, and so forms a huge circular bow] filled by the lower part of the Megaliesberg formation, some 60 miles across and 18,000 feet deep ; or if it has an outlet or an extension carry- ing these auriferous deposits into unknown areas trending towards the Atlantic Ocean. The foregoing observations establish a minimum thickness of 17,000 feet for the Witwatersrand series, uninterrupted by volcanic disturbances except so far as the intrusion of some few diorite dykes is concerned. To a later period of intermittent outbursts may be attributed the more crystalline nature of the rocks of the upper series, which encloses many interbedded trap-rocks extending over large areas; the first of these being the thick sheet of dolerite (d, of the map and sections) already referred to, which forms a distinct line of demarcation. Both series of strata are persistent over very large tracts, especially the auriferous conglomerates (‘ bankets ”),— a feature admitting of their identification from one side of the basin to the other. This fact, with the unbroken outcrops and inward dip around the basin, proves that each series of ‘“ bankets,” and apparently the individual deposits (with some exceptions, due to thinning-out), are continuous. As regards the agency by which such extensive beds of gravel, a few feet or even a few inches only in thickness, can have been so evenly distributed, I confess my in- ability to offer a satisfactory suggestion. Ferruginous shales occur amongst the sandstones (which are often false-bedded and ripple-marked), most frequently between what is called the “ Main Reef” and the lower beds of the forma- tion, one series of them invariably enclosing a band of magnetic iron-ore of peculiar character and appearance. By a process of chemical segregation, this iron ore has been concentrated along distinct bands, black, brown, or red in colour, leaving white streaks between, but not always following the lines of lamination; the banding may be straight, wavy, contorted, or folded; but there is neither much folding of the beds nor contortion. There are two, three, or more of these magnetic beds in some places, but the chief one, which powerfully affects the compass, is (or certainly appears to be) continuous around the end of the basin, having been traced all along the Rand, round by Heidelberg, and across the Vaal River 456 MR. W. H. PENNING ON THE into the Free State ; indeed, it has been the chief clue in following and identifying the “Main Reef” from the Rand, west, east, and south, for some 200 miles. There are talcose and chloritic slates, which, if not so persistent as the magnetic beds, maintain their relative positions for great distances, and micaceous sandstones, by which certain “ banket” series can be recognized,—also most of the ‘‘ reefs” contain pebbles of a distinct character, and one set of conglomerates differs from the others in enclosing numerous angular fragments of shale. (b) Lhe Klip-River Series—This series of rocks (¢), previously called by me the “ Megaliesberg beds,” proves to be part of a larger ‘‘ Megaliesberg formation,” which includes the equivalent of the *‘ Klip-River ” (or upper) series, and the ‘‘ Witwatersrand” (or lower) series already described. It consists of “shales and flag- stones, fissile and thin-bedded, which generally are grey, but weather to yellow, or dirty-yellow, or dirty-brown. In some localities, as along the valley of the Eland’s Spruit, there occurs a series of cherts and quartzites, which appear to replace the lower shales” *. These cherts and quartzites also form the lower part of the series around Pretoria and at the Klip River, the shales being in much greater force (where first observed) in the Lydenburg district; there occur throughout numerous interstratified traps, sometimes of great thick- ness. This series is at least 18,000 feet thick; and from its southern edge, by the granite near Pretoria, across the mountains from which the formation derives its name, it is actually seen, and its thickness can be determined, independent of any inference as to the curve formed by underground extension. Near its base, this series has a distinctive “ banket,” called the “ Black Reef,” amidst sandstones and quartzites, which again are overlain by a peculiar rock, which I described as chalcedolite T, in connexion with the Lydenburg district, and which confirms my opinion of that area forming part of the Megaliesberg formation. By weathering it often presents most fantastic characters. This chalcedolite, including the “ Black-Reef” banket, comes close to the granite south of Pretoria, the whole of the lower series being absent; and, as it seems impossible for the latter to have abso- lutely thinned away, and improbable for it to have been entirely overlapped against the mass of granite, I have come to the conclusion that it is faulted down, and must exist below in its relative position. The base of this series, generally conformable to that below, is traceable on the south along the valley of the Klip River (whence its name), then across the Vaal, here and there under the Coal- formation, into the Free State, and again into the Transvaal near Klerksdorp; here it is faulted down against the lower series of rocks which then dip westward. ‘This is perhaps a mere local disturbance, but it may indicate the commencement of another basin, in which, as previously suggested (p. 455), the Megaliesberg formation may be carried in that direction to a great distance, to reappear possibly even so far west as Bechuanaland. * Quart. Journ. Geol. Soc; vol. xli: (1885) p. 576. t Lhid, YY fae eS GEOLOGY OF THE SOUTHERN TRANSVAAL, 457 In view of the great persistence of certain beds over large areas in this region, I am inclined to think that, as already stated, the auriferous conglomerates of the Devil’s Kantoor, and the chaleedo- lites of the Blyde River and elsewhere in the Lydenburg district, establish the position of the rocks of that district as part of the upper series of the Megaliesberg formation. They need not again be described ; but it may be repeated that ‘the chief exposure of the chalcedolites is along the Blyde River, best seen on its western side or escarpment, where the rock occurs in two series,” as it does near Pretoria, “ the lower several hundred feet in thickness. ... . It contains fine gold in places, and, where in a decomposed state, it has been worked as so-called ‘rotten-reef’ to a considerable extent” *. . As previously mentioned, the De-Kaap-Valley beds are faulted, on both the north and the south, against the granite, the lines of fault probably coming together in the region to the eastward. In a similar manner the whole of the lower half of the Megaliesberg formation is let down against the north side of the granite S. of Pretoria, and against its east side by a fault at nearly right angles to the former. Another line of fracture branches off to the east, bringing the upper half of the formation directly against the lower, but with a possible intrusion of trap-rock along the line ; the geo- logy hereabouts is, however, very obscure, and the tilted rocks soon pass in under the newer formation of the next section. Jam in- clined to the opinion that the Pretoria E. and W. fault extends even down to Klerksdorp (across a district at present unknown to me), where, as in the preceding instance, the upper series abuts directly against the lower, as shown in the section (Pl. XV. fig. 1). Also, that another E. and W. fault will be found to occur north of Pretoria, letting down the beds that form the Megaliesberg Range against the granite which is known to appear at the surface not very far north of the mountains. Ill. Tae Hites-Vetpr Coat-FoRMATION. The newest formation of all here is: that which I described in 1884 as the High-Level Coal-fields of South Africat. It rests unconformably upon the older rocks, and, with local exceptions, © very nearly maintains its original horizontality. Some important coal-seams are now known to be continuous for many miles; and I feel still more convinced of the lacustrine origin of coal. The * Kimberley Beds” (g) make up the lower portion of this coal- formation on the west, but thin out eastwards, and are overlapped near Klerksdorp by the ‘“‘ High-Veldt Beds” (g') or upper portion, the estimated thickness of which is 2300 feet. The amygdaloidal dolerite (f) upon which the ‘“ Kimberley Beds ”’ rest, and which is met with at a depth of 300 feet in the Kimberley Diamond Mine, has now been traced eastwards to a great distance. It crops out at * Op. cit. (1885) p. 577. t Quart. Journ. Geol. Soe. vol. xl. p. 658. Q. J. G. 8. No. 187. 21 458 MR. W. H. PENNING ON THE the Vaal River near Barkly, and may then be followed up along the valley, past Christiana and Bloemhof, to Klerksdorp, where it over- lies the Witwatersrand series. Thence it rises to cap the Gats- Randt, where it rests on the Klip-River series, and its outcrop makes a clear line of demarcation between the High-Veldt coal-deposits and the Megaliesberg formation. Brief mention may also be made of a voleanic rock (h), of which but few small patches remain, overlying this coal-formation. East of Boksburg it caps a small hill as a bluish-grey basalt, which weathers into rounded lumps of a dark-brown colour. North of that place it is seen in section overlying grey and black paper- shales, sometimes micaceous, with coarser sandstone below; here the volcanic rock is weathered into a soft grey mass with numerous small concentric nodules. In this locality, and near the base of the formation, there is a bed of Joose, calcareous, sandy clay, enclosing many waterworn pebbles, some of large size, derived from the “bankets ” and quartzites of the subjacent formation; this is an old alluvium to which further reference will be made. A few additional sections may be enumerated as follows :— Kast of the Wilge River, the Holfontein mines give a good section :-— Coarse sandstone. ft. in. Shale. 21 O Coal, with a constant thin parting of shale about the middle; the best coal below. Sandstone. On the farm “ Witklip,” forty miles E. of Johannesburg, in the “Greville Colliery,” the following section is obtained :— ft<, im. 20 0 Loam, over calcareous clay. 4 0 Sand and pebbles. 11 O Shale, dark in colour. 3 0 Sand and pebbles. 19 O Shales. 4 0 Shales, with thin layers of coal. 15 0. Coal. The Boksburg Collieries have seams of coal of varying thickness. The “ Brakpan” colliery-shaft goes down 58 feet to coal, which is over 20 feet thick, including a bed of sandstone about a foot thick. South of the Zuikerbosch-Randt River, on the farm ‘* Modder- fontein,’’ two pits, 14 feet deep, show :— ft. in. Grey sandstone. 0 3. Shale. 6 0. Coal. Sandstone. South of the Vaal River, opposite Klerksdorp, the mine of the Kroonstad Coal-Company is in— GEOLOGY OF THE SOUTHERN TRANSVAAL, 459 Sa — Yellowish sandstone. Yellow and purple laminated clay. Micaceous shale. Shaly coal. Shaly coal, ferruginous. Good coal. Finely laminated coal. PHAR KCAOS IY. DENUDATION. As in all other parts of the world, there has been here enormous denudation of the more ancient rocks, resulting in the most marked unconformity between the Silurians(?) of the De-Kaap Valley and the Devonians (?) of the Godwaan Plateau. Some points in the later planings-down of this region are well worthy of consideration, both in that («) which preceded the deposition of the coal-bearing rocks, and in that (6) to which the present surface-configuration is due. (a) Before the High-Veldt Coal-formation.—In treating of the auriferous rocks I have not commented upon their characters as gold-producing rocks; but in one respect the occurrence or non- occurrence of that metal, or of any other metal or mineral, in them (or in deposits resulting from them) may afford strict geological evidence. Such evidence will be of even more value in connexion with gold, which under ordinary conditions cannot travel far, as its absence from rocks where its presence would naturally have been surmised renders more probable any explanation of its removal by extraordinary or unexpected agencies. The tilting of the Witwatersrand series began at that period of voleanic activity which gave rise to the outpouring of the dolerite (d) in an extensive sheet that can even now be traced for a distance of at least 300 miles. But this evidently slow upheaval still continued at the same centres, as the trap-rock also was turned up with the beds beneath it, and now forms a fringe to the Klip-River series. around the inner part of the Rand basin. Meanwhile the upper series of rocks were deposited, and in some places overlapped the bent-up, faulted, and locally-denuded edges of the Witwatersrand series, as for instance some twenty miles west of Johannesburg. Even then there must have been extensive denudation going on over the areas upheaved, that is to say above the granite bosses south of Pretoria and at Parys in the Orange Free State ; a similar argument to that which follows, although in reference to a still earlier period, in connexion with the upheaval of the beds of the De-Kaap Valley, might also apply. } It might naturally be expected that large alluvial deposits of gravel, as such or as old conglomerate, would result from the removal of the masses of rock that once overlay these granite centres, espe- cially as many of the beds consisted of conglomerates, and were made up, to a great extent, of pebbles of hard quartz and quartzite. But, as a matter of fact, the country is almost destitute of alluvial deposits worthy of the name. There are some few of recent origin, 212 a Hh} iW 460 MR. W. H. PENNING ON THE and these generally contain nuggets of gold; but both their mass and their auriferous value are very small in proportion to the rocks which have been removed. The point, however, is this: admitting the power of the streams to have carried away the sand and clay resulting from ordinary disintegration of these rocks, the gold must have remained behind with the heavier minerals that cannot be transported far by ordinary agencies. Yet the absence of alluvial gold is remarkable, and leads to the conclusion that the denudation of the period in question was not due to rain and rivers. Equally difficult is it to believe that it could have been marine; and, at first, it may seem still more difficult to account for the absence of gold-bearing alluvia, as I now suggest, by the action of ice, in a region lying between the 25th and 27th parallels of south latitude. Previous observers have mentioned traces of glacial action in South Africa, and Mr. E. J. Dunn has described and mapped the Dwyka conglomerate as glacial; but, although always on the look-out . for them, I have seen such traces only in two instances. It has even been asserted that the whole of the western slope of the Drakensberg shows signs of glaciation. This is certainly not correct so far as the Transvaal is concerned, whatever may be the case farther south along the extension of that range in the Free State and the Cape Colony. What traces there may once have been on the mountains have long since disappeared; and, as I am now inclined to believe, those at lower levels are covered over by the coal-forma- tion deposited long after the (now assumed) glacial period of this region. ‘Tn 1881, when visiting the River-Diggings, where the first diamonds were found, I made the following note :—‘‘ Roches mou- tonnées, striated, in short valley running in at lower end of Winter’s Rush,” a point not far above the junction of the Vaal and Hartz Rivers. The strize follow the direction of the little valley, due west, towards the Vaal River, the recent valley thus following the line of the old, which here is not an uncommon occurrence. Crossing the strip of land between, one comes to another small valley, tributary to the Hartz River; here also were seen strize running north and. south, and thus again coinciding with the direction of the valley. I have another note of an ice-marked boulder (?) near the Modder River, some thirty miles 8. of Kimberley, and there are “ boulder- clays” (glacial?) near Pietermaritzburg in Natal. It may be admitted that these few observations of glacial markings, 300 miles or more from the locality in question, form a very small basis on which to found a theory of a glacial period in the Transvaal; but it must be remembered that those made near the Vaal River are at 2000 feet less elevation. The basis may be insufficient, but the hypothesis will go a long way towards explaining some minor difficulties (e. g. the formation of “ pans,” thus = “ rock-basins ”) that seem to me otherwise insoluble, as well as the almost total absence of auriferous alluvium in regions containing rich gold- bearing deposits that have been reduced to an enormous extent by erosion—glacial, fluviatile, or marine. Therefore, 1 am convinced GEOLOGY OF THE SOUTHERN TRANSVAAL, 461 that this region was under glacial influences at some time during the long period which intervened between the deposition of the Megaliesberg formation, in probably the Devonian era, and of the coal-bearing rocks of the High Veldt, the age of which is certainly Oolitic. Mr. Dunn places the Dwyka conglomerate just above the Carboniferous. (b) Denudaton of the present Surface.—Whatever agencies may have effected the denudation of the older rocks of the Southern Transvaal before the imposition upon them of the Secondary High- Veldt formation, those of later date were decidedly fluviatile. The noteworthy feature of this denudation is its apparent continuity from the close of the Oolitic period until now. There are evidences of an enormously prolonged period covered by the last subaerial denudation. (c) During the early Part of the High-Veldt Coal-formation.— Intervening between the two periods of great erosion (a and 3}, above) there was another denudation of this area, naturally fluvia- tile, anterior to and during the deposition of the lower beds of the coal-bearing formation, The alluvial deposits containing gold, re- sulting from this minor degradation, will be found (I venture to affirm) in the valleys of that period, many of these valleys being more or Jess coincident with the present depressions. In their lower portions there are still tongues or branches from the main mass of the High-Veldt Coal-formation which, in some cases, may overlie ancient alluvium ; indeed, that is the character of the “ sand and pebbles” mentioned in one of the preceding coal-sections (p. 458); such deposits would somewhat correspond to the ‘ deep leads” of Australia. The last, greatly prolonged, fluviatile denu- dation must surely have left numerous relics of its ancient river- courses in the shape of gravels, which will be found to occupy the summits and flanks of the hills within the larger valleys, although very few have yet been detected. Those acquainted with the Quaternary gravels of Europe cannot fail to be struck with the large areas in South Africa, both hill and valley, exhibiting bare rock, only here and there covered by a few inches, or possibly a few feet, of blown sand. Sometimes there are deep patches of loamy soil, but these, when cut into, have more the appearance of rainwash than of fluviatile accumulations. The De-Kaap, Blyde, and Olifants Rivers have some good gravels along their present courses, and the headwaters of the Limpopo very few, whatever may be the case away to the north of the Megalies- berg Range. Theupper portion of the Vaal has muddy banks, with gravel here and there, until it leaves the Transvaal; then there are large masses of terrace-gravel in which were first found the. South- African diamonds. ~ Puate XV. Geological: Map of the Area between Klerksdorp and the De-Kaap Valley, and three sections. 462 MR. W. H. PENNING ON THE Discussion. Mr. Ginson maintained that the thickness of the Witwatersrand beds, which was stated by the Author to be 17,000 feet, could not be determined, the strata being so greatly faulted and so similar in composition that a complete sequence could not be obtained. So far as he knew them, the rocks mentioned as granite in the paper were schists, gneisses, some granites, and various other highly altered crystalline rocks. The Witwatersrand beds appeared to have been thrust over these crystalline rocks. He considered the Author’s view that the coal-bearing rocks had covered so wide an area to be doubtful. Mr. Atrorp thought the geological section exhibited fairly correct, but where it showed the De-Kaap Valley as a denuded anticlinal it was certainly open to question. There were very evident signs of folding in the beds of the Makongwa Mountains, which lie to the south-east of the De-Kaap Valley; also farther north, across the Crocodile River. He could not think that sufficient evidence had yet been met with to justify the use of the names of geological systems such as Oolitic, Devonian, and Silurian. He had seen the ‘‘ corals ” alluded to, and very much doubted that they were corals at all. They occur in a bed of steatite which comes in between the granite and the schists of the Makongwa Mountains, and they are ex- ceedingly obscure. Excepting only a few coal fossils, no organic remains had to his knowledge as yet been discovered in the Transvaal. The fossil fishes, of which a photograph was shown, come from the Ladybrand District, in the south-eastern part of the Orange Free State, and are not in any way connected with the Coal-formation. There are some good specimens of these in the Bloemfontein Museum. The sandstones, quartzites, siliceous schists, and conglomerates of the Witwatersrand form a vast series of rocks which are recognizable under varied conditions over almost the whole of the Transvaal. It appears, therefore, curious to bracket them with the small local beds of the Klip River, which are probably only the denuded and altered remains of the same, and to give the whole the name of a small range of hills such as the Megaliesberg. It remains to be seen how far the series may be capable of subdivision. The Black Reef is a small and very local series of highly ferruginous deposits, which haye become notable on account of some parts of it having been found auriferous. The difference of dip in the rocks of the Wit~ watersrand is interesting, and more so when noticed in relation to the gold-bearing value of the beds. Prof. Rupzrt Jonzs congratulated the Society on the accumulating knowledge of South-African geology, although the several published accounts of observations are imperfect and more or less contradictory. The Author’s section, though apparently generalized in character, evidently contains distinct information on some points in the geo- logical structure of the region concerned. From what the speaker had gathered from his friends, the rocks in the Transvaal are much Quart. Journ.Geol. Soc. Vol. XLVIIL.PLXV. Quart. Journ.Geol. Soc. Vol. XLVILLPLXV. Pie Poe os a Ot > INDEX. ai Grosses off a See HIGH VELDT De- ~Kaar ALE Magwasst Matjes Schoon 5: Moot GATS RANDT Ea ine oi? he Steentivel Olifants Gopwaan Javelburg [97 |HighVolde Beas ‘t oper Karoo" (Stow) Spriuak — Spriut. Spruit. es p 4 het Sprud , Ree. ? Plateau 2% » ee > Crocodile ; Lat MED ect y Owe Hugh: Level | Goodfield eae tt OT EET ION Groin ol River Beogi Olive Shales” (Stow ) i ' Hate ee. Seties Srifha 77 k — ae eure = GY GB NN AS Wwe : Dolerite me ZIG ah Bede uncer late ee a ANOS NED, | Dolepite = LLL, LON EASA NIE [eed Sous) Seege LiL ie Ae Sa the base. Magaliesberg a [GZ] Witeratersrend Beds SECTION ALONG LINE A.B. 4 (Bankets etc.) Fig. 20 3" Siete s — — 25a Makonuwa M™S® Crosses De-Kaap Beundarycf Secliorrline witwaters ex] Granite VALLEY. ms 4H0 A.B. RANDT as E online | Loma Rava ercrog Dolertte CIEE, ; { GATS RANDT. Sac Mile NW Fotasp |i Pewee SE- 5 tomo “oe Min ae le N, Le] oerte ate Reman 4 ; Ny Ta Posterior to the Megalies H \. & al (native tte aN aAiP wie \ High Wldt Formation. He oe 0 Postertor to theltighVeldt SECTION ALONG LINE E..F. SECTION ALONG LINE C.D. es (st Ea Fig.3. Fig. 2. ee ia y } L255 32! ay} 25 26 SS a eA SOARS SEER ore = j; A 7, GEOLOGICAL MAP OF THE AREA BETWEEN KLERKSDORPanp tHe DE-KAAP VALLEY we PR StANDERTON : SHOWING THE RELATION OF THE z Be Il H,0 lost on ignition ... 13°70 12:00 8:69 15:52 15:52 13:90 12:09 12'35 12:82 7-93 7-98 817 2-91 H,0 lost at 100° C...... 0:46 0:55 1:85 inzoluble in WEL OLE esters spate srr |e ese lee 1:37 MRotalireetseaterseeseencterres 100:30 99:97 100:00 98:12 100:58 100:00 100:00 100:00 100:00 100-00 99°85 100-00 98°68 SIDS ERS. spedonéconsosccoastnd i ite Papeete ||. Gan seee 2587 2:65 2-56 2545 2644 274 2:85 ails iD eo tease I. and II. Ornamental serpentine from the Lizard. Dark green to black, thickly spotted with red ; enclosing imperfect crystals. [‘Dhis analysis may be taken as repre- senting a very common and characteristic variety of Lizard Serpentine. ] (Of the water, 2:06 was lost in the water bath.) Analysis by J. A. Phillips, Phil. Mag. 1871, (4) vol. xli. p. 101. III. “Clouded reddish brown and greenish serpentine containing small grains of diallage” [bastite], said to be from the Lizard Point, [Probably from near Pen Voose.] The FeO includes also oxide of chromium. Analysis by 1. S, Hunt. Am. Journ, Sci. vol. xxvi. (1858) p. 239. TVA “The red, earthy sometimes semi-crystalline base of the Serpentine porphyry of Kynance Coye.” Analysis by S. Haughton. Phil. Mag. 1858, (4) vol. x. p. 254. Vv. Matrix of black serpentine from near [i, ¢. to the north of] Cadgwith. Contains porphyritic crystals of bastite. ‘The SiO, includes traces of TiO,. The rock was dried at 100° C. The H,O includes traces of CO, ; and traces of sulphuric acid are also noted as present. Analysis by Mr. Hudleston, in Appendix to paper on serpentine of the Lizard district, by Prof. Bonney. Quart. Journ. Geol, Soc, vol, xxxili. (1877) p. 925. VI. Greyish green granular serpentine. Porthalla, ) Vil. Dark oil-green serpentine. Porthalla. | mheeiG, includes also insoluble residue. The i Z alkalies include also loss, Analyses b Mr. VIII. Reddish brown granular serpentine. Porthalla. ; | ; Collins. Quart. Journ. Geol, She at x1. IDG Dull dark red serpentine. Porthalla, [@his, I think, must be rather an abnormal specimen, for evidently it (1884) p. 467. is almost a picrite.—T. G. B.] ) Xe The Rill, Supeaved to contain felspar, See Geol. Mag. (1887) pp. 69, 137, 880. By Mr. M. W. Travers, of University College, London. A partial analysis is given in Geol. Mag. (1887) p. 380. XI. Gew Graze. Red serpentine, minutely granular in texture, Described in Quart. Journ. Geol. Soc, vol. xxxiil. (1877) p. 918. By the same, Xil. Quarry near Lower Predannack, Reddish serpentine with numerous crystals of white hornblende, [Almost a picrite, though I have Neyer seen any fe — 4: G.B.] Described in Quart, Joum, Geol. Soe. vol. xxxiii, (1877) p. 918, and vol. xxxix, (1883) p. 99, artes Pree an yet clepar. XIII From a weathered portion of the samejhand-specimen, By the same. Analyses of Soapstones by S. Havauron. (Phil. Mag. 1855, (4) vol. x. p. 255.) Ho | SiO, Al,0, Mgo Total. | | See = ———_] From the vein at Kynance.................. 42°47 6:65 28:83 19:37 97:32 | From the vein at Gew Graze 42:10 767 3057 | 1846 9ss0 | ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 467 sess py ) | (8 Sls Localities Ae 8 s\k = Remarks : Bs |Slere Os |) m| oo Bla S18 S\8\5\ 9/5 / a | aA o\- PB eee [+e] # Pa ? |} |? Red (2) op hiaag (above Frying Pan)...) * ...... *)...) 4%) % Red. oO. , Quarry N. of ......... 1... a |.00} #|...] @ | 9 (Red. Painys Head, near .............0. 0. |) a | Pls | Pix |* Black, abundant in this district (2). oe ai avai paaks xi wabibs |b foo sbeor| wah. aeih Black, calcite or magnesite in veins. arn Spernic (quarry W. of) .... *| «| #|.../? | * [Red (3). do. shore beneath the same sefone] -oc]eecfaoal | ? |Red (3, two normal, one more compact). Lankidden Cove ................../... % ../..+/...]#| ? \Greenish-purple (3, one rather streaky, a 7 am | | one with veins as above). werack (by road above village) * | * | *|.. |...| % | « |Red. Be TROVE 6.650 incccnass.| * | | * |...|...| # |.../Red (3, two normal, one more compact). do. Quarry W. of village .)...| ? |... ce x | |Dark, much disseminated opacite, ? af- ) fected by contact. ES OS eae See Pe lsacicd |...| % | * [Reddish. Porthalla (quarry by path)...... / Heecsfoasl Phage (ae Greenish-purplish. do. shore (ordinary) ...... Joe] Pleo.) | ae] ae |... (Id. do. do. (atcontact) ...|.../... +:| * |...[%|...|[d., approaches Mullion type, but more compact. do. do. (streaky) ...... vss) Plas.) #) 4] /.../Id., colours more separated. ) 10 speci- wee le yaa dense Jeasjsee]ee * |x| /.../Id., colours in bands. mens. roo y Downs Quarry . — do. do. do. | le lel eda a Streaky, reddish-—a brown hornblende. Helston-Coverack road... | | : delston-Lizard road (first out-| ?) 2?) ?/* ...|%!% Dark, much white hornblende, a little crop). / | ) brown. LS) % |ocelooe] Isa *|x|Nearly -black, much white horn- ; blende (2). Lower Predannack Quarry ...... |x| ?/...,%#!%/« * [Reddish, much white hornblende (4). }Ogo-dour Cove (near yunetion) |...) 0.210. -1.5:)...1 x Black, rather compact, streaky, much eq chrysotile. SE a on * x |...|*/...)%)* (Striped, dark purple-grey, and light ( | greenish-grey; (3), sphene in one ; specimen. EINE os. cnnghcccnsenciavacea: * |? | *|% ...)%)/*/Reddish, a little brown hornblende; A | (?) trace of felspar. OEE are ee *)*%/*)/* ?)x/.. Dark, white spots; felspar or chlori- toid (?). *|x*|...)Rather dark; some of the chlorite is probably a bleached biotite. ...| * | * ‘Striped reddish-brown and light green- | ish-grey. fayke:. MN.) -of the...) fda ...(*|...{[Red, compact ; ? fluxion structure. J Awarnick Pit (W.N.W. of # | # ik * Kynaunce). MONE aes cide eecdsas.|... | wes] en _ Steeple ”). . Hear iawgis hacs nc ieces.|.cclycchetoes ..-|*|...|Greenish-grey, rather — decomposed, | | veined with steatitic minerals, ‘ Mtreath Beach ..................]... fae io ol Calcite or magnesite veins. . | 1 " - a nee OE es _ Serpentine more common on parts of the west coast and on the _ horth; while XII. is a variety almost confined to the west coast, rich in small crystals of colourless hornblende; XIII. is from a weathered portion of the same specimen as XII. It was made by 468 PROF, T. G. BONNEY AND MAJOR-GEN, C. A. MSMAHON reason of a misunderstanding, but the difference is so remarkable that it seemed worthy of preservation, if only to indicate the caution which must be exercised in selecting a sample for analysis, (2) Relation to the older Rocks——The serpentine of the Lizard district, as stated by Prof. Bonney, is associated with the sub- divisions of the crystalline schists which were named by him the Granulitic Group and the Hornblendic Group, but it has not yet been detected among either his Micaceous Group, or the gneisses of very ancient aspect which were discovered by Mr. Fox in the islands fringing the south coast *. It will suffice for the present to say that both the Granulitie and the Hornblendic Group exhibit. marked structural characteristics, In the former a dark dioritic rock is sometimes veined, sometimes banded, by one of lighter colour, which often closely resembles a granite ; the latter is very frequently so regularly banded as to suggest an original stratification. Whatever be the significance of these structures—a question which we reserve for the moment— neither rock, in its present state, can be the result of a single operation. It has indeed been suggested that all-the rocks of the Lizard district are the result of some sort of segregation from one magmay. We do not propose to treat this hypothesis seriously ; but there are three other hypotheses which call for discussion, and these. appear to cover the field: that (a) the serpentine (with some of the later rocks) and the older series form an igneous complex which has been afterwards profoundly affected by earth-movements—as if a heterogeneous mass had been passed between a pair of rollers; (>) the serpentine is really intrusive in the older series, but the relations of the rocks have been so far masked by subsequent earth- movements as to obliterate any conclusive evidence of the intrusion ; (c) the serpentine (as maintained by Prof. Bonney from the first) is intrusive in the older series, and the amount of sub- sequent disturbance has not sufficed in most places materially to disturb their relations. Along the eastern coast many sections can be found which exhibit the serpentine and the rocks of the Granulitic Group in intimate association, from Compass Cove to Polbream Point, also about the Frying Pan, and again from the north side of Polbarrow to the Balk. We do not deny that occasional sections may be found in which the present relations of the two are the result of faulting, or which, did they stand alone, might seem to support the view that the granulitic rock was intrusive in the serpentine ; but, after again examining all the sections on both coasts described by Prof. Bonney in 1877, we have not the slightest doubt that the serpentine is intrusive in the granulitic rock, which was, at that time, sub- stantially in its present condition, and that, as a rule, their relations have not been disturbed subsequently to any noteworthy extent. The serpentine has broken through the granulitic rock, some- * Quart. Journ. Geol. Soe. vol. xliv. (1888) p. 309. + Geol. Mag. (1888) p. 554, (1890) p. 505. ON TILE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 46) times apparently twisting up the ragged ends into dyke-like masses, sometimes perhaps breaking off and carrying up huge fragments. Here a dyke of serpentine parts two masses of granulitic rock, the outer margins of each being in contact with serpentine, which con- tinues for some distance ; there a tongue of serpentine is forced like a wedge into a banded mass of the granulitic rock, or is protruded between two of the layers. Here, in a similar mass, the bands are nipped up or cut off obliquely by the serpentine (fig. 1); there, in Fig. 1.—Section in Quarry near Kildown Point. 1, Serpentine. . 2. Reddish granitoid rock, with dark bands (Granulitic Group). one of less regular structure, the granitic veins may be seen to run up to the serpentine, and be, as it were, cut off by it (fig. 2). So far as structure goes, the relations of the Granulitic Group and the serpentine are identical with those which are exhibited by banded gneisses or schists, and granites intrusive into them. Fig. 2.—Section near Cavouga Rocks. 1. Serpentine. 2. Granulitic Group. The darker part represents the dioritic, the lighter the granitic rock. The serpentine, in the tongues, and near the granulitic rock generally, exhibits little indication of having been crushed: though now and then a faint resemblance to a foliation—recalling to some 470 PROF. T. G. BONNEY AND MAJOR-GEN. C. A. M°MAHON extent a fluidal structure—may be detected. The rock, however, is usually rather rotten; very commonly there is a dusty-looking, pale-coloured, more or less chrysotilic layer between the two rocks, which is continuous with the serpentine, but does not adhere to the granulitic mass. Indeed we have never found the two rocks actually welded. Signs of crushing and slickensides may no doubt be not unfrequently seen at junctions. This is only to be expected, because the tenacity of the two rocks is so different that, even if welded, they would have parted here under strains from ordinary earth-movements. The serpentine also at the margins is often decomposed, and its structure is obscured by secondary chrysotile, steatite, and other like minerals. ‘The serpentine, which seems to include these gneissoid masses, sometimes exhibits a slight streaki- ness, the significance of which will be considered presently. Some- times this may be parallel to the apparent bedding of the granulitic rock, but at others the two structures are almost at right angles, so that evidently they are not necessarily connected. The relations of the serpentine to the hornblende-schist differ somewhat from those with the Granulitic Group. Masses of the latter, apparently included in or pierced by the serpentine, may be found by scores; with only one or two slight intervals, they literally fringe the eastern coast for a distance of three miles, mea- sured along its curve, and they are not unknown, though much less common, on the western coast *. But the Hornblendic Group is not often seen actually cut by or entangled with the serpentine. ‘This, however, may also be said in regard to the other intrusions. Dykes of any kind, so far as we know, are not very common init. Still, though the serpentine and hornblende-schist in some cases may be faulted together, the relations in a few are clear. At Henscath, just north of Mullion Cove, we find it impossible to explain by earth-movements, or by any theory but that of intrusion, the position of the appa- rently insulated mass of serpentine on the little headland, and the two rocks, in at least one case, are still welded together. Again, the same holds good in regard to the junction exposed in the upper rocks near Pare Bean Cove (north of Ogo-dour Cove +). Strips of the hornblende-schist, regularly banded, are split off by or included in the serpentine, and the two rocks in more than one place are welded together, though the rottenness of the latter makes it im- possible to detach and bring away specimens exhibiting the junction. Moreover, the hornblende-schist, thus included, exhibits * The masses at Kynance Cove, mentioned in my paper, Quart. Journ. Geol. Soc. vol. xxxiii. (1877) pp. 884-928, certainly in some cases, and perhaps in all, are better referred to the Granulitic than to the Hornblendic Group. The road descending to the cove (p. 888) crosses a mass which, when less clearly exposed, was mistaken for a granite vein. There are other characteristic masses along the shore N. of Pentreath beach ; some of these are mentioned as hornblende-schist in the above paper, in which no division of the schists is attempted ; they were not again examined for the purpose of writing the second paper, the precise reference of every block being unimportant.— BG. +t Bonney, Quart. Journ. Geol. Soc. vol. xxxix. (1883) p. 22 (referred to as being ‘to the north of Ugethawr’). ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT, 471 some peculiarities *, which we consider to be the result of contact- metamorphism. Again, in a little pit on Carnbarrow, at the top of the cliffs, a “* slabby ” piece of banded hornblende-schist, about 8 inches thick, is completely surrounded by serpentine. Lastly, there are the junctions at Porthalla, which must be described in rather more detail, as Prof. Bonney’s interpretation has been questioned in the pages of this Journal by Mr. J. H. Collinst, who maintains that there is a gradual transition from the hornblende-schist to the serpentine, the latter being regularly interbanded or interstratified with the former. We find, instead.of the orderly arrangement depicted in his published section ¢, that a mass of serpentine breaks through the hornblende-schist, and runs diagonally, roughly in a westerly direction, up the craggy face of the hill§. Above it is a great mass of hornblende-schist, generally with little banding or foliation, and thus dioritic in character. Ata short distance below it are exposed the rather fissile schists referred by Prof. Bonney to the Micaceous Group, the intervening space being concealed by débris. But in three or four places on the rocky shore, between tide-marks, or just above high water, the relations of the serpentine and schist are well displayed. For instance, rather on the western side of the serpentine, slabs of banded hornblende-schist occur in that rock, the lower about three feet thick, the upper about one foot; the intervening space, rather more than two feet thick, being occupied by serpentine || (fig. 3). Fig. 3.—Section at the foot of the cliffs, Porthalla. : @ aenu4+o SS, sites atae o. « ech . Stevan pe weer sere: See eS 1. Serpentine. 2. Hornblende-schist, about 3 ft. 3. Serpentine, about 23 ft. 4. Hornblende-schist, about 14 ft. 5. Serpentine. * See below, p. 473. : t Quart. Journ. Geol. Soc. vol. xl. (1884) p. 458; Geol. Mag. (1885) p. 298, and (1886) p. 359. t Quart. Journ. Geol. Soc. vol. xl. (1884) p. 461, fig. 2. § Since the publication of Mr. Collins’s paper, I have been thrice at Porthalla. On one occasion I had with me a copy of Mr. Collins’s published section, and we sought to reconcile it with what we saw. After using our best endeavours, we were obliged to abandon the task as hopeless. Gen. M*Mahon had also visited Porthalla in 1887, and had met with no better success. The junction of the serpentine and hornblende-schist, which I saw in a quarry in 1882, had disappeared by 1886, and cannot now be found.—T. G. B. || Observed and drawn in 1886.—T. G. B. 472 PROF, T. G. BONNEY AND MAJOR-GEN, C. A. MSMAHON These intercalated pieces of schist cannot be traced far. Again, near the eastern side of the serpentine, apparently enclosed in it, a mass of regularly ‘‘ bedded” hornblende-schist * may be seen at the foot of the cliff. This is split by a wedge-like mass of serpentine, which narrows down to less than a foot across, on the western side of which comes a single “bed” of schist about 5 inches thick. Next comes a sheet of serpentine of about the same thickness, followed by another block of hornblende-schist. In this, if we mistake not, may be seen the end of another wedge of serpentine, coming in the opposite direction to the former. Again, with regard to the asserted production of the serpentine from a stratified rock, bearing more relation to the hornblende-schist than to a peri- dotite, by some kind of transmutation not easy to understand, Mr. Collins’s own analyses ¢ indicate that the serpentine varies from a perfectly normal example of an altered peridotite to one which exhibits some approach to the picrites §, a variation which is far from rare. It is true that near the incre the serpentine sometimes becomes streaky or even somewhat banded in structure; the horn- blende-schist also departs a little from the normal type, as at some other junctions, and is covered with a thin film of a steatitic mineral. Thus the two rocks occasionally so closely resemble one another that by the unaided eye alone they can hardly be distin- guished; the difference is, however, quickly perceived by continued scratching with a knife, or by a few blows with the hammer, and is obvious 0D microscopic examination. In other cases, however, the two rocks are so clearly distinguished that the point of a knife may be placed on their junction-plane ; while the changes in the horn- blende-schist, which we pause for a moment to describe, are sug- gestive here, as elsewhere, of contact-metamorphism. The hornblende: schist, in the localities where we examined it in contact with the serpentine, is considerably altered. Macro- scopically, it assumes a rather grey and slightly ‘‘dusty” aspect ; microscopically, it exhibits marked changes. We have examined slices from specimens obtained in three localities. (1) Specimen in contact with serpentine at Porthalla. This rock exhibits a banded structure, and consists mainly of a colourless micaceous mineral, the flakes varying in size in different bands, which, with crossed nicols, gives low tints (white to milky grey) and straight extinction, and of hornblende, which, in form andstructure, either resembles the last-named mineral or is somewhat acicular, the hue varying from a pale brown to almost colourless. The * The ‘beds’ are almost vertical, and strike between W.N.W. and N.W. + The serpentine is rather rotten; the section is, in places, obscured by shingle, &c. ; and part of the hornblende-schist is stained of a reddish colour, so that it resembles the serpentine. The mass of the former on the one side of the first. ‘wedge’ cannot exceed about four yards, and on the other about three from this distance brings us again to serpentine. t See tables of analyses, facing p- 466. § The term is used, not for a variety of the true peridotites, but for one in. which some felspar is commonly present, ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 473 former mineral is probably a chlorite, similar to that described by Herr Weigand. (2) From the intrusive junction in Pare Bean Cove (2 specimens) : one, “ 2 or 3 inches from contact-surface,” consists of roundish or rather oblong grains of pale-brown hornblende and colourless augite in a minutely granular matrix; this is composed of a filmy brownish mineral, resembling the hornblende, and a colourless one (? felspar), together with some scattered granules of brownish iron- oxide. Some transverse cracks are filled by secondary minerals, in part a flaky hornblende. The other specimen, “ from a slab about 8 inches thick, enclosed in serpentine,” is very like that from Porth- alla, except that two or three bands in it are chiefly occupied by a dull green chlorite, which in places is iron-stained, and is associated with grains of magnetite; also one or two grains of decomposed felspar can be detected. (3) From a piece about the same thickness, similarly enclosed, from the pit above Carnbarrow. This, to some extent, combines in one specimen bands which respectively resemble parts of each of the above, but other bands are characterized by a fair amount of decomposing felspar, and thus in structure it more closely resembles the normal rock. In two of these the above-mentioned mineral changes might possibly be set down to ordinary decomposition, though we do not so interpret it; but this could hardly be asserted in the first and last specimens. The alterations, especially in the hornblende, so far as our experience goes, resemble those which have been pro- duced in “greenstones” by contact-metamorphism. In each case, as it happens, the hornblende-schist is a well-banded variety. Mr. Collins states that “‘ the massive serpentine of Porthalla does not present that reticulated structure which has been regarded as the result of the alteration of olivine rocks” *. Hither he must have examined a yery small number of specimens and been singularly _ unfortunate in collecting them, or have had a limited experience in _ the structure of altered peridotites. It is quite true that there are differences between this rock and the serpentine at Coverack or farther to the south ; the latter contains more bastite, and as it was once rather coarsely crystalline, the characteristic reticulated structure can be seen at a glance. The Porthalla rock, however, was evi- dently at the first a fine-grained peridotite}, but, making allow- _ ance for this, it presents no difficulties to an experienced eye. Occa- _ sionally the structure is as characteristic, though on a scale of about 3, asin the normal rock to the south. In some parts, where the banding is most marked, there is considerably more variation in the mineral constituents ; a white micaceous mineral in small flakes is occasionally rather abundant—possibly the variety of chlorite _ investigated by Herr Weigand in the Rauenthal serpentine—larger flakes of this are sometimes associated in nests, with grains of mag- * Geol. Mag. (1885) p. 300. + Probably very like that from St. Paul’s Island, described by Prof. Renard. Q.J.G.S. No. 187. 2x 474 PROF. T, G. BONNEY AND MAJOR-GEN. C. A. MCMAHON netitie, or lie nearly parallel (with inclusions of the latter) so as to suggest the bleaching of a biotite *. Occasionally the remains of a pyroxenic constituent can be detected; sometimes it is probably an enstatite, but sometimes the oblique extinction of a remnant sug- gests a colourless hornblende. In some cases a considerable quan- tity of a semi-transparent mineral is present in minute granules, occasionally aggregated, which suggests the presence of an alumina- silicate. ‘The banding is seen to be caused by the variation in the amount of this mineral, the opacite, and the chlorite. The most strongly banded variety, as we believe, is rather rareyv. The structure in both is better seen in slightly weathered specimens than on freshly broken surfaces. No sign of crushing can be discerned in these specimens. Both the variety with thin streaky lines and little rounded eyes of a mica-like mineral, and that with distinct bands of different colour and texture ¢, present macroscopically a very close resemblance to the fluxion-structure of a felstone or a rhyolitic rock. A very fine specimen exhibiting this structure (which is much more conspicuous on slightly weathered surfaces) is figured on Plate XVI. | The microscopic examination of thin slices shows that the appa- rent foliation is due to the streaky condition of the parent rock prior to its serpentinization—differences in the original composition of the streaks being now represented by slight mineral and struc- tural differences in the resulting serpentine. In the opinion of the authors, the structure can only be explained as a fluxion-structure ; that is to say, as being the result of traction acting on either an imperfectly blended mixture of two magmas, differing slightly from each other in composition, specific gravity, or fluidity, as in the case of a banded felsite or rhyolite, or on a mass, in which complete crystallization had been arrested by subsequent motion at a time when only a portion of the constituent minerals had separated themselves out from the magma. The eruptive character of the serpentine (peridotite), which has been described above, as well as the microscopic evidence, shows conclusively that the original rock cannot have had a sedimentary origin. (3) The Structures of the Serpentine—A somewhat similar structure is exhibited by the serpentine in other parts of the Lizard district. Along the western coast this rock very commonly exhibits some approach to mineral banding. For instance, at Mullion sometimes, and rather more markedly at Lower Predannack, the crystals of colourless hornblende tend to lie parallel. But, on microscopic examination, we cannot discover that the constituent minerals exhibit any indication of having either been crushed or suffered any mechanical disturbance which cannot be explained * Asin the scyelite of Caithness, Judd, Quart. Journ. Geol. Soc. vol. xl. (1884) p. 406. t One specimen, picked up on the shore by Prof. Bonney, suggests the possibility of the one variety being intrusive in the other (see Plate XVI. at the jine AB). + They are sometimes 3'' wide. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT, 475 by slight strains, either in cooling or in the alteration of the olivine constituent into serpentine. At Lawarnick Pit, near Kynance Cove, a faint banding is often perceptible in the compact serpentine, and a Jike structure is often developed rather conspicu- ously on the weathered surfaces of the rock, both in the neighbour- hood inland and for some distance along the coast to the north. This structure commonly has a roughly uniform strike, and thus might naturally be interpreted as a result of pressure ; but we noted variations in this district from a little W. of N., round by W. to W.S.W. The rock also is not rendered fissile by it. So, if a pressure structure, it is certainly anterior to serpentinization. On the east coast this structure is much more rareand local. It may not seldom be detected on slightly weathered surfaces in the black serpentine 8. of Kennack Cove, but the rock under the micro- scope does not give any indication of having suffered from a general crushing. Near Compass Cove we observed a sheet of compact- looking serpentine, from 4" to 8" across, in the ordinary serpentine ; the latter being one of the usual red serpentines with fairly conspi- cuous but rather altered bastite-grains. This mineral also occurs in the former, but much more sparsely (being sometimes absent), and in grains less than half the diameter of the other. The compact rock under the microscope does not, however, show any sign of crushing, and presents the usual structures; indeed, the presence or absence of bastite is the main difference between the two slides. On the E. side of Lankidden Cove, a rather compact serpentine exhibits grains of an iron oxide arranged in lines rudely parallel, and in the middle is a band about 4!’ wide, with a slightly streaky structure, containing bastite ; the dominant colour in both being a greenish grey. Neither the grains of iron oxide, probably chromite, nor of bastite,in the latter, show any signs of crushing. The former indicates, by the parallelism of the “ strings” of opacite, and the arrangement of the “rootlets” of flaky serpentine, that _ there has been originally a somewhat parallel arrangement of the olivine grains, but the resemblance is far greater to a fluxion- than - to a crush-structure. A similar structure was noted in a second - locality, but the serpentine generally is normal. The following extract from Prof. Bonney’s diary, describing a visit to the lherzolite of the Lac de Lherz, written in 1876 (June 27th), indicates the existence of a similar structure in a _peridotite, which certainly appears quite free from the effects of _ dynamo-metamorphism :—*“* Occasionally also a sort of stratified appearance comes out in weathering, just as I have observed in some of the Lizard serpentines. I could not see that this corre- _ sponded with any marked internal structure.” He states, in regard to this, in his account of his visit printed in the ‘ Geological Magazine’ for Feb. 1877 *, that the structure, in his opinion, has, like that at the Lizard, “some connexion with an internal paral- |elism,” and thinks “ it will prove to be connected with a fluidal structure.” * Dec. ii. vol. iv. p. 60. ox 2 A476 PROF. T. G. BONNEY AND MAJOR-GEN. 0. A. M°MAHON A structure which would certainly leave its mark, and probably produce a serpentine, such as some of those mentioned above, is described and figured by Prof. Renard in his description of the peridotite of St. Paul’s Island *. The rock is rather fine-grained and contains “ eyes” of enstatite. The author, by an admirable piece of inductive reasoning, proves that the rock exhibits a fluxion structure and is of igneous origin, but abstains from accepting the conclusion because it has been asserted by certain authorities that some peridotites are metamorphosed stratified rocks. III. Tue Rocks otpER THAN THE SERPENTINE. These were divided by Prof. Bonney, in 1882, into three groups, between which, however, no sharp line of demarcation was supposed to exist. Of these he considered the Granulitic to be the © upper, the Hornblendic the middle, and the Micaceous the lower. He thought the series, as a whole, had been sedimentary in origin, but that the hornblendic rocks were probably altered tuffs, and, in some cases, might even have been basic lavas. In 1888, Mr. Fox tT communicated to the Society the results of his examination of the islands fringing the south coast of the Lizard, together with petro- graphic notes by Mr. Teall; this was followed in 1889+ by a paper from General M*Mahon on the granulitic and hornblendic rocks. It will, therefore, suffice on the present occasion to refer for details to these and other papers, and indicate in general terms the problems which are presented for solution. (1) The Granulitic Group. This group, as has been said, is characterized by a dark dioritic rock, veined or interbanded by a lighter one, which resembles a rather fine-grained granite. The former is not seldom porphyritic, the structure setting in and disappearing in an irregular way, so that a mass is “spattered” with felspar crystals, as is the face of a rock by the marks of a charge of shot. A slight foliation can often be discerned in the groundmass. Porphyritic felspar, as we now know, makes an igneous origin at least probable; but the matrix (which consists mainly of rather rounded or slightly elongated grains of felspar and hornblende, with more or less biotite §), so far as we are aware, has not disclosed any characteristic structure. The lighter rock is very closely allied to a vein-granite. The felspars are not idiomorphic, but occur, like the quartz, in rather rounded or elongated grains. The history of the rock, so far as regards its macroscopic character, can be best studied in Kennack and in Pen Voose Coves. It will suffice to recapitulate the principal facts of which any theory as to the origin of the group must take account. * ‘Challenger’ Reports, Narrative, vol. ii. tT Quart. Journ. Geol. Soe. vol. xliv. (1888) p. 309. t Zbid. vol. xlv. (1889) p. 519. § Apatite, sphene, and magnétite are present in nearly all slices. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 477 (1) In some cases the dioritic rock is pierced by veins of the granitic, which may be of any thickness from a few inches to a few lines; occasionally the former is completely brecciated and the pieces are separated by the latter rock, the intervals also varying in thickness in like way. Thus in the Granulitic Group we find sections which closely resemble those where an igneous rock breaks up and includes another igneous* or a massive sedimentary rock. (2) In other cases the two varieties, for considerable distances, appear perfectly interstratified, and exhibit regular bands of the one or the other which vary in thickness from several inches to a small fraction of an inch, with occasional layers of a rather intermediate character}. In the latter case the lines of junction, though fairly sharp, do not resemble ordinary intrusive junctions—there are no indications that the one rock has been broken by the other. The structures of the two are similar, and the one seems to pass into the other by a very rapid mineral change. (3) The thin slices under the microscope do not exhibit either that mixture of larger and smaller grains, or the peculiar minute “mosaic” structure, which commonly occur when a rock already crystalline has been crushed. The structure is not that charac- teristic either of the “ newer gneiss” series of Glen Logan, or of - one of the crushed granitoid rocks common in the Central Alps; though to these it occasionally presents a very faint resemblance. Nor is it that of the Saxon granulites. The present structure, whether original or secondary, seems to have been assumed in situ. (4) Between the two extremes mentioned in (1) and (2), every intermediate form can be discovered. ‘The angular dioritic frag- ments appear to be gradually flattened or elongated till they become lenticular streaks or even bands, and the vein-like intercalations of granite appear to be drawn out with them into similar bands, very much as a mixture of glass of two colours can be drawn out when it is heated until it becomes viscous. * These conditions appear to be best fulfilled by the following hypo- thesis :—that into a basic magma, which at any rate was sufficiently solid to break into fragments, an acid magma, at a very high tempe- rature, was injected,—that either the more basic material was still somewhat plastic when this intrusion took place, or it was, by this | aecession of heated stuff +, so far softened that it was drawn out into streaks, and was even sometimes slightly mixed with the other _ by actual fusion, when movements occurred in the mass; and * This, so far as my experience goes, is rather rare and local in its occurrence. The most remarkable instance which I have seen of the brecciation of one _ igneous rock by another was at the Corporation quarries, Montreal, where e nepheline syenite is shattered by and embedded in a rather compact dark rock, perhaps a tephrite.—T. G. B. TL have seen some striking illustrations of the complete brecciation of gneiss _ by granite in Spiti—C. A. MM. ___ + This structure is rather more conspicuous in Pen Voose Cove; the former _ in Kennack Cove. ‘t Probably the temperature of solidification in the basic rock would be considerably lower than that of the acid rock. 478 PROF. T. G. BONNEY AND MAJOR-GEN. C. A. M°MAHON that afterwards, as the temperature gradually fell, the whole mass became crystalline *. Thus the banded gneissoid rock of the Granulitic Group is an example of a kind of flow-structure on a large scale, wholly or (more probably) in part antecedent to crystalliza- tion. As this rock, in its distinctive characters, agrees with a large number of ‘‘ banded gneisses,” in which the ordinary symptoms of pressure-modification cannot be detected, and which appear to have at any rate completed their crystallization in situ, this hypothesis may prove to be of wide application. If, however, the bands do not differ very materially in their mineral composition—as is often the case with the banded Archean gneisses—the hypothesis may assume a simpler form, and with them it may be only necessary to suppose that, as in the case of many rhyolites, some differentiation of consti- tuents had been set up in the magma, the one part becoming slightly harder than the other, though still capable of being drawn out, so that the whole mass assumed a coarse fluidal structure, and subse- quently, since its environment was widely different from that of a normal igneous rock, took on a holocrystalline, yet still a peculiar structure, different from that usually found in granites and diorites. (2) The Hornblendic Group. That this group underlies the Granulitic, is, in most places, a probability rather than a certainty, the two commonly being sepa- rated by a fault or a mass of serpentine. But, assuming the Granulitic Group to occupy a definite horizon, its relations to the Hornblendic are suggested in more than one place, and are, we think, clear in the crags on the south side of Cadgwith Cove, where the latter rocks may be seen gradually rising up from beneath the former. The Hornblendic Group exhibits structures curiously imitative, if not actually indicative, of stratification, certainly over a larger area and probably through a greater thickness than the Granulitic f. It includes fairly well banded schists almost everywhere from Porth- alla to the Lighthouses on the one coast, and from Polurrian Cove to near Old Lizard Head on the other. Their structures have been described by the present authors, who have regarded them, though from somewhat different points of view, as indicative of stratifica- tion in the original materials. WAY This group has been again examined with considerable care. From the chemical analysis as well as the mineral composition it seems clear that its rocks must originally have been of igneous origin ; the more massive may represent altered basaltic lavas, the more * Probably it was a mixture of crystalline grains and half melted stuff rather than a true liquid, so that it was difficult for any mineral to assume an idiomorphic form, The larger porphyritic crystals in the diorite were probably anterior to the epoch mentioned above. t+ The hornblende-schists are displayed, practically unbroken, in cliffs some 200 feet high, between points nearly a mile apart as the crow flies; the Granu- litic Group, so far as we remember, seldom occurs without a break from top to bottom of such a cliff, or for more than a few dozen yards at most. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 479 banded altered tuffs of similar composition. As regards the former, some of the “ eyed” hornblende-schists mentioned by one of us, as for instance on the north side of Porthoustock Cove, may be a porphyritic dolerite which has been modified by pressure *, and converted into a slightly foliated epidiorite. Other masses again, as in the upper cliffs at Porthalla, are not at all banded and are even practically without foliation. It is, however, difficult to attribute the mineral banding and other structures in most parts of the mass to the erushing or shearing of a holocrystalline rock. Is it then to be ex- plained as a kind of fluxion-structure, as we have already done in the case of the Granulitic Group? Some of the hornblende-schists present a very close structural resemblance to certain hornblendic bands in the latter group, and to some other rocks, hereafter to be mentioned, which are undoubtedly igneous. Moreover, the mineral banding—stripes consisting mainly of felspar or epidote alternating with those mainly of hornblende—as at Cadgwith or to the S. of Church Cove, would lend itself very well to this explanation. Indeed, where the bands attain a considerable thickness, it is not very easy to explain them by segregation during metamorphism f. One case, indeed, where the thickness of the bands is perhaps at a maximum for the district, seems to require the former explanation. A pit has recently been opened by the side of the road leading down to Mullion Cove. The rock excavated is partly a coarse saussurite- hornblende rock, without definite structure, partly a well-banded variety of the same, some of the bands being full -3” thick; the one clearly passes irregularly into the other. The former under the microscope presents considerable resemblance to one of the east-coast gabbros, for it affords the remains of plagioclase felspar, indications, and in one case at least a remnant, of diallage, and even a sugges- tion of the former presence of olivine. ‘The banded variety contains the same minerals, and bears considerable resemblance to the flaser- gabbros described hereafter ; its structure does not suggest crushing, and one or two of the thinner bands, where the two minerals are smaller in size, present a very close resemblance to an ophitic structure, in which there is a slight orientation in the felspars. We do not think it possible to explain this structure by the shearing of a coarse holocrystalline mass. At any rate the rock must originally have been a variety of gabbro +. But in some members of the HornblendicGroup we have to explain, not only a banding, but also repeated resemblances to slightly irregular deposition, or even to ‘‘ false bedding.” We did our best, during our study of these rocks, to apply either * The larger felspars, which are rendered distinct by saussuritization, are occasionally partly idiomorphie and seemingly crushed out. t Icollected a specimen, in 1888, from the hornblende-schists of Sark (which are practically identical with those of the Lizard), in which the bands attain a thickness of °5'.—T. G. B. t It might be urged that this rock did not belong to the hornblendic schists as here defined. It agrees, however, macroscopically with them, and differs rather markedly from the altered gabbros of the east coast, of which, moreover, we have not seen any instances on this side of the district. . 480 PROF. T. G. BONNEY AND MAJOR-GEN. C. A. M°MAHON of the above explanations to them. The ordinary cases, where the rock is foliated rather than banded, and the slabby bedding, which is commonly so marked a feature as to be the first thing that attracts the eye as the mass is approached, might be accounted — for, like the structures in the “ newer gneiss” series of Glen Logan and its vicinity, by the effects of shearing movements during a long continued process of thrust-faulting ; but, in applying this hypo- thesis to some of the structures which are more especially suggestive of stratification, we were always encountered by diffi- culties which we failed to overcome. Again and again the gliding- planes, which we had devised in order to explain the oblique disposition of the apparent stratule in the rock, were interrupted by some unbroken band which either forbade the idea of any dis- placement, or demanded for its manufacture a contradictory set of movements. In these cases the “ fluxion hypothesis” also landed us in similar difficulties. Thus, although our reasons cannot be fully appreciated by those who have not followed our steps, we are at present unable to suggest any form of mechanical disturbance as - a complete explanation of the more banded members of the Horn- blendic Group, and think that for these the stratification of an ash (perhaps by the intervention of water*) is the better “ working hypothesis.” | Subsequently, of course, there must have been almost complete, if not quite complete, rearrangement of the constituents. The ash originally must have consisted of more or less fragmental felspar, augite, iron oxide, and possibly olivine, with bits of more or less scoriaceous tachylyte or magma-basalt. Of the latter, at any rate, every trace has disappeared, the constituents have separated as from a molten mass, and the whole is a crystalline mixture of felspar, horn- blende, &c.t In this hypothesis there are also difficulties, so that, until further evidence be discovered, it must be regarded as only tentative, for we now feel convinced that some members of the group were originally dolerites, and some structures are due to fluxion. Moreover, it must not be forgotten that the basic member of the overlying Granulitic Group often ‘differs little from the more hornblendic part of the present one. The former appears to have been raised to a high temperature after it had at least begun to consolidate. It is then probable that the underlying rock was not less affected, and important changes may thus have been brought about. (3) The Micaceous Group. This—the talco-micaceous group of De la Beche—was retained — by Prof. Bonney because of the presence of a mica-schist, and some other non-hornblendic rocks, among the green schists. The last * Because by the action of currents the materials would be to some extent separated in accordance with their specific gravity, and their deposition at any spot would be varied by the constantly chang!ng velocity of flow. + See, for a suggestion of the process, M‘Mahon, Quart. Journ. Geol. Soe. vol. xlv. (1889) pp. 525-531. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 481 are, indeed, the most abundant rocks, but they differ much in appearance from the normal hornblende-schists, in the minuteness of their constituents and in the presence of a more acicular variety of that mineral *. But the advance of our knowledge during the last few years leads us to doubt the advisability of making any definite separation. Three bands of brownish mica-schist, like that at Polpeor, have been discovered by Mr. Fox (and visited under his guidance by the authors), intercalated with the normal hornblende- schist in Polledan Cove, E. of Housel Bayt; and the character of the green schists may be accounted for by subsequent pressure and shearing. The whole region has evidently been greatly modi- fied since its constituent rocks first crystallized. The mica-schist at Polpeor is crumpled, and the green schists often exhibit structures resembling the ‘‘mylonite” of the Highland thrust-fault region ; tongues of a porphyritic diabase may be seen on the shore, so erushed and sheared as to be barely separable from the green schists 7. On Old Lizard Head §, a cleavage foliation may be seen traversing the corrugated banded greenish schist at a high angle, and the rock below is in places a breccia of a gneissoid rock and of a rather soft *‘ereen schist” in hopeless confusion, very suggestive of faulting. Polkerris Cove (S. of Porthalla), in which some serpentine and a little gabbro occur, affords evidence confirmatory of this view. On the northern side we find a flinty-looking schistose rock (very similar to one variety at Polpeor), and can identify in one place a porphyritic dyke, rather like that named above. The flinty rock is sometimes porphyritic, and may be a modified dyke, but other parts suggest affinities with the hornblende-schist, into which there is a passage on the southern side of the Cove. A specimen from the northern side, which in the field seemed more nearly related to the hornblende-schist than to the diabase, has been examined. It shows marked indications of crushing and shearing; fragmental “eyes” of rotten felspar or of hornblende (sometimes very like altered diallage) occur in a sort of mosaic of minute hornblende and felspathic grains (possibly also of quartz), with a sort of ‘ fluxion structure.” One part of the slide is coarser and still retains traces of a fragmental structure; another consists of thin bands of a mosaic, in which this or that mineral predominates. Thus the passage of the normal hornblende-schist into a rather flinty-looking schistose rock in consequence of shear seems to be demonstrated |j. * See description of the principal varieties, Quart. Journ. Geol. Soc. vol. xxxix. (1883) p. 12. t Described by Mr. Fox in Trans. Roy. Geol. Soc. Cornw. vol. xi. pt. v. (1891). Asimilar mica-schist occurs near Pisti! Ogo, but here in the ‘green schist.’ ¢ A larger and Jess disturbed mass occurs a little farther east, and is described in Quart. Journ. Geol. Soe. vol. xxxix. (1883) p. 4. § Name on the six-inch map; called ‘The Quadrant’ in Prof. Bonney’s paper. That name is now applied to an island below. | The cliffs and shore do not afford a continuous section, so that a fault may escape notice. As there isa fault at Porthalla and must be one at Porth- oustock, this, the only intermediate cove, may be also determined by a fault. 482 PROF. T. G. BONNEY AND MAJOR-GEN. C. A. MCMAHON We therefore think that the rocks along the coast, from the coast east of Polpeor to Old Lizard Head, owe their peculiarities mainly to subsequent mechanical disturbances, probably the result of an over- thrust, so that the distinctive name had better be abandoned. The coarsely crystalline gneissoid rocks discovered by Mr. Fox in the outlying islands, and so admirably described by him and by Mr. Teall, are situated, in our opinion, below the thrust plane, so that we have here an association similar to that which occurs in some parts of the N.W. Highlands of Scotland. At Porthalla, between the great fault and the typical hornblende-schist (with serpentine), a band of mica-schist occurs associated with fissile green schist. To this mass we should apply a similar explanation, and no longer desire to separate it from the Hornblendic Group *. IV. Ienrous Rocks NEWER THAN THE SERPENTINE, (1) The Troctolite. This rock was described by Prof. Bonney under the name of ‘“‘the older gabbro,” and its resemblance to the troctolite of Vol- persdorf was pointed out. Subsequent analysis indicated that it might be thus named, though it was not so typical an example. It has also been described and figured by Mr. Teall 7. Thus there is little left to be said. It has been found only at Coverack Cove, where it occurs both in irregular masses and in thin veins, with little difference in the texture of the rock, and very perfectly welded to the serpentine, which is practically unaffected by it. We cannot, however, regard the association of the two rocks as a case either of segregation or of veining strictly contemporaneous, for the ser- pentine occasionally has been completely brecciated. For instance, one block on the shore, which measured about 2’ x 14’, consisted of about equal parts of the two rocks, the serpentine being mostly in rectangular pieces, the largest about 8” x 5”, the smallest about 13" x 2’, the thinnest vein of the troctolite being about 7” thick, yet nearly as coarse as the rest. * Prof. Bonney is now convinced that Mr. Collins was right in regarding the gneissoid band (described by him) as only a pressure-modified granite vein, but both the Authors fail to understand on what grounds Mr. Collins separates the hornblende-schists at Porthalla from those in other parts of the Lizard, and considers them to be metamorphosed Lower-Silurian rocks. He states (Quart. Journ. Geol. Soe. vol. xl. (1884) p. 466) ‘ that the hornblende schist of Porthalla is a very peculiar rock indeed.’ We cannot understand how any one well acquainted with the hornblende-schists of the Lizard could make this statement. An exceptional specimen might be found anywhere, but speaking of the general character of the Porthalla schist, which it must be remembered cannot be sepa- rated from the mass which extends to Porthoustock Cove, we unhesitatinglyaffirm that we cannot detect in it any valid distinction, macroscopic or microscopic, from much of that which occurs in other parts of the Lizard peninsula. t ‘British Petrogr.’ pl. viii. fig. 2. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 4533 (2) The Gabbro. The principal mass of gabbro, as stated in a former paper, is rudely oval in form, the longer axis measuring full four miles, and the shorter about two. It rises in Crousa Down to a height of nearly 300 feet above the sea, by which it is washed for a con- siderable distance north of Coverack Cove. In the Survey map it is represented as giving place to greenstone in the little cove opposite to the dangerous skerries called the Manacles. This, however, is hardly correct, for though dykes of the latter rock become rather more frequent on this part of the coast, and perhaps ultimately occupy as much space as the gabbro itself, that rock continues to Porthoustock Cove, on the southern slopes of which it may be seen ; though, as will be hereafter noticed, it does not, so far as we know, descend to the water’s edge. This mass of gabbro evidently throws off many veins on its south- ern flank, which cut both the troctolite and the serpentine in Cover- ack Cove. There is also the great dyke-like mass, nearly two miles long and about a furlong wide, according to the Survey map, which runs inland roughly in a N.W. direction from the skerries of Carrick Luz, and approaches at nearest within about a third of a mile of the former mass. On either side, in Lankidden Cove on the east and towards Compass Cove on the west, dykes are numerous, doubtless in some way connected with it. They disappear in Kennack Cove, but are found again about Enys Head, and then, after a consider- able interval, at Polbarrow, becoming ultimately very numerous around Pen Voose. It is, however, only at the two first-named localities that the rock is found in masses of considerable size; generally it occurs in dykes or veins (at most only a few yards, and commonly only a few feet thick) which not seldom ramify and terminate in veins sometimes less than an inch in thickness. The mineral composition of the rock and its changes have already received full attention ; therefore it may suffice to say that in its normal condition it varies from a plagioclase-olivine-augite (or diallage) rock to a saussurite-hornblende rock, the last mineral being partly actinolite, and always one of the distinctly green varieties. It is impossible in this case to prove that olivine was an original constituent, but inasmuch as it is present in certain masses, which exhibit a transition from the normal gabbro te the ordinary saus- surite-hornblende rock, there is no reason for supposing it to have been originally absent from the latter. In the great mass at Crousa Down the gabbro is often comparatively unaltered. In the dykes, including the large one at Carrick Luz, it is generally more or less altered. The olivine has usually disappeared, though occasionally its position is indicated (as is rather common at Coverack) by a blotch of hematite ; the augite occurs as diallage, and every stage of the change from this mineral into hornblende can be observed. The felspar in like way passes gradually into saussurite ; in most cases it appears to be less stable than the diallage, for a saussurite- diallage rock is common. The change to saussurite does not appear 484 - PROF. T. G. BONNEY AND MAJOR-GEN. C. A. MCMAHON to be connected with any mode of dynamo-metamorphism. It is quite true that it is very characteristic of the foliated masses, but it may also be observed, as, for instance, at Coverack, in rock of the most normal character. It is evidently due to the action of water, and might more correctly be designated meteoric metamorphism, for it evidently proceeds inwards from the exterior of the mass; pro- bably being produced when this is at no great distance from the surface. This gabbro occasionally is distinctly foliated or even banded, a structure which during the last few years has given rise to much discussion. Prof. Bonney, in describing it, regarded the structure as the result of crystallization under a pressure (or resistance) definite in direction. Mr. Teall ascribed it to pressure subsequent to solidifi- cation *, and compared it with the flaser-gabbro of the Germans, which has been similarly explained. But to this view, as pointed out by the former, and subsequently confirmed by Gen. M*Mahon ¢, the absence of all signs of the effects of pressure in the associated serpentine seems a fatal objection. One of the chief objects of our visit in 1890 was to study afresh this very remarkable structure, and the conclusion at which we arrived will be most readily indicated by giving a brief summary of our observations, though this may involve some slight repetition of statements already published. The foliated and banded structure in the gabbro is most con- spicuous in the Carrick-Luz dyke; it is also locally very well developed in the neighbouring dykes, especially on the western side, in the neighbourhood of Pen Voose and at Polbarrow. It may be observed, though it is not common, in the Crousa-Down mass, and elsewhere. Every variety may be found, from a slightly streaky or wavy foliation § to a distinct mineral banding, and not seldom the pyroxenic crystals appear as “eyes.” Hach of these two types is excellently figured by Mr. Teall |], so that on the present occasion it is needless to do more than refer to his plates and to the descrip- tions already published. The results of our investigations may be tbus summarized :— (1) As will afterwards be more fully explained, the gabbro had assumed its foliated structure before it was cut by the later intrusives (dykes of diabase, &c.), which probably are of more than one age. (2) Whatever be the origin of the quasi-foliated structure in the serpentine, this cannot be connected with the foliation of the gabbro. The former is most marked on the western coast and at Porthalla; but these cases cannot be cited in support of the pressure-hypo- thesis, because gabbro does not, so far as is known, occur in either * Geol. Mag. (1886) p. 481. t Ibid. (1886) p. 575. t Ibid. (1887) p. 74. § We use this term as indicative of orientation rather than of linear aggre- gation of constituents, producing at most a slight and interrupted ‘streakiness,’ from which every stage exists to distinct bands mainly of different minerals, sometimes over a quarter of an inch thick. | ‘ Brit. Petrogr.’ pls. xxvi. and xliii. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 485 district. The most streaky serpentine on the eastern coast, south of the Manacles, is the dark variety on either side of Caerleon Cove, but here the gabbro is not markedly foliated. The serpentine at Pen Voose, except for slight and very local crushing near faults, is perfectly normal. So it is in the neighbourhood of the great Carrick-Luz dyke, where the foliation in the gabbro is at a maximum, About Enys Head streaky serpentine and gabbro, sometimes foliated, occur together, but there does not appear to be any necessary con- nexion between the structures. (3) The gabbro and serpentine are sometimes welded together, sometimes separated ; the former condition is perhaps more common in Coverack Cove than elsewhere, and is more usual with the thin veins than the larger masses. (4) The gabbro is often rather variable in texture. The great mass of Crousa Down appears to be the most uniform in this respect, and it consists of medium-sized grains, though occasionally small patches of coarser varieties occur along the coast. The Carrick-Luz mass is more coarsely crystalline. All the smaller dykes, as a rule, are coarse-grained, and even in the thinnest veins the rock generally does not become fine-grained, but maintains a medium texture. Some of the largest crystals of diallage occur in masses less than a foot thick. Even in the same mass the gabbro not seldom exhibits considerable variation in texture, the ordinary coarse kind being streaked or mottled with vaguely-defined patches of finer grain. Fig. 4.—Gabbro veins near a ‘natural arch’ on the shore, west of the Carrick-Luz mass. 1. Serpentine. 2. Coarse gabbro. 3. Moderately foliated gabbro, 4. Very foliated gabbro, (5) The foliation sets in and disappears in a most capricious fashion. Part of a dyke, an arm of a vein, may be foliated (fig. 4), without any apparent reason or connexion with any structure in the adjacent rock (especially when this is serpentine), and the rest may be normal. The significance of the relations of the ordinary and the foliated 486 PROF. T. G. BONNEY AND MAJOR-GEN. C. A, MCMAHON or banded structure, one with another and with the exterior of the mass, will best appear from a series of examples. (a) In the Carrick-Luz dyke the strike of the structure appears to be fairly steady *, running rather W. of N.W., which is also the average direction of the dyke. It appears to dip at a rather high angle, 60° or more, on the northern side—that is, in the probable direction of the fissure. The mass varies somewhat in coarseness, and in the amount of foliation and of banding, the latter being more distinct on the western side, but here and there it seems in- conspicuous. The felspar occasionally, the diallage frequently, are unaltered. Now serpentine yields readily to pressure. When this is moderate in amount the rock brecciates; when the pressure is more severe the fragments take a lenticular shape and become slickensided ; when it is extreme, as may often be seen in the Alps, the rock assumes the appearance of a slaty schist, with correspond- ing changes in its microscopic structure t. Any conspicuous grains of bastite, augite, chromite, &c. are more or less crushed out. But in the Lizard serpentine these minerals, like the matrix, are in a normal condition, so that we are forced to conclude, if we adopt the pressure hypothesis, that the gabbro, one of the toughest of rocks, has been crushed into a kind of schist, while its comparatively brittle associate has undergone no structural change. (6) Sometimes the foliated structure in the gabbro occurs near to and parallel with the edge of a dyke, but at others, though rarely, it is inclined at a high angle to it; it is also found in wisps or streaks in a non-foliated mass, as may be seen, for instance, at Crousa Down. The annexed diagram (fig. 5), representing part of a vein on the eastern side of Compass Cove, shows a wedge of serpentine Fig. 5.—Foliation of gabbro dyke east of Compass Cove. " a aoe tls Weer D. Serpentine. A, B. Moderately fine gabbro, with foliation; from A to B about 42 inches. C. Coarse Gabbro. * We can answer only for the shore-section. + Bonney, Geol. Mag. (1890) p. 533. ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 487 splitting an intrusive dyke of gabbro and enclosed in it. Of this rock the main mass is coarse, and in it foliation is either extremely indistinct or absent. In the arm we find on one side medium- grained gabbro, well foliated, passing into a compact gabbro, which is but slightly streaked with a foliated structure, as indicated. The serpentine, whether in the included block or in the main mass, shows no sign of crushing. (c) Yet more significant is another mass nearer the Carrick-Luz dyke, the more important portion of which is represented in the annexed diagram (fig.6). The face of adyke of gabbro forms a crag about ten feet high, the lower edge resting on serpentine. Above this the mass for about two feet consists of a rather fine-grained gabbro, Fig. 6.—Foliation in gabbro between Compass Cove and Spernic Cove. 4 “1 PSAs ah. “> ) aa rahanias ah wy cu AY age ry ye °K Ve. e ”* < pf ep bp va 7 i bad t. « ¢ Ae o re of WV. ee 3 7 % ay Dipak as Wy _h 3 ast LLL ee eS ge Rene tie Sass tse sess | AS i ie SR aeevcecece#veaoe#ted 1. Serpentine. 2. Foliated gabbro. 3. Partially foliated gabbro. 4. Unfoliated gabbro. foliated or finely banded, “ not unlike a piece of hornblende schist,” - the bands being so thin that the mass, as a whole, is rather foliated than striped. The next two feet consist of ordinary and foliated gabbro, very irregularly mixed, but the streaks, as shown in the diagram, have in places a distinct tendency to sweep round into the fine foliated mass below. Lastly comes some half-dozen feet of sporadically coarse or slightly foliated gabbro, in which occurs now and then a thin wisp-like band of the fine foliated rock, resem- bling that at the bottom, but not parallel with it. Microscopic examination of these foliated gabbros has not led to any very definite results. The constituent minerals have been so much altered since the structure was produced as to obliterate any distinct indication of the agent by which it was caused. The original plagioclastic felspar has been almost wholly replaced by secondary products. Occasionally some diallage may be detected. In the augen-flaser gabbros the larger grains of diallage still remain comparatively unchanged, though they also often have a 488 PROF, T. G. BONNEY AND MAJOR-GEN. C, A. MSMAHON d border of, or are partially replaced by, secondary hornblende. Here and there, in other parts, and in the ordinary flaser-gabbros, some trace of the original diallage may be found amid a crowd of horn- blende grains *; the former occasionally exhibit some traces of mechanical disturbance, such as a slight bending of the cleavage- planes or pinching up of an end. These indications, however, are suggestive of a strain, due to a tensile movement of the mass rather than of a crushing down of the grains. The cleavage-planes in the different grains generally, but not always, exhibit a tendency to parallelism. The saussuritic constituent is at times fairly clear and transparent, at times brownish, varying from moderately translucent to almost opaque. The former, with crossed nicols, appears as aggregates of rather bright-coloured specks, and the original mineral seems to have occurred in rather polygonal grains, often about ‘01 inch in diameter; traces of this structure also can be occasionally de- tected in the more opaque patches. It might be argued that this ‘‘mosaic” is a proof of crushing, but the uniform general distri- bution of the structure appears adverse to this idea, and it might be explained either as the result of secondary change in an original larger felspar grain +, or as an original microgranular structure f. This more opaque part occurs in irregular rounded patches, in rudely rough-edged oblongs, or in streaky clots. Assuming them to represent a felspar of slightly different composition, as is rendered probable by their mode of occurrence, these patches on the whole are not at all suggestive of crushing, for though the last mode of arrangement might be so interpreted, it would be equally possible with a fluxion-structure. (d) Near the Spernic Arch there are several thin veins of com- pact diorite intrusive in gabbro. One of these veins, about two feet thick, splits up into minor veins a few inches thick, which run with the foliation of the gabbro in a way that reminds one of the alter- nating white and black parallel bands of the granulitic series ; but when the dioritic veins are followed up, they are seen to cut obliquely across the foliation of the gabbro at a low angle. Yet though the gabbro is intensely foliated, the compact trap does not give, under the microscope, any indication of crushing or any more parallelism of structure than is usually presented by the flow of * A white or very pale augite in roundish grains is present in some examples. t+ Prof. Judd on the replacement of labradorite by scapolite, Min. Mag. vol. viii. p. 186. t I collected, during a visit to the “ norite region,” N. of St. Jerome in Canada (in 1884), a specimen of a fine-grained norite, which showed on weathered sur- faces a faint structure much resembling a fluxion-structure. Microscopic examination shows that the felspar (which is well preserved) occurs chiefly in small polygonal grains (about the above-mentioned size), mixed with larger grains, often about three times the diameter, but sometimes more. The mass does not give the slightest hint of having been crushed, and we appear to have a record of crystallization in situ, analogous to that of a microgranite. The pyroxenic constituent, which is not abundant, is less well preserved, and irregular in outline, but appears to have formed, as best it could, im setu.— EG. e: ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 489 igneous rocks. The diorite is certainly not derived, in this case, from the gabbro by shearing, and its intrusion was subsequent to the epoch when the gabbro had attained a maximum foliation. (e) One more typical case may be given. On the top of the cliffs above Polbarrow there is a boss of gabbro perfectly unfoliated. JE ua oun 4 ‘4 BANDED SERPENTINE FROM PORTHALLA. ea ~@ ON THE CRYSTALLINE ROCKS OF THE LIZARD DISTRICT. 499 olivine rock, of igneous origin; and, secondly, the ophicalcite, con- sisting of calcite and serpentine intermixed and generally banded. He was glad that the Authors maintained the original igneous origin of the Cornish serpentine, which most resembled the former variety in the West of Ireland. With regard to the banding of the dykes of gabbro where they were in contact with the walls, he observed that this was a structure not uncommon amongst dykes of igneous rock, and he believed it to have originated during the cooling process. These bands were, in fact, planes of cooling, and the structure of the rock along the walls of the dyke contrasted with tie central portions, where the cooling process was slower, and allowed the formation of a more crystalline rock in which these planes were absent. The Presrpenr remarked that the questions discussed in the paper had far more than a mere local interest. In particular, the problem ot the banded structures among crystalline schists touched some of the profoundest difficulties of the theory of metamorphism. There was ground, he thought, for believing that mechanical deformation had been rather too freely appealed to as an explanation of the general banded and schistose structures of the older rocks. This cause had unquestionably been largely instrumental in the pro- duction of such structures ; but, as he had stated in his Anniversary Address, there were features of the more ancient gneisses which it was hard to imagine could be due to anything else than some original variations in the arrangement of the materials of the rock before soli- dification. He had been much struck with the extraordinary way in which some of the Tertiary gabbros of Skye simulate the rudely- parallel wavy lenticular banding of different materials in many gneisses; and he thought it was rather among such examples of flow- structure in eruptive rocks that the analogies of some of the struc- tures of the gneisses were to be sought. The Authors had, there- fore, in his opinion, done a service ‘in recalling the attention of geologists to this view of the subject. General M°Manown said that, as those who had taken part in the debate appeared to agree generally with the conclusions arrived at by the Authors, he would confine himself to calling special attention to one of the specimens exhibited, and to a brief description of a section which he thought had an important bearing on the subject under discussion. Prof. Bonney stated that the theory suggested by Prof. Hull had been, in substance, formerly held by himself, but that he had found cases for which it did not suffice. The case quoted by the President was of great interest, and he might add that since the paper was written he (the speaker) had seen others. He could not sit down without testifying to the value of Mr. Teall’s work at the Lizard, and begged the Society to remember that General M*Mahon was the originator of the right idea (as the speaker believed it to be) as to the foliation of the gabbro. Q.J.G.S. No. 188. Qn 500 PROF. H. A. NICHOLSON AND MR. J. E. MARR ON 27. The Cross Fert Intrer. By Prof. H. A. Nicnorson, M.D., D.Sce., F.G.S., and J. E. Marr, Esq., M.A., Sec. G.S. (Read April 8, 1891.) [Prats XVII.) ConrTENTs. Page S$), L. Entroductory ) sec Aewided. ciesege stereo ba 500 II. General Description of the Inlier ..................... 501 III. Detailed Description of the Strata....... co (naw tee 501 (a) The eastern portion of the Inlier. (0) The western portion of the Inlier. IV. Ages of the different members of the Coniston Litiestone Series? \<...).:..c.9edeanes ss agaseviess case ae 509 We Conclusion . a0 2.9.2 0c akc -caton te Cee eee ee 512 Appendix I. By Alfred Harker, Hsq. ...............¢--0.- 512 Appendix IL... ‘By A, Hl. Footd,; Hkq. ..2.4.-2..c-0e meee 526 § I. Lyrropucrory. Since the “ Description of an Insulated Group of Rocks of Slate and Greenstone in Cumberland and Westmoreland, on the East side of Appleby,” given by Dr. Buckland in the Transactions of the Geological Society (ser. 1, vol. iv. (1817) p. 105), a considerable number of papers have been devoted to the Lower Paleozoic rocks of that region, but as references to these are given in Mr. Whitaker’s list of papers bearing upon the geology of the Lake District, published in the late Mr. Clifton Ward’s memoir on “ The Geology of the Northern Part of the English Lake District,” it is needless to insert here an account of the bibliography of the area. An excellent summary of the work which has been achieved is given by Mr. J. G. Goodchild in a paper published in the Proceedings of the Geologists’ Association for 1889*, In this paper also considerable additions are made to our knowledge of the rocks of the Inlier, and to these we shall have occasion to refer. In the present paper, we have attempted to fix the ages of the various formations of Lower Paleozoic rocks in the Cross Fell district, to determine their organic contents, and to compare them with the corresponding rocks of other areas, rather than to give a detailed description of the general structure of the region. This latter will no doubt be done by the officers of the Geological Survey who have been engaged in mapping the district, with far fuller evidence than is at the disposal of those who have not examined the region in an exhaustive manner. Nevertheless, as a general knowledge of the structure of the region is necessary to understand the details which follow, we have drawn up a rough map which will serve as a guide until such time as the official Survey map is published, and we append a description which may serve to render the principal features intelligible. * Proc, Geol. Assoc. vol. xi. p. 258. THE CROSS FELL INLIER. 501 § IZ. Generar Descriprion or tHe INiIER. The Cross Fell Inlier of Lower Paleozoic rocks is marked by the occurrence of a group of pyramidal hills, stretching in a band from half a mile to a mile in width, for a distance of about 16 miles in a general N.W. to 8.E. direction on the west side of the Pennine escarpment, from a little north of the village of Melmerby on the north to the south-west flanks of Roman Fell on the south. It is bounded by two great faults, which enclose it as an elongated spindle-shaped mass. ‘The eastern fracture, which may be termed the Escarpment Fault, brings the Lower Carboniferous rocks against those of Lower Palwozoic age, whilst the western one, the Pennine Fault of the older writers, and which Mr. Goodchild terms “the Outer Pennine Fault,” places the Lower Paleozoic rocks in juxtaposition with the New Red Sandstone for many miles. The lenticular inlier between these faults is furthermore broken by another great N.W. and S.E. fault, bringing Lower Ordovician rocks on the east side against the Higher Ordovician and Silurian strata of the west. This is “the Middle Pennine Fault” of Mr. Goodchild; but as a great part of the displacement here was produced at a much earlier date than that due to the other two faults, we consider it better to speak of it as the Knock Pike- Flagdaw Fault, as it is well seen between the two hills bearing those names. It is true that subsequent movement has occurred on this line along part of the course of the fault, so that at the north end of the Inlier the Lower Carboniferous rocks are includea between the Ordovician and New Red Sandstone deposits, but to the south the fault is seen to pass under the Carboniferous con- glomerates of Roman Fell with little or no disturbance of those rocks, and emerges again on the south-west side of the hill where the Lower Paleozoic rocks are developed. By means of the Knock Pike-Flagdaw Fault the lenticular inlier is divided into an older eastern and a newer western portion, and it will be convenient to describe these separately, commencing with the older rocks which lie to the east of the Knock Pike-Flagdaw Fault. Unfortunately none of the rocks which occur on that side of the fault are seen to the west of it, and consequently a considerable gap occurs in the succession here, though how great it is hard to say. § III. Derarrep Descriprion oF rue Srrara. a. The Eastern Portion of the Inlier.—Most of the sedimentary rocks on the eastern side of the Knock Pike-Flagdaw Fault have been referred to the Skiddaw Slates, but as only few fossiliferous localities have been detected in this area, the correlation has been made to a large extent from similarity of lithological character. As the evidence furnished by these rocks is at present insufficient to establish a detailed sequence, we shall content ourselves with a wery brief notice of the deposits, for it will be necessary to devote 2N 2 502 PROF. H. A. NICHOLSON AND MR, J. E. MARR ON considerable attention to the Skiddaw Slates of the more central portion of the Lake District before their minute subdivisions can be satisfactorily determined. We believe that the oldest rocks of the Cross Feli Inlier occur in the extreme north-eastern portion of the inlier in the neighbourhood of Cuns Fell, where they are probably separated by yet another N.W. to S.E. fault from the newer rocks to the west. Here, in the course of Dry Sike and Hungrigg Sike, a series of greenish shales are seen dipping in a general south-westerly direction at a high angle. ‘These shales furnished the obscure fossil described by one oi us in the Geol. Mag. for 1869 (pl. xviii. p) as possibly of vege- table origin; but this affords no clue as to the precise age of the series. They are succeeded to the south by blacker shales and grits which bear considerable resemblance to the older Skiddaw Slates of the Lake District, and they are probably contemporaneous with these, though no fossils are recorded from them in this area, and we have found none. Similar beds are again seen farther south at Brownber, and in the streams adjoining it, and they continue over a considerable part of the ground east of the Knock Pike-Flagdaw Fault, as far south as Roman Fell. At Brownber (and in a few other places such as Murton Pike and the neighbour- hood of Keisley) they have undergone great disturbance, and are penetrated by numerous quartz-veins, which are folded with the strata, causing considerable changes in the rocks, as notified by one of us at the Newcastle meeting of the British Association in 1889. Mr. Alfred Harker has kindly examined sections of these rocks for us, and has furnished us with an appendix to our paper, giving descriptions of these and other rocks of this area. Next in order of age we would place the black shales, which occur repeated thrice, firstly in Ashlock Sike and the neighbour- ing tracts east of Ousby, next in Ellergill and the adjoining ground under Cross Fell, and lastly im the course of Knock Ore Gill. These strata—which we term the “ Ellergill beds ”—have yielded an abundant harvest of fossils, of ahaa a list is given in Prof. Lapworth’s paper on “ the Geological Distribution of the Rhab- dophora ” *, most of them being well-known Upper Arenig forms. At the summit of the beds referred to the Skiddaw Slates, Mr. J G. Goodchild locates his ‘‘ Milburn Group,” consisting, as he has pointed out, of a series of slates alternating with submarine tufts. These are excellently displayed in Wythwaite Hole, and in the streams to the south of it, where they are also intercalated with lavas. Few fossils have yet been discovered in them, but Mr. Good- child records Didymograptus Murchisoni, Boeck, and we have found — Diplograptus dentatus, Brongn., so that the beds are probably closely related to the Ellergill Group, from which they differ in the occur- rence of the volcanic material. Mr, Goodchild rightly insists upon the importance of his discovery as throwing lhght upon the vexed question of the relationship of the Skiddaw Slates to the volcanic * Ann. & Mag. Nat. Hist. ser, 5, vol. iii. p. 23, sep. cop. THE CROSS FELL INLIER. 503 rocks of the Borrowdale Series, but into this question we cannot enter here. Mr. Harker has examined slides of rocks from the Milburn Group of Wythwaite and the neighbourhood, and his description will be found in Appendix I. to this paper. Though all the bedded rocks on the eastern side of the main inlier are either shales or lavas and ashes interstratified with shales, we do get another group of volcanic rocks on the east side of the Knock Pike-Flagdaw Fault in a subsidiary inlier which lies east of the village of Melmerby, and which is separated from the main inlier by a band of Carboniferous conglomerate, a few score yards in width, abutting against the New Red Sandstone. At the south end of this small inlier a group of basic rocks, consisting of ashes and porphyritic and vesicular lavas, is separated from a group of rhyolitic rocks by the above-mentioned fault. Unfortunately, owing to the intervention of the Carboniferous beds, which are here faulted down, the relationship of the basic lavas and ashes to the shales farther south is not seen. That these basic volcanic rocks are the equivalents of the Eycott lavas was recognized by Mr. Goodchild *, and a porphyritic rock, of which a beautiful specimen from Rake Brow is preserved in the Museum of Practical Geology (London), is quite similar to one of the well-known porphy- ritic Eycott lavas +. Unfortunately this group of rocks is flanked by Carboniferous rocks or by faults on all sides, so that its true relationship to the other rocks of the district is not shown. The rocks of the eastern half of the inlier are also marked by the intrusion of a considerable quantity of igneous matter, and as an examination of the intruded rocks is of importance as throwing light upon the general sequence of events in-the district, we may give a brief account of their development. The principal masses occupy the prominent ridge of Cuns Fell, and the slopes of Thack Moor. Cuns Fell is formed mainly of diabase running in a general north-east to south-westerly direction, and sending off a considerable tongue to the south. On the east side of the hill, in Ousby Dale, much felsitic rock is associated with the diabase, under such conditions that it is difficult to make out the relations of the two rocks, though on the summit of the hill a felsitic dyke is undoubtedly intrusive in the diabase. Whether or no the felsitic rocks are the newer, they probably belong to the same general period. ‘That they were intruded before the end of Silurian times is rendered probable by the absence of cleavage in the highly baked shales below the. mass, and by the existence of a schistose structure in the diabase, seen at the spring in Ousby Dale. This appears to indicate that they were intruded previously to the exertion of the pressure which has folded and cleaved the rocks, and we are inclined to believe that the masses are of the same general age as the volcanic material which forms the Borrowdale series, and are related to the volcanic outpourings of that group. * Trans. Cumb. and Westm. Assoc. vol. ix, (1884) p. 183. + For notes on these rocks see Appendix I. 504 PROF. H. A. NICHOLSON AND MR. J. EB. MARR ON It is noticeable in this connexion that we get several complexes of acid and basic intrusive rock in the North of England, such as might well supply material for the formation of the Ordovician lavas. The rock of Thack Moor is chiefly felsitic, and though it covers a considerable amount of ground, it is usually much decom- posed at the surface. ‘Two other felsitic masses occur to the south of this, one parallel with the Maiden Way and the other beneath Cocklock Scar. The only other intrusive rock to which we would call special attention is a broad mica-trap dyke seen in Dry Sike, east of Melmerby, and which is of interest from the greater metamorphism produced along its margin than is usual with this class of rock in the North of England. b. Vhe Western Portion of the Inlier.—The rocks on the western side of the Knock Pike-Flagdaw Fault consist exclusively of the upper portion of the Ordovician and the lower part of the Silurian rocks, so that along the line of the above-mentioned fault the greater part of the Borrowdale series of rocks appears to have been cut out. The general strike of the rocks is N.W. and S.E., showing a marked divergence from that which characterizes the beds of the more central portions of the Lake District. One of the principal features of this half of the Inlier is the existence of a set of N.E. to S.W. faults, which causes frequent repetition of the Upper Ordovician and Lower Silurian rocks. By these faults the west side of the Inlier is divided into a series of rectangular or triangular blocks, with a similar succession in each, and it will save space if we describe in some detail the block which exhibits the most complete section, and afterwards note similarities and discrepancies. The rhyolitic pyramidal hill known as Dufton Pike is separated from the similar hill of Knock Pike to the north by one of the above-mentioned transverse faults, which may be spoken of as the Cosca Fault. To the N.E. of this fault, the stream called Swindale Beck runs along the south-east flank of Knock Pike to the village of Knock, and in it the following section is displayed (Pl. XVIL.). The line of the Knock Pike-Flagdaw Fault is here marked by an intrusive mass of mica-trap noticed in Appendix I. The shales on the east side of the fault, the dyke itself, and the rhyolite are all exposed in a small quarry by the side of the moorland road, N.W. of the beck. Below this a capital exposure of the rhyolite (2) which forms Knock Pike is seen in the course of the stream. It is succeeded by thin layers of rather fine, apparently unfossiliferous ashes, which seemingly pass up into the remarkable beds numbered 3 in the section. These consist of calcareous shales, with nodular masses of limestone, crowded with fossils, some of the calcareous bands being exclusively composed of the valves of Beyrichia. One bed of the series has been spoken of by. Professors Harkness and Nicholson * as the “ Discina-corona bed,” and we propose to name * Quart. Journ. Geol. Soc. vol. xxxiii. (1877) p. 463. THE CROSS FELL INLIER. 505 the series the “‘ Corona Series.” The fossils contained in the series are of great interest, and altogether different from those embedded in the overlying strata. We append a list of those which have been found in the stratum in this stream :— Beyrichia Wilckensiana, Jones. Primitia semicircularis, Jones and Holl. Lingula tenuigranulata, M‘Coy. Strophomena grandis, Sow. The series (4) consists mainly of black and blue shales, often calcareous, interstratified with bands of calcareous rock which have _ undergone considerable disturbance. These are the well-known lufton Shales. Their fauna is quite similar to that of the main division of the Coniston Limestone of the Lake District, so that whether we suppose that this is a more specially shaly base of the Coniston Limestone series, or an argillaceous representative of the whole of that series, we are fully persuaded that ‘these Dufton Shales are of the age of the Coniston Limestone, and not an under- lying deposit as has been previously asserted. We have found in the Dufton Shales of this stream the undermentioned fossils :— Dicellograptus complanatus, Lapw. (?). Diplograptus socialis, Lapw. (?). Calymene senaria, Conr. Cybele verrucosa, Dalm. Illenus Bowmanni, Salt. Phacops Brongniartu, Portl. Remopleurides Colbu, Portl. At the point where a tributary stream (Rundale Beck) enters Swindale from the east, the Dutton Shales are succeeded by a very calcareous deposit, numbered 5 in the section. This consists of thick beds of whitish lmestone with peculiar ashy-looking green shales. Though this deposit at first sight bears far stronger resemblance to the ordinary Coniston Limestone than do the Dufton Shales, an examination of the fossils indicates clearly that it is not Coniston - Limestone, but is the equivalent of a thin band at the base of the Ashgill Shales in the Lake District, which one of us has previously referred to (Quart. Journ. Geol. Soc. vol. xli. (1885) p. 487) as the ** Staurocephalus zone.” We may speak of this limestone as the “ Stawrocephalus Limestone” ; in Swindale it has yielded the following fossils, which are mostly found in the interstratified calcareous green shales :-— Echinospherites arachnoideus, Forbes. Turrilepas. Acidaspis. Illenus Bowmanni, Salt. Lichas laciniatus, Wahl. Phacops Jukes, Salt. Phillipsinella parabola, Barr. (?). Staurocephalus globiceps, Portl. a06 PROF. H. A. NICHOLSON AND MR. J. E. MARR ON Trinucleus seticornis, His. Orthoceras sp. No. 6 of the section consists of blue shales with Strophomena siluriana, Dav., and the undermentioned fossils, all found in the corresponding Ashgill shales of the Lake District :— Phacops mucronatus, Brongn. (?). Orthis biforata, Schloth. elegantula, Dalm. —— protensa, Sow. Orthisina sp. Strophomena silurtana, Dav. Above the Ashgill Shales, a strike-fault cuts out the Skelgill beds in the main stream, but they are found in the tributary stream, Rundale Beck. These and the deposit No. 7 of our section (the Browgill series) have been previously noticed in our paper ‘‘On the Stockdale Shales,” and we have nothing to add to the description given therein. The Browgill beds pass up as usual into the blue flags with Monograptus vomerinus (No. 8), which apper- tain to the Lower Coniston (Brathay) Flags, and which represent the Wenlock Shales of other areas. These, as shown in the map and section, abut against the New Red Sandstone (No. 9) which is thrown against them by the Pennine Fault just east of the village of Knock. In continuing our description of the rocks it will be convenient to consider the blocks into which the west side of the Inlier is divided by the cross-faults. The block to the south of the one iast described is about two miles iong, and is bounded on the south by the Harthwaite Fault. It is occupied by a greatly disturbed syncline having the rhyolites of Dufton Pike to the north-east, and those of Wharleycroft with a thin band of andesite (the latter probably the oldest rock seen in the western half of the Inlier) to the south-west. Between these the moory country is mainly occupied by the Dufton Shales, well exposed in Hurning Lane, Pusgill, Dufton Town Sike, Billy’s Beck, and Harthwaite Beck. ‘That they are much disturbed is shown, not only by the great crumpling which they have undergone as seen in actual section, but by the occur- rence of lenticular outliers of higher strata on the 8.W. slope of Dufton Pike, at Pusgill House, and near the head of Billy’s Beck, and of an inlier of lower deposits in Harthwaite Beck, against the Harthwaite Fault. The principal point to be noticed concerning the rocks of this block is the very fossiliferous character of the “ Corona-beds” of Pusgill, of the Dufton Shales of Pusgill, Dufton Town Sike, and Billy’s Beck, and of the Stawrocephalus Limestone near the head of the latter, containing, amongst cther fossils, Stawrocephalus globiceps, Portl., and a new species of Skenidium. South of the Harthwaite Fault is a triangular block apparently extending to Murton, though no exposure has been seen by us in THE CROSS FELL INLIER. 507 its southern portion. In this block comes the Keisley Limestone, to the consideration of which we must devote a few words. It is well known that the mass of limestone at Keisley, which has been frequently described, is a white or pink crystalline rock, often crowded with fossils. It occupies the southern part of Keisley Bank, and appears to be of considerable thickness, having a general southerly dip at high angles. That it is faulted against the rhyo- litie series of Keisley Bank, Harthwaite, and Gregory is evident, for it rests on an ash at the east end of the block, whilst to the west it reposes on the rhyolite which forms the summit of Keisley Bank and widens out westward. ‘This fault is also apparently one of low hade. The limestone itself shows signs of much disturbance : it contains twisted wisps of shale in places, and the occurrence of beds containing numerous /é/eni with their convex surfaces uni- formly pointed downwards indicates inversion. A list of the fossils has been previously published, but as additional forms have been found and corrections must be made in this list, we here append a fresh one :— Halysites sp. Lindstremia sp. Primitia Maceoyu, Jones. Ampyx tumidus, Forbes. Cheirurus bimucronatus, Murch. cancrurus, Salt. clavifrons, Dalm. (?). Cyphaspis (?) cf. triradiatus, Tornq. Cyphoniscus socials, Salt. Cytheropsis phaseolus, His. Homalonotus punctillosus, Torngq. Illenus Bowmanm, Salt. cf. contfrons, Billings. sp. Lachas laciniatus, Wahl. lawatus, M‘Coy. Remopleurides, cf. longicostatus, Portl. Sphereaxochus calvus, M‘Coy. Atrypa expansa, Lindstr. (?). Orthis Actonice, Sow. porcata, Sow. testudinaria, Dalm. -—— vespertilio, Sow. Strophomena corrugatella, Dav. deltoidea, Conrad. expansa, Sow. rhomboidalis, Wilckens. Loxonema obscura, Port. Orthoceras *, cf. elongatocinctum, Porth. * For notes on this and other Cephalopods, see Appendix II. yerive supplied to us by A. H. Foord, Esq., F.G.S. 508 PROF, H. A. NICHOLSON AND MR. J. E, MARR ON It has been generally recognized that this list indicates the Coniston-Limestone age of the deposit. Nevertheless, unequivocal Dufton Shales are found only one-third of a mile away in a stream between Keisley and Wharleycroft, and probably approach close to the limestone. We believe that the Keisley Limestone may have been thrust in a north-easterly direction for some distance, and that the limestone bands have been thickened by folding during this process, whilst the shales have been to a large extent squeezed out. On the other hand, the Dufton Shales have probably had their argillaceous mem- bers largely repeated, and the limestones pulled out into lenticular masses. This seems to us the mode of explaining the great dif- ference in the lithological characters of two deposits occurring close together and containing fossils of the same age, which accords best with the observed facts, though before finally accepting it we would advocate a closer study of similar lenticular masses of lime- stone which occur elsewhere, and, so far as we are aware, always in disturbed districts. Another block occurs to the south of the one just described, and is terminated by a fauit on the S.W. flank of Roman Fell, east of the farmhouse of Fell Dikes. It is also about two miles long, and is remarkable chiefly on account of the interesting development of the ‘* Corona-beds” shown on the western slopes of Roman Fell, which have been noticed by Mr. Goodchild*, who rightly records the existence of Lower Bala fossils in that locality. These “Corona- beds” are situated above the rhyolite of the flanks of Roman Fell, and doubtless pass under the Dufton Shales of the Hilton Beck Smelt Mill. Three principal subdivisions are noticeable in Lycum Sike and the streams to the south. Resting on the rhyolite are pink ashes with Orthis testudinaria, Dalm., in abundance. Above these are pink shales with fine examples of Trematis corona, Salt., and Lingula tenuigranulata, M‘Coy, whilst the highest beds seen, which abut against the Pennine Fault, are very calcareous ashy beds crowded with gasteropods and lamellibranchs, and containing occasional specimens of the two above-mentioned horny brachiopods. The pink staining noticeable in these rocks is no doubt due to percolation from the overlying conglomerates. One more block is found south of this: a cross-fault once more brings up a mass of rhyolite, which runs southward for one third of a mile, when it is cut off at Howgill Fold by the great fault which brings down the Carboniferous conglomerates as vertical beds converted into quartzite (see Appendix I.). This block is chiefly noticeable on account of the re-appearance of the Knock Pike-Flagdaw Fault from under the conglomerate, so that a small triangular patch of leaden-grey shales appertaining to the rocks of the eastern side of the Inlier is seen on the hillside north of Howgill Fold. The last block to be noticed occurs in the extreme north of the * Proc. Geol. Assoc. vol. xi. (1890) pp. xev & 263. ee Ee ee THE CROSS FELL INLIER. 509 district, forming a great part of the subsidiary inlier which has been already noticed as furnishing representatives of the Eycott voleanic rocks. These basic rocks are faulted against the rhyolitic lavas and ashes of Shield Green, east of Melmerby. Above the highest lava is an ash apparently unfossiliferous, and the possible equivalents of the “* Corona-beds” succeed it in the form of green ashy fossiliferous shales, which are exposed on a fell-road leading out of the Alston Moor road. These however, may belong to a somewhat lower horizon. They contain the following fossils :— Prasopora Graye, Nich. & Eth. Jun. Amphion pauper, Salt. (?). Cyphaspis megalops, M‘Coy (?). Trinucleus Goldfussi, Barr. (?). Orthis testudinaria, Dalm. Above these calcareous ashes are greatly-disturbed calcareous ashy shales, which belong either to the ‘ Corona-beds” or to the base of the Dufton Shales. They are seen on the high road, and have yielded :— Prasopora Graye, Nich. & Eth. Jun. Callopora pillula, Nich. & Eth. Jun. Diplograptus sp. Agnostus sp. Trinucleus seticornis, His. (?). Lingula tenuigranuldta, M‘Coy. Orthis elegantula, var. (?), Dalm. plicata, Sow. Triplesia (2) spiriferoides.. Strophomena rhomboidalis, Wilckens. Above and beyond these are representatives of the Stockdale Shales. One more subsidiary inlier north of this shows the highest Lower Palseozoic beds seen in this district. These are the Coniston Grits which are seen dipping in a southerly direction in Limekiln Beck. A specimen of these grits has been long exhibited in the Museum of Practical Geology, and the deposit has been recognized by the officers of the Geological Survey as belonging to the Coniston Grit Series. § IV. AGEs or THE DIFFERENT MEMBERS OF THE CoNISTON Limestone SERIEs. We have already compared the earlier and later accumulations of the Cross Fell area with their equivalents in the main part of the English Lake District, and we need only add a few words con- cerning the rocks which lie between the rhyolites and the Stock- dale Shales, for these are more fully developed here than in other parts of the North of England. 510 PROF. H. A. NICHOLSON AND MR. J. E. MARR ON (a) Corona Series.—The principal variations in the lithological characters of this group were described when discussing the succes- sion in Swindale Beck and on Roman Fell. They appear to con- sist essentially of calcareous ashes, with their limestones, the ashy matter becoming more abundant. and coarser as we go southwards ; though if the beds on the Fell Road at Melmerby actually belong to this division, this statement must be modified, for they also contain much ashy matter. It is, however, possible that they are on a somewhat lower horizon than the Corona-beds proper, as has been already suggested. They somewhat closely resemble the Balclatchie beds of the Girvan district, but a much larger collection of fossils than that which we have acquired must be made before a definite opinion as to their age can be offered. We have already mentioned the fossils found in the Corona-beds at Swindale Beck, where the fossils are not very numerous. In other localities the yield has been far more abundant, and we append a list of the forms we have obtained from these beds:— Monotrypa sp. Pusgill. Conchicolites gregarius, Nich. Pusgill; Roman Fell. — Ateleocystites sp. Roman Fell. Beyrichia Wilckensiana, Jones. Pusgill; Roman Fell. Primitia semicircularis, Jones & Holl.. Pusgill. Homalonotus rudis, Salt. (?). Roman Feli. Lingula tenvigranulata, M‘Coy. Pusgill; Roman Fell. Orthis testudinaria, Dalm. ‘ Roman Fell. Trematis corona, Salt. Pusegill; Harthwaite Beck ; Roman Fell. Ambonychia gryphus, Portl. Pusgill; Roman Fell. Bellerophon acutus, Sow. (?). Roman Fell. bilobatus, Sow. Pusgill; Roman Fell. Actinoceras Pusgillensis, n. sp. Pusgill. | Cyrtoceras (?). Roman Fell. The gasteropods and lamellibranchs which occur so abundantly on Roman Fell, and less numerously at Pusgill, would require the attention of a specialist for their satisfactory determination. - The genera Ctenodonta and Pleurotomaria appear to be represented by several species. These Corona-beds seem to be older than anything which has been referred to the Coniston Limestone Series in the main Lake District. Their fauna is a very marked one, and is entirely different from that of the ordinary Coniston Limestone; and we are not aware of any similar fauna having been recorded in the British area, though it is probable that when the fossils of the Ardwell Group of the Girvan district are described they will be found to present considerable affinities to the forms which we have found in Westmorland. . Abroad, we have two calcareous deposits whose faunas are closely related to that of our Corona-beds, viz.: the Beyrichia-limestone of THE CROSS FELL INLIER. 51] Scandinavia, and the Trenton Limestone of North America. As the overlying Z'rinucleus-shales of the former region, and the Utica Slates and Hudson-River beds of the latter, can be closely paralleled with the succeeding Dufton Shales of the Cross fell district, it seems highly probable that the Corona-beds may be on about the same horizon as the Beyrichia-limestone and the Trenton Limestone. When the beds of Wales and the Welsh border-land have been worked out in greater detail, we may expect to find a similar fauna recorded from beds situated between the Llandeilo and Bala Limestones. (b) Dufton Shales.— We have already given a list of fossils from these beds as developed in Swindale. ‘The annexed list shows those which we have obtained from other localities :— Diplograptus truncatus, Lapw. Hurning Lane. Acidaspis, . sp. Pusgill. Ampyzx tetragonus, Ang. Pusgill; Billy’s Beck. Calymene senaria, Conrad, Pusgill; Hurning Lane ; Dufton Town Sike. Cybele Loveni, Linnrs. Dufton Town Sike. verrucosa, Dalm. Pusgill ; Dufton Town Sike. Homalonotus bisulcatus, Salt. Pusgill. Lichas laxatus, M‘Coy. Pusgill. Trinucleus concentricus, Eaton. Pusgill. seticornis, His. Pusgill ; Hurning Lane. Youngia trispinosa, Nich, & Eth. Pusgill; Hurning Lane. Leptena sericea, Dalm. Pusgill; Hurning Lane. transversalis, Wahl. Hilton Beck. Lingula ovata, M‘Coy. Hilton Beck. Orthis biforata, Schloth. Dufton Town Sike. testudinaria, Dalm. Pusgill; Harthwaite Beck. vespertilio, Sow. Dutton Town Sike. Strophomena expansa, Sow. Harthwaite Beck. Most of these fossils are common in the Coniston Limestone, the Bala Limestone, and the Z'rinucleus-shales of Sweden, and there is no doubt that the Dufton Shales, if not actual representatives of the Coniston Limestone, are far more closely allied to it than to the underlying “* Corona-beds ” with which they have hitherto been associated. (c) Keisley Limestone-—As to the general age of this there is no doubt. The group of fossils is essentially that of the Coniston Limestone. At the same time, the occurrence of some forms which have not been found nearer than the Chair of Kildare may indicate that we have here a fossil zone which is not represented by fossili- ferous beds in the central part of the Lake District. (d) Staurocephalus-Limestone, and (e) Ashgill Shales.—As these beds are quite similar to the corresponding strata of the Lake District, it is unnecessary to say anything further about them. 512 MR. A. HARKER ON ROCKS § V. ConcLusion. Although the rocks of this district are so greatly disturbed, an examination of the richly fossiliferous deposits indicates beyond doubt, in most cases, the original order of succession of the strata. We find that, though in general the deposits are similar to those of the adjoining Lake District, there are important variations in detail, especially with regard to the Coniston Limestone group, and we feel convinced that a study of this Cross Fell area is absolutely neces- sary to the right understanding of the sequence of events which marks the history of the Lower Paleozoic rocks of the North of England. EXPLANATION OF PLATE XVII. Map of the Cross Fell Inlier on the scale of one inch to a mile. A portion of the same on the scale of two inches to a mile. Section in Swindale Beck on the scale of six inches to a mile. Fig. 1. Fig. 2. 3. Fig. APPENDIX I.—Perronoercat Notes on Rocks From the Cross Feta Inter. By Atrrep Harxer, Esq., M.A., F.G.S., Fellow of St. John’s College, Cambridge. 1. SxippAw SLATEs. It will not be practicable here to deal with more than the leading rock-types of a rather complex area. Further, as my own acquaint- ance with the district is slight, and most of the specimens studied were collected by Prof. Nicholson and Mr. Marr, these notes will not aim at being more than merely descriptive, and must be regarded as only supplementary to the field-work embodied in the foregoing paper. The general character of the Skiddaw Slates is known from the descriptions of numerous writers on the Lake District. Jn parti- cular, the occurrence in the group of subsidiary volcanic rocks has been indicated on the Geological Survey maps and in Mr. Clifton Ward’s Memoir. It will be sufficient here to notice certain modifi- cations of the slate-rocks due to metamorphic agencies, and to describe two or three examples from the volcanic portions of the group. The evidences of dynamo-metamorphism in the Skiddaw Slates of Brownber, &e., have already been pointed out by Mr. Marr. Judging from his specimens, the rocks thus modified still part along the surfaces of original deposition, as marked out by alternations of purely argillaceous and more gritty bands; but on the wavy FROM THE CROSS FELL INLIER. 513 divisional surfaces thus obtained are seen minute wrinkles having the same direction as the broader undulations, while a cross-section shows that these small folds have for the most part passed into little parallel faults making a high angle with the bent surfaces of lamination. In a thin section |913]* the gradual passage of the minute folds into reversed faults is beautifully exhibited, and many more are brought to light than can be detected in a hand- specimen. ‘There are sometimes as many as two or three hundred in an inch, or even more. All the appearances recall the micro- structure of the *‘ gnarled” beds near Amlwch, &c., in Anglesey f. In the specimens from Brownber the wrinkled lamination-surfaces present a dark and glossy aspect, which the microscope shows to be due to the development of a chloritic or micaceous mineral in the rock. The chief secondary product is a flaky mineral showing the strong cleavage of the micas, chlorites, &c., and giving sensibly straight extinction. The flakes vary from pale greenish-yellow to colourless, the absorption being stronger for vibrations parallel to the cleavage-traces than for those perpendicular. The least axis of the ellipsoid of optic elasticity is at right angles to the cleavage. The birefringence, roughly estimated by comparison of the polariza- tion-tints with those of quartz, is usually about 0°012, but some- times as much as 0:014. These figures correspond in the table of Lévy and Lacroix to clintonite and delessite respectively. Further, there are in places small colourless flakes giving much higher double refraction and agreeing in character with muscovite. It is evident that, besides the dominant chloritic mineral, a micaceous one is also present, and the appearances suggest that the latter represents a further stage of metamorphism than the former. The brilliantly polarizing mica appears only on planes of actual discon- tinuous movement in the slate or in little isolated flakes in the gritty bands, and these are evidently the places where the mechan- ical stresses developed would reach a maximum. It appears that the discontinuous movement in the mass of the rock has been effected after the production of the chloritic mineral which almost completely pseudomorphs the original argillaceous material, and the flakes, except where they have been dragged along in the slipping, lie obliquely to the little faults. The gritty bands in the rock sometimes retain their clastic appearance, but in some cases their appearance suggests recrystal- lization in situ. The constituents are quartz and felspar, among which occur sparsely flakes of the chloritic mineral and the colour- less mica. The quartz often shows something of the ‘ undulose ” or “ spectral ’’ polarization indicative of a condition of strain. The felspar is frequently twinned, and seems to embrace both orthoclase and an-acid plagioclase. The perfectly pellucid character of the little erystal-grains and, in some places, the fashion in which they * The numbers in square brackets refer to the microscopic rock-sections in the collections of the Woodwardian Museum, Cambridge. +t See Rep. Brit. Assoc. for 1885, pp. 859, 840. 514. MR. A. HARKBPR ON ROCKS fit into one another can scarcely be explained except on the sup- position that they have been recrystallized Buaer the influence of mechanically produced stress. These metamorphosed slates of Brownber contain pseudomorphs of limonite, about a tenth of an inch in diameter, evidently replacing cubes of ‘pyrites. The pyrites has been formed in situ, for occasional grains of quartz, &c. are enclosed. Moreover, its decomposition has been subsequent to the crushing of the rock, for the cubes are not sensibly distorted. Indeed, the movement of the rock about the pyrites-crystals has been such as to leave vacant spaces, after- wards filled by secondary quartz. This quartz has a rather fibrous structure, and is arranged at right angles to the faces of the pyrites cubes. Itis found only on those parts of the cubes where the pressure would be relieved by the flowing movement of the rock- mass, and the phenomena are precisely similar to those which I have elsewhere described as common in pyritous slates *. As to modifications produced by thermal agency, a few words will suffice. The Skiddaw Slates show some degree of metamorphism near their contact with the Cuns Fell diabase. A slice in Prof... Nicholson’s collection is a well-laminated rock, in which numerous minute grains of clastic quartz are mingled with the argillaceous material. It is marked throughout with irregularly ovoid spots, one-fiftieth to one-hundredth of an inch in diameter. Along certain bands these spots are merely clear patches due to the dusty (carbonaceous ?) matter having been expelled, to collect just beyond the margin. In other bands the clear spots thus left behave optically in a different way from the surrounding ground, being mostly dark between crossed nicols. The quartz-sand occurs indifferently inside and outside the spots, and the grains have lost nothing of their sharpness of outline. Near the large Jamprophyre dyke in Dry Sike, again, the Skiddaw Slates appear highly metamorphosed, being converted into a very compact black rock with a certain degree of lustre and a conchoidal fracture, like some varieties of hornfels. A slice of this rock, which is rather a microscopic grit than a true slate, shows as the chief metamorphic product a rather obscure chloritic mineral. The numerous minute quartz-grains retain their angular outline [912]. Among the lavas occurring in the Skiddaw Slate group, an interesting rock was collected by Mr. Marr in the stream north- west of Master Sike [920]. It is an andesite consisting essentially of an isotropic base crowded with very minute felspar-microliths. These only occasionally show any parallel arrangement, although a streaky fluxion-structure is seen in the mass as a whole. There are a few small porphyritic felspars with good outlines. No augite is recognizable, though its former presence is probably indicated by the pale delessite-like substance filling some small ovoid vesicles in the rock. The interesting point isa vesicle about a twelfth of an inch in length, with a complex structure recalling in some respects * Geol. Mag. (1889) pp. 396, 397. FROM THE CROSS FELL INLIER. 515 that of the lithophyse in some acid lavas. There are, however, novel peculiarities. The cavity has been at one time lined with a thin coating of a pale-green chloritoid mineral, which for brevity may be called delessite. It has a fibrous structure, with partial fan-like grouping, roughly perpendicular to the surface on which it was deposited. From this surface it has for the most part broken away, so as to divide the cavity by partition-walls, not however continuous. A second coating of the same material has also to some extent become detached, and with it portions of the andesitic matrix itself. Subsequently crowds of minute but perfectly-formed felspar prisms have been formed, clustering especially, with a ten- dency to perpendicular growth, on the detached fragments of andesite where these were not protected by a coating of delessite. Finally, all the remaining space has been occupied by ciear crystal- line quartz. ‘The little felspar-crystals are clear, and. invariably have twin-lamellation. The birefringence is very near that of quartz, and sections nearly perpendicular to the twin-plane give extinction-angles up to about 18°. These characters do not distin- guish between albite and andesine. ‘The curious feature is the clear evidence that the felspar-crystals were formed within the vesicle subsequently to the deposition of the usual coating of green decom- position-product. A few days after the preceding paragraph was written, Mr. W. Maynard Hutchings informed me of his independent discovery of felspar within the vesicles of some Lake District rocks, and the specimen which he kindly lent me showed relations in some respects analogous to those briefly described above. The subject is one which will no doubt repay further investigation, and we may expect that Mr. Hutchings’s work will throw hght on this curious mode of occurrence of felspar. The rock exposed in Wythwaite Hole seems to be a contempora- neous lava of more crystalline type (dolerite), but is too deeply altered for minute study. Besides evident spherical vesicles, there are seen under the microscope little irregular spaces occupied by - quartz-mosaic, but the manner in which the lath-shaped felspars project into these renders it doubtful whether the spaces were originally vacant [1321]. A very singular rock occurs on Wythwaite Top. To the eye, it appears a coarse ash or fine breccia. Besides minute glistening felspar-crystals in the general mass, there are little fragments which themselves enclose felspars.. In aslice [1322] the fragmental character is scarcely apparent. Idiomorphic felspars are scattered through the rock, showing twin-striation of the ordinary kind, occasionally crossed by pericline-lamelle. Rarely there is a grain of quartz of clastic appearance, or a green pseudomorph which seems to come from a rhombic pyroxene. ‘The general ground of the rock appears in ordinary light partly turbid, partly clear, the two occurring in intermingled irregular patches. The turbid por- tion presents a finely “ felsitic” appearance, but the clear ground consists almost entirely of a mass of perfectly pellucid small crystals Q.J.G.S. No. 188. 20 516 MR. A. HARKER ON ROCKS and grains of felspar. Most of these have twin-lamellation and often an imperfect prismatic shape, so far as their crowded occur- rence permits. Others are only once twinned, and some shapeless simple grains, with a tendency to occur interstitially, are perhaps quartz. The structure of these patches and the limpid nature of the felspars are characteristic of metamorphic rocks, and point unmistakably to recrystallization of the rock i situ. Any doubt is removed by an examination of the porphyritically disposed felspars already noticed. These are frequently bent and broken, and there seems to be some secondary twin-lamellation induced by stress. The most striking feature, however, is the replacement of the original turbid crystals by new felspar-substance exactly similar to the little striated felspars in the general ground of the rock. Some of the original crystals do not show this alteration ; others are partially changed; and others, again, are totally converted into clear crystalline aggregates, preserving only the outline of the crystal from which they have been formed. The newly-formed triclinic felspars, cut perpendicular to the twin-plane, give a maxi- mum extinction-angle of about 18°. This agrees with albite, though a certain variety of andesine would give the same value. A curious point about the dynamo-metamorphism is the seemingly capricious manner in which it has affected the mass of the rock. As an example of the ordinary fine ashes of the Skiddaw Slate group, a rock from Burney was sliced [921]. It shows plenty of felspar-crystals, or more frequently broken portions of crystals, some showing twin-lamellation, others not. The general mass of the rock has probably been a felspathic dust, but now contains a quantity of quartz and calcite, besides pyrites and yellow opaque spots of ferruginous matter. Of these the quartz seems to be the. latest-formed, occurring in little continuous patches of ragged out- line and enclosing other decomposition-products. Little, if any, of this mineral is in original grains, and there is no indication of detrital material mingled with the volcanic. The grits in this group of rocks are derived from the disintegra- tion of igneous rocks of more than one kind. A specimen from the north side of Mudgill Sike [964] was found to consist essentially of grains of quartz, unrolled crystals of plagioclase, and rather rounded fragments of a microlithic andesite. A few rolled granules of decomposing augite occur, and a fragment of quartz-porphyry showing a microgranitic groundmass and a portion of a porphyritic quartz. The quartz-grains, which are the most abundant consti- tuent, are mostly subangular to rounded, but some of the smaller ones are quite angular. They are for the most part clear, though some © contain rows of fluid-pores. The felspars are sometimes penetrated by apatite needles, and resemble those which occur porphyritically in many andesites. In sections perpendicular to the albite twin- plane the extinction-angles range up to 13° or 14°. The rock con- tains a little calcite, partly in pseudomorphs after felspar. Other authigenetic constituents are pyrites and a little quartz, the latter sometimes forming narrow veinlets. FROM THE CROSS FELL INLIER. 517 2. Bastc Lavas or Metuersy. The porphyritic lava of Melmerby presents in hand-specimens a striking appearance. Glassy-looking felspars, up to half an inch in length, showing to the eye both Carlsbad and albite twinning, are embedded in a dark compact ground, One is strongly reminded of the well-known porphyritic lavas of Eycott,Hill, and closer exami- nation leaves no doubt as to the identity of the rocks from the two localities. The type is a rather unusual one, the composition being decidedly more basic than that of normal andesites, while on the other hand olivine is wanting, being in a sense represented by a rhombic pyroxene with abundance of free iron-ore. Perhaps the best name would be hypersthene-basalt. In specimens from Eycott, Mr. J. Hughes * found the silica-per- centages 53°30, 52°60, and 51:10; Mr. T. Cooksey + found 53°40 and 52°73. The Melmerby rock is so similar that an analysis is scarcely necessary. A determination of its specific gravity gave 2-753, the figure for the Eycott rock being 2°754 (Cooksey, loc. cit.). The porphyritic felspars are often grouped in such a fashion as to interfere with one another’s growth, while always presenting crystal- faces to the surrounding matrix. They are on the border-land between labradorite and bytownite, being apparently a little more basic than the type hafnefjordite, Ab,An,. In sections perpendi- cular to the lamellee of albite-twinning the extinction-angles range up to 37°, reckoned from the twin-line. This is for the greater part of the crystal: the border gives a rather lower angle. As regards habit, extinction-angles, zonary structure, inclusions, &c., Mr. Teall’s + accurate description of the Eycott felspars may stand equally well for these. The chief feature of the groundmass is the plentiful occurrence of a rhombic pyroxene, almost often to the exclusion of augite. This is also the case in the Eycott rock, as was first pointed out by Prof. Bonney (loc. cit.). The mineral is, however, invariably replaced by bright-green fibrous bastite. ‘The strong colour and pleochroism of © this substance, with the occurrence in it of abundant secondary iron-ore, point to a ferriferous variety of pyroxene, hypersthene rather than enstatite. The groundmass of the rock contains plenty of magnetite. In tolerably fresh specimens [1250, 1251] this shows the outlines of octahedral crystals, and is clearly original ; but in more weathered examples the whole mass of the rock is rendered almost opaque by the separation of secondary magnetite with other decomposition-products. The felspars of the ground- mass are probably not very dissimilar in nature to the larger felspars; they occur in small slender prisms, but always show albite-lamellation, and do not sink to mere microliths. There appears to have been some augite, in small granules now repre- sented by calcite and a pale almost isotropic material; and there * J. Clifton Ward, Monthly Microscop. Journ. (1877) vol. xvii. P- 246. t+ T. G. Bonney, Geol. Mag. (1885) p. 80. t ‘ Brit. Petrogr.’ (1888) pp. 225-227. sag 02 518 MR. A. HARKER ON ROCKS must have been a considerable amount of unindividualized base. The general characters of the groundmass bring the rock nearer to the normal basalts than to the andesites. 3. Ruyoxrric Rocks. The normal rhyolitic rocks do not differ in any marked way from those underlying and intercalated in the Coniston Limestone of the Lake District. Both lavas and ashes are found. The lavas are not conspicuously porphyritic, though little felspars are often scattered through the rock: these, as usual, are plagioclase. The rocks show a generally compact appearance, light grey or cream- coloured when not stained by iron oxide. The groundmass has probably been to a great extent glassy, when the rocks were fresh, but this has not always been the case. A rock from the beck north of Keisley, for instance, has a microlithic character. The specimen just mentioned [919] shows a kind of flow-breccia- tion, which I believe is not an uncommon feature. Very similar appearances are seen in some of the Caernarvonshire rhyolites. In the Keisley rock the fragments of the original lava, probably the broken-up crust of a coulée, are divided by a matrix, or system of irregular branching veins, which makes up quite half of the whole mass. This matrix seems at first sight to consist of rather finely crystalline quartz; but closer scrutiny serves to detect in some of the clear grains the rectangular outline and the twinning of felspar- This matrix, therefore, must be regarded not as an infilling of vein- quartz entirely subsequent to the formation of the rock, but rather as an inflowing of the highly acid mother-liquor from which the earlier portion of the rock was formed, and so as representing only the latest phase in the consolidation of the lava. In one place the matrix contains an amygdaloidal cavity, some twentieth of an inch in length, on the border of which crystallization is rather better developed, and the felspar-twinning of some of the crystal-grains is well seen. The slide shows some genuine vein-quartz which occupies little cracks traversing the microlithic fragments and their matrix alike, and the contrast of these with the latter can be easily observed. Other rocks in the neighbourhood of Keisley are ashes, and one from Harthwaite Beck is a vesicular andesite with finely micro- lithic ground [1283]. It is noteworthy that here, as elsewhere among our Ordovician lavas, a vesicular structure is much rarer in the rhyolites than in the andesites. A well-marked type of acid lava is exemplified by a specimen from Gregory Hill near Dufton, ‘‘the second rhyolite below the Keisley Limestone.” A slide of this rock in Prof. Nicholson’s collection [L. D. 28] shows a groundmass enclosing a few scattered telspars, which, when not too much decomposed, give faint indica- tions of twin-striation. The ground has in natural light a mottled appearance owing to numerous clear spots, with a tendency to rounded outline, from which some dusty material seems to have FROM THE CROSS FELL INLIER. 519 been eliminated. ‘These spots consist of crowds of exceedingly fine microliths, doubtless of felspar, embedded in quartz, which in each spot behaves as a single crystal. ‘The darker portion of the ground has similar microliths, but in a matrix which remains dark between crossed nicols. The spots, about one-hundredth of an inch in dia- meter, make up most of the rock, which has a superficial resem- blance to certain “spotted slates.’ In describing a somewhat similar structure in the rhyolites of Penmaenbach *, I was inclined to regard the crystallization of the quartz in the spots as an original character, but the point is not quite clear. The rhyolitic rocks exposed in Swindale Beck show some remark- able features. The dominant type, as seen in the field and in hand- specimens, is a compact pale-salmon or cream-coloured rock, in which darker grey patches, with sharply defined outlines, indicate included fragments. Under the microscove [822] the greater part of the groundmass is obscure, owing to secondary quartz. There is a well-marked, rather wavy parallel structure which might be the lamination of a fine ash, though itis more like the flow-structure o/ a lava; and the scattered felspar-crystals, rarely broken, lie with their long axes in the same direction. There are numerous little included fragments of a microlithic andesite, and these would naturally cause the rock to be regarded as an ash, were it not for a special structure well shown in the slide. This is the occurrence of discontinuous bands or narrow streaks, following the flow-lines in the rock, in which a thoroughly crystalline texture is developed. These crystalline streaks consist of clear felspar, often in lath-shaped crystals showing twin-striation, some quartz, and, between the felspars, little areas of pale decomposition-products, such as usually indicate vanished augite. This Swindale Beck rock is, therefore, a “eutaxitic” lava, which has caught up fragments of the rocks, andesitic and others, through which it has broken out. A eutaxitic structure, though of rather different type, has been noticed in certain Caernarvonshire rhyolites (op. cit. pp. 21, 22). 4, Actp InrrustvE Rocks anp TransrtionaL Types. The acid intrusive rocks of the district are quartz-porphyries, which do not call for much remark. The best known is the so- called ** Dufton granite,” which forms a small boss to the west of Dufton Pike. Itis a rock of red colour, resembling some of the “‘oranite-porphyries ” in general appearance. Besides abundant red felspars, it shows quartz-grains about a quarter of an inch long. and small flakes of black mica. Scattered through the rock are hexagonal plates of white mica, wp to an inch in diameter; while occasionally is seen a colourless felspar-crystal, perhaps an inch and a half long, with the markedly tabular habit, the glassy lustre, and the longitudinal fissures (following an orthopinacoidal cleavage) of sanidine. ‘These crystals are twinned on the Carlsbad law, and recall similar ones in a dyke on Stakeley Moor, south of the Shap * + Bala Vole. Series of Caernarv.’ (1889) pp. 22, 23. 520 MR. A. HARKER ON ROCKS Fell granite*. The whiter and finer-grained rock which crops out just to the eastward of the foregoing is probably only a marginal modification of the main mass. Here the plates of white mica are rarer, and the most striking feature is the occurrence of long narrow blade-like crystals of dark mica, precisely like some found on the margin of the Shap Fell granite and in the dyke on Potter Fell, which seems to be connected with that mass‘. A slide [842] of this Dufton Pike rock shows plenty of porphyritic quartz, in clear idiomorphic crystals with only a few glass-cavities or small inclusions of groundmass. Among the porphyritic felspars, a plagioclase with Carlsbad-, albite-, and pericline-twinning pre- dominates. The light mica is perfectly clear and colourless, the dark decomposing with a green colour. Rarely the two are inter- grown. Both micas recur in small flakes with a rough parallel disposition, and these must be regarded as part of the groundmass. An occasional hexagonal prism of apatite is seen. The ground of felspar and quartz is of the microcrystalline or ‘“ microgranite ” type. : A specimen from a dyke north-west of Cuns Fell differs somewhat from the preceding, especially in the absence of white mica, and probably represents the usual type of the district [918]. The porphyritic crystals of quartz have their edges rather rounded, and are sometimes broken, but the fragments are not far separated. A flake of dark mica is sometimes enclosed in the quartz, as well as in the porphyritic felspars. On the hillside north-west of the “ Spring” in Ousby Dale the rock shows some remarkable modifications, which can be referred only to intense dynamic metamorphism. In the field it shows only a slightly different appearance from other examples of these quartz- felspar-porphyries, except that it has a general yellow iron-stained colour. A section, however, shows that a large part of the rock consists of colourless mica [1319]. The porphyritic felspars without losing their form are completely replaced by minute scales of this mineral, the scales in any one pseudomorph having a very general, though not uniform, orientation parallel to the length of the original felspar-crystal. Similar scales of mica occur in great quantity in the general mass of the rock, together with large flakes, which are rather ragged and wavy, and do not give very precise extinction between crossed nicols. ‘The rounded and corroded por- phyritic crystals of quartz are only occasionally cracked and broken. Dark mica is absent, but is perhaps represented by the larger flakes of colourless mica, each of which encloses a shapeless patch of limonite. Little flocculent patches of yellow ferruginous matter occur also in the groundmass. LHxcept for these and the minute scales of mica, the ground consists of a clear microcrystalline mass of quartz and felspar, without trace of crystal outlines, and highly suggestive of recrystallization im situ. It is not easy to * See p. 288 of this volume. T See p. 277 of this volume. FROM THE CROSS FELL INLIER. 521 distinguish the felspar from the quartz; the great majority of the grains are simple, but here and there one shows twin-striation. It will be noticed below that the diabase near this locality gives evi- dence of great crushing. The large igneous mass on Thack Moor and the intrusions in Ousby Dale, apparently offshoots of it, belong to the same general type as the Dufton intrusive rock, but lack white mica. In all these rocks the closeness of the porphyritic crystals, obscuring the groundmass, gives a very crystalline appearance in the field or in hand-specimens. Certain intrusive rocks, such as the large mass to the south of Cocklock Scar and some dykes, e.g. at Maiden Way in Ardale Beck, show a considerable departure from the foregoing and an approach to the characters of the lamprophyres described below. In hana- specimens they have a less crystalline appearance than the quartz- porphyries, and quartz is not recognized. The general aspect is that of some so-called porphyrites, the colour being brown with little stained felspar-crystals and facets of dark mica and augite. The slices [916, 922} bear out the idea of a transition to the lam- prophyre type. Besides the greater abundance of dark mica, mostly undergoing a greenish alteration, we notice the coming in of abun- dant apatite and magnetite, and especially of augite in perfectly formed crystals, now completely pseudomorphed. ‘The groundmass of these rocks is much decomposed, and it is difficult to decide whether any part of the free silica is original. 5. Tur LAMPROPHYRES. It would not be easy, and perhaps not very profitable, to attempt any such division of the “ mica-traps ” of the North of England as that between minettes and kersantites. A distinction founded on the crystallographic systems of the felspars is more futile than usual in this case, since those minerals are usually too far destroyed for recognition. Analysis might, of course, show the relative pro- portions of potash and soda present; but as this does not seem to hear any relation to the percentage of silica, &c.*, the character of the original felspar in any case would import little as regards the essential nature of the rock. The family-name “ lamprophyre” is therefore sufficiently precise. In some of the rocks, as in Rosen- busch’s “ vosgesite,” idiomorphic augite accompanies or to some extent takes the place of mica. — Like most lamprophyres, these rocks are very prone to decomposi- tion by weathering agents, and are often impregnated with secon- dary carbonates. The freshest examples have a dark grey ground plentifully spangled with flakes of dark brown or nearly black mica, sometimes as much as an inch in diameter, but usually much less. In more altered specimens the mica takes a deep brown colour, with * See Bonney and Houghton, Quart. Journ. Geol. Soc. vol. xxxv. (1879) p- 165. 5a . MR. A. HARKER ON ROCKS submetallic lustre, and the groundmass, also stained with brown, has a dull appearance. Visible porphyritic felspars are found, but only very sparingly. The boundaries are often rounded, as if by magmatic corrosion. Occasionally a crystal of orthoclase shows a quite glassy lustre, but the felspars are usually dull and semi- opaque. Certain rocks, such as the large dyke in Dry Sike and the boss-like mass in Swindale, enclose grains of quartz of consider- able size, the boundary rounded, or rarely with indications of the dihexahedral form. ‘The grains are commonly surrounded by a thin coating of a dark green substance. ‘Included fragments of partially vitrified grit, &c., are found in the large. Swindale boss. Vesicles, usually filled by calcite, also occur. The mica of these rocks is a brown biotite in tabular erystals more or less regularly bounded in the usual pseudo-hexagonal fashion. The plane of the optic axes is perpendicular to the basal cleavage and parallel to the clinopinacoid. ‘The bisectrix makes a very sensible angle (3° or 4°) with the normal to the basal cleavage, so that in cross-sections of the flakes the frequent twin-lamellation parallel to the base is easily detected. Indeed, owing to the pleo- chroism, this can be seen with a single nicol. J uxtaposition-twins parallel to the prism-plane * are also common. In almost every case the deep brown colour characteristic of biotite i in most rocks is here confined to the border of each flake, the interior being much paler, or, indeed, for vibrations parallel to the q@-axis, sensibly colourless. Whether this is an original zonary structure or a result of internal bleaching is not quite clear. Rosenbusch T apparently takes the former view for the mica of lamprophyres generally. It is noticeable that, although in cross-sections of flakes the border is usually very sharply defined, basal sections show a more gradual passage from dark to pale. Streaks of darker colour, following the basal cleavage-direction, are occasionally seen passing through the pale interior, and this seems rather to favour the idea of secondary bleaching. Much rarer than the dark border is a. dark nucleus, always sharply defined by crystallographic planes, with a paler margin [915]. The dark and pale mica-substances in these rocks possess very different powers of birefringence. ough measurements gave the figures 0:06 and 0-04, which, according to Lévy and Taeoix, correspond to brown biotite and splourleee meroxene respectively. Resorption-phenomena, with a separation of iron ore, are some- times seen on the edge of a flake [914]. Again, owing to mechani- cal forces, the flakes have in some cases yielded along “ gliding- planes,” as in the artificial twin-lamelle produced in calcite, Xe. The gliding-planes do not coincide with the basal cleavage [91+]. The mica occasionally encloses grains of magnetite, or is penetrated by slender hexagonal prisms of apatite. Again, the edge of a fluke * See Lévy &, Lacroix, ‘Minéraux des Roches ’ (1888), p. 289, fig. 137. t ‘ Mikr. Physiogr. d. mass. Gest.’ 2nd ed. (1887) p. 310. FROM THE CROSS FELL INLIBR, 523 sometimes shows numerous very minute needles (? rutile) arranged along the basal cleavage, parallel to the sides ofthe hexagon [915]. The most characteristic inclusions, however, are very minute erystals, probably of zircon, which are invariably surrounded by strongly pleochroic halos. ‘These halos are very conspicuous, re- producing in the pale interior of the flakes the dark colour of the brown border. A common decomposition-product of the mica is calcite, or possibly dolomite, often forming lenticles or plates along cleavage-planes [445 | *. In addition to mica, some at Jeast of the rocks, such as those of Dry Sike, have contained augite. ‘The mineral is completely destroyed, but the pseudomorphs show the characteristic octagonal cross-section [915]. Original magnetite is not plentiful in our specimens. Most of the slides contain little crystals of pyrites, which moulds the biotite, and is perhaps a secondary mineral. The porphyritic felspars, when recognizable, are seen to be orthoclase and a plagioclase with rather low extinction-angles ; but most of the felspars are deeply altered, and some are entirely replaced by calcite {914}. The quartz-grains, already alluded to, present in section an ovoid outline, or, more rarely, rounded crystal-forms [915]. The bordering ring is seen to consist of minute crystals of fibrous horn- blende, pale green in a slice. These are partly moulded by the quartz, while, on the other hand, they are not distinctly separated from the mass of the rock. The appearances suggest a reaction between the quartz-grains and a corroding magma, and recall the primary quartz-grains with a coating of augite-pranules described by Iddings f in certain basalts. The groundmass of these lamprophyric rocks is for the most part too much decomposed for precise study. It has probably been microcrystalline throughout. In some cases the mica seems to belong to two generations, of which the later, in smaller flakes, forms part of the ground [445,446 a]. The bulk of the ground has consisted of felspars, which sometimes sank almost to microlithie dimensions [915]. The most abundant decomposition-products are calcite and quartz. Where these occur collected in distinct patches, the quartz is often idiomorphic and moulded by the calcite. 6. Basro Iyrrusive Rocks. The chief intrusive rocks of basic character in the district are the diabases of Cuns Fell. These show considerable variations in texture. In the coarse-grained type it is easy to recognize the rectangular crystals of striated felspar and the dark green cleavage-faces of augite, or, again, black lustrous plates of hornblende. As will appear, this last mineral is not an original constituent. In the * This is the slide figured in Teall’s ‘ Brit. Petrogr.’ pl. xxxii. fig. 2 t Amer. Journ. Sci. (3) vol. xxxvi. (1888) p. 208; Bull. U.S. Geol. Surv. (1890) No. 66. 524 MR. A. HARKER ON ROCKS specimens of finer grain the individual minerals are scarcely to be detected by the unaided eye. In mineralogical constitution the Cuns Fell diabases present no special peculiarities. Apatite prisms occur plentifully, though only locally [924]. Crystals and rods of magnetite are always present, but ilmenite is not found. The bulk of the rocks, now considerably decomposed, has been built of felspar and augite, the former in idiomorphic crystals, the latter mostly in ophitic or semi-ophitic plates. The Cuns Fell rock has been termed a gabbro, but its structure is that of a typical diabase, with even an occasional approach to the doleritic type in the development of a few shapeless felspars of later consolidation. The ordinary felspars show albite- and pericline-twinning, and, so far as can be judged from their extinction-angles in rock-slices, may be referred to the border-land between andesine and labradorite. The augite, when unaltered, is colourless in sections, but this mineral is frequently quite destroyed, the common decomposition-products being pale-green delessite, clear quartz, calcite, and opaque dust (kaolin?). It is, however, fre- quently replaced by hornblende; sometimes pale greenish-yellow with a fibrous structure and inclusions of secondary magnetite, but more usually clear yellow-brown and pleochroic with a compact. structure and good prismatic cleavage. ‘There can be no doubt that this mineral is derived from the augite: the process of conversion, beginning at the margin and along cleavage-cracks, is seen in various stages [925]. The hornblende, as in some other rocks of this kind, gives rather high extinction-angles (¢ y=about 20°); and it is to be noticed that, in an augite-plate partly transformed into hornblende, the extinctions for the two minerals are, as usual, on the same side of the vertical axis. Another common feature of these diabases is the occurrence of a fringe of colourless hornblende growing in crystallographic relation with the crystal-plates, but outside them, and clearly formed at the expense of other minerals, to which it presents a very ragged edge. Such fringes surround not only the uralitic hornblende, but also patches of delessite which are evidently the relics of vanished augite. The order of the several changes indicated is therefore :— (i.) partial or total replacement of augite by hornblende, the uralitic hornblende being perhaps a stage in the conversion to the compact ; (ii.) growth of colourless hornblende-fringes about both hornblende and augite, this proceeding concurrently with alterations in the felspar, &c. ; (iii.) conversion of much of the remaining augite into delessite and other weathering-products. The hornblendic rocks examined are from the west and north- west sides of Cuns Fell. On the south-east side, near its junction with the Skiddaw Slates, the diabase takes on a finer grain, and, in particular, the felspars occur in long narrow prisms only once or twice twinned on the albite law, without, however, any marked parallelism of disposition [928]. At Dale Beck the diabase has a very crushed schistose appearance, and a slice shows that it consists mainly of calcite and delessite [926]. Another specimen a a FROM THE CROSS FELL INLIER. 5Sp from this neighbourhood (marked ‘Spring’ on the six-inch map) has in addition a quantity of finely crystalline to cryptocrystalline silica, sometimes with a spherulitic structure [1320]. In the field this rock presents a singular appearance owing to the numerous parallel narrow veins of calcite which traverse its dark mass. A specimen of diabase from a dyke in Rake Beck differs in some respects from the Cuns Fell rocks. The felspar seems to approach typical labradorite, and encloses occasional crystals of light-brown sphene [923]. At Deep Slack Wood, close to the exposure of the basic lavas of Melmerby, occurs a fresh-looking, finely crystalline dolerite, very different from the neighbouring rocks. It consists largely of little . lath-shaped striated felspars, besides a few larger individuals with broader lamellz and an occasional shapeless felspar with strong zonary structure, of the kind so characteristic of doleritic rocks [1323]. The augite forms ophitic plates each enclosing many of the small felspars ; magnetite occurs interstitially among the felspars in some abundance. The felspars are always quite clear; much of the augite is fresh, but part is replaced by brown and green decom- position-products, and a similar greeu substance fills the few scattered vesicles. The rock is unlike any known Ordovician lava or intrusion in the district, and suggests a dyke of much later age, post-Carboniferous or even perhaps Tertiary. 7. QUARTZITE oF Roman FELL. An interesting modification of the Carboniferous sandstone is represented by some of Mr. Marr’s specimens collected on Roman Fell. The sandstone here has been locally converted into a compact vitreous-looking rock by the deposition of secondary quartz. Under the microscope [911] the rock exhibits all the characters of a quartzite such as those of Hartshill and the Stiperstones *, ‘he original grains are almost all well rolled. Quartz largely pre- dominates, mostly with a very dusky appearance due to crowds of inclusions, often ranged in parallel lines. A few of the quartz- grains are composite. In much less quantity occur weathered telspar-fragments and little rounded pieces of a microlithic andesite. The interstitial quartz occurs entirely as a ‘secondary enlarge- iment,” easily distinguished from the original grains, with which it is in crystalline continuity, by its clear appearance. More rarely the felspar grains show a similar secondary growth, a phenomenon observed elsewhere by Van Hise and others. It occurs both on orthoclase and plagioclase-fragments. * Compare, ¢. g., the Lickey quartzite figured in Mr. Teall’s ‘ Brit. Petrogr.’ plates xly., xlvi. 526 MR. A. H. FOORD ON SOME APPENDIX II.—On some Cernatoropa from the Cross Fett Lyxigr. By A. H. Foorp, Esq., F.G.S. 1. Orrnoceras, cf. ELONGATOcINCTUM, Portlock. Portlock, ‘Geology of Londonderry,’ (1848) p. 372, pl. xxvii. figs. 2a, 2b. Blake, ‘ British Fossil Cephalopoda,’ pt. 1. (1882) p. 119, pl. xin. figs. 7, 8,8 a. Description. An elongated, cylindrical species with a very slow rate of increase ; the septa, as seen in fragments, distant from each other rather less than one-third the diameter of the shell. The siphuncle nearly central. The test ornamented with regular, transverse, thread-like strie, which do not appear to undulate ; * about nine of them occupy the space of a line. Abundant in the limestone of Keisley, near Dufton. Remarks. This is probably the species referred to by Profs. Harkness and Nicholson * under the name of Orthoceras vagans, Salter. The present species differs, however, from the latter (as interpreted by Prof. Blake) in its more approximate septa, and in the well-defined character of the ornaments of the test, whereas Salter’s species is described as smooth (Salter) or showing only lines of growth (Blake). All the specimens are more or less covered by the matrix, but the ornaments of the test are shown on several of them. The present species differs from Portlock’s in respect of the septa, which are wider apart than they are stated to be in his species. Orthoceras sodale, Barrande§, presents some resemblance to the present form in its slow rate of increase, the position of the siphuncle, and the surface-ornaments; but the septa are more remote in O. sodale, which is also a larger and more robust shell than the one here described. There appears to be another species associated with this one in the same matrix, but it is too fragmentary for description. 2. ORTHOCERAS, Sp. Description. An eroded fragment of a large Orthoceras from the Staurocephalus zone of Swindale is too imperfect for specific identi- fication. ‘The four chambers preceding the body-chamber are alone preserved. The shell may have been cylindrical or possibly elliptical in section; but the weathering it has been subjected to has de- stroyed its original form, causing one side to be flat while the other remains rounded. The septa are deeply concave; rather distant from each other, 7. ¢. about seven lines apart where the diameter of the cast is about 1 inch 10 lines, The last two septa are, however, * Quart. Journ. Geol. Soc. vol. xxxiii. (1877) p. 461, ‘On the Strata and their Fossil Contents between the Borrowdale Series of the North of England and the Coniston Flags.’ +t In Sharpe, ‘On the Geology of Oporto,’ Quart. Journ. Geol. Soe. vol. v. (1849) p. 153. + ‘ British Fossil Cephalopoda,’ pt. i. (1882) p. 141. § «Syst. Sil. de la Bohéme,’ vol. ii. pt. iii. (1874) p. 453, pl. eecexvil. (excl. figs. 12, 13). CEPHALOPODA FROM THE CROSS FELL INLIER, 527 searcely 6 lines apart, the diameter here having increased to about 2 inches. Two sections have been made, the one transverse, the other longitudinal, but without disclosing any trace of the siphuncle, which must have been contained in the part removed by weathering ; if so, it could not have been far from the margin. No trace of the test exists. Remarks. The general features of this fossil recall those of Orthoceras ludense, J. de C. Sow. (=O. columnare, Boll, O. tem- perans, Barr.), from the Upper Silurian rocks of Britain, Sweden, Bohemia, &c., and in its distant septa it is also comparable with Orthoceras omissum, Blake*. Both O. ludense and O. omissum, however, are found upon a much higher horizon than the present fossil. 3. ORTHOCERAS PUSGILLENSE, Sp. nov. Description. Several fragments of the septate part of a species with a bulbous siphuncle were collected in the Corona-beds at Orthoceras pusgillense. wel Abia Hn a, cast, showing septa; 0, base of septum with siphuncle ; ¢, polished section ; d, polished section, showing siphuncle pushed out of position. Pusgill. The shell is cylindrical when uncompressed, and tapers somewhat rapidly, that is, at the rate of about 1 in 74, in a fragment 2 inches in length. The septa are very numerous, being about 13 lines distant from each other where the diameter of the shell is 11 lines; they are shallow, and are pierced by a nearly central siphuncle, having beaded segments which have a width equalling nearly one-fourth the diameter of the shell. The test is quite smooth. Some of the specimens have been crushed laterally * ‘British Fossil Cephalopoda,’ pt. i. (1882) p. 169, pl. xv. figs. 9, 9a. 528 CEPHALOPODA FROM THE CROSS FELL INLIER. in such a manner as to displace the siphuncle, bending it and forcing it towards the opposite side. No estimate can be made of the size of this fossil, the body-chamber being absent in all the specimens collected. Three crushed and weathered fragments from the same locality and horizon, but somewhat larger than those just described, have similarly close-set septa and a beaded siphuncle, and they most probably belong to the same species. The present form agrees with Orthoceras (? Actinoceras) mendaa’, Salter *, in the close proximity of its septa, but differs therefrom in its more rapid rate of tapering. The test not being preserved in the specimens from the Durness Limestone, upon which Salter’s species was founded, no comparison can be made between the two species as regards this feature. It will be better therefore to regard it, provisionally, as new. 4, Cyrroceras (?). Fragmentary casts of the peripheral part of a small curved shell from Roman Fell exhibit septal characters similar to those of Cyrtoceras (Phragmoceras, Portl.) inequiseptum, Portlock t, sp., from the Bala Beds of Desertcreat, Co. Tyrone. These specimens are too imperfect for description. Discussion. Prof. Bory Dawxtns said that the case cited by Mr. Marr of a fault having been in course of development at different geological periods is by no means an isolated example. Many faults show signs of movement at different ages. The Thousand-yards Fault, for example, passing up the valley of the Irwell to the N.W. of Manchester, shows a throw of 1000 yards in the Coal-measures, but very much less in the Permian and Triassic strata thrown down to the north. With regard to the phyllites, the phyllites of the Isle of Man form a link between the clay-slates on the one hand and mica-schist on the other. Dr. Hicxs asked whether the term “ Bala,” as used by the Authors, included any typical Llandeilo rocks; or whether it was confined, as he thought it ought.to be, to such rocks as are classed under that name in North Wales. The section on the wall appeared to show a continuous succession from the Skiddaw Slates to Upper Silurian. He would be glad to know whether there was clear evidence of continuous deposition in the area generally ; and whether there was no evidence of a break between the beds which had been recognized by Mr. Goodchild as of Tremadoc age and the ovérlying Arenig beds. With regard to the faults shown on the map, he would like some explanation as to how the Authors accounted for the fact that they appeared repeatedly to cross the higher beds without affecting the * Quart. Journ. Geol. Soc. vol. xv. (1859) p. 374, pl. xiii. fig. 24, a, d. t ‘Geology of Londonderry’ (1843), p. 382, pl. xxviii. A. figs. 4a, 4 6. Pa Nee ear Quart. Journ.Geol. Soc. Vol XIVII.PLXVIT. i Od Meee 0U0UlUlllllt~C~C~CE RN ON xk 7 4 - RY ee AA ’ 4 ‘ : ie) ‘ c ~ P , fi . é : . a - ~~ ‘ 4 sy E===| Dusion Shales = S) Gronw Beds | oe bX Thyolitie Croup 3 ; 3 fe + cl: [EEE hycon Group “| Laas in ett ; Limestone +++] ftresive Rocks | ; Fig.3.SECTION IN SWINDALE BECK Z Six INCHES TO A Mice. 1 Shetddlaw Slates 7 Stockdale Shales 2 Khyclitic Croup 8 Coniston Flags 3 lorona Beds 9 New Red Sandstone 4 Dution Shales F Faults 7 5 Staurocphalus limestone P Pennine Fait 6 Ashgill Shales L Lomprophyre bana) REGORY ;, _ 3 i oJ 29 aN tye + __* e aA aliaK.ano 4 A ¥ Knock [pores] New Red Sandstone E==) Dution Shales ae eae 4 Car bonuterous (= S! rons Beds 21°] Gruston Grits IN) Phyolie Croup MAP of tHe CROSS FELL INUER WD toniston, Flags (EEE Aycow Group ONE INCH TO A Mite. 3 Snchadele Shales [ESS Stealalaw Slates Fe=9| Asher: Shales MV Zavas in alette fig. 2.08 a portion of the sume ore the Scale 07 two wiches to & rile. fee Staurocephabus Limestone L¥ 4] Jntresive Rocks Kewsley Limestone Milfs. on rn err ee : > ye, A, a, eo ay sc" ' pa ety nt tg a Wh Bi : “ue mitre Y % Whe aoe Se eine srt Pend , ade ‘ 7 ‘ . as : ary te ; bh ae J oat . j aie pone ’ ' : j i a ee a 2 ; : ; ns cad] ahh ig et, ee w tlie hasty Sila de File Sweat { ; ee 6 ta : 7 ay == - ~ ond \ \ F A » ; ae i re ; Weer as | bgt ian 2 wT ~Lay 4 . 4 a 4 ar) t end oe Oy rr a - ‘14 ita +, it | Ta) 0 el Caen : owe, wap Baca nein % zs ht ae rs) AA 4 iat 9: h; . ik ) aa) i ; ie ee ee ee ee ee i oe ee at : y : F . ‘ i i ce an Tike | gtd CEMA NN te Nv ey Sak ne a i} i ‘ rf i i" ihe. d 4 , vi ; BAN A Eat CORY NN anita, fv aa patho ren tea shan f Pan taper ie Leow a0 er eapetacn CA Ty emmy ae A pe ¥ » : * » a sen } r{ ; ¥ . ., oe eee lly phil a jt trl Mh wy a aliieh satel : ; . Cp aS as ~ » © = wok at a . ne a . . Sn ‘ ; oi . ‘a ‘ \ A iP A) ae 4 it WAC oper | ae ay ) ttt) ot el) ee i ii i el THE CROSS FELL INLIER, 529 underlying and adjoining older beds. He thought it would be well if the Authors could supply further evidence bearing on the direction of the earth-moyements which had produced such results, and on the physical changes generally which had affected the area under discussion. Mr. Rurrey asked what was the breadth of the dykes represented in the section; and whether, in any case, they were found to follow the directions of faults. He approved of the application of the term ‘* lamprophyre ” to these rocks. Mr. Huptesron said that the district was one of the most inter- esting from a geological point of view in England, and he expressed regret that so few persons who had been over the ground were at the meeting. It would have been an advantage to have heard Mr. Goodchild’s views. Nevertheless, they had seen that it was possible to criticize effectually without special local knowledge. An intimate acquaintance with this Inlier would help to explain some of the difficulties experienced in the geology of the Lake District ; the distribution of the Eycott-Hill type of rock was a case in point. There could be no better proof of the importance of the argument from palzontology than the correlation of the Dufton Shales with the Keisley Limestone, so unlike in lithclogical character. Perhaps the explanation of that difference as having been mainly due to sub- sequent deformation was open to some doubt. There could be no question as to the value and general interest of the paper. The Cuarrman congratulated the Authors on the reception of their paper by the meeting, and pointed out that contributions of this class, in which stratigraphical details were illustrated by careful palzontological research, had become comparatively less common in the Society’s Journal than they formerly were, although the value of such papers, as was well illustrated by the present example, had by no means diminished. Mr. Marr, in reply, explained that the term ‘‘ Bala” was used by them as synonymous with Caradoc, so that their Lower Bala of this district was not Llandeilo. He remarked that the apparent conformity between the Skiddaw Slates and the Rhyolitic Group was illusory, and that the newer beds abutted discordantly against the fault. He justified the absence of cross-faults in the Skiddaw Slates on the ground that the faults were only drawn where actually observed, but pointed out that Sedgwick had long ago shown how it was perfectly possible to get faults which did not affect older rocks in immediate vicinity to the newer rocks which were profoundly affected. He stated that Mr. Harker had in his appendix entered into some detail concerning the lamprophyres. He observed that the Authors had not given any definite expiana- tion of the difference between Keisley Limestone and Dufton Shales, as evidence for such was not to be gained from this isolated region. All that they maintained was that the Keisley Limestone and Dufton Shales were referable to the same subdivision, which was older than the Staurocephalus-Limestone and newer than the Corona-beds. 530 MR. F. RUTLEY ON A SPHERULITIC AND 28. On a SeueErvwiric aid Perurrte Ossrpran from Prias, JALisco, Mexico. By Frank Rurrey, Esq., F.G.8., Lecturer on Mine- ralogy in the Koyal College of Seience, London. (Read April 22, 1891.) [Puare XVIII.) Tue specimen upon which the followig observations were made is a greyish-green or leek-green obsidian with a waxy lustre, contain- ing numerous deep brownish-red spherulites, ranging from the size of peas to smaller dimensions. It was collected by Dr. A. E. Foote, to whose kindness [ am indebted for it. Under the microscope the thin section shows banded and perlitic structures, the former being delicate and even, while the latter is developed with a perfection which I have never seen equalled, Hach perlitic area, as a rule, includes several centres, sometimes five or six, or more, around which perlitic structure has been developed (Pl. XVIII. fig. 1). A similar complex perlitic structure has already been noted in certain dacites by Prof. Judd*. The glass of the obsi- dian contains numerous globulites and longulites, which are densely crowded together along the fluxion-banding. The perlitic fissures appear in all cases to have been filled up with siliceous matter, showing, in some instances, minute crystalline rods or fibres, passing from the walls towards the middle of the fissure, where a distinct line of arrest parallel to the walls may be seen. This secondary crystalline matter exhibits double refraction, so that the perlitic structure appears brightly illuminated and in strong relief against the glass of the obsidian, when viewed between crossed nicols (P]. XVIII. fig. 2). It is only here and there, however, that these somewhat regular growths from opposite walls of the cracks are to be seen. At times the minute crystalline rods or fibres occur in little divergent groups, and occasionally the growth has also passed from the walls of the crack into the glassy substance of the obsidian, thus giving rise to very minute pellets or spherulites of chalcedony which, when the section is viewed between crossed nicols, cause the perlitic fissures to appear broader than they do in ordinary light. These divergent groupings or spherulitic bodies are very minute as a rule, but in one or two instances, where they are sufficiently large to admit of the determination of their optical sign, they are found to be positive. The longulites which follow the direction of the fluxion-banding seem to pass through the material which has sealed up the perlitic fissures ; but this appearance is, of course, deceptive, and is due to the inclination of the cracks to the plane of section, a longulite when partially overlying or underlying an obliquely- inclined crack appearing to penetrate the transparent matter with which the latter is sealed. * See ‘The Volcanic Rocks of the North-east of Fife,’ by J. Durham, with an appendix by Prof. J. W. Judd, Quart. Journ. Geol. Soc. vol. xlii. (1886) p. 429. PERLITIC OBSIDIAN FROM PILAS, MEXICO. 531 . In addition to the ordinary circumferential perlitic fissures, radial fissures may sometimes be observed, but they do not pass uninter- ruptedly through a series of circumferential cracks, being merely continuous from one crack to the next. The dark brownish-red spherulites appear reddish or yellowish brown in thin section, when viewed in transmitted light. Their outlines are sharply defined, and the perlitic structure accommodates itself to their boundaries. From this, and from the fact that no trace of perlitic structure is to be discovered within them, it is evident that the spherulites were formed before the perlitic structure was developed. It might be urged that molecular rearrangement (induced during the formation of the spherulites) might have obliterated any perlitic structure, had such existed, in the areas now occupied by those bodies; but such an hypothesis cannot be enter- tained, since the delicate fluxion-banding of the obsidian is clearly visible—passing uninterruptedly through the spherulites ; and it is evident that, since this structure is preserved, so also would the perlitic structure have been, had it been developed prior to the for- mation of the spherulites. The sequence of the structures in this rock admits of no question. The fluxion-banding was developed first, the spherulites were sub- sequently formed, then perlitic structure was set up, and these fissures were finally sealed by the introduction of chalcedonic matter. At the point or points (for there are sometimes more than one) from which the spherulites originated, a confused microcrystalline structure may usually be seen, and from this point, or from these points, divergent bundles of delicate fibres or crystalline rods have been developed. Within these fasciculi, finely puckered or wavy transverse banding may be noticed, indicating slight pauses in the crystalline growth, and occasionally the fluxion-bands appear to have offered a temporary check to the development of crystalline bundles which were growing approximately at right angles to the direction of the fluxion-bands. Im such cases the band which caused the check has served as the base from which a fresh crop of crystalline ~ bundles has grown (as shown in Pl. XVIII. fig. 3), and in one instance the development of a spherulite has been completely arrested along a fluxion-band, as represented in Pl. XVIII. fig. 4. In another case a spherulite has been developed prior to the for- mation of a similar but larger spherulite which encloses it. Some of the spherulites envelop small crystals of triclinic felspar, as shown in the upper half of fig.4, Pl. XVIII. They present the appearance either of imperfectly-developed or of corroded crystals, probably the former, and delicate fringe-like processes from the enclosing sphe- rulite may be seen penetrating them for a short distance beyond their margins. Fig. 5, Pl. XVIII., represents globulites, longulites, and minute pellets of chalcedony occurring in the glassy portions of the rock. It seems very probable that this obsidian has been subjected to hydrothermal agency since its solidification and the development of its perlitic structure. The siliceous matter with which the perlitic Q.J.G.8. No. 188. 9p 532 MR. F. RUTLEY ON A SPHERULITIC AND fissures are filled, and which also occurs in the obsidian itself, either fringing the perlitic cracks or disseminated in small spherulitic bodies, indicates this, while the numerous globulites which are present are probably due to a like cause. Whether solfataric action gives rise to the formation of globulites in vitreous rocks is a question which has not, I think, as yet, been demonstrated ; but I may mention that a specimen of spherulitic obsidian, collected by Mr. G. F. Rodwell some years ago in the crater of Vulcano at a point where a powerful jet of steam issued and where sulphur was deposited, no longer pre- sents the bright glassy lustre so characteristic of fresh obsidians, but appears perfectly dull and has a stony instead of a glassy aspect. In thin section under the microscope the rock is seen to be tra- versed by a network of fine, irregular cracks, and its loss of lustre is found to be due to the development of innumerable globulites, as shown in fig.6, Pl. XVIII. The distinctness of the spherulitie structure has also suffered considerably, owing to the devitrification which, in this case, appears very probably to have been engendered by the action of steam. EXPLANATION OF PLATE XVIIL. . Fig. 1. Spherulitic and perlitic obsidian from Pilas, Jalisco, Mexico, showing complex perlitic structure and delicate fluxion-banding. % 18 linear. Ordinary transmitted light. 2. Ditto, showing perlitic fissures filled with doubly-refracting siliceous matter (chalcedony). x 140 linear. Crossed nicols. 3. Ditto, showing portion of a spherulite, in which the growth has been temporarily arrested along a fluxion-band, the latter having served as a basis for subsequent growth of the spherulite. x 18 linear. Crossed nicols. 4, Ditto, showing portion of a spherulite, the development of which has been permanently arrested along a fluxion-band. This figure also shows that the perlitic structure does not traverse the spherulite, but that it accommodates itself to the contours of the latter. The fluxion- banding passes through the spherulite. > 18 linear. Ordinary transmitted light. 5. Ditto, showing globulites, longulites, and pellets of chalcedony, occurring in the glass of the obsidian. The fluxion-banding passes diagonally upwards from left to right in this figure. x 250 linear. Ordinary transmitted light. 6. Devitrified spherulitic obsidian from the crater of Vulcano, Lipari Is. Collected at a point from which a powerful jet of steam was issuing. This obsidian is completely, or almost completely, devitrified by the development of globulites. In many parts of the section the globulites are much more densely crowded than in the portion here figured. x 250 linear. Ordinary transmitted light. Discussion. Prof. Jupp asked whether the Author had any information as to the locality of this interesting specimen, and especially as to its relation with any of the well-known varieties of Mexican obsidians. Mr. G. F. Kunz stated that the locality in Jalisco was about one hundred miles west of Mexico city, and one hundred and fifty miles north-west of Pachuca (‘* Navajas”’), the Hill of Knives. Se eae PERLITIC OBSIDIAN FROM PILAS, MEXICO, 533 Mr. Ruvey, in reply to Prof. Judd’s remarks, stated that, although the specimen resembled a pitchstone in lustre, he was inclined to believe this due to partial devitrification, as indicated in the paper. It was, however, difficult at times to say positively whether minute bodies, such as globulites, longulites, &c., present in a vitreous rock were developed at the time of solidification or at a subsequent period; but he was disposed to regard them in this case as secondary products. It was fortunate that Mr. Kunz was present and able to describe the locality from which the specimen was derived. bo ky lo 534 THE MELAPHYRES AND FELSITES OF CARADOC. : : 29. On some of the Mutaruyres of Caranoc, with Norms on the Assocratep Fetstres. By Frank Rvurrey, Esq., F.G.S., Lecturer on Mineralogy in the Royal College of Science, London. (Read June 24, 1891.) [Prats XIX.] CONTENTS. . Page T., Trio wetagit sv! j? ” 536 MR. F. RULTLEY ON SOME OF THE A dark bluish-grey, vesicular rock, the vesicles being mostly filled with pale greyish-white or dark green substances. The weathered surfaces are rusty-brown and the amygdules at the surface have usually decomposed, leaving empty pores which impart a somewhat scoriaceous aspect to the rock. Under the microscope, in thin section, it is seen to consist of once glassy matter, now more or less devitrified by the development of microliths and globulites, and crowded with small granules and crystals of magnetite, together with little needles and skeleton crystals, often twinned, and these from their extinction- angles appear in many instances to be labradorite. There are also many small lath-shaped crystals of triclinic felspar often considerably corroded (Pl. XIX. fig. 2). One or two of the skeleton crystals are shown in fig. 7, magnified 75 diameters. Fig. 6in Pl. XIX. re- presents what appears to be a diminutive pseudomorph of magnetite after olivine, magnified 140 diameters. Taking the boundaries of the crystal-section as the brachydome (021) and the brachypinacoid (010), the angle for the faces 021: 021 approximates to 80°. Minute pale greenish grains are also present which may consist of a dusty admixture of chlorite with some other mineral. The devitrified glass in the thinnest parts of the section appears nearly colourless ; where thicker it is of a pale brown tint. The vesicles in this rock are exceedingly numerous and very irregular in form (Pl. XIX. fig. 1). They are filled with quartz, a chlorite-like mineral (apparently delessite in part), calcite, and some- times chalcedony and a substance resembling felsitic matter, The larger grains in these microcrystalline aggregates usually show a positive uniaxial interference-figure, but occasionally the emergence of a negative bisectrix may be noted in convergent light, indicating that a biaxial mineral (probably felspar) is also present. Quartz, chalcedony, or felsitic matter frequently borders the vesicles, the interior being filled with calcite only, with quartz only or with chlorite (delessite?) only, but often with an admixture of chlorite or delessite and quartz. The skeleton crystals are mostly of the H-shaped or swallow-tailed types. They very commonly show straight extinction, but in many cases the direction of maxi- mum extinction makes angles with the axis of elongation varying from 12° to over 15°. The rock appears to be closely allied to certain Bohemian melaphyres described by Boiicky *. It doubtless represents the once vitreous, superficial portion of an old lava-flow of basalt or andesite, and is probably one of the most ancient examples of such a rock with which we are yet acquainted. No. XIV. S.W. side of Caer Caradoc, at the top of the hill. (Melaphyre Tuff.)—The fragments composing this tuff consist of a rock similar to that just described. They are mostly small, ranging from three or four millimetres in diameter to very minute dimen- sions. ‘The sections of the fragments frequently show concayities on * « Petrographische Studien an den Melaphyrgesteinen Bohmens,’ Archiv fir Naturw. Landesdurchforsch. vy. Bohmen, Bd. iii. Geol. Abth., Prag (1876). MELAPHYRES AND FELSITES OF CARADOC, 537 their boundaries, such as are invariably seen on the surfaces of the fragments of tuffs formed from vesicular or scoriaceous lavas. (Pl. XIX. fig. 4.’ The general aspect of the hand-specimen is that of a dark iron- grey or bluish-grey rock, compact in texture, but appearing slightly vesicular when examined under a pocket-lens. It also shows some irregular spots of yellowish-white to pinkish-white crystalline matter, which effervesces briskly when touched with a drop of acid. In thin section, under the microscope, the fragments composing the rock are seen to vary considerably in translucency, the ground- mass in some being a more or less completely devitrified pale-brown glass, containing opaque matter which appears merely as fine dust under a magnifying power of 250 linear. In other fragments the opaque particles in the groundmass are larger, and the rock of which these fragments consist seems to differ in no appreciable respect from the melaphyre previously described which occurs in the imme- diate vicinity of this tuff. In many of these fragments the magnetite grains are seen to be more closely massed along the margins of the fragments, giving rise to a narrow and perfectly-opaque black border, while in some instances the entire groundmass of the fragment has been rendered absolutely opaque through development of magnetite. It has already been shown that a basic, vitreous lava, such as the basalt-glass of Kilauea, when heated for 260 hours at a temperature ranging from 700° to 1200° C., becomes strongly magnetic and absolutely opaque through separation of magnetite *. Bearing this fact in mind it may, I think, be inferred that the fragments con- stituting the melaphyre tuff of Caer Caradoc have not resulted from the mere crushing of a lava, but that they were ejected from a crater as lapilli and volcanic sand; that their surfaces, in many instances, were sufficiently heated to give rise to the formation of an opaque superficial crust of magnetite, while, in other cases, a more protracted roasting carried this process to its extreme limit, so that all the iron present in the lava in the protoxide state became converted into the magnetic oxide. The action exerted upon a magnetic needle, when a specimen of the melaphyre lava (No. XII.) is brought near it, is exceedingly slight compared with that produced by a considerably smailer specimen of the melaphyre tuff (No. XIV.). Fig. 4 in Pl. XIX, represents part of a section of this tuff, magnified 18 diameters. The fragments and portions of fragments here shown exhibit a perfectly opaque groundmass, while the small felspar-crystals and skeletons lying in this groundmass appear unaltered and remain perfectly translucent. The broad light band passing diagonally across fig. 4 represents cementing material which, in this rock, consists of chalcedony. Irregularly shaped cavities occur in this cement, and these have been filled partly with pale green serpen- * ‘Notes on Alteration induced by Heat in certain Vitreous Rocks,’ Proc. Royal Soe. vol. xl. (1886) p. 437. 588 MR. F. RUTLEY ON SOME OF THE tinous matter, probably derived from the decomposition of the pyroxenic constituents of the melaphyre fragments, and partly with calcite. Judging from tbe texture of this tuff, it is probable that the vent from which the lapilli were ejected was not very far distant. No. VIII. $.W. side of Caradoc, about 80 feet below the Camp. (Amygdaloidal Melaphyre.)—A dark grey rock with small vesicles, some of which are filled with white and others with dark-green matter. Under the microscope it is seen to consist of small lath-shaped crystals of triclinic felspar, usually corroded and occasionally bent, together with magnetite (sometimes in octahedra, but mostly in irregular grains), and a considerable amount of green matter, which appears in some cases to be chlorite, but is not improbably a pala- gonitic substance resulting from the alteration of interstitial glass. The felspar-crystals lie irregularly in all directions. The vesicles are very irregular in form and some are filled with chlorite or delessite (the optical sign in the direction of the fibres is positive). Others are filled with calcite or quartz, while, at times, a little chalcedony is present. The section is stained in places by | ferric oxide. No. X. S.W. side of Caradoc, about 100 feet below the Camp. (Amygdaloidal Melaphyre.)—A compact dark brownish-grey rock with very small vesicles, some containing quartz, others.a dark green substance. Under the microscope it appears, when examined with alow power, to consist of a felted mass of very minute felspar crystals with opaque interstitial matter. The rock, in fact, shows the “ pilotaxitic ” structure of Rosenbusch, unless, indeed, the black interstitial matter represents a once glassy groundmass now ren- dered opaque by separation of magnetite, in which case the structure would once have been “ hyalopilitic.” The vesicles contain chlorite and quartz, and are very irregularin form. ‘The section is traversed. by some delicate fissures now filled with quartz. Fig. 3 in Pl. XIX. represents portion of a section of this rock. No. VII. 8.E. side of Caradoc, low down, perhaps 150 or 200 feet below the Camp. (Melaphyre.)—A rather pale greenish-grey to brownish-grey, finely-crystalline rock, presenting no striking pecu- liarities to the unassisted eye. Under the microscope it is seen to be more coarsely crystalline than any of the preceding. The felspars, which constitute a large proportion of the rock, he in all directions and, from their extinction-angles, appear to be labradorite, but in many cases they are partly converted into kaolin. Much brownish or greenish matter is present, often filling vesicles, in which case it usually forms divergent fibrous growths and minute spherulitic aggregates which have a positive optical sign. The section also shows rusty-brown patches of limonite, minute specks of pyrites and apparently a little unaltered magnetite. The felspars are often corroded and sometimes bent. No. XV. Little Caradoc, N.W. side, about 100 feet above the MELAPHYRES AND FELSITES OF CARADOC, 539 road toComley. (Dolerite.)—A greenish-grey holocrystalline rock of rather coarse texture, mainly consisting of dark-green augite and greyish felspar. Under the microscope the rock is seen to consist of crystals of perfectly unaltered augite, labradorite often partly converted into kaolin, magnetite, and interstitial patches of chlorite. The augite crystals, as a rule, appear to range from more than three millimetres to about one millimetre in length. The crystals of labradorite are frequently of somewhat larger dimensions. IIL. Concrusions WITH REGARD TO THE MELAPHYRES. The foregoing descriptions show that, within a very limited area, the melaphyres of Caradoc differ considerably in texture and in structure. Some have once been basalt-glass or andesite-glass, such being the superficial portions of a lava-stream ; others have pos- sessed a certain amount of interstitial glass which has subsequently been rendered more or less opaque by the development of magnetite, while, at times, it appears to have been converted into a substance possibly allied to palagonite. In some of these rocks the crystalline texture is very fine (pilo- taxitic), while in the case of the dolerite trom Little Caradoc it is comparatively coarse. Furthermore, near the summit of Caradoc we have a basalt-tuff or andesite-tuff. These rocks are spoken of as altered basalt or andesite, since any pyroxene or olivine which they may once have contained is in most cases so completely replaced by alteration-products that it is im- possible to define their original mineral constitution with precision, This is also partly due to the allotriomorphous character of those minerals which have undergone decomposition. ‘The felspars and magnetite are, as a rule, the only original constituents which remain unaltered, and these at times have suffered very considerable change, The term melaphyre, as indicating an altered basalt or andesite, seems perfectly applicable to these old lavas. The dolerite of Little Caradoc differs from these rocks, in that the augite remains perfectly fresh, or is only altered along minute fissures, and the felspars are more or less turbid and altered, while in the lavas of Caradoc proper it is the pyroxenic constituent which has undergone decomposition, but the felspars remain fresh and, as a rule, unchanged. Whether the dolerite of Little Caradoc may be regarded as a volcanic neck or plug, from which the basic lavas lying to the south- west of it emanated, is a point which field-work can alone demon- strate or disprove, but, taking into consideration the gradations of texture which these rocks present at different levels, such a sup- position comes within the range of possibility. 540 MR. F. RUTLEY ON SOME OF THE IV. Tue Fetsiric Serres. The Felsitic Series in the Caradoc area is a very important one, but extremely difficult to work out under the microscope, since the structures characteristic of rhyolites are, in most cases, obscure, so obscure, in fact, that, unless exceptionally thin sections are ex- amined, they may often completely baffle detection. After careful examination, however, it has been possible to recognize not only spherulitic structure and occasionally bands of spherulites, but also perlitic structure. ‘The latter is very obscure in the best examples, but is sufficiently marked to prove that the structure is present. It may be fairly well seen in a section made from a specimen collected a little above Caradoc Coppice, near the southern,.end of the hill and on its north-western flank. Very faint indications of the structure were first seen in this and in one or two sections from other spots in the neighbourhood. The sections were then reduced in thickness, and, in the thinnest portions of them, perlitic structure was found to be unquestionably present although still obscure. In ordinary transmitted light it is less easy to detect than between crossed nicols, and in the latter case it is rendered more apparent by a rapid rotation either of the section or, if a Dick microscope be used, of the nicols. The reason of this appears to be that the crystalline grains which lie in or along the perlitic fissures are, as a rule, slightly larger than those which constitute the main mass of the rock and that, although the optical orientation of the different grains along any one perlitic fissure is very diverse, yet on rapid rotation either of the section or of the crossed nicols the maximum illumination of one grain in the series is so quickly followed by the maximum illumination of each succeeding grain that the retina retains these impressions sufficiently long to receive the general impression of a narrow ring or of a number of narrow rings more brilliantly illuminated than the remainder of the section. Without having recourse to rotation these rings can, however, still be seen CEE XTX. fig. 5). Spherulites are somewhat plentiful in these rhyolitic rocks. They are usually small, but not difficult to detect even under low powers. For the most part they are irregularly distributed, but in some of the sections examined they are more closely massed and occasionally coalesce in irregular bands. Setting aside the devitrification which these rocks have experienced, we have their exact counterparts in many spherulitic obsidians of comparatively recent date, notably in those of the Yellowstone District, especially in some which occur near the Madison River. Ordinary fluxion-banding appears to be very poorly represented in the Caradoc rhyolites. | Evidence of such structure has been better seen on the ground than under the microscope. On the S.E. flank of Caradoc, a little to the south of the Camp and about half way up the hill, there is, for instance, an outcrop of felsite, on the weathered surface of which there is a well-marked banding visible. At this MELAPHYRES AND FELSITES OF CARADOC. 541 point the strike is 8. 80° W.and the dip about 55° towards the north. A microscopic section of this rock, however, shows scarcely any traces of fluxion-banding,. Irregularly-shaped vesicles, filled with quartz &c., are often present in these rhyolites, and they occasionally show a tendency towards elongation in a definite direction. This is well seen in some of the exposures of rhyolite occurring at the top of Caradoc close to and within the Camp ; but the forms of these vesicles, when viewed under the microscope, are often remarkably irregular, throwing out processes in all directions, the latter frequently con- stricted to mere threads near their points of origin from the main vesicle, then expanding and finally tapering to sharp points. In spite of this irregularity, however, they may be seen to have a rudely linear arrangement in the rock. Fluxion structure, as evidenced by streams of microliths, is not to be detected in rocks which have undergone such complete devitrification, save perhaps in hazy banding, preduced by slight differences in crystalline texture. For the purpose of ascertaining how far certain structures in vitre- ous rocks may be destroyed or rendered invisible through subsequent alteration, I have examined a number of sections of unaltered ob- - sidians and pitchstones in search of perlitic and fluxion structures, so delicate that devitrification would almost infaliibly obliterate them or render them so indistinct that they could no longer be recognized with any certainty. The following are a few notes on the subject, which may possibly be of some interest :— 1. In a perlitic obsidian (from Schemnitz, Hungary) the breadth of the perlitic fissures ranges from about 5255 to yyy of an inch. Such a continuous perlitic fissure, when seen in section, may often be observed to thin away from the larger dimension to nothing. One can easily imagine that the devitrification of such a rock would result in the total obliteration of the more delicate fissures. The fluxion-banding in this section consists of streams of micro- liths. A portion of one of the broadest lines in one of these streams measures +5) Of an inch. Devitrification might not obliterate such a band, but it might easily render the recognition of its component microliths impossible. 2. In a section of a Mexican obsidian no fluxion-banding what- ever is visible under the microscope, the only indication of flow consisting in the uniform direction of elongation of included gas- pores. The numerous microliths present in the section he with their longest axes in all directions. There are, therefore, in such a rock no structures which would bear testimony to its origin after devitrification. The result of such change would be simply a felsite without fluxion-banding and with- out perlitic structure. In cases such as these, mode of occurrence and associations in the field could alone give a clue to the original nature of the rock. The best example of fluxion structure which I have yet met with in the Caradoc district is in a rhyolite-tuff occurring at Bowdler’s Chair at the southern extremity of the Gaerstones ridge. The 542 MR. F. RUTLEY ON SOME OF THE specimen collected, which is merely a small surface-chip, is a brown to dark purplish-grey rock, showing a brecciated appearance when » examined with a pocket-lens. The section has been taken at a depth of about an inch from the weathered surface and shows, under the microscope, that the rock is in great part composed of fragments of rhyolite (devitrified obsidian) which exhibit a delicate, well-defined fluxion-banding. This banding is frequently sinuous, but some of the fragments show markings which approximate to damascene structure. The fragments are completely devitrified, displaying a micro- to erypto-crystalline structure when viewed between crossed nicols. There can, I think, be no doubt that they are fragments of deyi- trified obsidian. With regard to the material in which they are embedded a more guarded opinion should be given. It is somewhat darker in colour than the rhyolite fragments and contains numerous little crystals and fragments of crystals of felspar, some of which show the re- peated twin-lamellation of plagioclase and, in one good example, the extinction-angle clearly indicates labradorite. Occasionally small grains of quartz may also be detected on employing convergent light. eee cdwnceseraneneseeee 2 0 24. Yellow-brown: Arenacoous Bed.ss <,..<.0e<-s-sn as 4-nnaeeeys seek 2 3 oy ATOUMCHOUS .DGU. . ction: Socpercnastc. ence Gee cance clea eemeeee 3 0 22. Yellow Limestone. Small oolitic granules ............... 5 0 Ah Herd Oolitic Pamestome: i255 200s.) eked: th eee 6 20. White Limestone. Oolitic\granules ..../....<.-4-ceeese-aae 3 0 19. Dy pieal Pea Grits ie pacts baa ranks «yn eta deca ee 3 6 18. Yellow Argillaceous Partings. Pisolites .................. 10 17. White Limestone. The topmost8 inches contain pisolites 7 6 16: Wellow-arey Dimestone o2.) 0-0 on. 5.debeoutds s aanc Ae 6 15., Typical. Pea Gurite sisy. 418.0 plese «} ncn pe ah es + 6 14. Grey Limestone. Some pisolites...........0..-.cssssneeares 3 + 13. Grey Limestone with shell-fragments and a few pisolites. 1 6 12. Hard “Brown Limestone: 70: ctencures+-n teens aetna 10 1d} Coarse: Wimestone | 0.308: e1. is sob ee see 5 2k ae eae i 4 10. Light Brown Limestone. ennaeeeremereee 7 8. Brow Lamestome oie co ccenn cies na fone eea ee nnel eee ae 8 1 elt ‘Brown Enmestone 02) 2c. 9. ose geting eee 2 6G. Grey WiniestOne 2.5.0. c.rcses ven serce-2.~ceneieueeek eee ee 3 5. Light Limestone with yellow patches ...............s.ss00.+ 7 4. Brown Limestone. Small pisolites ............0.s-s5eenwsses 7 3s Coarse Limestone, - sha. 74 ' MR. R. LYDEKKER ON LOWER JAWS OF PROCOPTODON. the molars and the ascending ramus. Moreover, the surface of the jaw external to the last molar is separated from the surface above the ‘‘ angle” by a much less sharply defined ridge than that which occurs in P. Rapha. An imperfect but rather larger right mandibular ramus (B. M. No. M. 1897) is figured by Owen in the Phil. Trans. for 1874, pl. lxxx. figs. 1, 2, where it is referred to P. Goliah. The dimensions of these two specimens are as follows :— Bingera sp. No. M. 1897. inches, inches, Length of last 2 molars........ 1:8 1°83 SAW ioe ial ReMi Mi at | Mech ere a 345 9) on cay SOROK=8OTICS ew cle ek 3d Greatest depth at symphysis.... 1:9 1:85 WIGLMAOE Aa” oa au ire ape are 29 It may be added that the immature mandibular ramus figured in the Phil. Trans. for 1874, pl. lxxx. fig. 7, and referred to P. Goliah, differs from the specimens described as P. Rapha by the greater antero-posterior length of the fourth premolar, which is seen am alveolo. The foregoing description leaves therefore no doubt as to the exis- tence of two large-sized species of Procoptodon in the Australian Pleistocene, which are markedly distinguished from one another by the characters of their lower jaws. Both species were subject to considerable variations in size, which may have been either sexual or racial. Since the stouter and deeper type of jaw has been referred to the species described as P. Rapha, while one mandible of the more slender type has been described as P. Goliah, it seems advisable to allow these two types of mandible to be so named, until we obtain decisive evidence as to which really belongs to P. Goliah ; if, indeed, the type of that species affords any characters by which it can be distinguished from the second species. The possibility that the types of those two species are really specifically identical must also be borne in mind; andif this should prove tobe the case the name P. pusio might be the one which would have to be adopted for the second species. Here, however, the uncertainty again arises as to whether the type of that species indicates a small race of P. Goliah or of the second species, EXPLANATION OF PLATE XXI. Fig. 1. Inner view of the left ramus of the mandible of Procoptodon Rapha. 2. Inner view of the right ramus of the mandible of Procoptodon Goliah. 2a. Aperture of masseteric fossa of ditto. 26. Oral aspect of last two molars of ditto. an. angle of mandible; ca. aperture of dental canal ; sy. symphysis ; p. 4, premolar ; m. 1-4, molars. With the exception of fig. 20, which is }, the figures are } nat. size. R.Mintern del.et lith , Quart .Journ.Geol.Soc.Vol. XLVII. Pl XX]. Mintern Bros -imyp. JAWS OF PROCOPTODON. | RECENTLY-EXPOSED SECTIONS IN GLACIAL DEPOSITS AT HENDON. 575 33. On some Recentty-Exposep Secrions in the GuaciaL Deposits at Hrnpon. By Henry Hicks, M.D., F.R.S., Sec. Geol. Soc. (Read May 27, 1891.) [Puare XXII. | ConTENTS. Page is PP PRAROUT EG THON hc lcecabe cockcvgng ceute se eee eens tH CETL teehee sic gin’ TD II. Some Recently-exposed Sections SasiMeon vah bh eat CLA MaReRN a step blzenc.) EO IIT. Distribution of the Glacial Deposits................c:seeeeeeeeeeeeeeeeees 581 TISAI Src ee Oe is ARSENE of QO ye Se ie | ys 582 I. Inrropvucrion. For the past twenty years I have watched with interest the numerous exposures which have been made in the Glacial deposits at Hendon and in the adjoining areas, in making foundations for buildings, in digging for gravel, and in excavations in the course of Jaying down sewers. Some of the observations which I have made have been incorporated in the writings of authors who have referred to this area*, but hitherto I have avoided publishing the general results owing to the fact that new exposures which added fresh information were constantly being made. The recent com- pletion of the main sewering of most of the parish of Hendon, however, has furnished data for giving a fairly complete summary which I now place before the Society. Twenty years ago the nature, extent, and boundary of these deposits were very imper- fectly known, and doubt existed as to how they should be classified. When Mr. Henry Walker was making his researches in the Glacial deposits at Finchley, the results of which he communicated to the Geologists’ Association in 1871, I expressed to him the opinion, which he mentions in his paper, that the sands and gravels at and near Hendon ought to be classed with the so-called Middle Sands and Gravels of the Eastern Counties. This view, which I still hold, is vow generally adopted; but the clay-with-flints which overlies these beds on the Hendon Plateau has only recently been satisfac- torily demonstrated to be, in part, the equivalent of the Chalky Boulder-clay at Finchley, Whetstone, &c. IL. Some Recentty-Exposrep SxEcrions. When the sewer was being carried through Parson Street, directly opposite the Vicarage, 368 feet above Ordnance datum, a considerable thickness of Boulder-clay, resting on sand and gravel, was exposed in the excavation. It was of a bluish-grey colour, * H. Walker, Proc. Geologists’ Assoc. (1871) vol. ii. p. 288; Belt, Quarterly Journ. Sci. (1878) ; Whitaker, ‘Geology of London’ (1889). ce ee ee eS te eee 576 DR. H. HICKS ON SOME RECENTLY-EXPOSED SECTIONS and it contained, in addition to numerous unrolled white-coated flints, many distinct bits of chalk. Until recently this was the only perfectly satisfactory instance which had come under my observation in Hendon of a typical chalky Boulder-clay similar to that which is so well known at Finchley. That the Upper Boulder-clay must have been spread out very generally over the Hendon plateau, and by the same agency as that which deposited it on the areas farther north, is now certain. What is here mostly found, however, is a brown or yellowish- brown clay like that which at Whetstone and Finchley usually occurs below, but also frequently passes into and occupies the position of the more distinctly chalky clay. In some places it is very stony, but more often the fragments and pebbles are only scattered aboutin it. In parts also there is an almost entire absence of pebbles and rock-fragments. In a drain recently made in the Green Lane on the Hendon slope, looking towards the Brent Valley, and 175 feet only above Ordnance datum-line, a brown clay was exposed which yielded many fragments of a hard white chalk, the largest piece being about 2 inches long. The clay was also seamed in all directions with decomposing carbonate of lime (“race”). The occurrence of this Chalky Boulder-clay at so low an horizon is interesting also as proving conclusively that the Brent Valley had been in the main scooped out previous to the deposition in it of the newer Glacial deposits. Moreover, on the Finchley side of the valley there is equally good evidence, for the Boulder-clay is found there also at many points considerably below the 200-feet contour-line. In an extensive pit, opened in the year 1889, near the centre of the plateau upon which Hendon is situated, and about 4 mile 8.E. of the exposure referred to in Parson Street, an average thickness of 7 feet of a brown clay with bluish streaks was exposed, resting upon about 8 feet of sand and gravel. This pit is about 245 feet above Ordnance-datum, and has been dug in a field opposite West View, between New Brent Street and Heriot Road. When the underlying floor of London Clay was reached it was found to ‘be very irregular in character, and resting on this floor were large masses of angular sarsen-stones, in some cases showing what appeared to be distinct traces of ice-markings. As the pit was enlarged, it was seen that the Upper Boulder-clay sometimes reached downwards through the Sands and Gravels until it had completely penetrated them and touched the underlying floor of London Clay. Ina deep drain which was carried from this pit in a S.W. direction for about 130 yards for the purpose of drawing off the water, several of these channels filled with Boulder-clay were met with, the Sands and Gravels lying between in a com- paratively undisturbed condition. These channels varied from a few feet to several yards in width. In the pit and drain the floor of London Clay was found to be very uneven, rising up here and there in hillocks, and as the Gravels were seen to pass horizontally across these banks, it became evident that such irregularities must IN THE GLACIAL DEPOSITS AT HENDON, 577 have been produced by ice-action prior to the deposition of the Gravels in them, At the eastern end of the pit the Upper Boulder- clay, like that mentioned as occurring in the Green Lane and 70 feet lower, contains many decomposed patches and concretions of carbonate of lime resembling the ‘‘ race ” found in some brick-earths. Here the “race” occurs most abundantly near the base of the clay. Mr. Whitaker, in describing,the Glacial deposits at Church End, Finchley*, refers also to the fact that the brown clay there found underlying the Chalky Boulder-clay contains much “ race.” The resemblance between this brown clay, when, as is often the case, it contains scarcely any pebbles, and the London Clay is particularly striking, and leads one to the conclusion that much of the brown clay must have been derived by denudation from exposed surfaces of London Clay during the Glacial period 7. There are several patches of sandy gravel enclosed in the Upper Clay in this pit which, it is clear, must have been torn off as frozen masses from underlying beds and re-deposited as boulders in the clay. The sandy gravel in these patches oftens exhibits a rough oblique stratification, and the materials seem to be identical with those in the underlying gravels here and elsewhere in the neigh- bourhood. Over the surface of the clay, and filling some depres- sions in it, a rough gravel is frequently found, the result probably of the subsequent denudation of the Boulder-clay by sub-aerial action and flood-waters at the close of the Glacial period. This gravel has been spread out very generally as a thin coating over the neighbourhood, and extends beyond the line on the Map, which is intended to indicate the boundary of the more typical Glacial deposits. As this gravel contains many white-coated fresh-looking flints and subangular fragments and some northern erratics in association with the well-rolled pebbles from the Tertiary beds, it is evident that it must have been derived immediately from a Boulder-clay which had been spread out very universally over the area, especially as it is now found not only on most of the hills and slopes east and west of the Hendon plateau, but also between it and the Thames Valley. In the section (fig. 1, see next page) which was recently exposed at the 8.E. corner of the pit A, the rough gravel at the base contained many angular masses of sarsen-stones and large flint blocks, some of the latter being but little worn and haying white-coated surfaces. The majority of the flints, however, here as elsewhere, are well-rolled pebbles, and must have been derived by denudation from Eocene beds in the neighbourhood. There were also numerous fragments of chert and ferruginous sandstone, evidently derived from the Lower Green- sand, and also some pebbles of quartz, quartzite, ironstone, &c. In this pit the lower rough gravels vary in thickness from 2 to 4 feet, according to the irregularity of the floor. Between the sandy gravels * ‘Geology of London’ (1889), vol. i. p. 311. + Mr. H. B. Woodward refers also to this resemblance, and says = this clay is “ often very like London Clay,” op. cit. vol. i. p. 509. Fig. 1.—Section in West View Gravel-pit. [About 20 feet in length. ] MNLATAFUS UMMA aca TE K) C28 © Ge & asker 5 a. Surface soil. a', Gravelly soil. : 6. Brown clay, with a few flints, patches of gravel, and in places much ‘raée.’ ¢. Laminated sandy clay. d. Grey sandy gravel. J. Yellowish sand with seams of clay. y. Gravel with angular masses of sarsen-stone, subangular flints | and flint-pebbles, chert, quartzite, &e. h. London Clay. Fig. 2.—Section on the W. Side of West View Gravel-pit. sow ew =. we? » ere a. Surface soil. a'. Gravelly soil with white-coated flints, &c. J 6. Brown clay, with blue streaks and witha few flint-pebbles (6-15 feet). | ¢. Laminated sandy clay (6 inches). | d. Ochreous sandy gravel. | | \ Irregular. e. Bands of grey subangular gravel. J. Ochreous sand with seams of clay. g. Coarse gravel (2-4 feet). h. London Clay. 5 feet. RECENTLY-EXPOSED SECTIONS IN GLACIAL DEPOSITS AT HENDON. 579 and the Boulder-clay there is usually from 6 to 12 inches of a sandy laminated clay, almost entirely devoid of stones*. In one part of the pit the laminated clay and an underlying band of gravel were seen to have been bent down into one of the channels in a rather remarkable manner (fig. 2); and as there is no indication of a sinking of the surface, or of any increased thickness of soil over the Boulder-clay, it seems evident that this effect must have been produced contemporaneously and mainly by pressure, when the overlying Boulder-clay was deposited, especially as the other beds which underlie the laminated clay are also in a disturbed and contorted condition. Section 3, which has been exposed during the last two years in a field at Hendon Grove near the western edge of the plateau, Fig. 3.—Section in Upper Gravel-pit, at Hendon Grove. [About 35 feet in length. ] a. Surface soil. a'. Gravelly soil (with flints). Irregular. b. Yellowish-brown clay, in places passing downwards through the underlying sands and gravels (2-7 feet). d. Loamy sand with ochreous bands (1 foot). e. Gravel (1 foot). J. Yellowish sand, sometimes showing current-bedding (2 feet). g. Ochreous gravel, with large pieces of sarsen-stone, rounded and subangular flints, chert, quartzite, quartz, ironstone, &c. (2-4 feet). h. London Clay, much disturbed, of a brown colour, and forming a very uneven floor under the Glacial deposits. offers evidence very similar to that referred to as occurring in the West View pit, the chief difference being a diminished thickness of Upper Boulder-clay. Within a very short distance, however, from this point, in the excavation made in laying down the sewer in the road leading to St. Mary’s Church, about 285 feet above O. D., from six to eight feet of clay were passed through before the sand was reached. The London Clay was touched at from 14 to 16 feet from the surface. It must be understood that, in the scores of sections exposed in pits and deep drains at various points, consider- able variability has been frequently observed in all the deposits. * A similar laminated clay has been mentioned by Mr. H. B. Woodward as oceurring frequently under the Chalky Boulder-clay, and I have also recently seen it occupying the same position in a large gravel-pit at Whetstone and in pits at Finchley, &e. Q.J.G.8. No. 188. re 580 DR. H. HICKS ON SOME RECENTLY-EXPOSED SECTIONS In some cases there was an almost complete absence of gravel and a preponderance of fine sand, whilst in others the whole thickness was found to consist of a sandy gravel. Everywhere, however, except where the furrows containing the overlying Boulder-clay reach downwards to the London Clay, some amount of sand or gravel is present under the clay, usually in a stratified condition, and showing well-marked lines of current-bedding. When there were but few pits opened, and no continuous sections had been seen, I was led, on one or two occasions, to form a mistaken idea as to the thickness of the Glacial deposits in some parts of the district, by calculating, from some excavations which happened to have been made in one of the channels containing Boulder-clay (without reaching gravel at the base), that an average thickness of the sands probably occurred below. From the excavations which have been made since then for the sewers, and other more continuous sections, I have obtained sufficient evidence to show that, where well- sinkings had exposed an unusual thickness (2. ¢. 15 or more feet) of the Boulder-clay, they had been sunk in the depressed masses, and there was but little, if any, gravel separating it there from the London Clay. As demonstrating the practical utility of some acquaintance with the physical conditions under which these deposits were accumulated, I may mention that in this district, where sand and gravel command a very high price for building purposes and laying out of paths, the search for gravel has often been discarded, because it was not known that over most of the plateau there was a considerable thickness of clay overlying the sand, and where test-holes had been made they had frequently either not been carried deep enough, or had pierced one of these channels filled up by the Upper Boulder-clay. From finding a certain amount of washed gravel on the surface of the clay, it had also frequently been supposed that it occurred only in pockets in the clay, whilst a further boring of two or three feet would often have revealed the presence below of 7 or 8 feet of good sand and gravel, equal, if not superior, to that which for building purposes would have to be carted from the Thames Valley or brought from Hertfordshire or Bedfordshire. On several occa- sions also I noticed that where the lower gravels had been re- moved all further search had been suddenly abandoned, in con- sequence of meeting with a depressed mass of Upper Clay, whilst had the mass been crossed the gravels would have been again reached within a few yards. The largest gravel-pit at present in the neighbourhood, the one from which fig. 1 is taken, was opened at my suggestion by the men who had been digging gravel for my own use in my grounds, and it was with difficulty that I could persuade them that the gravel would be reached below the 7 or 8 feet of the stiff brown clay which is found in that pit. As showing how the previously-formed hills of London Clay have been, as it were, mantled by the Glacial deposits, the sections at Hendon Grove are particularly instructive. At the highest point on the edge of the plateau, about 280 feet above Ordnance datum, : IN THE GLACIAL DEPOSITS AT HENDON. 581 an excavation extending to over 60 feet has been made, exhibiting the conditions witnessed in fig. 3. Towards the north end of the pit B, the floor was seen to rise up rather suddenly, and the sand and gravels diminished toa few feet. ‘Cowards the south end a mass of Boulder-clay, about 12 feet across, dipped down suddenly till it rested on the underlying London Clay without the intervention of any sand or gravel. Between these points there was an average thickness of 6 to 7 feet of more or less well-stratified sand and gravel, and at the base, resting on the London Clay, were several masses of sarsen-stone, 2 to 24 feet in length. The lower gravel also contained, scattered about in it, several large angular masses of sarsen-stone, large but slightly-worn flints, and sometimes masses of a brown-clay much resembling the underlying London Clay. On the westward slope, at a distance of about 150 feet from this pit, and some 25 feet lower in horizon, another pit (C) was opened. Here it was found that the lower rough gravels, which averaged about 4 feet in the upper pit, were greatly diminished in thickness, and were replaced by clean white sand, the beds above being on the whole almost identical with those in the upper pit. A wide depressed mass of the Upper Boulder-clay was also found to come in at the south side of this pit. In some deep drains, which were made for the purpose of carrying away the water from these pits, similar varying conditions were witnessed *. It will be seen, there- fore, that these Glacial deposits do not, as has been usually supposed, merely rest on a plateau of the London Clay, but they lie on a very irregular surface and descend everywhere along slopes of previously formed depressions and valleys in the London Clay. I have care- fully traced the sections along the slopes, and I find that there is sometimes a difference of at least a hundred feet from the highest point at which London Clay has been touched on the ridge, to that in which it is found underlying almost identical sections on the slopes. IIL. Disrrrpution oF THE GLACIAL Deposits. On the accompanying map (Pl. XXII.) I have outlined the margin of the deposits, so far as they have been made clear through pits and in sewering the district, and I have indicated some of the spots where the deposits have been weil exposed. It will be observed that the patch occupies an area more than three times the size of that shown in the Geological-Survey map of surface deposits. Many years ago I mentioned to Mr. Whitaker and to Mr. Horace B. Woodward that the deposits extended beyond the limit shown, and in the recent excellent Memoir of the Geological Survey they have fully referred to the facts I then communicated to them. Much additional evidence has, however, been since obtained, and the boundary has been further extended. The nature and contents of * A test-hole was recently dug at the bottom of this field, about 65 feet below the horizon of the upper pit, and similar sand and gravel was inet with, covered by about 3 feet of Boulder-clay. Here more ‘ race’ was found in the clay than im the excavations higher up in the field. 28 2 582 DR. H. HICKS ON SOME RECENTLY-EXPOSED SECTIONS the deposits also have been far more completely made out in the more recent exposures referred to in this paper. In the patch directly north of the Hendon plateau, upon which no pits are indicated, so few exposures have been made that the boundaries of the Glacial Drift can only be approximately given, and the evidence of its character has been derived mainly irom shallow drains at various points, and from the railway-cutting and some deep wells in Page Street, just beyond the limit of the Map accompanying the. present paper (Pl. XXII.). What has been exposed in the shallow drains was mainly a yellowish-brown clay containing many flint-pebbles, but some gravel was met with in the railway-cutting, and it must also be present where the wells have been sunk. That at the Tithe Farm is stated to be about 20 feet deep. In the two other patches along the W. side of the Brent Valley the deposits have been exposed in gravel-pits and deep wells and in excavations for the main sewer. At Holder’s Hill the Boulder- clay was found to attain to a considerable thickness, and it resembled that found on the Hendon plateau. The area on the Finchley side of the Brent (indicated as covered with Glacial deposits) contains in places the typical Chalky Boulder-clay, but at most points along the sides of the Brent Valley a thick coating of brown clay overlies the sands and gravels. In the patches at Temple Fortune and Golder’s Green the clay is also mainly of a brown or yellowish-brown colour, and no typical Chalky Clay has hitherto been discovered there. Sands and gravels were exposed at several points in the sewer excavations in these areas, and much sand and gravel was obtained some years ago in a field on the south side of Bridge Lane, between the Decoy Farm and Temple Fortune. The Golder’s Green patch might well be extended to the west side of Hamp- stead Hill, for Glacial deposits have been exposed at several points between the Brent Valley and the slope of that hill. IV. ConcLusiIons. There can now be no doubt that Glacial deposits similar to those found at Finchley and Whetstone on the N.E. were spread out in a S.W. direction across the Brent Valley and over the Hendon plateau, reaching downwards on the slopes to below the Ordnance datum-line of 200 feet. There is good evidence also to show that they passed across the valley separating Hendon from Kingsbury, and that they now occur on most of the heights in the latter parish. They are also found at Dollis Hill, and at some other points in the parish of Willesden. It is certain, therefore, that the physical features of this part of N.W. Middlesex were moulded at a very early stage in the Glacial period, or clearly previous to the deposition of the so-called Middle Sands and Gravels, and of the Upper or Chalky Boulder-clay. At this time there could have been no barrier of any importance to prevent these deposits from extend- ing into the Thames Valley, and the evidence clearly points to the conclusion that the implement-bearing deposits on the higher horizons in the Thames Valley should be classed as of contempo- IN THB GLACIAL DEPOSITS AT HENDON. 583 raneous age with these undoubted Glacial deposits at Hendon and Finchley, which they so closely resemble. This necessarily indicates that man lived in the neighbourhood of the Thames Valley at least during a portion of the Glacial period, if not, as is highly probable, in pre-Glacial times. PLATE XXII. Map showing the Distribution of the Glacial Deposits in and aroun 1 Hendon. Discussion. Mr. H. B. Woopwarp said the chief interest of the sections related to the brown clay that occurred between the two layers of gravel. When he surveyed the Hendon outlier (in 1869) the lower gravel was not exposed, and he took the brown clay to be London Clay. Dr. Hicks had clearly proved that this brown clay belonged to the Glacial Drift. It could hardly be regarded as true Boulder- clay, for although patches of gravel had been caught up in it, there were no true erratic boulders, so far as he was aware, and the “ scratched stones ” mentioned by Dr. Hicks afforded no convincing evidence of glacial striae. The brown clay, however, behaved much like Boulder-clay in the way in which it had here and there cut abruptly into the beds below, and it might be regarded as a recon- structed mass, a kind of boulder, in fact, of London Clay. The “race”? was a secondary product, due perhaps to the decomposition of fragments of septaria derived from the London Clay. The features of the district, as Dr. Hicks maintained, were to a large extent of pre-Glacial origin. He had come to this conclusion while at work in Essex, where, as near Brentwood, the Glacial Drifts smothered up some of the old features and abutted against pre-existing outliers of Bagshot Beds. Mr. J. Atten Brown regarded the paper as of much importance to those who, like himself, had been engaged in the investigation of the Quaternary deposits. in Middlesex. He pointed out the simi- larity in many respects between the sections now shown and those at ‘The Mount,” Ealing, described and figured by him in the Pro- ceedings of the Geologists’ Association. In an excavation made lately on Castlebar Hill, Ealing, of which he had made sections, as yet unpublished, there was the same abrupt cutting off of the gravel beds as shown in one of the present sections and the lke infilling of brown (probably) Boulder-clay; he considered that such brown re- deposited London Clay was the equivalent of Boulder-clay, although the boulders and chalky material were often absent. ‘The matter left by ice traversing a clayey country would necessarily be chiefly composed of clay, and what chalk might have been originally in it would probably be dissolved out or redeposited as “‘ race.” There was more chalky matter in such deposits the nearer the Boulder-clay was found to the outcrop of the Chalk. He had seen chalk, sometimes in large masses, with the stratified implementiferous gravels bent and contorted between them, in the high-level drift-gravel near Langley (Bucks), showing the passage of very large masses of ice during the period when man was existing there. 584 RECENTLY-EXPOSED SECTIONS IN GLACIAL DEPOSITS AT HENDON. He desired to know whether the Author had analysed the gravel seen in these sections and whether he could distinguish any differ- ence between the constituents of the higher bed and that at the base, - and what proportion of Southern Drift he found there, this being a matter of considerable importance, now that Professor Prestwich’s valuable series of papers had appeared. With regard to the ana- logous deposit on Castlebar Hill, Ealing (about 160 O.D.), he had found in it about 8 per cent. of ragstone and chert pebbles ; 7 per cent. of white quartz, of which the largest was but little bigger than a pea; no northern rock but a small pebble of light quartzite ; about 12 per cent. of black Tertiary pebbles, and the rest flint- pebbles and subangular pieces of flint more or less stained, with some blocks of sarsen-stone much eroded. Dr. G. J. Hine called attention to the fact that a notable pro- portion of the fragments from the Hendon Beds, exhibited by the Author, consists of cherty rock, similar to that forming the Sponge- beds of the Lower Greensand in Kent and Surrey, from 20 to 30 miles to the south of London; and the mode of their occurrence in their present position had not yet been explained. Mr. Moncxton remarked that the Glacial Gravels were very largely composed of material derived from the near neighbourhood, and this favoured the view that the valleys in which the gravels are found were not previously formed, but were contemporaneous with the formation of the gravels: that is, the portion of the valley in which the eravels occur. In reference to Prof. Prestwich’s suggestion (Quart. Journ. Geol. Soc. vol. xlvi. (1890) p. 136) that the Hendon Gravels might be Westleton, he considered that the presence of a large proportion of subangular flints and the absence of any great quantity of quartz-pebbles showed they were not Westleton but Glacial gravels. He did not think that they were Southern Drift. The Avrnor, in reply to Mr. Woodward, said he did not refer to the seams of clay as having been torn off from the floor by ice, but to some distinct masses which were found enclosed in the Lower Gravels. Replying to Mr. Monckton, he stated that when he said that there was no barrier of any importance between Hendon and the Thames Valley, he meant no continuous hills like those of Hampstead and Highgate. The Brent Valley, which undoubtedly had been scooped out before the Upper Boulder-clay was deposited over the district, especially offered every facility for the extension of the material in that direction. It was interesting to know that Mr. Monckton and Dr. Hinde recognized a similarity between the chert fragments and the chert of the Lower Greensand in the south ; but he thought it probable that similar beds must have been exposed somewhere to the north, as about an equal proportion occurred in the gravels underlying the Chalky Boulder-clay as far north as Whet- stone. The Author was very glad to find that Mr. Allen Brown had obtained further evidence to show that Glacial deposits occurred near Ealing, and he had no doubt that ere long it would be possible to show that they extended almost continuously from Hendon to that area. t. Journ. me zee iE XIV. ae xan : | // Sr ‘ CA $3 Shieaphouse Farm :|'15° ot i 4% : a ee ij! - «i P fa re "ME yY Noy fist ; ile _ r “ A Uede. a ee b \ _ MAP SHOWING THE DISTRIBUTION OF THE GLACIAL DEPOSITS IN HENDON AND THE ADJOINING AREAS Scale six ches to one Mile INDEX. Glacial Deposits consislirg of Boulder -llay, sand and yravel, London lay covered wn places | 4y gravelly sod, rain and river Waste. Positions of Sewers and waits, A where the Glacial Deposits have beer, eazp0sed. Positions of \Wells &pils where the Seas statin lucia Dewosils hae dosed Sections becty cappostil, A West Vew Pit,(Sections 7&2.) B Upper Pit Hendon brove (Section 3) C Lower Pit, Herndon rove. __-.. LOO lorloiy-lene ——_ beolagical boundaries. O Ae a = Guar: Hate Faris S Or A | REV. E. HILL ON WELLS IN WEST SUFFOLK BOULDER-CLAY. 34. On Wetts in West Surrotk Bounper-cray. By the Rey. Epwin Hitt, M.A., F.G.8. (Read June 24, 1891.) On taking up my residence in the heart of the Suffolk Boulder-clay, I at once began to enquire the depths of wells, in order to learn the thickness of its mass. The nature of the answers was unexpected. The depths at which water is obtained appeared altogether irregular and capricious; ultimately it seemed necessary to conclude that in some cases, possibly in many, the water is met with not below but in the interior of the Clay itself. Here is an instance. A well in an outhouse of this Rectory (Cockfield, seven miles 8.8.E. of Bury St. Edmund’s) gives water at a depth of 35 feet. Another, outside its grounds, 120 yards W., is 7+ feet deep, and'I am told that in sinking it no water was obtained till this depth was reached. But another, the same distance N.N.W. of the first, is only some 8 or 10 feet in depth. Boulder-clay forms the subsoil to within a few inches of the surface, which in the case of each of these wells is about 300 feet above Ordnance datum. Other wells have the following depths, the distances and directions being measured from the Rectory :—One half a mile N.N.K. (surface about 290 feet) * 50 feet ; another a quarter of a mile N. (300 feet) 70 feet. In the opposite direction, at Cockfield New Hall, three quarters of a mile 8S. W. (255 feet), water was found at various depths down to 35 feet, but at the New Barn, a quarter of a mile farther S.W. (about 280 feet), the depth was 126 feet. At the Post Office, about a mile W.N.W. of the Rectory (about 270 feet), a shaft was sunk 83 feet and abandoned, nothing beyond surface-water having been obtained, while a few hundred yards off in several directions supplies are furnished by wells of varying but moderate depths. It may be asked whether these irregular depths may not be due to the irregular surface upon which the Clay is lying. But the water-levels in the three wells first named, so near, yet of such different depths, appear to have no connexion. Again, such an irregular subsurface must certainly be intersected by any long trench if of any serious depth. Such a trench exists in the valley of a brook which runs from N. toS., passing about half a mile west of the Rectory, and more than 70 feet below the levels there. This valley taps a few springs, and here and there in the bed of the brook a little gravel may be seen. But nowhere do I see any indication that it has reached an extensive underlying formation. The conclusion I come to is, that the water must be contained in permeable seams, included in the mass of the Boulder-clay. The Clay itself appears absolutely impervious to moisture. In a neighbouring field water stood for several weeks in a small pit by the side of a deeper trench not two feet off. A shallow ditch recently dug along a fence a few inches into the Clay lay * Numbers between parentheses indicate height of surface above sea-level. 586 REV. FE. HILL ON WELLS IN WEST SUFFOLK BOULDER-CLAY. perfectly dry until a fall of snow. When this melted and filled it, the water stood there day after day till it disappeared by evapora- tion. But then, this clay being thus impervious, the question arises, how does the rainfall find its way down to the water-bearing seams? In the Geol. Surv. Mem. on quarter-sheet 51 8.E. (1886), which includes the west boundary of this parish *, it is said that ‘** owing to the boulders and stones it [the clay] contains, and also to occasional seams of sand, wells are often made in it and a fair supply of water obtained.” But the boulders and stones do not make it pervious: yet these sand-seams must obtain their water from the surface. I conclude then that this Boulder-clay is not an uniform homogeneous mass; it must contain seams or beds of gravel and sand, and these must rise to the surface, or in some way communicate with it. This conclusion was altogether contrary to my preconceived ideas. I knew that many geologists regarded this Boulder-clay as formed beneath a glacier by the friction of a mass forced forward over its bed. I expected therefore to find a clay perfectly imper- vious to water, and with any appearance of bedding which it might present parallel to the direction of motion. I have looked attentively at all visible sections. Those in the immediate neigh- bourhood are only such as ditches can give. They showed pockets and patches containing sand or silt of irregular and fantastic shapes, and though, by reason of denudation, it is probable that this is no superficial structure, they are necessarily very limited and unsatis- factory. As yet I have been able to examine more extensive sections only in two localities, namely, Bury St. Edmund’s, seven miles to the N.N.W., and Sudbury, ten miles to the & Near Bury, just beyond Horringer (Horningsheath) Red House, a large pit in the fields shows about 8 feet of Boulder-clay resting on gravel, and below the gravel, chalk. Here then we have the base of the Clay shown for several hundred yards. It is typical Boulder- clay of a reddish colour, full of chalk and flints. Butin the middle, at my first visit in Nov. 1890, was a mass about 12 yards long clearly distinguished by its bluish tint from the rest; on one side it abutted against this vertically, on the other overlay it obliquely. At Sudbury Boulder-clay is seen in several pits, one of which contains the remarkable mass described by Mr. Marr in the Geol. Mag. for 1887 (p. 262). The best section is that in the Ballingdon Brick-yards, across the river, about half a mile N.W. of the Sudbury Railway Station. Here, below the thin surface-soil, is some 10 feet of yellow Boulder-clay, in which a sandy streak, visible at some distance, runs for several feet horizontally, then turns up and bends back till it reaches the surface. Below the clay is from 12 to 20 feet of much contorted stratified gravel. This lies on the irregular surface of a second Boulder-clay, very dark in its upper portion, but light with dark patches below, and containing irregular patches and seams of yellow sandy clay. About 20 feet of this lower mass was * Cockfield is mainly included in quarter-sheet 50,8.W. The memoir on this map was published in 1881. REY. E. HILL ON WELLS IN WEST SUFFOLK BOULDER-CLAY. 587 seen, but not what it rested on, except that at one spot the work- men had struck an excessively hard gritty bed, apparently cemented by carbonate of lime. Both upper and lower clay were well shown by freshly cut faces, and both were by no means homogeneous masses. The appearances both at Bury St. Edmund’s and at Sudbury are quite different from what I should expect in a mass resulting from the abrasion by a glacier of its bed. As no one (so far as | know) has ever seen a so-called ‘ ground-moraine” beneath an existing glacier, we can only argue by analogy. According to the supposed method of formation I should expect this to possess a structure analogous to that produced when a sledge is dragged over soil. No section shows such appearances. I think it worth while to call attention to these facts, as a satisfactory theory of the origin of this vast mass must give a satis- factory account of its phenomena. I have no theory of my own at present to propound, but I hope to continue observations on this interesting and difficult subject. Note on the Cockfield Post-Office Well. Locality.—Cockfield Post Office, rather more than half a mile N.W. of Cockfield Railway Station. Height above O.D. about 270 feet. The bed of the brook is about a quarter of a mile east, less than 40 feet below. feet. Ci eiaitow.clay (Tniek Garhler ete eviieectsssscs.cc-.c.c2.03... 000s about 8 (2 bod preuvel, writly Maree Sait sep wceese onc glen ete oases neneswaensine 7 (3) Yellow clay with much small chalk ....,.....50...0c02...eseeesecees 3 (4) Blue Boulder-clay with much chalk in well-rounded pebbles... 40 (oO) aeroken fumos Gf chalk ACG srcinc cece n ccc. sc ssea ce sescessacenes 5 (6) Blue clay with sub-rounded chalk fragments, base not reached. The well was continued in (6) to a total depth of 83 feet, and then abandoned, no water having been met with below the gravel. In (4) were masses of dark calcareous clay, full of ammonites, &e. (Kimmeridge Clay ?), often scratched. At the depth of 22 feet a chalk boulder at least two feet long was met with. When (5) was reached it was at first supposed to be the surface of the Chalk itself. In (6) the chalk fragments were mostly as large as pigeons’ eggs ; flints were certainly less numerous than in (4), both according to my own observation and the opinion of the well-sinker. The well was begun in March 1890, and abandoned in May of the same year. Discussion. Prof. Presrwicn remarked that intercalated beds of gravel and sand were common at different levels in the more northern Boulder- clay, and that in parts of the Eastern Counties a bed of gravel, from 1 to 20 feet thick, generally occurred in the centre of the Boulder-clay. These formed small water-bearing beds, but the 588 REY. E. HILL ON WELLS IN WEST SUFFOLK BOULDER-CLAY. main sources were usually at the base of the Clay—a base which was extremely irregular. He asked the Author how, as the wells stopped at the water-bearing stratum, he could be sure that this was, in all instances, intercalated and not an underlying bed. It was essential to know the level of the ground at the different wells, and this would no doubt be given in the paper. ‘The component beds of the Boulder-clay would vary according to the surface passed over by the ice, and may, therefore, include long trails of sands and gravels, and are necessarily local and irregular. He hoped that the Author would continue his observations. Dr. Evans agreed with the Author in regarding the mixed character of the Boulder-clay of Suffolk and some of the features that it presents as being hardly consistent with its being merely the result of a coating of land-ice. In illustration of the perme- ability of the beds at certain spots, he cited the deep circular pits or meres in the neighbourhood of East Wretham, Thetford, which were due to the dissolution of the underlying Chalk by water charged with carbonic acid having forced its way through the Clay. The level of the water in these meres depends upon the saturation of the Chalk, and the bottom of what in one year was a deep pool might in another year be cropped with turnips. Mr. .CLement Rei observed that intercalations of seams of sand — were almost universally characteristic of the Boulder-clay, and helped to render it somewhat pervious to water. He was unable to follow the Author’s argument, that irregularities in the deposits proved that the Boulder-clay could not have been formed under ice. Mr. CHariEsworts said that at Saffron Walden in Essex, on the borders of West Suffolk, the Boulder-clay is now being quarried on an extensive scale for the purpose of making in combination with Chalk what is there called ‘‘ Portland cement.” The denudation of the Chalk and Oxford Clay has largely contributed to this Boulder-clay at Saffron Walden; and sections of the deposit are displayed of extreme geological interest. Mr. Tortery called attention to the researches upon the glacial geology of the Eden Valley carried on by Mr. Goodchild, who believed (as does Mr. Reid for Norfolk) that the irregular beds of gravel and sand occurring in the Clay were formed within or under the ice-sheet, the gravel, &c., having been washed out of the Clay into hollows of the ice during partial or local melting of the ice-sheet. Mr. Goopcuitp said that similar intercalations of sands and gravels in the Boulder-clay were common in the North. He re- minded the Society that he had proposed an explanation of the origin of such deposits many years ago in the Society’s Journal and elsewhere (Quart. Journ. Geol. Soc. vol. xxxi. (1875), and Geol. Mag. for 1874). The Prestpenr referred to his own early work in the Boulder- clay and the abundant evidence which he had everywhere found of intercalated nests and layers of sand and gravel in that deposit. He had always been accustomed to regard these intercalations and REV. E, HILL ON WELLS IN WEST SUFFOLK BOULDER-CLAY. 589 their singular contortions as affording some of the strongest proofs of glacier-action, and though he admitted that the Boulder- clay still presented many unsolved difficulties, he had never seen what he could regard as a valid argument against the view that the true typical Boulder-clay is essentially a product of land-ice. The AvrnHor, in answer to Prof. Prestwich, said that he had taken into account the variations in surface-level of spots where wells existed. Dr. Evans’s instances of permeability in Boulder- clay were a valuable corroboration. ‘The appearances of sections did not to himself suggest an origin such as erosion by subglacial streams. He would be very glad to study the sections at Saffron Walden and those in the Eden Valley described by Mr. Goodchild. He was not aware of any case in which a “ ground-moraine” had been seen in actual process of formation, but he imagined that any structures possessed by a mass so formed would be horizontal in their general direction. The appearances described in the paper were not of that character. 590 MR. J. J. LISTER ON THE GEOLOGY 35. Norrs on the Guotoey of the Tonga Istanps. By J. J. Lister, Esq., M.A. (Read June 24, 1891.) [Communicated by J. E. Marr, Esq., M.A., F.R.S., Sec. G.S.] {[Puare XXIII. ] ConTENTS. Page 35. PRY RODUOGEION: 2000. os cacctorec ee eee eee 590 Tf, TorocRaray or Tam GROUP... 20.02)... dds see laced cess eeses 590 IIT. Grorogican CHARACTERS OF THE GrRovur. (a), Ve. Volaaniot slats asus: suse cpcnewaaedsecudad 591 (6) Islands formed of Sedimentary Volcanic Wistorial) OS oon oes.cuaseme ote eeeateece te eeeee 595 (c) Islands formed entirely of Limestone ............ 606 EV: Comeniaiont 23 Ge ibe es Pee 616 1. Inrropvction. Very little of a detailed character has, so far as I am aware, been published on the geology of the Tonga Islands. Prof. Dana makes a brief mention of some of the islands in * Coral and Coral Islands’ *; and in the edition of ‘ Coral Reefs’ published in 1874 Darwin gave a summary of what was known of the group ft. I spent several months in Tonga in 1889 and 1890, and had opportunities of visiting some of the islands in H.M.S. ‘ Egeria,’ which was engaged, for part of the time, in making a survey. Such observations as I made on the geology of the group are embodied in the following notes. I am indebted to Mr. Alfred Harker, M.A., F.G.S., who has examined the volcanic rocks which I obtained t, and also to Dr. John Murray, who has looked over sections of the calcareous rocks of Eua. Il. TorpograPHy oF THE Group. The Tonga or Friendly Islands are situated in the Pacific Ocean between 18° and 22° 30’ lat. S., and on either side of the 175th degree long. W. Tongatabu lies a little over 1000 miles to the N.N.E. of Auckland in New Zealand. The large islands of Fiji are about 400 miles to the W.N.W., and those of Samoa about 300 miles to the N.N.E. The greater number of islands fall into three main groups. The northernmost is Vavau, which consists of one moderately large island (about 13 miles in its longest diameter), giving its name to * 2nd ed., pp. 288 & 289. + P. 212 and elsewhere, 3rd ed. (1889). +t Mr. Harker has published notes on these specimens in the Geol. Mag. for June, 1891. OF THE TONGA ISLANDS. 591 the group, and a number of islets lying to the south and south-west of it. Rather more than fifty miles to the south is the Hapai group, containing the high volcanic islands of Tofua and Kao on the west ; on the east a chain of low, flat, coral islets; the Kotu group in the centre ; and on the south a number of isolated islands, the largest of which is Nomuka. Some fifty miles to the south and somewhat to the west of Nomuka lies Tongatabu, the largest island of the group, measuring over twenty miles in its longest diameter. A channel about ten miles broad separates Tongatabu from the small and high island of Eua to the south-east, while eighty-five miles to the south-west lies the little island called Ata by the natives, the Pylstaart of the charts, the most southerly outlier of the Tonga Islands. (See Map on p. 594 and Pl. XXIII.) Lying far to the north of the islands mentioned are Niuafou (=Good Hope, or Proby Island), Keppel, and Boscawen Islands, which, though politically belonging to Tonga, occupy an inter- mediate>position between this group, Fiji, and Samoa. ITI. Geonoeicat CHARACTERS OF THE GROUP. According to the characters which they present, the Tonga Islands may be arranged in three divisions, viz.:—(a) purely volcanic islands; (5) islands formed of volcanic materials laid out beneath the sea, and since elevated, with or without a covering of reef-limestones; and (c) islands formed entirely of reef-limestones. (a) The Volcanic Islands. The purely volcanic islands form a line running parallel to the long axis of the group, but a little to the west of it. The direction of this line is approximately N.N.E. and 8.8.W. At the southern end is the little island of Ata or Pylstaart, which, judging from the extremely steep and rugged outline that it presents, may possibly be a member of this series. I have, however, merely seen a photograph of the island, and it can therefore only provisionally be reckoned a member of the volcanic series. The islands of Honga-tonga and Honga-hapai, lying north of the west end of Tongatabu, are I believe the fragments of an old erater. They stand out of the sea about a mile and a half apart, and reach a height of some 300 and 200 feet respectively. Each island has a high vertical face on the side turned towards the other, and on all sides hard black layers of rock are seen alternating with softer ones, and slope away from a point situated between the two islands. Fifteen miles to the north of these islands is the recently-formed volcanic mound named Falcon Island. In the year 1867, H.M.S. ‘ Falcon ’ reported a shoal in the position of the present island, and ten years later smoke was seen issuing trom the sea at this spot by H.M.8. ‘Sappho.’ In October 1885, a submarine volcano suddenly burst into activity, and built up a 592 MR, J. J. LISTER ON THE GEOLOGY mound. The volcano was still in eruption the following year. In 1887 the height was estimated as 290 feet by a French man-of- war; and from a sketch taken in that year, it appears that the summit of the island presented a rugged, uneven outline *. The ‘ Egeria’ visited the island in Oct. 1889. In the four years’ that had elapsed since its formation, a large portion had been removed by the action of the sea. Only a part (perhaps one third) of the original wide-based mound remained. This was limited for half its circumference by a cliff which was highest in the middle (153 feet), where it faced south, and subsided gradually at either end. From the summit of the cliff the surface of the mound sloped evenly downwards until it reached the level of a wide flat marked by tidal ridges, which extended round that part of the circumference of the mound not limited by the cliff. Landslips from the cliff were of frequent occurrence at high water; as many as twelve were seen in one afternoon, and the ground round the edge of the cliff was traversed by concentric cracks, showing where slices had _ already become loosened preparatory to their descent. Capt. Oldham erected a line of cairns at equal distances, extending from the edge of the cliff—down the slope of the mound and over the flat—as a means of measuring the rate at which the island is reduced. The mound consisted of layers of finely-divided volcanic ash. On the face of the cliff the layers were marked by the salts which had crystallized at the surface more abundantly in some than in others, and it might be seen that each layer was thickest at the highest part of the mound, and gradually thinned out towards the periphery. Numbers of volcanic bombs were scattered over the slope of the mound, being largest and most numerous at the highest part. These appear to have been ejected at the close of the eruption, as none appeared in the cliff-section; and the even, unscored outline of the slope negatived the idea that they had accumulated at the surface by the removal of any considerable quantities of finer materials. The lavas presented all stages of vesicular structure, and some of the bombs displayed a distinct spiral twist in the surface ridges. Mr. Harker has examined the lavas and finds that they are basic augite-andesites of specific gravity 2-708. The temperature of the interior of the mound was still high at the time of our visit. In a hole 7 feet deep a thermometer rose to 100° Fahr., while it registered 77°°5 at the surface. To the south of the island there was an extensive shallow area with 3 fathoms of water over it. It appears that this was, in part at least, occupied by that portion of the original mound which has * The above details are taken from an account of the island by Capt. Wharton, R.N., F.R.S., published in ‘Nature’ for Jan. 23, 1890 (pp. 276- 278). This account is accompanied by figures and a map. t Geol. Mag. for June, 1891, p. 250. OF THE TONGA ISLANDs. 593 been removed by the action of the waves. Much of the material so removed has been thrown up on the north side of the island, forming the flat above mentioned, and an extensive shallow area beyond it. It seems clear that unless another eruption occurs the island will in a short time be reduced again to the condition of a submarine bank formed of volcanic ashes rearranged by the action of the waves *, It is remarkable that the depth of 1021 fathoms has been found between Falcon Island and Nomuka, the nearest of the Hapai Islands. Next in the series come the high volcanic islands of Tofua and Kao. .Tofua is a volcano in a state of intermittent activity. It is marked in the Admiralty chart as attaining a height of 1890 feet above the sea. The summit presents a fairly even outline when seen from a distance, and the sides slope steeply in all directions. The crater is situated on the northern side. Kao is-said to attain a height of over 3000 feet. It presents a singularly-perfect conical outline, from all points of view. It has not been in activity within the period of native tradition. Continuing the line are Metis Island, which appeared a few years before Falcon Island, and is stated in the Admiralty chart to have been still in activity in 1886; and the volcano Lette, which is figured in the chart of the group (No. 2421). In 1866, when this island was visited by Lieut. Creak, R.N., in H.MLS. ‘Esk,’ vapour was seen issuing from the crater f. To the north of Vavau is the volcano of Amargura. An explosion occurred in 1847, when the island was in part “destroyed by the eruption of its crater” and “ ashes were thrown in large quantities on passing ships 500-600 miles to the N.E.” t. Northwards the line passes through Boscawen and Keppel Islands, which lie close together, halfway between Amargura and Samoa. Boscawen Island (=Niua-tabu-tabu) is described in Findlay’s Directory as one entire mountain about 2000 feet in height ‘resembling the Moluccas;” it is therefore probably a volcanic island. Niua-fou (or Good Hope Island), lying considerably to the west- ward of these islands, is probably a member of the same series. It is described as a volcanic island with black lava rocks all round the shores, and with a crater in the middle containing a brackish-water lake. An eruption occurred in 1853, when a village was destroyed and many lives were lost. Another eruption occurred in 1867 §. In the continuation of the direct line northward is the large voleanic island of Savaii, the most westerly of the Samoan islands. The Samoan volcanoes, however, form a series having a direction approximately W.N.W. and E.S.E., which crosses that of which the * For afuller account (with figs.) of the present condition of the island, see a paper by the present writer in Proc. Roy. Geograph. Soe. vol. xii. (1890) p. 157. + Findlay, ‘South Pacific Directory,’ 3rd ed. p. 456. t Findlay, op. cit. p. 456, § Findlay, op. cit. p. 558. [ia0* = © Relumah Foluna Wola ANiualor Metis £ 6) Fatcon [@y6 alcort 8 0 SHonga tonga Sengu Hapee: 2900 MAP of PART OF THE SOUTH PACIFIC OCEAN. THE GEOLOGY OF THE TONGA ISLANDS. 595 Tongan volcanoes are members at right angles. It is significant that Savaii is the largest of either series, and is situated at the point where they intersect. Savaii ends the series in a northerly direction. The soundings which have been taken in this part of the Pacific show that the Tonga Islands are situated on a ridge, which rises to within 1000 fathoms of the surface and is prolonged in a south- westerly direction for half the distance between this group and New Zealand. To the south of the Tonga Islands the top of the ridge reaches the surface of the sea, or comes close to it, at the Minerva Reef and three shoal patches indicated in the chart. On the eastern side, where soundings have been taken, the slope descends to depths of over 2000 fathoms, and at two points, one to the north-east of Vavau, the other to the east of the southern extension of the ridge, the very great depths of 4530 and 4428 fathoms have been sounded. On the west the slope appears to be less steep, but a depression (with a depth where it has been measured of over 1500 fathoms) lies between this ridge and the Fiji group. The Kermadec Islands, lying between the southern part of the ridge and New Zealand, are the highest points of a large area which is also within the 1000-fathom line. It is, however, separated from the Tonga ridge to the north and New Zealand to the south by depths of over 1500 fathoms. Both elevations are traversed by a line of volcanoes, which, as Mr. 8. Percy Smith * has pointed out, continues the direction of that of the Taupo Zone of New Zealand. Mr. Smith suggests that the volcanoes are situated in the course of a great fissure which extends from Ruapehu, at the southern end of the Taupo Zone, through the Kermadec and Tonga groups to Samoa. ‘This view receives some support from the fact that the eruption of Falcon Island, beginning in Oct. 1885, was contemporaneous with the increased activity (in November) of the geyser ot the white terrace of Lake Rotomahana, which preceded the great explosion of Toran in June of the following year. [In the map facing this page the dotted areas hetween New Zealand and the Tonga Islands indicate depths of less than 1000 fathoms. (The eastern limit of the Kermadec area is undetermined.) The thick cireles indicate active or efiaGh volcanoes, the broken circles islands whose volcanic nature is doubtful. | (b) Islands formed of Sedimentary Volcanic Material. Belonging to this division are several islands in the Hapai Group, and the island of Kua. . The Nomuka Group, which forms the southern division of the Hapai Islands, consists of a number of small islands which are the highest points of an extensive plateau. Over a large area the top of this plateau is within 50 fathoms of the surface of the sea. * «Geological Notes on the Kermadec Group,’ Trans. N. Z. Inst. vol. xx. p. 335. Q.J.G.8. No. 188. 2Q7 596 MR. J. J. LISTER ON THE GEOLOGY Nomuka itself is a limestone island, and will be described later ; but Mango, Tonua, Nomuka-iki, Tonumeia, and probablv Kelefasia are formed of volcanic tufts. The island of Mango lies about six miles E.8.E. of Nomuka. The main part of the island is some two miles in length, and of an oval shape, with the long axis running east and west. Projecting to the south of the west end of the island there is a small peninsula connected with the main mass by a narrow neck. At the eastern and western ends of the island there are rounded hills attaining a height of about 150 feet above sea-level. This island is composed for the most part of layers of volcanic tuffs which vary widely in character. The summit of the eastern hill is formed of thick and approximately horizontal layers of fine white material which is largely calcareous, but contains a fine scoriaceous residue. The western hill is composed of layers of a coarse sandstone dipping at an angle of about 2° or 3° towards the W.S.W., and con- sisting of rounded fragments of lava embedded in a calcareous matrix. Fragments of coral, some of them 6 inches in diameter, are mixed with the volcanic fragments. Between the two hills iayers of the two kinds of rock alternate with one another. The southern peninsula is formed of a mass of breccia traversed in various directions by cracks, but presenting no regular stratifi- cation. The fragments are embedded in a calcareous matrix and vary greatly in size. Lying on the surface are some large boulders, some of coral, some of volcanic rock, which appear to have been isolated by the gradual removal of the finer material in which they were embedded. The largest of these was a rounded mass of coral which measured 10 feet in length, 7 feet in breadth, and 34 feet in height. In the cliff which surrounds this part of the island, frag- ments of coral were thickly scattered among the volcanic con- stituents. I counted 43 pieces exposed in one square yard. Though Mango is surrounded by a broad fringing reef, I failed to find any raised coral-rock upon it. The island appears to have been originally formed as a sub- marine bank, probably of volcanic origin, on which corals grew. On a return of volcanic activity the coral-reef was broken up, and the fragments of it mixed with the volcanic materials have formed the breccias of which the island is composed. lt appears that the southern promontory was the nearest point of the present island to the centre of the eruption, for there the con- stituents of the breccia are largest and mixed indiscriminately without stratification. The nearly horizontally stratified rocks on the eastern and western hills were laid out now in finer, now in coarser layers by the action of the water. The island has subsequently been elevated to its present height, the rounded outline of the hills being due to subaerial denudation. It is remarkable that while the island is now surrounded by flourishing coral-reefs, there should be no old raised reef upon it. The most probable explanation appears to be that elevation took place too rapidly to allow of the growth of reefs of any considerable extent. oe OF THE TONGA ISLANDS. 597 Some miles to the westward of Mango is the small island of Nomuka-iki (=little Nomuka), to the south of Nomuka, and separated from it by a narrow channel. This island is about 60 feet high, and one mile in its longest diameter; it aiso is formed in part of a hill of volcanic tuff, the remainder consisting of a flat of calcareous sand. The layers are horizontal and are made up of finer and coarser beds of brown and grey ashes alternating with one another. Here there are no fragments of coral mixed with the ashes. A fossil univalve of tho genus Pyrula occurred in these beds, and some of the finer ones contained vertical burrows (of an annelid?) which were filled in with the coarser material of the overlying layer. As at Mango, there are no raised coral-rocks overlying the voleanic beds, although broad fringing reefs surround the shores. The beds have evidently been laid out by water, but under quieter conditions than those of the neighbouring island, and the mound kas since been elevated to its present height. A. little to the south of Mango are the two islands Tonumeia and Kelefasia. Capt. Oldham visited Tonumeia when engaged in the survey of the group in 1890. He found it to consist of volcanic tuffs forming a cliff 80 feet high, and dipping at an angle of 3° to the south*. Mr. Harker finds that a specimen of these beds consists of “ fine volcanic ash compacted by a calcareous and ferru- ginous cement into a yellow-brown rock.” Capt. Oldham also obtained some black nodules two inches in diameter, with portions of a calcareous matrix still adherent, and which had presumably weathered out of the layers of tuff. Mr. Harker describes these as consisting of oxide of manganese having ‘“‘the general characters of psilomelane, but soft and of a low specific gravity, and therefore perhaps altered.” The presence of these nodules, supposing them to have occurred naturally in the beds of which Tonumeia is composed, is very re- markable. They are generally supposed to be formed only in deep water ; but the situation of the island on the same shallow plateau as Mango and Nomuka-iki, and the general correspondence of its formation with theirs, point strongly to the conclusion that these islands were all formed under similar conditions. The presence of coral and other organic remains amongst the volcanic constituents of these islands shows that they were formed in shallow water. Judging by its appearance as seen from the sea, I have little doubt that Kelefasia belongs to the same class of islands. The island Tonua, to the north-east of Mango, was also examined by Capt. Oldham. It is composed of yellowish-white rocks which form a cliff twenty feet high, and consist of volcanic ashes with little calcareous matrix, though shells of pteropods and the cast of a gasteropod (Murex?) were found among the volcanic fragments *. In the central part of the Hapai group there are two islands, * The substance of Capt. Oldham’s report on this island and Tonua is given in Mr. Harker’s paper in the Geol. Mag. for June, 1891. t Harker, op. cit. 272 598 MR, J. J. LISTER ON THE GEOLOGY Kotu and Matuku, which, as seen from the sea, appeared to me to be of the same nature as those above described in the Nomuka group. ‘The only confirmatory evidence that I can offer is a remark on the island of ** Kotoo” in the narrative of Capt. Cook’s Voyage to the Northern Hemisphere (vol. i. p. 271), where it is stated that this island ‘‘ terminates in reddish clayey cliffs.” Leaving the Hapai Group, I will now attempt to describe the island of Kua, lying to the south-east of Tongatabu. (For the map of this island, see Pl. XXIII.) Eua is somewhat oval in shape, but pointed at the two ends; the long axis is directed almost exactly north and south. It is about 12} miles long, rather more than 4 miles broad, and attains a height of 1078 feet. The island is formed of a basis of volcanic rock, in great part covered by lime- stone. >) es = ° S (2) = oS x] s 2° ° ® a a E L —- aoe -- . ; + —— = = sy ‘ ‘| “S1g [3 JO uontsod oy “qu pur SI UOTjoes 9Y} JO “KN 0N4 9 of} pus “A'S 04} IV nog} “AA pue gq Surssed uoroas “UdYB} WOM OLZT PU GOTT ‘SON suounidads OSvq BSOYM WOAJ O[OVUUTG ot MOYs 04 ‘pazvor 0} sur, Axyuno -uins ey) Ys ‘D IVIL, “NOILLVNVTdX@ “A FTVIS LYIA = i 'D VOBAI9 J, “S96T ‘meyy nu ‘T “Suy ‘(spumsy vbuoy,) vng ssouan suonoag aynununsborg 602 MR, J. J. LISTER ON THE GEOLOGY The most striking feature of the formation is the existence of well marked terraces on the lower slopes. The terrace a, as above stated, extends along the whole length of the western side of the island, at an elevation of about 250 to 365 feet above sea-level. The outer, western border of the terrace is for a great part of its length higher than the interior. Beginning at the northern end, we find the terrace forming a narrow band, so narrow that it makes only a slight interruption in the sloping side of the island. To the south it gradually widens out, being at first level from side to side, but soon the interior becomes depressed below the margin. ‘The depression is greatest at one-third of the length of the island from north to south. Here the margin is about 65 feet higher than the interior. As we advance southward the depression of the interior becomes less marked, until at the junction of the middle and southern thirds of the length of the island the terrace is again level, and remains so to the southern end. Passing across the terrace from east to west, in the region of the depression we descend a very gradual slope to near the outer border. From the lowest part the ground rises to the margin of the terrace, in most places by a gradual ascent, but at one point the rise is abrupt. Opposite the town of Ohunua there is a gap in the high margin, the bottom of which is at the same level as the depressed inner part of the terrace. Through the gap there runs a steep-sided ravine, cut in the limestone rock; it is formed by the junction of two of the overflow channels of the streams of the upper part of the island, and there is water in it only after heavy rains. Having passed the gap the ravine pursues a straight course down the sloping side of the island to the sea, and at its mouth there is a break in the reef and an anchorage for small vessels. On the margin of the terrace to the south of the gap the lime- stone rock forms projecting masses, and in some of these definite Astrea-like coral structure is to be seen. Other specimens present a homogeneous structure to the naked eye. Dr. Murray has ex- amined sections of these *, and finds that they are ‘‘ chiefly made up of calcareous alge, together with fragments of coral, molluses, echinoderms, and foraminifera” +. He concludes from the cha- racter of the organisms that the deposit was “formed in quite shallow water.” With regard to the peculiar conformation of the terrace, it appears that the depression in the interior must either have existed at the time that the terrace was elevated above sea-level, or have been excavated since by the denuding action of rain or streams. Slide No. 1271. Among the foraminifera the following kinds were recognized :— Miliolina, sp. Planorbulina larvata (P. & J.). Orbitolites duplex (?).. Carpenteria monticularis (Carter). Textularia trochus (d’Orb.). Polytrema mintaceum (Linn.). Globigerina bulloides and Gl. rubra(?). | Calcarina hispida (Brady). —— OF THE TONGA ISLANDS. 603 In ordinary weather the streams flow beneath the limestone rock, and it is only after heavy rains that running water is found in the ravines which traverse the terrace. Moreover, the sides of these slope abruptly down from the level of the terrace, and their course is for the most part quite independent of the direction of the depres- sion. Finally, it may be asked, Why should the drainage from a flat area excavate the interior and leave the margin comparatively untouched along almost the entire length of the terrace? It appears, then, that the terrace had much the same general conformation which it now presents when it was upraised from the sea ; in other words, that the elevated border constituted a small barrier-reef lying nearly a mile from the shore of the island as it then existed, with a break in it at one point, and enclosing a lagoon with a depth of 6 fathoms over a considerable area. At its northern and southern ends the reef joined the shore and became fringing. A similar relation of reef to shore occurs in the reefs now forming the harbour of Suva”in Viti Levu, the capital of Fiji, where the barrier which protects the anchorage is continued into a fringing reef along the southern shore of the island. As above described, there is a narrow limestone-terrace running along the eastern side of the island at-about the same level as terrace a on the western, and interrupted in the middle by a deep gap, the head of which is formed by the cliffs of nearly the highest part of the island. On the north of the gap the terrace ends abruptly in a cliff, facing south, and standing cut beyond the cliff is an isolated pinnacle of limestone rock, which I estimated as measuring about 300 feet from top to bottom. The rock of the base of the pinnacle presents to the naked eye a compact homogeneous structure, except that in one specimen there is a cavity which was occupied by a univalve of the genus Cerithiwm. Dr. Murray has examined sections of the rock * and finds that it is made up prin- cipally of calcareous algee, together with fragments of echinoderms, molluscs, and a number of foraminifera 7}. The limestones of terrace 6, which stands at an elevation of about 500 feet, are also for the most part homogeneous compact rocks. A section from the edge of the terrace ¢ is found by Dr. Murray to be “chiefly made up of calcareous organisms. Fragments of molluscs, echinoderms, polyzoa, and calcareous alow, together * Slides 1269 and 1270. t Viz. — Spiroloculina, sp. Polytrema miniaceum (Linn.). Bulimina (ef. compressa). Rotalia, sp. Bolivina textilarioides (Reuss). Tinoporus bacculatus. Textularia trochus (d’Orb.). Cymbalopora Poeyt (d’Orb.). Cristéllaria, sp. Gypsina inherens (Schultze). Globigerina (rubra ?). Cycloclypeus Carpenteri (Brady). Truncatulina, sp. Heterostegina depressa (d’Orb.). Planorbulina (ct, larvata). Nummulites Cumingii (Carpt.). t Slide No. 1272. 604 MR. J. J. LISTER ON THE GEOLOGY with a large number of foraminifera” *. True coral does, how- ever, exist on this terrace, as I found one large mass clearly exhibiting a structure like that of a Pordtes. The limestones from the higher part of the island present for the most part a compact structure. Some of them are crowded with foraminifera, but large masses of coral appear to be absent. With regard to the nature of the limestone formations as a whole, as indicated by the sections above described, Dr. Murray writes to me that they ‘‘may all have been formed in or about reefs, and true coral-reefs may have been living quite close to them when they were laid down ;” but from the scarcity of examples of the larger corals he concludes that ‘‘ it is more likely that they were laid down in depths of from 30 to 100 fathoms, and that they formed the base of true coral-reefs.” The conformation of the surface of terrace a points to its having formed a true reef, and having been elevated without much denu- dation ; but in the case of terrace 6 and the higher limestone rocks, the extent to which they have evidently been denuded and their nearness to the volcanic basis bear out Dr. Murray’s view, founded on the nature of the organisms they contain. Some features of interest are presented by the shores of the © island, Om its western side, in the neighbourhood. of the village of Ojvunua, the island is bordered by a narrow fringing reef some 50 to 100 yards in width. There are clear traces here of recent eleva- tion. ‘To the south of the village there is a low shore-cliff under- mined by the action of the sea, but separated from the present shore by a strip of sand some 50 to 60 yards wide, covered with cocoanut palms and other trees. From a little to the north of the village the island is bordered by a cliff some thirty feet high, which extends to the northern end, The foot of this cliff is washed at high tide by the sea, and is deeply. undermined, the upper limit reached by the action of the sea being sharply defined. Above this present line of undermining, and sepa- rated from it by a projecting ledge of rock, there is another and older one, the upper limit of which is sharply defined and separated from that of the present line of wave-action by a distance of seven feet. When looking at the cliff from the sea, I thought I detected a third line of wave-action at a still higher level. From the foot of the cliff the shore-platform extends some fifty to a hundred yards, and is bordered by growing coral. Over the greater part of its extent this platform is flat, and at high tide is * Miliolina, sp. Polytrema miniaceum (Linn.). Orbitolites complanata (?). Gypsina inherens (Schultze). Textularia trochus (d’Orb.). Rotalia, sp Polymorphina, sp. Calcarina hispida (Brady). Globigerina (bulloides ?). Cycloclypeus Carpenteri (Brady). Truncatulina, sp. Heterostegina depressa (d’Orb.). Planorbulina larvata (P. & J.). Nummulites Cumingii (Carpt.). Carpenteria monticularis (Carter). oe Ow % OF THE TONGA ISLANDS. 605 covered by about two feet of water. In some places, however, there are the remains of the shore-platform which was formed when the island stood at a lower level, and corresponds with the higher line of undermining seen in the cliffs above. Its upper surface is above the level of high water, and is worn into irregular shapes. It is overgrown in some places by a small shrub (Pemphis acidula), These features point clearly to recent elevation, and the distance between the two lines of undermining seen on the cliff shows that the extent of this was 7 feet. Further, it is clear that the eleva- tion took place rapidly, at least as measured by the rate at which the sea is wearing back the cliffs, for if it had been accomplished slowly, in comparison with that measure, the rock which forms a projecting ledge between the two levels of excavation would have been worn back during the elevation. It is interesting also to note that if the island were to remain at the level it occupies at present, all traces of the recent elevation would in time be obliterated, for as the cliff was worn back the rock bearing the impress of the action of the waves when the island was at the lower level would be removed, while the remnants of the older reef-platform would be worn away to the level of the present one. It may further be observed that a reef-platform such as this consists of two distinct parts—an outer formed by the growth outwards of: the existing reef, and an inner formed of the base of the older and elevated reefs, whose higher portions have been removed by the action of the sea, At the south end of the island there is no reef; deep water extends to the foot of the cliffs. A narrow fringing reef, some hundred yards wide, extends along the eastern side of Eua. Its upper surface is above the level of high water and partly worn away by the sea, offering evidence of the recent elevation of the island, corresponding to that found on the western shore. The margin of the reef is exposed to the full force of the rollers coming up before the trade wind, and presents a remarkable basin-like growth of corallines similar to that found on the southern side of Tongatabu, to be described later. At the south end of the eastern shore of Kua there is a remark- able instance of the different manner in which the volcanic and limestone rocks are excavated by the action of the waves. A narrow terrace, perhaps a quarter of a mile broad, runs round this end of the island, standing at a level] of some 240 feet above the sea. It is bordered by cliffs which go sheer down into deep water. Limiting it on the land side there is a steep ascent to a higher terrace above. The limestone rock here lies on the steeply sloping surface of the volcanic basis of the island. The outer or sea border of the terrace and nearly the whole face of the cliff are formed of lime- stone; but at the inner border of the terrace, and here and there in the lower part of the cliffs, the volcanic basis is exposed. At ‘one point there is a large circular chasm excavated in the terrace which descends to sea-level and communicates with the sea by a 606 MR. J. J. LISTER ON THE GEOLOGY wide opening, overarched by a natural bridge of the limestone rock, the under surface of which is festooned with stalactites. The sea entering beneath the arch washes the bases of the nearly vertical sides of the chasm. The place is called by the natives Matalanga Maui (or the hole made by the implement for planting yams belong- ing to the god Maui). The manner in which the chasm has been excavated appears obvious. At the foot of the cliffs there are several caves in ditferent stages of excavation, which are formed by the sea wearing back the volcanic rock where it is exposed beneath the limestone. The Matalanga Maui has been formed by the removal of the volcanic rock to such an extent that the cave has approached the surface of the terrace and the roof has fallen in. I may here recapitulate the main features which seem to be indicated by the structure of Kua. The presence of the plutonic rock points to the existence of an ancient and much denuded mass of such a character in the vicinity. In the neighbourhood of this mass, and at a considerable depth below the surface of the sea, a volcanic eruption occurred, forming. a mound, some of whose upper layers were mingled with pelagic organisms as they were laid out. It appears that the mound was elevated above the surface, con- siderably denuded, and afterwards depressed to such a depth, at least, as to submerge its summit, A covering of calcareous organisms was deposited on the volcanic basis while it was submerged; but so far as the evidence goes the formation is a shallow-water one. ‘The island has since been again elevated by stages to the present height. During periods when the elevatory forces were in abeyance, the sea wore back the already emerged limestone-deposits, forming terraces bounded by lines of cliff, but in one case at least (namely, the western terrace a) it appears that a true coral-reef was formed in the neighbourhood of the shore. Dr. Murray finds that it is impossible to assign the organic deposits of the island to a definite geological epoch. Though he is of opinion that they are old, there is no satisfactory evidence to refer them to the Tertiary period. (c) Islands formed entirely of Limestone. The islands forming the remarkable group of Vavau differ widely in their shape and arrangement from coral islands in general. Vavau consists of one large island which gives its name to the group, and of a number of smaller islets (see Pl. XXIII.). Vavau Island presents a high and fairly even coast-line to the north and north-east, bordered by limestone cliffs from 300 to 500 feet in height. On the south and south-west the general lie of the land is lower than that on the northern coast. Long promontories stretch out from the main mass of the island, separated by narrow and deep iment ng: \ Land OF THE TONGA ISLANDS. 607 arms of the sea. The islets are grouped about this coast, in some cases continuing the lines of the promontories. The beautiful and secluded harbour of Neiafu is bounded in part by two of these pro- montories, and protected from the sea by one of the larger islets situated between them. It forms a long basin more than 20 fathoms deep over the greater part, and almost entirely surrounded by land over 100 feet in height. The surface rock all over the group consists, so far as I am aware, of limestone. J was unable to find the underlying basis at any point. Even at the foot of the high cliffs on the northern shore the rock, at the two points at which I saw it, was of this nature, and L found no fragments of volcanic or any other formation than lime- stone on the beaches. Although I saw no vertical cliff at this part of the island, extending from the top to the shore, the coast-line is in many places so steep that I have no doubt that the formation attains a thickness of at least 300 feet. It consisted for the most part of a hard homogeneous material, but at one point, in the lower part of the cliff, numbers of large bivalve shells were embedded. The main mass of the island presents an undulating surface, gradually sloping from the high northern border towards the south. Though there are valley-like depressions in several places, I came upon none which contained streams or even ravines. Springs of fresh water issue from the rock at many points on the shore. The outlying islands to the south of the group, as may be seen by the Admiralty chart (No. 2357), are surrounded by coral-reefs, which in many cases extend as broad expanses far from the island with which they are connected. As the inlets penetrate among the islands and promontories, and their shores are cut off from the open ocean, the reets become narrower or disappear ; though in some cases (as in the inlet to the north of the promontory in which Neiafu is situated) there are coral reefs of considerable extent growing far up in the sheltered channels. There is one feature which at once strikes a visitor approaching the group by the usual steamboat course from the south-west and passing up to the harbour of Neiafu, namely, that the off-lying islands and projecting points of the main island, on either hand, are flat-topped when seen in profile, and that the majority of them stand approximately at one of three levels of elevation above the sea. Corresponding in level with the tops of the lower islands there are, moreover, terraces on the sides of the intermediate and highest ones ; and in the same way there are, on the sides of the highest points of land, terraces whose level corresponds with the tops of the islands of an intermediate height. The islands and promontories may thus be described as being one-, two-, or three-storied (see fig. 3). On first entering the group, the flat one-storied islands Hunga and Nuipapa lie on either hand: their sides drop vertically into the sea and are deeply undermined below by the action of the waves. Right ahead stands up the mass of Mo’unga Lafa (= Flat Mountain), whose densely wooded sides present two well-marked terraces in *1TVAATRT SSS ———————— a = | / ‘mpany fo YINnoy ay2 02 SpunpsyT ay? buown ma.4 —'G “BIT THE GEOLOGY OF THE TONGA ISLANDS. 609 their descent to the sea. Opposite it, on the right of the passage is the island of Falevai (called “* Kopa” in the Admiralty chart), which consists in part of a two-storied mass, in part of a one-storied tongue of land which extends out towards Mo’unga Lafa, while a terrace at the same level interrupts the slope on the side of the higher part of the island. Small islands are dotted over the sheltered expanse of water, two of which, A’a (called “ Koto” in the chart) and Langito’o, present features of interest which will be alluded to below. As we advance, the hill Talau comes into view, exactly repeating the shape of Mo’unga Lafa, in the level outline of the summit and the two lines of terraces on its sides. The rough aneroid measurements that I obtained give about 140 feet as the upper limit of the first ‘story,’ 260 to 350 feet as that of the second, and 420 and 520 for the summits of Talau and Mo’unga Lafa, the two highest points of land in this part of the group. Although these numbers show that the levels of the terraces at different points do not agree accurately, the conclusion was strongly impressed upon me on the spot, that the terraces and islands at each level do in fact correspond, that’ is, that at successive stages in the elevation of the group, the summit of each “story” has stood at sea-level. The differences in their level may be in part accounted for by the unequal action of the elevatory forces at different points of the area acted on. Thus, on the hills Talau and Mo’unga Lafa, which are between three and four miles apart, while the summit of the lowest story is at about the same level in each (140-150 feet), the two upper ones stand at about 270 and 420 feet in Talau as against 360 and 520 feet in Mo’unga Lafa. From this it appears that the second elevation was of greater extent in the neighbourhood of Mo’unga Lafa than to the eastward near Talau. One remarkable feature in the contour of the land is the fact that in some cases the summits of the islands, though level in profile, are excavated by well-marked depressions, The most perfect example of this structure is presented by the little island of A’a (the Koto of the chart) above menticned. The island is of an oval shape, being about three quarters of a mile long and half a mile broad. It rises out of deep water, and is not sur- rounded by a reef. In the chart a sounding of over 40 fathoms is shown close to the southern shore, and except for a small shallow area at the eastern end the other soundings in the neighbourhood indicate nearly as great or greater depths. The sides slope steeply up to a height of from 85 to 95 feet (an elevation considerably less than that of the top of the first “ story ” in the surrounding islands), but the interior of the island is de- pressed 50 or 60 feet below the level of the margin, being occupied by a flat which stands at a height of some 36 feet above the sea. The margin is of a nearly uniform height and forms a complete unbroken ring. On its summit rough masses of limestone rock stand up, but the flat interior is formed of brown earth. It seems clear that when the rock forming the edge of the island stood at sea- level a small lagoon some 10 fathoms deep occupied the centre. 610 MR. J. J. LISTER ON THE GEOLOGY I was led to visit A’a by seeing its depressed interior from the top of Mounga Lafa. From that position the neighbouring island Langito’o, to the west of A’a, which rises to about the same height, appeared to have a similar depression in it. Unfortunately I had no opportunity of visiting this island. Fig. 4.—Sketch of the contour of the island of A’a (Vavau Group). The numbers indicate heights, in feet, above the sea. The summit of the three-storied hill Talau, near Neiafu, presents similar features. Here, too, a high margin surrounds a central depression, but the margin is complete for only three quarters of a ring. For the other quarter the level of the margin descends to that of the interior. The accompanying rough map of the summit of the hill shows Fig. 5.—Sketch of the contour of the summit of Talau ( Vavau Group), Tonga Islands. The numbers indicate heights, in feet, above the sea. the contour. From this it is seen that the bottom of the depression lies some 40 to 50 feet below the higher part of the margin, OF THE TONGA ISLANDS. 611 The summit of the other three-storied hill, Mo’unga Lafa, is, I believe, as the name implies, flat. The uniform covering of bush obstructs any general view of it; but in traversing it I was unable to find any depression corresponding to that in the summit of Talau. In the short visits that I paid to the Vavau Group, I landed on a few only of the islands, and had not the means of making satis- factory observations of contour. I have drawn attention to the features above described, with the hope that when the survey of the Tonga Islands which was begun by H.M.S. ‘ Egeria’ is continued, the heights and contours of the islands and terraces occurring at different levels in this remarkable group may receive special attention. The peculiar shape of the group, penetrated by long, narrow, and deep inlets of the sea, may perhaps be accounted for by supposing a much-denuded basis, probably of volcanic nature, whose sur- face was deeply scored by valleys previous to its subsidence. While the island was submerged the deposits which formed the limestone rock were laid down, and the lower layers would have already begun to accumulate during the downward movement. But, though the base of the limestone deposit may very probably have been formed during subsidence, the upper part and the contour of the land must be chiefly referred to the succeeding period, which has, in the main at least, been one of elevation. During intervals in the elevation, the terraces and plateaus which constitute so marked a feature have been formed, first on the highest and later on the lower peaks and ridges, while the intervening valleys still penetrate far among the islands and promontories. It is interesting that, both at Vavau and at Kua, the reefs which have been formed during intervals in the elevation of the islands have in some cases acquired atoll- or barrier-like contours. The larger islands of the Hapai Group lie in a line extending from Haano in the north to Uiha and Alefa in the south. They are indeed the elevated parts of a long reef, which, after extending in a south-westerly direction for some 30 geographical miles, sweeps round in an even curve to the west and north-west, and is finally lost among the reefs and islands of the Kotu Group. At two points in the northern part of the reef, gaps are indicated in the chart. The northern of these is bordered by reefs which are continued for a considerable distance in a westerly direction. To the south of the southern part of this reef there is an outer and more interrupted line of reef concentric with it. On the convex or weather side the margin of the reef is even, but on the opposite side the coral grows in irregular patches and the margin is ill-defined. The island of Lefuka may be taken as a type of the islands situated on the reef. It is about three quarters of a mile in breadth, and some three or four miles in length. From the beach on the western side there is a very gradual slope towards the eastern side of the island, which is about 20 feet above high water. Near the Q. J. G. 8. No. 188. 2uU 612 MR. J. J. LISTER ON THE GEOLOGY eastern side there is a rather rapid descent, terminating in a litile cliff some 10 feet high, whose base is much excavated by the waves. The rock of which the cliff is made consists of masses of coral of various sizes, cemented together into a conglomerate. Its structure, in fact, resembles that of a shore-platform. Extending from the foot of the cliff some 300 yards to seaward is the present shore-platform, on which, at low tide, there is about a foot of water near the shore, while the outer part lies bare. The cliff of raised rock extends along the whole of the eastern shore of the island. Some distance from the concave side of the long Hapai reef there are numerous shoals and small islands. The island of Fotuha, north of the Kotu Group, displays the abrupt undermined cliffs of a raised coral-island, and stands at an elevation of about 100 feet above the sea. The few soundings that are given in the chart show depths of 10 to 24 fathoms along the greater part of the western boundary of the reef, though 30 fathoms is marked opposite the northern break in the reef above alluded to, and 56 fathoms within its northern border. The Kotu Group consists of a number of small islands, from whose shores considerable tracts of reef extend at sea-level. | The only island of this group on which I landed was Hafeva, I found no raised coral-rock, but a loose sandy formation which attained a height of some 12 feet at the shore. The interior of the island is depressed and occupied by a marsh. There is a wide gap among the reefs on the western side, with a depth of some 8 fathoms, forming an anchorage which was at one time used by whalers. Of the other islands of the Kotu Group, three show the deeply undermined cliffs characteristic of raised coral-islands; but two of them, Kotu and Matuku, as already stated, recall, in their reddish- brown cliffs and peculiar outline, the tuff islands of Mango and Nomuka-iki of the southern part of the Hapai Group (see p. 598). It seems clear that the long reef of the Hapai Group must be regarded as a barrier-reef or an imperfect atoll. The shoals and islands situated within the concavity of the reef, both the northern ones and those of the Kotu Group, probably either rest on or are parts of the mass on whose eastern and southern slopes the reef has erown, and the volcaric character which I thought I recognized in two of the laiter offers some evidence as to the kind of formation of which the basis is composed. The outer line of reefs, lying some four miles to the south of the main one, and concentric with it, isa very remarkable feature. Its relation to that reef seems to imply that it is of an older date, and accords very well with the view that the reefs are slowly growing outwards on the slopes of a submerged mound. It must be ob- served, however, that in the northern part of the great reef, in the neighbourhood of the larger islands, there appears no reason for supposing that the reef is moving away from the centre of the group, for it is bordered by flourishing coral-reefs on the western as well as on the eastern side. OF THE TONGA ISLANDS. 613 The raised reef-platform of which the islands to the north are formed, as well as the raised coral-island Fotuha, and those of the Kotu Group, show that elevation has taken place, though only to a comparatively small extent, in the northern and western parts of the group. On the other hand, the absence of islands of any consider- able size from the southern parts of the reefs appears to indicate that this region was not affected by the elevation. The island of Nomukain the southern part of the Hapai Group is triangular in shape, the sides, which are about equal, measuring some 2 miles in length. ‘To the south lies the little island Nomuka- iki, separated by a channel a mile and a half wide, and 15 fathoms deep. A large shallow lagoon occupies a great part of the larger island ; it is less than a fathom deep, and its level does not rise and fall with the tide, while the water it contains is much denser than sea- water *. At the bottom of the lagoon there is a thick layer of black mud, which gives out a strong smell of decomposing matter when it is stirred. The land border is broad on the west and north- west of the lagoon, but narrow on the other sides. A ridge on the western side attains a height of over 160 feet at two points. On the east the land is much lower, varying from 15 to 96 feet above the sea, and on the south-west it is often hardly higher than the top of the beach. The formation consists of coral rock. Nomuka and Nomuka-iki are surrounded by fringing reefs, which in the case of the latter stretch out a considerable distance (in one part over a mile) from the shore. The twenty-fathom line encloses both islands ; on the eastern and north-western sides of Nomuka it is about half a mile from the edge of the reef, while on the west and south-west of Nomuka-iki it is a mile away and encloses some outlying coral patches. Outside the twenty-fathom line the slope is very gradual. Eight miles from land on the west and south-west (the only directions in which distant soundings have been taken) the depth is not greater than 50 fathoms f. Nomuka is evidently an atoll which has been formed in shallow water, though it is only in the last stage of its elevation that the ring has been completed ; and here the shallowness of tie surround- ing sea appears to exclude subsidence as a factor in its formation: for the summit of the elevation on which, on that hypothesis, the limestone formation rests would lie beneath the lagoon; but the floor of the lagoon is only some 9 fathoms higher than the bottom of the surrounding sea outside the reef. To the east of the N omuka Group, a reef dotted with islands is * On comparing the density of the lagoon-water with that of sea-water and distilled water by means of a hydrometer, the readings were :—- Distilled water ...........00.0ee- — 2 a a ee. ke ee +25 Lipfoon~Waber... cassie ccassonsss +39 + Opposite the western end of the channel between the islands I obtained a branch of a living madrepore at a depth of 31 fathoms, and a living fragment of one of the Poritide came up on the ‘ Egeria’s’ sounding-lead from the same depth. 202 614 MR. J. J. LISTER ON THE GEOLOGY marked on the chart, resembling on a small scale the great Hapai barrier to the north. Tongatabu is the largest island of the group. It is some 22 miles in greatest length, and of an irregular crescent shape, with its convexity presented to the south. Considerable tracts along the northern shore are not above the reach of the highest tides ; but the level gradually rises towards the convex border, and the highest part is at the south-east, opposite Hua. An extensive shallow area lies to the north of the island. Opposite the middle of the northern shore, and extending for about a third of the whole of its length, there is a large basin, which forms the harbour of Nukualofa. It has an average depth of 15 fathoms, and communicates with the outer water by a deep channel on the eastern side. A wide shallow area, having a depth of from 6 to 9 fathoms, extends from the western point of the island eastward, bounding the basin and eastern channel on the north. Along the outer or seaward margin of this area coral reefs grow up to the surface, forming a more or less continuous barrier on the east and north-west. To the north, however, the line of reefs is interrupted, and here shallow banks (though apparently without reefs) extend for some 15 miles in the direction of the Nomuka Group. Several islands, some of sand, some of coral rock elevated about 15 feet, are dotted along the reefs on the north of the harbour and eastern channel. The interior of Tongatabu is partly occupied by an irregular and shallow lagoon, which communicates with the sea on the north. Regarding the island and the shallow tract to the north as a whole, we find an area whose circumference, represented by the convex border of the island on the south, and the shallow 9-fathom area, on which reefs have grown, to the north, stands at a higher level than the central part, represented by the lagoon and the 15-fathom basin. In mentioning this island, Darwin pointed out* that “it re- sembles either an up-raised atoll, with one half originally imperfect, or one unequally elevated.” The gaps in the line of reef along the northern margin show that the atoll was imperfect: but the gradual rise of the land towards the convex side, and the fact that it attains its highest point where it most nearly approaches the greatly elevated island of Kua, appear to point strongly to the con- clusion that it has also been unequally elevated. The island is formed of coral limestone throughout, and in many places this rock lies at the surface. In the grassy roads of the town of Nukualofa one often meets with flat patches of coral, many feet in diameter, on which the lines of astreids or madrepores are seen radiating from a common centre. Elsewhere the rock is covered with a layer of reddish-brown clay, which in some cases attains many feet in thickness 7. The contour of Tongatabu is flat or presents gentle undulations, * ‘Coral Reefs,’ 3rd ed. p. 177. + A similar clay overlies the limestone formation of Vavau. 7 SE a OF THE TONGA ISLANDS. 615 but there are no streams or stream-channels on the surface. The rain soaks through the porous rock, and finds its way to the sea by underground channels excavated in the limestone. The caverns and subterranean galleries which haye been thus hollowed out by the water are in some places very extensive, and in walking about the island one is often reminded of their presence by the hollow sound of one’s footsteps. Some remarkable features are presented by the narrow reef fringing the southern and eastern shores of Tongatabu. The land gradually slopes towards the ocean, and ends abruptly in a low cliff of coral rock. From the foot of the cliff there extends towards the sea a narrow reef-platform, whose outer edge is raised, forming a barrier* several feet in height. This barrier is in the position of the mound formed of nullipores which is commonly found near the seaward edge of a reef-platformy. It presents a steep surface towards the platform, and a vertical one towards the sea. In many places there is on the platform a depth of five feet or more of water, whose level is above that of the sea, and which is prevented from flowing away by the barrier on the seaward side. The upper surface of the barrier is formed of a number of shallow basins—the upper ones several feet in width—whose rims are per- fectly level and which stand at different heights, the lower basins surrounding the upper, stage below stage, strongly recalling the structure which was presented by the well-known pink and white terraces of Lake Rotomahana in New Zealand. ‘The outer part of the barrier is penetrated by fissures which open above by rounded apertures left between the margins of the basins. The rollers, coming up from the open ocean, break against the steep face presented to them by the reef, and are thrown aloft in clouds of spray, while great jets and spouts of water are forced up through the fissures. The water falls into the basins on the barrier, streaming over their level edges along the whole of their circum- ference, to be caught by the basins beneath, and so down from ee to basin. The edges and outer sides of the basins are covered with Live nullipores, which grow in encrusting lamine, and flourish in the freshly aerated water which streams over them. Wherever the water flows, the nullipores grow; and thus the edges of the basins are kept level, and are constantly growing upwards and outwards. It appears probable that such a barrier is a form of the nullipore mound. Its unusual height and the remarkable basin-lke formation may be due to the steepness of the outer slope of the reef, causing the waves which break against it to be thrown upwards, and thus the water in which the nullipores flourish always falls on them from above. The reef on the eastern shore of Eua presents similar features. The little island of Kua-iki lying opposite the mouth of the eastern * This term is used without any reference to a barrier-reef. t+ The nullipore mound at Keeling Island is described in Darwin’s ‘ Coral Reefs,’ 3rd ed. pp. 13, 14. 616 MR. J. J. LISTER ON THE GEOLOGY channel is composed of a higher central area, separated by a rapid descent from the lower peripheral part. It has evidently been elevated in two stages. LV. ConcLusion. The view of the mode of formation of the Tonga atolls suggested in Darwin’s ‘ Coral Reefs,’ is that they grew during a period of sub- sidence, when the volcanoes of the group were inactive, and that elevation has occurred simultaneously with a recent volcanic out- break *. Dana also considers that the islands were formed during subsidence, and regards the elevation which has taken place as merely a local phenomenon 7. The additional information on the group which we now possess appears to make a different view of the formation of these islands more probable. It is clear that elevation has been in progress for a long period, and has been renewed again and again after intervals during which reefs of considerable extent have grown; it has to a greater or less extent affected all the larger divisions of the group, while at Vavau and Kua reef-structures have been raised to heights of over 500 and 1000 feet. Such an elevatory movement is quite in keeping with the pre- sence of the line of volcanoes which traverses the group, and I am not aware that there is any evidence that these have, as a whole, passed through a quiescent stage previous to their activity in recent times. With regard to atolls in general, one reason for considering them to have been formed during subsidence is the improbability of there being any other basis than a subsided island on which the reefs could have grown. In the Tonga Islands, however, the soundings of H.M.S. ‘ Egeria’ show that Tongatabu and Nomuka rest on shallow banks, and where the underlying basis of the islands is exposed, as in the Nomuka group and at Kua, it is seen that they consist of layers of volcanic materials ]aid out under water. The process now going on at Falcon Island illustrates the way ‘in which a volcanic mound, composed of soft material and originally standing above the sea, may be reduced to a submarine bank, and so form a basis for coral growth. In consideration of these facts it appears probable that the atoll- shaped islands of the southern part of the Tonga Group have grown on banks of volcanic material laid out in shallow water, and that there is no necessity to call in the hypothesis of subsidence to account for their formation. EXPLANATION OF PLATE XXIII. Map of the Tonga Islands. The map of Eua is prepared partly from the survey of H.M.S. ‘Egeria’ (1888). The outlines of the patches of volcanic formation in the central part of the island are not accurately drawn. * ‘Coral Reefs,’ 3rd ed. pp. 177, 186, 188. + ‘Origin of Coral Reefs and Islands,’ Am. Journ. Sci. ser. 3, vol. xxx. (1885) p. 99. uel is pia Sh — Suri uid y RueYY ——.22 a augsouny foe (1 Inw vonvnvany 3 — | VINO[— Pe | Ween — dV HIlSyiS — 69€ SL 101 - | ¥0% 962! €s : ps2" a ul 12 a 2 dl qnnHo wid an ool 28i y Z ‘ “ 2 ¢ \ , ’ * * a f ' ‘ Po A a hd ’ * ae i * - al o> SS oe | a aes ie — TONGA or FRIENDLY I$ — @) Wriilizales Volean0es Jalan \ About 500 Feet high Vavau {<== Mounga Lofa= ce) “Tee blands igs : — ” — Lamnestone Form Se Bei? fede ae 18.90 Feet high 19° —— - Metis | 15) Feet high . i) Kao 1? : 2235 ~ 30308 high sé l) ToFua 1° oe 7890 Feet highe 4 a Fouhe "= 2 Kotu I cee S) 2225 hold Qe Hialleva 6 20° oe ae ° S ae f x 252 = ae z Q x FaLcon 1° _ [021 | 1881. Dr. R..H. Traquair. 1853. Professor L. G. de Koninck. | 1882. Dr. G. J. Hinde. 1854. Dr. 8. P. Woodward. 1883. Mr. John Milne. 1855. Drs. G. and F. Sandberger. | 1884. Mr. E. Tulley Newton. 1856. Professor G. P. Deshayes. 1885. Dr. Charles Callaway. 1857. Dr. S. P. Woodward. 1886. Mr. J. S. Gardner. 1858. Mr. James Hall. 1887. Mr. B. N. Peach. 1859. Mr. Charles Peach. 1888. Mr. J. Horne. Professor T. Rupert Jones. | 1889. Mr. A. Smith Woodward. 1860. oe W. K. Parker. 1890. Mr, W. A. E. Ussher. 1861. Professor A. Daubrée. 1891, Mr. R. Lydekker. ANNUAL REPORT, 29 AWARDS OF THE MURCHISON MEDAL AND OF THE PROCEEDS OF “THE MURCHISON GEOLOGICAL FUND,” ESTABLISHED UNDER THE WILL OF THE LATE SIR RODERICK IMPEY MURCHISON, Barr., F.R.S., F.G.S. “To be applied in every consecutive year in such manner as the Council of the Society may deem most useful in advancing geological science, whether by granting sums of money to travellers in pursuit of know- ledge, to authors of memoirs, or to persons actually employed in any inquiries bearing upon the science of geology, or in rewarding any such travellers, authors, or other persons, and the Medal to be given to some person to whom such Council shall grant any sum of money or recompense in respect of geological science.” 1873. 1873. 1874. 1874. 1874. 1875. 1875. 1876. 1878. 1877. 1877. 1878. 1878. 1879. 1879. 1880. 1881. 1881. 1882. 1882. 1883. Mr. William Davies. Medal. Professor Oswald Heer. Dr. J. J. Bigsby. Medal. Mr. Alfred Bell. Professor Ralph Tate. Mr. W. J. Henwood. Medal. Professor H. G. Seeley. Mr. A. R. C.. Selwyn. Medal. Dr. James Croll. Rey. W. B. Clarke. Medal. Professor J. F, Blake. Dr. H. B. Geinitz. Medal. Professor C, Lapworth. Professor F. M‘Coy. Medal. Mr. J. W. Kirkby. Mr. R. Etheridge. Medal. Professor A.Geikie. Medal. Mr. F. Rutley. Professor J.Gosselet. Medal. Professor T. Rupert Jones. Professor H. R. Gdéppert. Medal, 1883. Mr. John Young. 1884. Dr. H. Woodward. Medal. 1884, Mr. Martin Simpson. 1885. Dr. Ferdinand Romer. * Medal. 1885. Mr. Horace B. Woodward. 1886. Mr. W. Whitaker. Medal. 1886, Mr. Clement Reid. 1887. Rev. P. B. Brodie. Medal. 1887. Mr. Robert Kidston. 1888, Professor J. 8, Newberry. Medal. 1888. Mr. E, Wilson. 1889. Professor James Geikie. Medal. 1889. Mr. Grenville A. J. Cole. 1890. Professor Edward Hull. Medal. 1890. Mr. E. Wethered. 1891. Professor W. C. Brégger. Medal. 1891. Rey. R. Baron. 30 PROCEEDINGS OF THE GEOLOGICAL SOCIETY, AWARDS OF THE LYELL MEDAL AND OF THE PROCEEDS OF THE “LYELL GEOLOGICAL FUND,” ESTABLISHED UNDER THE WILL AND CODICIL OF THE LATE SIR CHARLES LYELL, Barr., F.R.S., F.G.S. The Medal ‘‘ to be given annually” (or from time to time) “as a mark of honorary distinction as an expression on the part of the governing body of the Society that the Medallist (who may be of any country or either sex) has deserved well of the Science,”— not less than one third of the annual interest [of the fund] to accompany the Medal, the remaining interest to be given in one or more portions at the discretion of the Council for the encouragement of Geology or of any of the allied sciences by which they shall consider Geology to have been most materially advanced, either for travelling expenses or for a memoir or paper published, or in progress, and without refer- ence to the sex or nationality of the author, or the language in which any such memoir or paper shall be written.” 1876. Professor John Morris. | 1884. Professor Charles Lapworth. ' Medal. 1885. Professor H. G. Seeley. 1877. Dr. James Hector. Medal. Medal. Lea: 1878. Mr. W. Pengelly. Mr. G. Busk. Medal. 1885, 1886. Mr. A. J. Jukes-Browne. Mr. W. Pengelly. Medal. 1878. Dr. W. Waagen. 1886. Mr. D. Mackintosh. 1879. Professor Edmond Hébert. | 1887. Mr. Samuel Allport. Medal. Medal. 1887. Rev. Osmond Fisher. 1879. Professor H. A. Nicholson. | 1888. Professor H. A. Nicholson. 1879. Dr. Henry Woodward. Medal. 1880. 1880. Mr. John Evans. Medal. 1888. Mr. A. H. Foord. Professor F. A. von Quen- | 1888. Mr. T. Roberts. stedt. 1889, Professor W. Boyd Dawkins. 1881. Sir J. W. Dawson. Medal. Medal. . 1881. Dr. Anton Fritsch. 1889. M. Louis Dollo. 1881. Mr. G. R. Vine. 1890. Professor T. Rupert Jones. 1882. Dr. J. Lycett. Medal. Medal. 1882. Rev. Norman Glass. 1890. Mr. C. Davies Sherborn. 1882. Professor C. Lapworth. 1891. Professor T. McKenny 1883, Dr. W. B. Carpenter. Medal. Hughes. Medal. 1883. 1883. 1884, Mr. P. H. Carpenter. M. E. Rigaux. Dr. Joseph Leidy, Medal. 1891. 1891. Dr. C. J. Forsyth-Major. Mr. G. W. Lamplugh. ANNUAL REPORT, 31 AWARDS OF THE BIGSBY MEDAL, FOUNDED BY Dr. J. J. BIGSBY, F.R.S., F.G.S. To be awarded biennially “as an acknowledgment of eminent services in any department of Geology, irrespective of the receiver’s country ; but he must not be older than 45 years at his last birthday, thus probably not too old for further work, and not too young to have done much,” 1877. Professor O. C. Marsh. 1885. Professor Alphonse Renard. 1879. Professor E. D. Cope. 1887. Professor Charles Lapworth. 1881. Dr. C. Barrois. 1889. Mr. J. J. Harris Teall. 1883. Dr. Henry Hicks. 1891. Dr. G. M. Dawson. AWARDS OF THE PROCEEDS OF THE BARLOW- JAMESON FUND, ESTABLISHED UNDER THE WILL OF THE LATE Dr. H. C. BARLOW, FE.GS. ‘‘The perpetual interest to be applied every two or three years, as may be approved by the Council, to or for the advancement of Geological Science.” 1880. Purchase of microscope. 1884. Professor Leo Lesquereux. 1881. Purchase of microscope lamps. | 1886. Dr. H. J. Johnston-Layis. 1882. Baron C, von Ettingshausen, | 1888. Museum. 1884, Dr. James Croll. 1890. Mr. W. Jerome Harisron. VOL, XLVII. c 32 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. Estimates for INOOME EXPECTED. THOMAS WILTSHIRE, Treas. 4th February, 1891. fae dh £8 Compositions scm. selene iy keels iix'e n'0'e ay Binge io teinyelee en 189 0 0 Due for Arrears of Admission-fees .......... 9410 0 Admission-fees, 1891 ......... agains veers eee ; 252 0° 0 . 346 10 0 Due for Arrears of Annual Contributions ..........++000. 100 0 0 Annual Contributions, 1891, from Resident Fellows, and Non-residents, 1509 to 186) ss. sows euin cis oa wie aR 660 0 O Annual Contributions in advances c's s/s are ancares « ales nee 35 0 0 Dividends on Consolidated 22 per Cents. .........0.0006 . IOP Dividends on London and North-Western Railway 4 per cent. Consolidated Preference Stock .........+sseeee0 87 15 0 Dividends on London and South-Western Railway 4 per ) cent. Consolidated Preference Stock ........eeeiseeees 78 0 0 Sale of Quarterly Journal, including Longman’s Ego) Mes 4H an tps ace rua Cpe cca eis coe © 165 0 0 Sale of Geological Map, including Stanford’s SMECCOUME totes ave gahigaobinenvgamstenie aca 48 0 0 Charwoman and Occasional Assistance......... 2 0 0 Attendants at, Meetings ...i:.cceesevescosesvevseee oS. 8: 0 ee ninteanti EGR Sooo coc asece cance iw enve 10 10 O Official Expenditure : PIORUIONEEY a varns (Wee vdee tank ced es pode e sewers stokes 25 0 0 Biscallancous Printing ~ .,.<-.ivcqccesso dues Gots oe a7 1a 6 Investment in £2000 L. & N. W. Rail- way 4 per cent. Consolidated Pref. 7 eer EA LE BS | hae ae eng 2575 13 0 Investment in £250 L. & N. W. Railway 4 per cent. Consolidated Pref. Stock, COAT aiid oh ins in eld ae ea 322 17 6 Investment in £2000 L. G8. W. Railway 4 per cent. Consolidated Pref. Stock, OEE, Sits ices Cec aan ee ete 2563 2 3 Balance in Bankers’ hands, 31 Dec. 1890..415 15 3 Balance in Clerk’s hands, 31 Dec. 1890 .. 18 2 38 ——— 433 17 6 THOMAS WILTSHIRE, Treasurer. £8422 18 11 2 9 19¢ mooeo iio PROCEEDINGS OF THE GEOLOGICAL SOCIETY. i Ol 19s 36 9 ¢ Ics Pp 8 ST eoeeoeteeoeeeeeaeveeveeeeeeeeeveeseeeeveeee eee eee F070 poywp : -Ifostoy *yueo aed zg ul poJsoAUL puny oy} UO spueprArq "'* QG8T ‘toqureveq Tg ‘stoyueg ye oouRTeg |Z YL ye criti tt ttt ‘O6ST ‘K7enue I ‘sloyurg 7B couveg ‘SINGWAV oe ‘SLIGO “ENQOOOY dsayy, ,,CNOq NOSINveP-MOTUVG ,, 0 TL OGLF a 10R8T ‘reqmieoe(T Tg ‘sioyueg ye souryeq ‘aha Miccter arin aera reese ss TepaHy JO 980) 0 ZL QQ Tt teen ene eens eee e seen eee soOag qUE0 Byes eo SOU Tare Cy nel INt zed $e ueqtodosey Ul pojsoAUL pun,f 94} UO spuoeprlAT Tepe WTA ‘souor yrodny ‘7, ‘Jorg oy prway}/O 6 TG “lilt ttt OBeyT ‘Arenuere T ‘sieyueg 4v souRlTeg ‘SINGWAV ease *‘SLIIGO ay “ENNOOOY LSNUy, ~~ANOWY TVOINOTOUL) TIGA'T ,, 0 OL 8&5 oo 0GSt ‘req uLede(T Te {stoyueg 4e soured = Bente Trae gO SIP PeNT 39.1800 0 07 68 seeeeeessooqG omnqueqeg *ju00 red F ABATE ULE4Se AA gs press te 2: peTOUye AN OL LH “YON PUB Vopuor] UL peysoaut pung oy} WO SpueprAI sees Tepe UPA WMH A Jorg OF premy|O OL 6T “tt OGST ‘Arena T “stoxuvg ye couypeg ‘SINGNWAY q sp. age ‘SLAIGOMY “ENNOOOY LSOayT, _~aNOW TVOINOTOUL) NOSIHOUN] ,, ZL OL 19% rrtrese* Oger faqureseq [Eg ‘sreyueg 4v courpeq Rr: oes Lossy “GWM AW OF peAY |F T 62 sip char age a so a aan eo ee phere ors Gan mn “ce * =" “TOSUIENT TD AN -I[ostog ‘quedo wed $% ul poysoaul pun oy} UO spuoplalq jolg 0} popieae [epeyT ploy Surytys jo yon le @ BS Sl OBR ‘Arenuve | ‘sioyuvg 4v couvpeg "‘SINANWAV ie Spee ‘SLdI MO GY “INNOOOY Lsnuy, ,,°;adNO WY NOILVNOG NOLSVTIO(M ,, "L681 ‘Aumnune 17% ‘spa4J, ‘TMIHSLIIM SVNOHL ~ ~ a a en en ee — 8 FL S096F 8 FL S096F — [‘sworgnonqng pjosun fo = 7 oe yoo9s pun ‘ainjruoing ‘hiolguy *8Worg99]]0) 94? / fo anjDna ay2 apnjouL ZOU Saop an0gD IY, L— IN| 0 0 OOL ****(po08 palepisu0d) sMOTINGIIyWOLY [wNUUIY JO stvolTV O™oT re 7 peo PoeLepIsuod) seoj-UOIssIUpY Jo slvoIy 0 0 00S 0 0 000G “ZL FP YFG Jord PayBPLLOsAoy "FU99 rad 7 AVA TIBI W.19489 AA -YINOG 2 WOPUO'T 0 0 88S 0 O O8ZS ‘9648 1904S For PovwpHOsaoy “Fue rf . rad F ABMIVyY W.10}S9 AA -Y WO NY WOpuo'T gi OL OL 089s G-G- Sele: aor rod $G Poyepyostto(y : ps §F —; syrodotg popun & 62.8 Bp 2 40 8 ie ee ‘spuvy S101) Ul souvpeg 4 e- GT qTp i ORL OL 1S et eee < yh Hees seseeeeseeeees Gepy JO JUNODNB UO PLOFUVIG Worf ON(T 5 OI OL Z¢ EO oe aa ae ae a ee ‘OW ‘TA[X Tae yee eS Sa ‘£40t00G oY} JO INOAVJ UT oOUBleE TOA ‘TeuINoL Jo JANOINV wo “OD WW UVULOMOT WorF ON(T he eae BS ge ASS ‘ALUTAOUT | ‘OBST ‘taquiso9qq STE f ALUTAOUd § ALAIOOG IHL LO NOILVATV A ian 1 SL 6% L S&L 6F o bua 6.Ore 6 Ple OO owe awe OhM Oe ere. iN) lee ee eee see Se yo0}9 pozep -josuog *yue0 red £Z ul poysoaur puny oy} WO SpueprAt( Veseeeveeeceseress Qaor Kienuer [ ‘Sioyuvg 4B vouvpye "'SLAIGLO DY 0 cL 9 ay) ON, Ny mae en moiaee eet io | ‘aqumeoeg] Te ‘sleyuvg 7w oourreg | 4 9 F 7 i es 2 ‘SINAWAVG p 8 F ‘LNQOOOY isaUy, ,,,aNoOgq AMSOI_ ,, ie — eit 38 PROCEEDINGS OF THE GEOLOGICAL SOCIETY, AWARD oF THE Wotztaston MEDAL. In presenting the Wollaston Medal to Prof. J. W. Jupp, F.R.S., the PresrpEent addressed him as follows :— Professor J uDD,— The Council have awarded to you the Wollaston Medal in recog- nition of the important services rendered by you to Geological science, especially in the department of Petrography. In recalling for a moment the value and extent of these services, I am reminded that, after showing your powers by an excellent paper on the strata of the Lincolnshire Wolds, you began your geological career in the Geological Survey under Murchison, and that you had thus a favourable opportunity of acquiring that practical acquaintance with the details of geological structure which can in no way be so thoroughly mastered as by actual patient mapping. Your volume on the “Geology of Rutland” proved how’ well you had profited by the advantages which your official duties afforded you. From the Jurassic rocks of England, which you had studied in minute detail, you were led to undertake the investigation of those of Scotland, which you succeeded in reducing to order, bringing them into closer relationship with their equivalents in the southern part of the United Kingdom. Tt was in the course of those northern expeditions that you were drawn from the field of stratigraphy into the study of volcanic rocks, to which you have since devoted so large a part of your time and thought, and in the study of which you have journeyed far and wide in this country, and have extended your travels to the islands of the Mediterranean. The problems presented by these rocks in the field led you to seek the aid of the microscope, and to enter upon a course of distinguished petrographical research. I trust that the award of this Medal will be received by you as a mark of the estimation in which your work is held by the Society in whose Quarterly Journal most of it has been published. Prof. Jupp, in reply, said :— Mr. PresipENT,— It is a source of legitimate gratification to the student of science when a favourable judgment on his efforts is pronounced by his contemporaries and fellow-workers. In receiving this highly-prized ANNIVERSARY MEETING—-WOLLASTON DONATION FUND. 39 mark of your approval, I would fain forget for one moment, if that were possible, how far the work—of which you have spoken in such graceful terms—falls in amount below my hopeful anticipations of the past, how it fails to reach the standard of excellence of my cherished ideals. Any value which that work may be found to possess is undoubtedly due, in great part, to the fostering care of the Society which to-day so generously crowns my labours. To the Geological Society, in its corporate capacity, I am indebted for the reception and publication of the results of my studies ; to individuals composing that Society I owe more than I can ever express, for kind sympathy, warm encouragement, and friendly aid; and to both Council and members I shall always be deeply grateful alike for helpful suggestion and discriminating criticism. AWARD OF THE WoLLAston Donation Funp. The Presipenr then presented the Balance of the Wollaston Fund to Ricnarp Lypexxer, Esq., B.A., F.G.8., saying :— Mr. Lypexxer,— The Council has awarded to you the proceeds of the Wollaston Donation Fund in recognition of the value of your numerous con- tributions to Vertebrate Paleontology. We trust that you will continue these investigations and that, whether they appear in the publications of this Society or elsewhere, the results, like those which have preceded them, may tend to the steady advancement of our favourite science. | Mr. LypexKKer, in reply, said :— Mr. PResiDENT,— The particular branch of Palaeontology to which my own studies have been more especially directed is one which, from its nature, is so beset with difficulties that it is very apt to lead to misgivings as to whether any real good results from its pursuit. The assurance conveyed by the honour that the Council of the Society has con- ferred upon me, that such work as I have been able to do is not unappreciated, is therefore very gratifying. Although circumstances have rendered it almost imperative that I VOL. XLVII, d 40 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. should devote my time to literary work rather than to original scientific research, yet I still hope to do something in the latter field. Please accept, Sir, on the behalf of the Council, my thanks for the mark of distinction that they have bestowed upon me. AWARD oF THE Murcuison MepAt. In handing the Murchison Medal, awarded to Professor W. C. Broceer, of Christiania, to J. J. H. Tear, Esq., M.A., F.RB.S., for transmission to the recipient, the Presipent spoke as follows :— Mr. TEeatt,— The Council has awarded the Murchison Medal to Professor W. C. Brogger, of Christiania, and in asking you to transmit it to him T will request you also to convey to him an expression of the high estimation in which we hold his researches among the older rocks of Scandinavia. He is remarkable among the geologists of Europe for the great range of his acquirements. If we were to read only his descriptions of the Silurian fauna of Southern Norway we should, doubtless, believe him to be essentially a paleontologist. If we looked over his maps and sections of the Christiania district, we should think of him rather as an admirable stratigrapher and car- tographer. If, again, we began with his account of the eruptive rocks and their zone of contact-metamorphism, we should conclude that his chief studies must have lain in microscopic and chemical petrography, of which he is so accomplished a master. Or, lastly, if we knew him only by such essays as his late paper on garnets, we should regard him as preeminently a mineralogist, gifted with rare originality. He has swept a full chord on the geological lyre, and every note sounds rich and true. It gives me personally an especial pleasure to be the intermediary in conveying the award of the Council, for I have had the advantage of being conducted by Professor Brégger over some of his classic ground around Christiania, and I know from my own experience how accurate and exhaustive is the work; how courteous, genial, and helpful the man. He will, I trust, receive this Medal, bearing the likeness and the name of one of the great masters of British Geology, who was also a pioneer in the geology of Norway, as a pledge of our esteem and sympathy with him in the great work he ANNIVERSARY MEETING——-MURCIIISON GEOLOGICAL FUND. 40 has already accomplished, and in the long and brilliant career which we hope is still in store for him. Mr, Tuatt, in reply, read the following communication received by him from Prof. Briceur :— “T beg to express my hearty gratitude for the great and com- pletely unexpected honour conferred upon me by the Council of the Geological Society in the award of the Murchison Medal. “The Founder of this medal, almost half a century ago, classified the Silurian rocks of the Christiania district, and pointed out their relations to the corresponding strata of Great Britain ; so that, if the subsequent investigations of Norwegian geologists have furnished results of interest to the students of British Geology, this is only a slight repayment of an old debt. “‘Tn ancient times the mountain-ranges of northern Great Britain and Norway were probably connected, and in the Quaternary period the Scandinavian ice-sheet stretched across to England and deposited boulders of Norwegian rocks, some of which were derived from the Christiania district. Now, in recent times science has rebuilt the bridges which formerly connected the two countries, inhabited by closely related peoples of the Germanic race. ‘“*Tt will be an object of especial interest to me to contribute, as far as I am able, to the reconstruction of bonds of union between Great Britain and Norway, in grateful remembrance of the benefits which Norwegian geologists in general, and I myself in particular, have derived from the celebrated Geological Society of London, ‘¢ Allow me, in conclusion, to express the great satisfaction I feel at receiving this honour during the Presidency of so eminent a geologist as Dr. Archibald Geikie, who is personally acquainted with the geology of Norway.” AWARD OF THE Murcuison GEoLocIcoaL Funp, The Prestpent then handed the Balance of the Murchison Geological Fund (awarded to the Rev. Ricwarp Baron, F.LS., F.G.S., of Antananarivo) to Wm. Tortry, Esq., F.R.S., for trans- mission to the recipient, saying :— Mr. TorLey,— I have to request you to transmit to the Rev. R. Baron the Balance d2 42 . PROCEEDINGS OF THE GEOLOGICAL SOCIETY. of the proceeds of the Murchison Geological Fund, in testimony of the interest taken by the Council in the geological work which, amid so many discouragements, he is carrying on in Madagascar. We desire him to accept this Award as a mark of our hearty sympathy and of our wish to aid him in his researches. Mr. Torrey, in reply, said :— Mr. PresipEnt,— On behalf of Mr. Baron, who is now in Madagascar, I beg to thank the Council and yourself for the honour conferred upon him in the award of the Murchison Fund. As a Missionary in an area as yet but little known, Mr. Baron has exceptional opportunities for original research, and that he has not neglected those opportunities is evident from his papers already read to this Society and to the Linnean Society. The Award now made will, I am sure, be an incentive to further work in a most promising field of research. AWARD OF THE LyEtt MEDAL. In presenting the Lyell Medal to Prof. T. McKenyy Hveuss, F.R.S., the Presmpent addressed him as follows :— Professor Huenrs,— The Lyell Medal has this vear been adjudged by the Council to you in appreciation of the value of your investigations in various departments of Geology, especially among the older rocks. Your researches in Caernarvonshire and Anglesey formed the starting- point of those later enquiries which have done so much to clear up the earlier chapters of the geological history of Wales. You have not confined yourself, however, to the rocks of any one system or period, but have ranged freely from Archean gneiss to raised beach, hovering for a moment here and resting a little there, generally critical, almost always suggestive, and with that happy faculty of enthusiasm which, reacting on younger minds, “allures to older worlds, and leads the way.” As I place this Medal in your hands I cannot but recall the days of our early friendship, now faded so far into the dim past of life, when, as colleagues in the Geological Survey, we used to attend the meetings of this Society in Somerset House, taking seats on a back — ANNIVERSARY MEETING—~LYELL MEDAL. 43 row and gazing down upon the magnates of the science seated beneath. Little did either of us dream that the whirligig of time would eventually place us where we find ourselves to-day. It is thus no small gratification to me to be called upon to present to you this Medal, which will not only serve to mark the Society’s appre- ciation of your work, but which will connect you by another link with the memory of our friend and master, Lyell. Prof. Hueres, in reply, said :— Mr. PResipENtT,— I feel that I have, as the senior, been selected to receive this high recognition of the work being carried on by the Cambridge School of Geology. IJ have not myself been able to offer much to the Society of late, save occasional criticism, but my colleagues, Mr. Marr and Mr. Harker, Fellows of the Society, whose opinions are regarded each year with increasing respect, the one your Secretary, the other on your Council, have from time to time contributed valuable papers, while my other colleague, Mr. Roberts, has also laid before the Society the results of important original observations made by him. The Society knows that it is chiefly to the lecture-room, the museum, and the field-classes that it must look for men to carry on its work in the future. But I must acknowledge in this respect also, that the heaviest work has fallen upon my colleagues. ‘They know, however, that in the administration of the Department, and directly and indirectly in promoting the cause of Science, I help as far as I can. We all work well together, and I feel that they will rejoice with me now, will help to carry back the Lyell Medal in triumph to Cambridge, and will join with me in offering to the Society our warmest thanks for the honour that has been done us. We shall regard it as a stimulus to follow in the steps of the great teacher whose name is commemorated on the Medal, and try always to distinguish clearly between what is proved, what is disproved, and what remains, however plausible, ‘‘not proven.” I am glad that it has fallen to my lot to receive this honour from the hands of an old and valued friend, upon whom has fallen the mantle of Lyell, a mantle in which the warp of science and the weft of literature are so deftly interwoven. 44 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. AWARD OF THE Lypitt GroLocIcaL Fonp. | The Prestprnt then handed one half of the Balance of the Lyell Geological Fund, awarded to Dr. C. J. Forsyrn-Maysor, of Florence, to Dr. H. Woopwarp, F.R.S., for transmission to the recipient, and addressed him as follows :— Dr. Woopwarp,— In requesting you to transmit to Dr. Forsyth-Major one moiety of the Balance of the Lyell Geological Fund, I wish to express the Council’s appreciation of his researches and its hope that he will continue them. He has done much to increase our knowledge of the Pliocene Mammalia of the Val d’Arno, and he has recently extended his explorations among the younger Tertiary deposits of the Eastern Mediterranean. Dr. Woopwarp, in reply, said :— Mr. PresipEnt,— On behalf of Dr. C. J. Forsyth-Major, I have to acknowledge the honour conferred upon him by the Council of this Society in awarding him a moiety of the Lyell Fund. The work to which Dr. Forsyth- Major has devoted his life so entirely accords with the researches and labours of Sir Charles Lyell that I cannot’ doubt the appro- priateness of this Award. Dr. Forsyth-Major has devoted many years to the elucidation of the Pleistocene and Pliocene mammalian faunas of the Val d’Arno and Northern Italy, and his numerous memoirs attest the value and accuracy of his work. Lately he has devoted two years to the exploration of the Pliocene fauna of the Island of Samos, and has — obtained thence two very important collections (at present only partially examined)—one now in the Geneva Museum, the other in the British Museum (Natural History), Cromwell Road. Among these are a large number of forms specifically identical’ with the mammals from the equivalent deposits of Pikermi in Attica, Baltavar in Hungary, and Maragha in Persia ; and also several new types of much interest as showing a former wider distribution for existing forms. It is Dr. Forsyth-Major’s hope to spend the early summer months in London, to complete his descriptions of these fossil remains, which your Award will doubtless assist him in doing. 4 ba ANNIVERSARY MEETING—LYELL GEOLOGICAL FUND. 45 He writes as follows :— “ Would you kindly transmit to the President and Council of the Geological Society my grateful acknowledgments of the honour conferred upon me, which I value so much the more as coming from a scientific body of my own country, to which, owing to the fact that my family resides abroad, I have become nearly a stranger. ‘“‘ If I rightly understand the intention of the Council, this Award is given less as a mark of their approval of what I have already done than as an incentive to future labours. “In my paleontological work I have striven to follow the example of one of the masters of our science, the late Dr. Hugh Falconer, devoting myself more to the collecting of facts and observations than to their speedy publication. This reserve seems to be imposed upon us even more in our day than in that of Dr. Falconer’s.” In presenting the other half of the Balance of the Lyell Geological Fund to G. W. Lametveu, Esq., F.G.8., the Preszrpenr addressed him as follows :— Mr. Lampitvuex,— The Council, in awarding to you one half of the proceeds of the Lyell Geological Fund, desires to assure you of the estimation in which it holds your work, and of the pleasure it will derive from their further prosecution. Your researches among the Glacial deposits of Yorkshire have been followed with much interest, and we have rejoiced in the enthusiasm which not only carried you through these labours at home, but which impelled you to seek the solution of some of your difficulties by journeying to the far-distant shores of British Columbia. Your investigation of the Speeton | Clay affords a striking example of how our knowledge may be cor- rected and extended by the patient labours of an observer resident on the spot which he has to examine. I hope you will accept this — Award with the best wishes of the Council and of the Society. Mr. Lamptvues, in reply, said :— Mr. PREsIDENT,— That I should have been selected by the Council to receive this Award affords me the greatest encouragement, since it comes to me as a token that my geological work, in spite of its narrow and local character, has after all a certain value. It is scarcely possible that anyone who has any sympathy what- 46 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. ever with nature should spend much time on the Yorkshire coast without becoming more or less of a geologist, and for my own part I drifted almost unconsciously into these studies in my boyhood, and have ever since found therein my happiest and healthiest recreation. My pleasure in them is now redoubled by this proof that the time so happily spent has also been spent usefully. I thank you, and hope that, as a coastguard in the service of science, I may still occasionally be able to send to headquarters reports which may contain some items of interest. AWARD OF THE BiesBy MEDAL. The Presrpent then handed the Bigsby Medal, awarded to Dr. Gro. M. Dawson, F.G.S8., of Ottawa, to Dr. Hicks, F.R.S., for transmission to the recipient, and addressed him as follows :— Dr. Hicxs,— In asking you to transmit the Bigsby Medal to Dr. George M. Dawson, I request you to convey to him at the same time an assurance of how fully the Council appreciates the value of his researches into the geological structure of Canada, and how cordially we hope that he may live long to prosecute the explorations which have shed so much lustre on the Geological Survey of his native country. Dr. Hicxs, in reply, read the following communication, received by him from Dr. Dawson :— “Mr, PrestIpENT,— ‘