PALAONTOGRAPHICAL SOCIETY. VOL. XXi. PLANTS OF THE CARBONIFEROUS STRATA. Part I. | Paces 1—32; Pirates I—VI. | | SUPPLEMENT TO THE FOSSIL CORALS. Pore lV, No; 2. LIASSIC. Paces 45—73; Puates XII—XVII. CRETACEOUS: ~ ECHINODERMATA. Met. Se Parr 11. Pacers 65—112; PuateEs IX. X, XLI—XXI, XXIa, XXIz. FIS ERES OLD RED SANDSTONE. Parr I, Paces 1—33; Prarres I—V. PLEISTOCENE MAMMALIA. Parr II. Paces 29—124; Prates VI—XIX. Issurp ror 1807. California Academy of Sciences ) Presented by Paleontographical Society. December , 1906. kKeeldd Ft. PRI he “ . «4 i . c A ’ , t ‘ . ~ ‘ A . i ° . 1 —— ay _ . = P . a dl ‘) » 4 - — is 7 . a is + 4 - . , x « ° 3 ¥ } ‘ ’ . 4 ¥ ‘ n ! TOL CM Ee eee . a. i Pd PA — - o - , ; * " 7 = ¢ . ¢ & = » ws } b= | a Diet ie ® . $00. 1968.3." 664 : PTOGRAPHICAL' So} A rv / . _ — ver uw * a ~ ‘ — PS December, 1861. :. 1860 i - B May, 1863. » 1861 » 2 y May, 1863. » 1862 e ss PB August, 1864. » 1863 55 > ‘ June, 1865. » 1864 a of F April, 1866. » 1865 b i < December, 1866. » 1866 : - Py June, 1867. » 1867 »” »” »” June, 1868. S “CO8T-LE8T Pee *uorzvortqnd jo asanoo ut ou st guemelddng y ¥ ' *saquq JO SUOIJOOIIOD OMT, 2 ‘quamefddng £ { €9T 61 G 69 9. G 88 61 LI 1S 69 96 61S 6¢ 6€ FOE 8S 6 Gh g 8éT TIS 66 6L 61§ Lb 87 €8& 06 F6L L&S ST 6LYP 9F8 0s SPL 966 06 T9L 81g €& Esher 169 1g VIS 18¢ 1G Sst TSST (ai LST 606T 6S O9T Sc8T GP og ce GG FIL €0L 16 261 8ST g LG 9ZT L 9g €&6 9 T og 6 i OZ D 86 68s oS 66 81Z 61 Y60T VoL PV PV PPE i S6L T¥9 Ze 16 LLY LG 5618 008 6h &V T1zé V G GE 9 ae ‘snopoom, Jo| . 7 3 ac poop | tenant | “Seno a sawadg jo'oyy| PPUSPOUNT | sorera Jo ‘on “TITA gv EN TA 168‘E | 006‘9T | 916 ST0'9 99T “ydersouoyy youa ut | ssaad.taqyary Jo | sabeg Jo ‘ON A “WAI JOU SOW ye “pouoyaaa Tuo satads ysiqIg 2 *aqo,dutoo Hee TREO T. 898T ‘998T cost €98T ‘T9ST POST ‘TO8T ‘6S8T ‘QE8T ‘LS8T ‘Se8T “SS8T FOBT ‘TYST “6S8T ‘TS8T 6S8T ‘OS8T ‘6FST 898T SST ‘OS8T LEST ‘SSST ‘ESsT LOST ‘998T ‘S98T : S981 CCST ‘SST ‘OS8T FOSL ‘TOST T98T ‘LEST ‘SS8T ‘ZEST ‘6FST TOST ‘LEST ‘SS8T ‘TS8T LEST ‘SFST LOST “99ST ‘S98T ‘“FOST €98T ‘TOST ‘TOST “6S8T ‘8o8T GSST “SS8T ‘ZeST ‘SST E98T ‘SST LOST ‘998T ‘S98T ‘FO8T €98T OS8T LS8T. 998T TOST “SS8T ‘TEST 89ST ‘F98T 998T ‘98ST TOSL ‘6S8T ‘8S8T ‘ZEST GS8L 6S81 898T ‘L98T ‘998T Sgst “ES8T “ESST ‘TS8T ‘OSST 998T 898T *poystqud sn. Ydersouoyzy 2113 YM UI SILIK IYy Jo sayeq. UAT ‘soroads Mou yo soup ayy Surystyquqso IOJ ynyas n i *pouoxpat ATO satoeds ysiqtag ¥ “BOSNT[O]L BBQ) 0} OJON 2 SI YOIYA ‘UBIO 9y} SUTB}UOD p 898 ‘99ST LOST “POST C98 €98T | ZOST ‘T9ST O98T ‘6S8T POST “O9ST “6S8T GOST “8E8T ‘LEST ‘LEST OS8T “PSST “ES8T |9S8T ‘SesT ‘PSST “ESsT FIST ‘OIST ‘GEST ‘“TE8I | OST ‘SE8T “LEST ‘TS8T x0S8T “6FST SOS8T ‘GFST ‘SPST LO8T aPST ‘6PST GST “PSST ‘EcsT 998T “F9ST “E98T 898T x0S8T 9C8T “FEST “eSsT LOST “998T “S98T €98T T98T. FEST ‘SS8T “OSST PSSL ‘SST ‘OSST POST “TIST ZOST ‘6S8T 098T SSSI “QG8T ‘FEST ‘ZE8I ‘GFST | “SSST ‘FEST ‘ZEST ‘SPST OOST ‘9S8T ‘ES8T “OS8T | 28ST ‘SS8T “ESsT ‘OSST SPST QSS8I ‘LFST LOST ‘998T ‘GOST “F9ST | 99ST ‘OST ‘S98T ‘ZO8T + O98T. LOST ‘O9ST ‘BEST ‘LEST |‘6S8T ‘8S8T ‘LEST ‘P9SST POST “SSSI ‘ZEST ‘TS8T | FEST “ESsT “ZEST ‘OST ZO8T ‘LEST. O98T ‘9S8T LOST ‘998T “S98T “FO8T | 99ST “FOST “S98T ‘Z98T Z98T O98T 6FS8T 6FST 9S8T Scsl 998T COST * PSST ‘TSST PRSST “FEST ‘TSST 898 “FOST LOST ‘ZO8T 99ST ‘Z98T POST ‘TST O9ST ‘6S8T ‘9E8T ‘SS8T | SSST ‘LEST ‘9S8T ‘Sc8T SS8T SSSI 6S81 LEST S98 ‘LOST ‘998T LOST ‘998T “e98T SST FST “SCSI ‘ZEST ‘ISS ‘OSS8T |‘ES8T ‘ZEST ‘ISS “6FST 998T COST g98T LOST. ‘syed oy} dn Surpurq ur pujassy w *XOpuUy SJUBAA 7 *‘poinsy jou ynq ‘paqitosap are satoads oy yo Auvyy & *SOATBAIT) 07 ODud-aT4IT, g *‘xopul » { "M ‘TJX puv surpavqg pfog ‘AA ‘sassoyy Aq “erpeurmepy eus004sIe[q OU, f Seen wee e eens oe eee on . uorjepdimoo Jo esanoo ut ‘ueMG ‘Jorg Aq ‘suorywuNI0,, o1ssery oy} Jo eIT4doxy ony, eee eee eee eee eee eee ee uoryatdu09 jo asanood ur ‘MaMGO Jorg Aq ‘suoIywUIIOT o191[0Q oy} Jo vipydey oy, UPR UCU ee eee e Cee eee eee eee cere rere) Hen e een e een neee trusseseeseceteeevenessveccscaseansuseruseccesscesetsessse Togzaditog UIMGO ‘Jorg Aq ‘suoIyeUAO,T 3xOoqing pue uaprvaA, 9Y4 Jo vIpI4dayy ony, *"*tagajdwmoa ‘UuaMG ‘Jorg Aq ‘suoIyeUtIo,, snosovjerg oy49 Jo viptydexy oy, treeeeeereceseseesess Poagsdumoa ‘Taq pur taMg ‘sjorg Aq ‘[spag Arerqs0y, Ioy}O pus UVYseTyovrg oy} Jo pue] Av[gQ uopuory oy} jo vypydey oxy, eee uorjaldutoo jo asanoo ur “1o4soxUv'y Avy “ puv oramog ‘¢ ‘sassopy Aq ‘ouoyspuvg poy plo 24} JO saqsty ony, sreseees Goagndwuoa ‘Sury ‘Jorg Aq ‘Mo1yRUIAOT ULIUTIAg OT} JO spissoyq OU, mesa 2apdumoa odaeyg *q “ty Aq ‘epodoyeydag snoaoujeasy aeddg eyy, srreresereesees torqgrduioa Jo esanoo ut ‘sdypiyg ‘Jorg Aq ‘saytumejeg eq, tresses gaanduoa Qqaokry "aq Aq yuoutyddng “s ef ss ajajdwoa 44004] *f AT puR stIIOT “Jorg Aq “wosnq[OPY 241[0O verIy) OT, uorja[dut0d Jo astnod Ut ‘poor *A “S “AP Aq ‘seaTearg “vosnT[oyY eMe00q] OTL, OO EE EHO eee eee Hem em eee eH eee eee He Eee ee moryeldutoo jo asanoo ut ‘sprawm poy ‘a ‘aT ‘tyr Aq ‘soayearug pure evpodojeydey ‘vosnyjopy eue0q ety, oshansinanarvaueaesanslessatncesevssetttessssaten oop rduiog (SOATHAIG) “TT ‘JOA tenctanssenenearrenausaeatsasennsesensssures gageduoo “(SOATBAIU()) “[ “TOA —:poom ‘A ‘S “apy Aq “Seip oy4 Jo vosnioy oy, teeeeverscssaesaeeoeesssesesseses TOTaTdU0D JO OSAN00 UT ‘epodorporig uvIaIniIg pus URIMOAeg OUT, “TIT ‘OA . wee neeaes Dubecspacensrsieesinsa957 CUO) ‘epod -OIYOVIG ShosajIMoqriey puB ULIUIIOG ey, “IT ‘OA 88" a990duoo “MospIAvg “J, ‘tj Aq “epodorpovag OISSUITT PUB ‘OIFI[OG snOddNJaID “MreIQIOT, ey, “[ ‘TOA “epodoryovrg [Issoy “+ wo1atdmoo jo esanoo ut Taq ‘y, ‘Jorg Aq “(spuvsusery pue 4jnvy ‘Kelp uopuory oy} Jo ssoyy Sutsttdm0o) vaowqsnap snoovajsoouleyy ey, trereseeseseses TIOTIATA MOD JO ASINOD UT “MaqTVG “AA "fC “AT Aq ‘suoreUtIO,T JoY}O pu ‘UBLIN{Ig ‘UTUOADG, “OUOJSOUMTT]-UIeJUNOPY OY} JO sozrqo[iay, ayy, sreserseseseeses jaardumoo ‘souor yaradny ‘Jorg Aq ‘eteyysy [Isso] OUT, tresses goardumoa ‘sauopr qlodnyy ‘Jorg Aq ‘eovajysomozuy snosvye.1) oy, "+ a2a7dmoo ‘souor gxodny ‘jorg Aq ‘vovsysomoqugy Arerysay, eu, WOTJo[CuLoo Jo asanoo ut ‘parmpoony “F{ “ATT Aq ‘eqvutoqsoseyy [Isso,y ey, ‘ agajdmoo uIimaeg "9 ‘ayy Aq ‘sepedraary [isso,q ery, ogg gn cay uorjaydutoo jo osanoo ut ‘f[ ‘JOA “qustpy “aq: Aq “uquMsepourgoY snosdvjety oxy, teeseeeeoessess TIOTQgTAULOD JO BSANOD UI “TT “OA és s tereeeeseees goandwod JT JOA “FUuSttyy ‘aq Aq ‘eJvUtLopoutyoY o191[00 eT, GP ts9"5997dmoa ‘saqioq ‘Jorg Aq ‘eyeutsepouryomy Areysey, aq, treseesseneceseessecesse gagodulooa sng “4H ‘ayy Aq “sig ayy jo vozAjog ey, UoryeTdmoo Jo esanoo ut ‘uvoung “aq Aq ‘sjeao_ isso oy} 0} guometddng ce eneeaeeresenteneeeneeevereeeses (x) agajdwmoa ‘QUUIVET ‘f PUB SPIBAPT-9UT IPL ae Aq ‘sjetog uRIIntig pus ‘uvtuoAsG ‘o17I[0O ‘snosovjery ‘AavIqAaJ, : sseeeeeeeees orggrdutoo Jo esanoo ut ‘Apeig ‘gq ‘HH pure ‘oyreg “yy ‘AA ‘souor gaedny yz, ‘sussapy Aq ‘eaayruroindo,, Seip ey, a0 * uorjetduroo jo heen ut ‘couurg “AA ‘Ma “ay Aq “eqvayg snosezmoqivy oy} JO BACT] ON, “ec “ee ce Lia { 1 1 { { sneer Reet e eee eee ae eee nena teens { Seen ee ee ete re weeeeeeenee One e eee ee eee eee PER HEHEHE EHO O EH EHH S Eee EE OH EEO EER EE t ease ee senas *qaed Tova Jo aSed-ap11} ay} Uo UOLd -T1qud jo savaX oY} Jo sayeq “ut IIT *pansst sea ydersouoyy OY} SULUIE}UOD OMNIOA IT]} yoyn dof savax oY} JO $2}eq ‘IL “HAVHNONONW FO LOgrdas ‘I ‘sydnibouoyy quavaffip ay, ur pagriosap saroads pun ‘sainbyf‘soquyd ‘sabod fo waqunu ayz ‘suunjoo bumopof puy HLaIs ay; ur pun £ (auns ay, burpurg 07 apinb w sw) ydvebouozg wmnorpund yooa wanjuoa yoryn sawnjoa hrunah ay2 ‘uunjoo aXooas ey we fuoretdtoa Jo esanod ay2 we Wo ‘ajatdutoo . gq paysyqnd ojsaypy ydnuhouogy yona wayjayn uunjoo rsuia oy, ur Kumoys :(S9St ‘ANOL of de) SUTUWA TNL OF ATASSI SHAVUPONOT, AL JO AUVINNAG et te ae i 7 ean5 + i) THE PALMONTOGRAPHICAL SOCIETY. INSTITUTED MDCCCXLVIL. VOLUME FOR 1867. LONDON : MDCCCLXVIII. 1 hci) wo gts th Neer alld OBSERVATIONS ON THE STRUCTURE OF FOSSIL PLANTS FOUND IN THE CARBONIFEROUS STRATA. BY HK. W. BINNEY, F.R.S., F.G.S. PART’ I: CALAMITES AND CALAMODENDRON. Pacrs 1—32; Puates I—VI. LONDON: PRINTED FOR THE PALAONTOGRAPHICAL SOCIETY. 1868. PAGQUEAY i Chen ty \ : oe . Bri vievot 10 On ai oe y re | f . P | 7 3 EY rr Ueue : wo -EPARTS BU ORNTI“AOU YEO ‘ pe Pte 6.5) YY A , - , ies ne Paks | OMIA ONLA LA CVE: BANE AE EO - CON T BN ES. Intropuctory REMARKS . . 2 * ‘ CALAMITES AND CALAMODENDRON : ; : x : F I. BrpLioGRaPHy : 2 F : : : : II. ReMAgKs ON THE SPECIMENS 1. GoLoGgicaL Position or Specimens Nos. 1—11] , 2. Remarks on SpectmENsS Nos. 1, 2,and4—11 . : : : 3. Remarks ON SpEcIMEN No.3 . ‘ : ; 4. GENERAL REMARKS ON SPECIMENS OF CALAMITES AND CALAMODENDRON 5. REMARKS ON ASTEROPHYLLITES AND ITS FRUCTIFICATION : III. Description of THE Specimens Nos. 1—16 IV. Conctupinec REMARKS 19 30 if 1. eoeaeh Yaar ‘ . +, Saw wei AA) OR ae yi + WE aay: dul A arvni8 ct Aa eased] ‘ | thest eet Gaeawde wi Kopreeet ree bo E Lomonty Parken (SF gh vataieh wart urinal et sc ata a Oar wena 4 ue ar Camera h.i40) aon savinanatt, ‘ amriernyte WO Agius wanavad ee ; Pnreasny tone? vei aa fevidivaqouaten ye ena well aa ; bho a adipiawe an? a. diver mene we F ; Vere as I alana ee OBSERVATIONS ON THE STRUCTURE OF FOSSIL PLANTS FOUND IN THE CARBONIFEROUS STRATA. INTRODUCTORY REMARKS. WukEn we consider the great number of valuable specimens from the Coal-measures of Great Britain now in our public and private collections, and see what has been done to bring them before the world, we are led to believe that our Carboniferous Fauna and Flora, but more especially the latter, have scarcely had that attention devoted to them which their importance demands. If the curators of our public Museums would describe the specimens under their charge, private collectors describe theirs, and the Council of the Palzontographical Society lend its assistance in publishing, something useful might be effected. In addition, the aid of the colliery-proprietors should be solicited for the purpose of obtaining funds to enable the Paleeontographical Society to engage the best artists. When this is done we are likely to possess a literature on our Carboniferous Fossils worthy of the first coal-producing country. Having specimens of CaLamitzEs in my own cabinet, collected by myself, I have been induced to make a small beginning, trusting that some other more competent parties may be induced to follow my example. Knowing the great difficulties that have to be encountered in investigating the nature of the Plants which have formed our beds of coal, my object will be chiefly to describe them, without attempting to trace their analogies with living organisms. 1 2 INTRODUCTION. © Probably many Members of the Paleeontographical Society have specimens in greater number and more perfect preservation than those in my collection, especially as to the branches and roots of Calamites, the first genus of fossil plants which it is my intention to describe, but my specimens show structure in a state of perfection that has not often been met with. On a future occasion other genera of fossil plants may be described and illustrated should an opportunity be afforded me. My acknowledgments are due to Mr. Cuttell, lapidary, for his skill in slicmg and mounting the sections of fossil wood, and to Mr. J. N. Fitch, lithographer, for the care and truthfulness with which he has executed the plates illustrating the specimens. CALAMITES anp CALAMODENDRON. I. BrpuioGRaPnHy. § 1. Durtne many years the genus Ca/amites, so common in our Coal-measures, was generally considered to be a reed-like plant, and hence its name. Very excellent figures of the different species of this genus, with their roots and branches, are given by M. Adolphe Brongniart in his ‘ Histoire des Végétaux fossiles,’ and by Messrs. Lindley and Hutton in the ‘ Fossil Flora.’ All their specimens, however, gave little, if any, evidence of the internal structure of the plant. Afterwards Brongniart, in his ‘ Tableau des Genres de Végétaux fossiles,’ after reviewing the labours of Cotta, Unger, Petzholt, and others, thought it better to divide the genus Calamites into Calamitea and Calamodendron, evidence having been obtained of the outer woody cylinder of the latter, which was not believed to occur im the former. § 2. Afterwards Mr. J. S. Dawes, who obtained much more perfect specimens than the Continental authors appear to have possessed, gave a most useful paper on the subject, which is printed in vol. vii of the ‘ Quarterly Journal of the Geological Society,’ and he there states (p. 198) that “on lately examining a specimen of Styillaria reniformis, the tissues appear so much to resemble those of the Calamite as to prove the close connection of these two genera; in fact, all those fossils of this family with the broad outer zones of woody tissue, such as Calamitea striata of Cotta, will in all probability prove to be some species of small-ribbed Sigi/aria.” About the same time Dr. Dawson gave a description of upright Ca/amites found by him near Pictou, Nova Scotia,’ but he does not adduce any evidence as to their structure or nature. § 3. The two last-named authors did not appear to be aware of the publication of a paper by me “On Fossil Calamites found standing in an erect position in the Carboni- ferous Strata near Wigan, Lancashire.”” In that communication, after describing at length the specimens exposed in the railway-cutting at Pemberton Hill, about two miles 1 “Quarterly Journal of the Geological Society,’ vol. vii, p. 194, 1851. * Read before the Literary and Philosophical Society of Manchester, July 6, 1847, and printed in the ‘London and Edinburgh Philosophical Magazine,’ ser. 3, vol. xxxi, pp. 259-266. 4 FOSSIL PLANTS. west of Wigan, I stated, “In the course of his examination of upright stems of Sigdlarie in the Coal-measures, the writer has nearly always found Calamites associated with them. At St. Helen’s they were abundant, and their bases were found in contact with the main roots of Szgillari@. One of the authors of the ‘ Fossil Flora,’ Mr. Hutton, in describing the Burdichouse fossils, at page 24, vol. ii, of that work, states as follows :—Amongst vegetables the characteristic fossils of this deposit are Lepidostroii, Lepidophyllites, Lepidodendra, and Filicites; the rarity of Calamites, which occur but seldom and of a diminutive size, and the almost entire absence of Stigmaria, are very striking to those who are accustomed to view the fossil groups usually presented by the beds of the Carbo- niferous formation; whilst the profusion of Lepidostrobi and Lepidophyllites, of various sizes and various stages of growth, associated with the stems of Lepidodendra, and those of no other plant, is an additional argument for the opinion, which has always appeared highly probable, that they were the fruit, leaves, and stem of the same tribe of plants. Of Stgillaria, a plant which in the flora of the Carboniferous group generally is of so much importance, we could not observe a trace. “In the course of his own observations the writer has never yet been able to meet with a stem of Siglaria of so small a size as six inches in diameter,’ or a Calamites of so large a size as that. Doubtless there must have been young Sigil/arie, whether or not there were large Calamites. Now, what are young Sigillarie ? This is a question which yet remains to be answered. “Tt is now admitted that little is known about the true nature of the genera Stgillaria and Calamites, except that they were not the hollow succulent stems which they were once supposed to be. “The rootlets of Calamites, as previously shown, if not actually identical with, at least very much resemble, those of Stgi//aria. In some specimens of the former genus, especially of the species approwimatus, figured and described in pl. cexvi, vol. iu, of the ‘ Fossil Flora,’ and the cruciatus, figured in pl. xix of Brongniart’s ‘ Histoire des Végétaux fossiles,’ their rootlets were arranged in regular quincuncial order. In the largest Calamites that to my knowledge has been figured, namely, that called gigas, pl. xxv in Brongmiart’s work before alluded to, the ribs and furrows begin to appear very like those of Sigd/aria, and the joints show indistinctly. The termination of the root of a Calamites is exactly of the same form as the terminal point of a Stigmaria, both being club-shaped. “Tam not aware that up to the present time much, if anything, is known of the structure of Calamites ; but if it should resemble that of Stgzd/arza, it may tend to prove that Calamites are but young Sigil/arie. In our observations it must not, however, be lost sight of that no central axis or pith has, to my knowledge, yet been discovered in the stem of Calamites like that found in Sigil/aria. Both plants are proved to have 1 Since the writing of this paper the author has seen in the Museum at Dudley a stem like that of a Sigillaria not more than seven inches in diameter. CALAMODENDRON. 5 had similar Aadztats, and therefore it is very probable that they might have had rootlets resembling each other without being the same plant. Still, however, as Stgil/aria was so long considered as a separate plant from S#gmarza, it is unphilosophical to take no notice of the analogies of what are now considered distinct genera. Although it will not by any means be safe to affirm that Siyil/larta and Calamites are the same plant, from their analogies, still it is conceived that sufficient evidence has been adduced in this paper to prove that the latter as well as the former plants have generally grown on the places where they are now found, and that the reason why one is so much more frequently found in an erect position than the other arises from the circumstance of the stem of the one being much stronger than that of the other. A deposit of mud on the branches and leaves of a slender stem of a Ca/amites might weigh it down and prostrate it, whilst the stout trunk of the Sivil/aria would resist such action and continue erect.” As the specimens near Wigan showed the best roots of Ca/amites found standing as they grew that ever came under my observation, I herewith give in the annexed woodcut (fig. 1) a drawing (reduced one eighth the natural size) made on the spot by my friend =a Fie. 1. the late M. Jobert. ua ie mi § 4. Dr. C. von Ettingshausen, in 1855, published a t mi r most elaborate memoir on Calamites, and showed the state —_, i , of our knowledge at that time of the plant so far as its fruit, i, i branches, and stem were concerned. ‘This author greatly = / ais simplified the matter by classing twenty-three species of / Jf y 7 “1 . stems under Calamites communis, and subsequent investi- gations have, in great measure, if not entirely, confirmed his i \ views ; but he gives us little information on the structure Mi: of the plant, his specimens having been apparently only > casts. J) § 5. Dr. Ludwig has given to the world a very lucid description, illustrated by most beautiful plates, of what he considers to be the fructification of Calamites.? His speci- mens are in a remarkable state of preservation, and are found associated (but not connected) with a small stem resembling a Ca/amites ; this stem, however, scarcely reminds us of the plant as commonly known by that name and so plentifully found in our Coal-measures. The genus Calamites has, no doubt, been made to include several plants of very different structures, having external resemblance to each other in some of their characters, and Ludwig’s specimen must be taken as the fructification of one of them. 1 © Abhandlungen der Kaiserlich-Koniglichen Geologischen Reichsanstalt, Jahrgang,’ 1855. Vienna. 2 “Calamiten-Friichte aus dem Spatheisenstein von Hattingen an der Ruhr,” von Rudolph Ludwig ; ‘Dunker und von Meyer’s Palzeontographica,’ vol. x, 1861 to 1863. 6 FOSSIL PLANTS. § 6. Mr. J. W. Salter, F.G.S., in speaking of the Fossil Flora of the Lancashire Coal- fields, says,’ “The great Calamites are common. And this genus appears to have been the most hardy, persistent, and widely diffused of the Palzeozoic plants. It began with the Lepidodendron in the Devonian, perhaps a little later, and continued far beyond that genus into Triassic times. “What we see of Calamites has been shown by several authors to be only the fluted cast of the pith of Calamodendron. But there is much reason to believe that the outer coating was usually very thin, and that the great succulent pith occupied nearly the whole of the stem. “Pinnularia is the root of Calamites, as I have convinced myself by specimens during this Survey; and we have now also in Britain the fruit of this interesting genus, such as has been illustrated by Ludwig in vol. x of the ‘ Paleeontographica.’” Mr. Salter gives a drawing of the fruit of Calamites from the Lower Coal-measures, Rochdale. ‘The speci- men I have not seen; but in the woodcut it exactly resembles the restored one of Ludwig. § 7. Professor Schimper, of Strasburg, has been one of the latest authors on the Continent who has treated of Calamites.” His specimens do not appear to afford much evidence of structure, but his remarks will give us some idea of the opinions there current as to the nature of the plant, and furnish us with the state of knowledge at the time possessed by so distinguished a botanist, who has devoted much attention to the study of both recent and fossil plants, and whose sources of obtaining information on the subject are varied and extensive. M. Schimper’s description is as follows : “HQUISETACE &. “ CALAMITEM. “© CaLamites, Suck. “ Caulis cylindraceus, fistulosus, articulatus articulis clausis, sulcatis, sulcis continuatis vel in singulis articulis alternantibus, foliis in vaginam dentatam coalitis vel radiatim patentibus vel nullis (7), eorumque loco tuberculis ( foliorum deciduorum pulvinulis seu folits rudimentarus ?) minutis verticillatim dispositis. “Tige cylindrique ou a peu prés creuse, articulee, a articles fermés par un diaphragme, sillonnée régulicrement dans le sens de la longueur; sillons se continuant directement a 1 «Memoirs of the Geological Survey ;’ “The Geology of the Country around Bolton-le-Moors, Lanca- shire,” p. 43, 1862. 2 “ Description des Espéces de Plantes rencontrées dans le Terrain de Transition des Vosges ; par Wm. Ph. Schimper.” ‘Memoires de la Société des Sciences naturelles de Strasbourg,’ vol. v, p. 323, 1862. CALAMODENDRON. 7 travers les articulations ou alternants; feuilles verticillées, réunies pour former une gaine cylindrique ou étalée, plus ou moins profondément dentée, ou nulles (?) et remplacées par des tubercules disposés en verticille (coussinets des feuilles trés caduques ou feuilles rudimentaires ?). “ Subgen.—ASTEROCALAMITES, Schpr. “ Vaginis profunde dentatis, divisionibus radiatim patentibus ; caulis sulcis continuis. Gaines profondément dentées a divisions étalées ; sillons de la tige continus. “CALAMITES RADIATUS, Brg. “CALAMITES RADIATUS, Brongniart. Hist. d. végét. foss.,1, p. 122, pl. xxvi, 1, 2, 1828. — — Unger. Gen. et spec. plant. foss. Vind., p. 44, 1850. EQuISETITES RADIATUS, Sternb. Verst., II, p. 45, 1821-38. —- — Goeppert. Foss. Flor. d. Uebergangsgeb., 114, 1852. CALAMITES TRANSITIONIS, Goepp. Ueber d. foss. Fl. Schles., in WimMer’s Fl. Schlesiens, II, 1841; eusd., Fl. d. Uebergangsgeb., pp. 116—118, taf. 3, 4, 39, 1852. = — Geinitz. Verstein d. Grauwackenform. in Sachsen, Il, S. 28, taf. 18, fig. 6, u. 7, 1853; eyusd., Kohlenform. in Haynischen-Ebersd., S. 30, t. 1. CALAMITES CANN#FORMIS, Schith. RormMeR in DuNKER und v. Meyer’s Paleontogr., III, 1, taf. vii, f. 4, 1850. BoRNIA TRANSITIONIS. Roem. in Dunk. und y. Meryer’s Paleontogr. (pro specim. jun.), III, 1, taf. vii, 7, p. 45, 1850. “ Caulis simplex plus minusve incrassatus, centim. 1-10 in diametro metiens, articula- tionibus cent. 1-12 a se invicem distantibus, costis exacte parallelis, subplanis, continuis, vaginis majusculis, infra medium divisis, radiatim expansis ; rhizomate ad articulationes cicatricibus singulis rotundatis a ramis caulescentibus provenientibus notato. “Tiges simples, offrant un diamétre de 1 a 10 cent., a articulations distantes de 1 a 12 cent., cdtes exactement paralléles, presque planes, se continuant en ligne droite a travers les articulations; gaines assez grandes, divisées jusqu’ au-dessous de la moitié de leur longueur, rayonnantes, rhizome marqué aux articulations de cicatrices isolées, provenant de l’insertion des tiges aériennes. “Cette espéce de Calamite, la seule de genre qui ait été rencontrée dans le terrain houiller inférieur (grauwacke) des Vosges, se reconnait facilement a ses cotes passant sans interruption a travers les articulations, de sorte que ces derniéres ne se distinguent que par un sillon circulaire, accompagné des deux cétés, dans les échantillons un peu forts, d’un renflement plus ou moins évident. La grosseur des tiges rencontrées dans notre terrain varie de 0:01m.a0'10m. et la distance de leurs articulations se montre peu constante, tant sur les petits que sur les grands individus. La gaine qui, par son CO FOSSIL PLANTS. expansion horizontale et sa division’ profonde, rappelle un peu la verticille des Astéro- phyllites, et qui caractérise si bien notre plante, se rencontre trés-rarement, parce que, a la suite de sa position naturelle, elle doit toujours rester dans la roche aprés la sépara- tion de la tige et dans un sens contraire a la position de cette derniére. Pour la trouver il faudrait toujours examiner la roche ambiante et la casser perpendiculairement a la tige, juste a V’endroit qui porte la contre-empreinte d’une articulation, comme cela s’est fait pour l’échantillon conservé dans le Musée dhistoire naturelle de Strasbourg, quia été figuré pour la premiére fois par M. Brongniart, dans l Histoire des végétaux fossiles, pl. xxvi, et que j'ai representé de nouveau a la pl. i, fig. d et c, de ce Mémoire. “ Je crois n’avoir pas besoin d’entrer dans de longs détails pour justifier la réunion du Calamites transitionis, Goepp., avec la Calamites radiatus, Bret. Si le hasard n’avait pas fait découvrir la gaine de ce dernier, personne n’aurait songé a le séparer du premier et encore moins a lui assigner une place dans un genre distinct. Je ne saurais non plus me ranger de l’avis du comte Sternberg et de mon ami Goeppert, en transportant notre fossile dans le genre Lguisetites, dont il se distingue trop par la forme et la direction de la gaine, et par la mode de disposition des sillons. Je serais plutot porté a y voir le prototype des Asterophyllites, et c'est pour cela que je propose de le considérer comme type d’un sous-genre avec le nom d’Asterocalamites. On n’a aucun exemple qu’ un végétal ou animal une fois existant ait disparu pendant un long espace de temps, pour réapparaitre plus tard. Cela serait cependant le cas pour le genre Lguisetites, qui se serait dérobé pendant toute la durée de l’époque houillére proprement dite, pour se montrer de nouveau avec le commencement de l’époque triasique et se conserver de la jusqu’ a notre époque. “Comme en Silésie et dans d’autres localités, notre Cal. radiatus caractérise le terrain houiller inférieur des Vosges et y constitue un des fossiles végétaux les plus répandus, surtout dans la vallée de Thann, prés de Bitschwiller, d’ou nous possédons de nombreux échantillons de toute dimension (Voyez pl. i).” § 8. Professor Goeppert has described, under the name of dphyllostachys Jugleriana, the fructification of a plant allied to Calamites.' Specimens somewhat resembling those described and figured by this learned author are now in my cabinet, and were found by me in both the upper and lower parts of the Carboniferous series, namely, associated with the Spirorbis-limestone in the Manchester coal-field at Ardwick, in the lower Brooksbottom seam of coal,’ and in the Mountain-limestone of Holywell, in North Wales. These specimens will be described and figured. The second of the specimens above mentioned (No. 12), if not of the same species as Goeppert’s, is probably of the same genus. § 9. Herr R. Richter, in a Memoir on the Lower Carboniferous Rocks of 1 Hingegangen bei der Akademie am 11 Mai, 1864. Dresden. 2 T am indebted to the kindness of Mr. John Aitken for this specimen. CALAMODENDRON. 9 Thuringia,’ described the structure of Ca/amites transitionis (Goeppert) as consisting of a main woody axis of cellular parenchyma, and a lesser inner axis or hollow cylinder with transverse partitions and corresponding constrictions. ‘The first has a smooth out- side, and is formed of cono-cylindrical layers, with indistinct radial divisions, termi- nating inwards in longitudinal ridges, which give to the smaller axis a longitudinally furrowed exterior. The woody structure and a cone-like fruit of Calamites transitionis are figured by Richter. § 10. Dr. Dawson has lately published an elaborate paper ‘ On the Conditions of the Deposition of Coal, more especially as illustrated by the Coal-measures of Nova Scotia and New Brunswick,” wherein he gives his views on Calamodendron and Calamites in the following words (p. 134) :—* Calamodendron.—The plants of this genus are quite distinct from Calamites proper. A Calamodendron, as usually seen, is a striated cast, with frequent cross-lines or joints; but when the whole stem is preserved it is seen that this cast represents merely an internal pith-cylinder, surrounded by a woody cylinder composed in part of scalariform or reticulated vessels, and in part of wood-cells with one row of large pores on each side. External to the wood was a cellular bark ; and the outer surface appears to have been simply ribbed in the manner of Sigl/arca. “Jt so happens that the internal cast of the pith of Calamodendron, which is really of the nature of a Sternbergia, so closely resembles the external appearance of the true Calamites as to be constantly mistaken for them. Most of these pith-cylinders of Ca/a- modendron have been grouped in the species Calamites approximatus ; but that species, as understood by some authors, appears also to include true Calamites (see Geinitz’s ‘Steikohlenformation in Sachsen’), which however, when well preserved, can always be distinguished by the scars of the leaves or branchlets which were attached to the nodes. “‘Calamodendron would seem, from its structure, to have been closely allied to Sigillaria, though, according to Unger, the tissues were differently arranged, and the woody cylinder must have been much thicker in proportion. “The tissues of Calamodendron are by no means infrequent in the coal, and the casts of the pith are common in the sandstones ; but its foliage and fruit are unknown. (Fig. 31, Plate VII, a to c.) “ Calamites.—Nine species of true Calamites have been recognised in Nova Scotia, of which seven occur at the Joggins ; the most abundant being C. Suckovit and C. Cistiz. The Calamites grew in dense brakes on sandy and muddy flats. They were unques- tionably allied to Hyuisetacee, and produced at their nodes either verticillate simple linear 1 «Zeitschrift der deutschen geologischen Gesellschaft,’ Jahrgang 1864, p. 155. I am indebted to Professor Rupert Jones for calling my attention to this Memoir as these pages were passing through the press. * Read December 20th, 1865. ‘ Quarterly Journal of the Geological Society,’ vol. xxii, pp. 95—169, 1866. 9 ~ 10 FOSSIL PLANTS. leaves, as in C. Cistid, or verticillate branchlets with pinnate or verticillate leaflets, as in C. Suckovii and C. nodosus. The Calamites do not seem to have contributed much to. the growth of coal, though their remains are not infrequent in it. ‘The soils in which they most frequently grew were apparently too wet and liable to inundation and silting up to be favorable to coal-accumulation. I have elsewhere shown (‘ Quart. Journ. Geol. Soc.,’ vol. x, p. 34) that some of the species of Ca/amites gave off numerous adventitious roots from the lower parts of their stems, and also multiplied by budding at their bases” (p. 135). Again, at pp. 140, 141, in treating of discigerous wood-cells, Dr. Dawson says: “These are the true bordered pores characteristic of Sigillaria, Calamodendron, and Dadoxylon. In the two former genera the discs or pores are large and irregularly arranged, either in one row or several rows. In the latter case they are sometimes regu- larly alternate and contiguous. In the genus Dadozylon they are of smaller size, and always regularly contiguous in two or more rows, so as to present a hexagonal areolation. Discigerous structures of Szgillaria and Calamodendron are very abundant in the coal, and numerous examples were figured in my former paper. I have indicated by the name reticulated tissue certain cells or vessels which may either be reticulated scalariform vessels, or an imperfect form of discigerous tissue. I believe them to belong to Stigmaria or Calamodendron. (Figs. 57 and 68, Pl. XIL.)” § 11. Professor Dr. C. von Ettingshausen' has lately described and figured some interesting specimens of Calamites showing their branches and leaves, but not affording much evidence of their internal structure. § 12. When the present memoir was nearly all written, Mr. Carruthers, after seeing some of my specimens, gave a restored figure of Ca/amites in a paper published in December, 1866,? in which he concludes, “It is not easy to find anything analogous to Calamites among recent plants; nevertheless its structure does not differ so essentially from the vascular cryptogams as to cause any uncertainty as to its position. ‘The histological character of its wood, the absence of medullary rays, and the nature of its fruit, clearly establish that it was a true cryptogam; and while it differed in the arrange- ment of the parts of its stem, in its foliar appendages, and in its organs of fructification, from Lepidodendron, yet it is evident that these were both near allies, and both more highly organized than any of their living representatives.” 1 ‘Die fossile Flora des Nahrisch-Schlesischen Dachschiefers ; Denkscriften der Kaiserlichen Akademie der Wissenschaften,’ Wien, 1866. 2 «On the Structure and Affinities of Lepidodendron and Calamites,’’ by W. Carruthers, F.L.S., Botanical Department, British Museum ; ‘Seeman’s Botanical Magazine’ for December, 1866. In the ‘Popular Science Review’ for July, 1867, p. 295, &c., Mr. Carruthers has further described and figured the Calamite and its fruit. II. RemarRKs ON THE SPECIMENS. § 1. Geological Position of the Specimens, Nos. 1 to 11. The Specimens described in this Memoir are sixteen in number; and all those exhibiting structure (with the exception of No. 3, which came from South Owram) were found by me in the lower division of the Lancashire Coal-measures, imbedded in calca- reous nodules occurring in seams of coal. Specimens No. 1, 2, and 4 to 11 are from the same locality as the Zrzgonocarpon described by Dr. J. D. Hooker, F.R.S., and myself in a memoir “ On the Structure of certain Limestone-nodules enclosed in seams of Bituminous Coal, with a description of some T'rigonocarpons contained therein” ;' and the other specimen, No. 3, is from the same seam of coal in the Lower Coal-measures as that in which the specimens described in a paper entitled “On some Fossil Plants, showing Structure, from the Lower Coal-measures of Lancashire”? were met with, but from a different locality, namely, from the ‘ Halifax Hard Seam” at South Owram, marked ** in the section given below. Specimens of Calamodendron are also sparingly met with in the ‘ Bullion Mines” near Burnley, marked **,—and in the “ Upper Foot Mine,” near Oldham, marked ***, in the following vertical sections of Coal-measures. The position of the seams of coal in which the fossil woods were found in the Car- boniferous series is shown by the following sections of the Lower Coal-measures. In Lancashire. In Yorkshire. Yds. Ft. In. Yds. Ft. In. Arley or Royley Seam ... xis .- 1 1 0 Beeston or Silkstone Seam _... Seren CS Strata ... oy ar a6 . 69 Q OO» Ritatey AS re. Ay ga ae OU Seam Bs ae ae ... 0 0 3. Royds or Black Seam ... ane ene! erga Strata... nee es ee vx Oe QO! OF tenia ee st . if a Jae Seam 0 0 6. Better Bed Seam Pee a, ee ee Strata vas ao ee 20 O 0 CStetee Ace ie Be cog. Or Oele Upper Flagstone (Upholland) ... ... 00 0 O Upper Flagstone (Elland) a: or SO? ae Strata aa ai Y. 2/1520. 0/00 "Siege e.. 2s 63 of i 4000000 Seam (90 yards) x ad .. 0 0 5 Seam (90 yards) as ve ate, Ove RSG 1 ‘Philosophical Transactions,’ 1855, p. 149. ? «Quarterly Journal of the Geological Society of London’ for May, 1862, vol. xviii, p. 106. 12 FOSSIL PLANTS. In Lancashire. In Yorkshire. Yds. Ft. In. Yds. Ft. In. Strata... oi a oo Ve. 20P 20) FOP Strataveeee EA Sat a . 30°10 Seam (40 yards) a a. .. 0 1 6 Seam (40 yards) a a vo 0 a Strata... + a ce sno, (64° 40) (0) (Straten. a Mi a .. 89) 70eme *** Upper Foot Seam (Dog Hill) ... O 1 2 Strata ... ver SB aie ay los 10) 10 ** Gannister Seam co .. 1 0 0 ** Halifax Hard Seam... sie .. (OCT2ES Strata... be see £. i 1S 30.840 OeStratas a t eo .. 14305 Lower Foot Seam (Quarlton) ... 1 10) 20) Muddle'Seam =. a ae .. OPC Strata... se ve Se 7. sO) Oe Surata on... i a9 Mee ., 2400558 Bassy Seam (New Mills) ue .. 0 2 6 Soft Seam oer = Nae . OE Strata... i an a. Pe a0" OP uStrata ie a see of . SGD ae Seam... a as si LO OPO Strata... bee + ae 9 110160 Sand or Featheredge Seam__.... .. 0 2 0 Sand Seam oe a Ag | AO as “Rough Rock ” of Lancashire, “ Upper Millstone” of the Geological Survey 20 0 0 ‘“ Upper Millstone” of Phillips, Halifax. 36 0 0 Strata (Rochdale or Lower Flags) ... 120 0 0 Strata (Lower Flagstone) see : (a oe *Seam tl. a. ae BA SH uONTO. te" wiintile Seam 957)57) st att 1), (OR RORIES: Strata... aT a urs seed eet ORO Seam... fe a a =.) 00 wa0 LO Strata... Pe bos en ela OMp Seam... oe BOL AL ue “Upper Millstone” of Lancashire. In the Lancashire Coal-field all the seams of coal from the “ Forty Yards” downwards have at places afforded Aviculopecten and other marine shells in the roofs of black shale ; and these latter strata generally contain calcareous nodules. The nodules in the seams of coal commonly known by the name of ‘ Bullions” have chiefly been found in the beds marked *, **, and *** in Lancashire; whilst in Yorkshire they have as yet been only observed in the “ Halifax Hard Seam,” marked **. § 2. Remarks on Specimens Nos. 1, 2, and 4 to 11. The specimens showing structure intended to be described in this Memoir are (with the exception of No. 3) from the thin seam of coal marked * in the vertical section of the Lower Coal-measures of Lancashire previously given ; and are from the same “ mine” or bed from which the specimens described by Dr. Hooker and myself were obtained. ‘They were found associated with Halonia, Sigillaria, Lepidodendron, Stigmaria, Trigonocarpon, Lycopodites, Lepidostrobus, Medullosa, and other genera of plants not yet determined. The foregoing are mentioned in the order of their relative abundance. CALAMODENDRON. 13 A portion of one of the specimens of fossil wood on analysis' gave Carbonate of lime “eh 2 $3 re AMOS Carbonate of magnesia __... oe iv ay be 87, Sesquioxide of iron vee as wt hay p4 U0 Sulphate of iron ... ee bi ah a Oe Carbonaceous matter and sae Pe hee 14995 The stratum lying immediately above the seam of coal in which the specimens occurred generally termed the ‘roof,’ was composed of black shale, containing large calcareous nodules, and for a distance of about two feet six inches upwards was one entire mass of ~ fossil shells of the genera Goniatites, Orthoceras, Aviculopecten, and Posidonia. The beds in the vicinity of the coal occurred in the following order, namely, Yds. Ft. In. 1. Black shale, with nodules containing fossil shells at A i) JOT 6 2. Upper seam of coal, enclosing the nodules full of fossil wood ... 0 0... 3. Fire-clay floor, full of Stigmaria... oO. 2.0 4. Clay and rock Deals a) 5. Lower seam of coal 0 0 10 6. Fire-clay, full of Stigmaria. The fossil wood occurred in spherical, lenticular, and elongate and flattened nodules, varying from an inch to a foot in diameter ; the round and globular specimens being in general small, whilst the flatter nodules were nearly always of a large size. No fossil shells were met with in the nodules 1. 2 found in the coal itself (‘2’), although, as previously stated, 2. 7/7/22 they were very abundant in the nodules found in the roof * == (“1”) of the seam,which there rarely contained any remains of Plants. The large nodules of ten to twelve inches in diameter, when they occurred, swelled out the seam of coal both above and below, as in the annexed woodcut, fig. 2. § 3. Remarks on No. 3 Specimen. The third specimen intended to be described in this Memoir is from a small seam of coal, about two feet in thickness, in the Lower Coal-measures, and marked ** in the vertical section at p. 12; and is from the same seam that the specimen of Svgillaria vascularis described by me in the paper published in the ‘ Quarterly Journal of the Geo- 1 For this analysis I am indebted to the kindness of Mr. Hermann. 14 FOSSIL PLANTS. logical Society,’ previously quoted, came from, although from a different locality. This specimen occurred in the “ Halifax Hard Seam” or “ Gannister Coal,” at South Owram, near Halifax. It was associated with Sigidlaria, Stigmaria, Lepidodendron, Halonia, Diplozylon, Lepidostrobus, Trigonocarpon, and other fossil Plants not well determined. The above is about the order of relative abundance in which these plants occurred. A portion of one of the nodules gave on analysis'\— Sulphates of potash and soda ue vie san es 620 Carbonate of lime ... a re Re oe so EBPOIO Carbonate of magnesia a a se be S56) PAPO Bisulphide of iron... si oe of re Jace G50 Oxides of iron a nee ve Ja 38 ls 78. Silica... sae ie ee ee = bi son OPRID Moisture... Be foe Ae ae ne Oc 02 The stratum found lying immediately above the seam of coal in which the nodules occurred was composed of black shale, containing large calcareous concretions, and for about eighteen inches was one entire mass of fossil shells of the genera Aviculopecten, Goniatites, Orthoceras, and Posidonia. The beds occurred in the following (descending) order : Ft. In. 1. Black shale, full of fossil shells and containing calcareous concretions 26) MOG 2. Halifax Hard Seam, with the nodules containing the fossil plants... ew) 3. Floor of fireclay and gannister, full of Stiymaria ficoides. The fossil wood is found in nodules dispersed throughout the coal; some being sphe- rical, and others elongated and flattened ovals, varying in size from the bulk of a common pea to eight and ten inches in diameter. In some portions of the seam of coal the nodules are so numerous as to render it utterly useless; and they are found to occur over a space of several acres, and then for the most part to disappear, and again to occur as numerous asever. Forthe distance of twenty-five to thirty miles the nodules are found in this seam of coal in more or less abundance, but always containing nearly the same plants. Fossil shells are rarely met with in the nodules found in the coal ; but they occur abundantly in the large calcareous concretions found in the roof of the “ mines,” and are there associated with Dadoxylon (containing Sternbergia piths) and Lepidostrobus. So far as my experience extends, the occurrence of nodules in the coal is always associated with that of fossil shells in the roof, and therefore may probably be owing to the presence of mineral matter held in solution in water and precipitated upon, or aggregated around, certain centres in the mass of vegetable matter now forming coal before the bituminization of such vegetables 1 For this analysis I am indebted to the kindness of Dr. R. Angus Smith, F.R.S., who had it done in his laboratory by Mr. Browning. CALAMODENDRON. 15 took place.!. No doubt such nodules contain a fair sample of the plants of which the seams of coal in which they are found were formed; and their calcification was most probably due to the abundance of shells afterwards accumulated in the soft mud and then decom- posed, and now forming the shale overlying the coal. At present little is known of the process by which animal and vegetable bodies are decomposed, and the particles of which they were formed removed and exactly replaced by mineral matter. All observers have been struck with the wonderful perfection of the process by which the most microscopic parts of minute vessels and cells have been pre- served in form ; but no author could satisfactorily account for it until the wonderful dis- coveries in Dialysis by Professor Graham, F.R.S., H.M. Master of the Mint, showed us how crystalloids, such as carbonate of lime, could percolate through animal and vegetable membranes. It is probably by the laws of Dialysis that we shall be enabled to find out the process of the calcification of the specimens described in this Memoir. § 4. General Remarks on Specimens of Cauamites and CALAMODENDRON. For a long time I have devoted considerable attention to the genus Calamites, and have collected a tolerably good suite of specimens which show structure. There is no difficulty in obtaining any quantity of fossil wood, showing the wedge-shaped bundles of pseudo-vascular? structure, which, springing in radiating series from certain circular and oval orifices, next the central axis, are parted by wedge-shaped masses of very coarse cellular tissue, increasing in the opposite direction to those first mentioned. It is well known that the casts of the central axis, showing by their ribs and furrows the former position of the wedge-shaped bundles, are found in most of our Coal-measures ; but when we look for specimens affording outside characters of the woody cylinder, and examples of the pith or central axis showing structure, we find great difficulty in obtaining them. Most collectors select large specimens, and in these, although we may sometimes be so fortunate as to meet with evidence of the outside of the plant better than a mere carbonaceous film, generally we have not much chance of getting any indication of the pith or central axis. The wedges of pseudo-vascular tissue, proceeding from their radiating orifices, are usually found compressed close together, and the space formerly occupied by the pith or central ' For a very excellent account of the petrification of wood, see Dr. Goeppert’s ‘ Die Gattungen der fossilen Pflanzen,’ published at Bonn in 1841, a work of great research, and which does not appear to have been much known in England, judging from the few references’ made to it. This learned author long anticipated me in the discovery of the structure of the rootlets of Stigmaria, although I was quite unaware of it when I wrote my paper on the same subject published in the ‘ Quarterly Journal of the Geological Society,’ vol. xv (1859), p. 79. ? The term “ pseudo-vascular”’ is taken from Mr. Dawes’ paper on Calamites published in vol. vii of the ‘Quarterly Journal of the Geological Society.’ It may be a question whether “vascular’’ should not be employed. 16 FOSSIL PLANTS. axis has been assumed to be composed of lax cellular tissue, like Sigd/aria and Stigmaria were supposed to have been. Mr. Dawes, as well as Petzholt, had noticed the transverse divisions in the pith or central axis of Ca/amites at the nodes, and both authors con- sidered the pith to be composed of cellular tissue, with a few vascular bundles in it. Searchers after fossil plants showing structure must be aware, if they have any great experience, that large specimens have their tissue generally very much distorted and disarranged, and seldom afford any evidence of the central axis or pith, except a mere line, or a cast of mineral matter, in an amorphous condition. My endeavours have been to find very small specimens, for these appear to have undergone less alteration from pressure than larger ones. ‘They vary in size from ,8, of an inch in diameter to three inches ; and it is only in the very smallest specimens that we find every part of the plant preserved. ‘T'ransverse sections of small specimens are met with which show only a radiating cylinder of pseudo-vascular tissue of two tubes in breadth, whilst some of the . larger ones exhibit a radiating cylinder of upwards of one hundred tubes in breadth. In the small specimens we have not much chance of seeing the external characters of the stem, and for this we have to resort to larger specimens. In order to reconstruct the whole of the Plant, it is necessary to build up its parts from different individuals of various sizes. The outside of a small Calamites is ribbed and furrowed, and shows nodes or joints; whilst a larger specimen, in its decorticated state, is nearly smooth, or slightly marked with fine longitudinal strie. The cell-walls of the tubes composing the pseudo-vascular cylinder of the former are thin, and the oval openings not so well defined; whilst in the latter, the cell-walls are stronger, and the elongated openings in them are of a more distinct form. In both large and small speci- mens the central axis or pith is divided at the joints by horizontal diaphragms. Both have a thick carbonaceous bark, and their pseudo-vascular systems are wedge-shaped, springing from orifices’ or openings around the central axis. In this Monograph no attempt will be made to distinguish the genus Calamodendron from the old genus of Calamites; but, as all my chief specimens show structure, they will be named Calamodendron. It was formerly supposed that the larger specimens belonged to the latter genus, and the smaller ones were classed with the former: but my specimens, both large and small, afford only evidence of one kind of structure; and I am, therefore, induced to class both provisionally under one genus in this Memoir. Other observers may bring forward fresh grounds, from structure, to show that Calamites is a distinct plant from the Calamodendron herein described. My specimens I have named Calamodendron commune. ' As I know of no similar arrangement in living plants, I have used the words “ orifices’ or “‘ openings” in the descriptions, in preference to the terms nuclei or areola. CALAMODENDRON. 17 § 5. Remarks on ASTEROPHYLLITES and its FRUCTIFICATION. For many years Asterophyllites has been known as the leaves of Calamites, and numerous specimens have been figured and described by Lindley and Hutton, Brongniart, Ettingshausen, Presl, and others. From the joints of the stem of the plant, at each of the oval spaces around it, proceeded a number of branches, radiating m_ every direction. ‘These, at their joints, sent out smaller branches, which, at intervals, from lesser joints, again furnished whorls of leaves, with a length of an inch to an inch and a half. The leaves were of the shape of a Lepidophyllum, but scarcely so long or so broad ; and were marked in the middle by a strong midrib. In my cabinet are some specimens of Asferophyllites with leaves longer than have been usually met with; and one of these will be figured, although not from the same locality as the specimens which show structure came from. Unfortunately no well-recognised specimens of the external forms of leaves have yet been met with in the calcareous nodules which afford the fossil stems showing structure. The fructification of Asterophyllites is well shown in some specimens found by me in the red shales of the Upper Coal-measures of Ardwick, near Manchester. A specimen will be figured and described, which throws considerable light on the nature of the plant. An example of another plant, resembling the Volkmannia sessilis of Presl, and allied to Asterophyllites, and found in the Mountain-limestone of North Wales, will be given. By the kindness of my friend, Mr. John Aitken, of Bacup, I am enabled to give the figure of the fructification of another allied plant, found near the Lower Brooksbottom Coal (the lowest in the section previously given, p. 12) by Mr. Henry Stephenson, at Ewood Bridge. This, if not the same as the Aphyllostachys Jugleriana of Goeppert, is very nearly allied to it. The three last-mentioned specimens are all about the same size so far as their cones are concerned; and these agree pretty well with the small cones found in great abundance near the stems of Calamodendron, except in being about twice the bulk of the latter ; but all four are very short when compared with the long cones of Memingites described and figured by Mr. Carruthers.’| Most probably this last-named genus had whorls of leaves, like those of Asterophylites, proceeding from a jointed stem, very similar to, but not identical with, that of Bechera grandis, figured and described by Lindley and Hutton. The fructification of Calamites has long been supposed to be that known as Volk- mannia. Some years since, when examining my specimens of Calamodendron, showing 1 «On an undescribed Cone from the Carboniferous beds of Airdrie,” ‘ Geological Magazine,’ vol. ii, p- 433, October, 1865. 3 18 FOSSIL PLANTS. structures, I noticed numerous small cones lying detached around them in the stone. These I immediately recognised as resembling Vo//mannia, and showed them to Dr. J. D. Hooker, F.R.S., who also thought that they bore great resemblance to that genus of fossil plants. On further examination, from the structure of the central axis of the cone being the same as that of the stem of Calamodendron, it was evident to me that the cone was the fructification of that plant. This was before Dr. Ludwig’s paper came under my notice. My specimens of the organs of fructification will be described at length, as they show structure in all their parts, and exhibit the spores in the sporangia, which Dr. Ludwig’s do not, so far-at least as described by that author. The cone is one third of an inch long in my best specimens, although it may have exceeded that length. In one case there are eight sporangium-receptacles, placed one immediately above the other. These receptacles are of a crown-shape, formed by the scales which proceed from the central axis of the cone, at first at right angles, and then, when they reach the outside, taking a vertical direction, somewhat like the scales of Lepidostrobus, figured and described by Dr. Hooker’; but they are arranged one above another throughout the whole series, and not in a spiral direction. The sporangia are of an irregular egg-shape, slightly elongated, and are arranged in fours, symmetrically, around a thorn-like process or spindle coming from the axis; but I have not been able in my specimens to see them distinctly enveloped in a bladder-shaped bag so plainly as described by Ludwig: but there is clear evidence of such a covering. | In each receptacle there are six of these series of four, arranged radially with regard to the central axis; so there are twenty-four sporangia altogether in every receptacle. Hach sporangium has a covering composed of a single row of parallel cells, which generally shows evidence of some disturbance, so that the original form of the sporangium is not often well displayed. This is filled with numerous round spore-like bodies, some of them having apparently a tri-radiate appearance, and looking as if they had divided into three sporules. ‘These are not unlike similar spores seen in Lepidostrobus Brownt ; but are more transparent, not so dark in colour, and of smaller size. The attachment of the bladder-shaped bag, containing the four sporangia, to the spindle is not well seen; but the connection of the latter to the central axis of the cone is clearly shown, and is exactly the same as that described by Dr. Ludwig in his specimen. The outside of the central axis is composed of tubes of hexagonal and pentagonal forms, having all their sides marked by transverse openings of an elongated oval shape, similar to what are observed on the pseudo-vascular bundles of tubes of Calamodendron ; but blank spaces show that some portions of the axis have been of a more perishable nature. 1 * Mem. Geol. Surv. Great Britain,’ vol. ii, part 1, p. 449, plate 7, fig. 8. CALAMODENDRON. 19 The form of the sporangium, when it is of large size, and had attained a state of maturity, is generally of an irregular egg-shape, as before stated; but, in some small specimens, it has a cordate or pear-shape form in tangential sections, and is filled full of dark-coloured spores; and its external cover, composed of a single row of cells, is thicker and more substantial. This envelope appears to have expanded and become thinner as it grew older; and, at length, when the spores were ripe, it probably burst and dispersed them. Specimens of sporangia are to be found in all stages of their growth; but the cones, although generally showing most beautiful structure, have been very much dis- arranged, and it is extremely difficult to get true sections of them, especially in a longi- tudinal direction, parallel to their central axis. A great number of specimens had to be examined before the quadrate arrangement of the sporangia, as first shown by Ludwig, could be made out, as well as the occurrence of six processes or spindles in the stead of that author’s five, and the twenty-four sporangia against his twenty. III]. Description OF THE SPECIMENS. § 1. Tue Specimens (Calamodendron commune) Nos. | and 2. (No. 1, Plate I, fig. 1; Plate II, figs. 1—6. No. 2, Plate I, fig. 2.) Specimen No. 1 (Plate I, fig. 1) is about eight inches in length, and ten inches in circumference. Although no doubt originally cylindrical, it is now of an irregular pear-shape from the pressure to which it was probably subjected in the process of minera- lization. The outside of the fossil, which is in a decorticated state, is marked by fine longitudinal striae ; but it does not show the ribs and furrows so commonly found on what is generally termed the outside of Calamites, nor are there any joints or nodes apparent on it. There is some evidence of the outer bark in a thin coating of bright coal, the specimen having shelled out of the matrix in which it was imbedded, and left its bark in the stone. But in another specimen (No. 2; Plate I, fig. 2), from the same place as No. 1 came, we have an example of a Calamodendron which has been split through the middle, and which there shows, on the outside of the central axis, the usual ribs and furrows so long considered as belonging to the outside of the plant; and in an upper portion of the specimen, not shown in the Plate, is a joint. On looking at a transverse section of No. 1 (Plate II, fig. 1), the wedge-shaped bundles of pseudo-vascular structure are seen squeezed together, and divided by other wedge-like masses of coarse cellular tissue, of a lighter colour; but there is no trace lett of the central axis (or pith) of the plant. This, as has been previously stated, is nearly always the case with large specimens, so far as my experience goes. In the specimen now under consideration, owing to the position of its fracture, we cannot obtain any 20 FOSSIL PLANTS. evidence of the cast of the outside of the central axis; but if the specimen were fractured like No. 2, it would most probably in all respects be similar to that figured in Plate I, fig. 2, as to the usual ribs, furrows, joints, and nodes so commonly found on ordinary Calamites. The wedge-shaped bundles of pseudo-vascular tissue originate from a small circular orifice or opening, sometimes simple, as in the specimen now under consideration (Plate II, fig. 2), but in other instances, apparently divided into several parts, as shown in the annexed woodcut? (fig. 3); they are composed of quadrangular tubes, arranged in radiating series, and increasing in size as they approach the circum- ference. These tubes, but hexagonal in section, also extend in an imperfect radiating series some distance from the “ orifice” towards the central axis, and increase in size as they approach that pomt. They are divided by oblique dissepiments, and their walls are much thicker than ‘those found in the inner woody cylinder of Sigilaria; and, instead of being covered with fine transverse striae, both simple and anastomosing, as in that genus of plants, they are marked by oval openings in the sides, horizontal in the longitudinal section, Plate II, fig. 3, and termed by Dr. Dawson “reticu- lated tissue.”* ‘These openings are also shown, in a tangential section, in Plate II, fig. 4, taken near the commencement of the wedges of pseudo-vascular tissue. The diameters of the tubes increase gradually from their origin at the orifice, or opening, to their termination at the outside of the cylinder, where they are largest. The number of these wedge-shaped bundles in this specimen (No. 1) is seventy-three, alternating with an equal number of wedge-shaped masses of coarse and lax quad- rangular tissue, having their broadest side next to the central axis, and diminishing in size as they extend towards the circumference ; the size of the cells thus decreasing in the opposite direction to that of the decrease of the tubes of the pseudo-vascular cylinder, namely, from the central axis to the outside. ‘These can be seen with the naked eye to about :th of the distance from the central axis to the outside, where they cannot be recognised without the aid of a microscope; but tangential sections of this part of the pseudo-vascular cylinder show that it is traversed by bundles of tissue, oval in section (Plate I, fig. 6) not much unlike in shape to those seen in Svgilaria vascularis, but with 1 This cut is from a drawing made by Mr. Bone under the direction of Dr. J. D. Hooker, who after carefully examining these openings, I believe, came to the conclusion that they were passes for a peculiar kind of tissue which has unfortunately been destroyed, rather than the mere cavities which we now see in the specimens. 2 The openings in the walls of the tubes have a resemblance in shape to Dr. Dawson’s left-hand piece in fig. 67, pl. 12 (‘ Quart. Journ. Geol. Soc.,’ vol. xxii); but the walls of the tubes in my specimens are much thicker than those described by Dr. Dawson (op. cit., p. 140, 169). This structure, no doubt, is familiar to all who have carefully examined under the microscope the charcoal in coal. CALAMODENDRON. 21 the marked difference of being composed of cells and not of barred vessels, as is the case in that plant. The origin of the pseudo-vascular wedges is from the oval orifices, and they seem to have no apparent connection, so far as my observation goes, with the central axis or pith. In small specimens the number of these orifices had been noticed as far as six ; whilst in the large specimen now under description, as previously stated, they amount to seventy-five in number. Specimens of these wedge-shaped masses, similar in structure and external character to those of No. 1, can be seen of different sizes from some so small as to consist of only three tubes in breadth (from the orifice to the outside) up to larger ones, with more than one hundred tubes. The specimen shows what at first sight might be mistaken for annular rings, or depo- sitions of successive growth ; but when carefully examining it under the microscope, we do not find in it sufficient evidence to establish, with certainty any appearance of the cessation of growth, like that shown by the annular rings of an exogenous plant of the present day. Other specimens in my cabinet give more evidence of successive growths ; but still in my opinion scarcely sufficient to establish the former existence of distinct annular rings, showing the stoppage of growth, like that which now takes place in our hard-wooded exogenous trees ; and these appearances in the specimen may have been caused, at the time of the mineralisation of the specimen, by successive deposits of mineral matter. However they may have been produced, these rings appear to me to be nearly similar in Calamodendron to those usually seen in transverse sections of the external woody cylinder of Styil/aria, as well as those in the outsides of Dadoxylon. The tangential section Plate II, fig. 5, is taken near the commencement of the pseudo- vascular bundles, and shows them of small size and divided by broad spaces of coarse cellular tissue. That of Plate II, fig. 6, is taken nearer to the circumference, and shows the pseudo-vascular bundles separated by others of coarse cellular tissue, elongated oval in section. Neither of these sections, however, exhibits the oval-shaped bundles of vessels proceeding from the joints, and communicating with the branches. These will be shown in other specimens, hereinafter described. § 2. Tue Spectmun (Calamodendron commune) No. 3. (Plate III, figs. 1—6.) This is of small size when compared with Nos. 1 and 2 last described, and is ex- hibited in Plate III, fig. 1, displaying only one side and the top of the specimen. It is one inch in length, and three-tenths of an inch in diameter across its major axis. The outer bark has been converted into a film of bright coal, which adheres to the stony matrix, and thus leaves the outside of the stem in a decorticated state. This stem, unlike 22 FOSSIL PLANTS. Specimen No. 1, is marked with the usual longitudinal ribs and furrows, interrupted by jomts, so commonly met with on the exteriors of ordinary Calamites. This difference in the outside appearance of large and small specimens is probably due to the extremities of the pseudo-vascular bundles forming the ribs, and the lax tissue the furrows, in the young specimens ; whilst the older individuals, having an exterior composed chiefly of ' pseudo-vascular tissue without such marked divisions of cellular tissue, do not exhibit such decided ribs and furrows. However different in appearance the outsides of the specimens Nos. 1 and 3 are, we shall presently see that their structure is the same in nearly all respects. Fig. 2 represents a transverse section of the stem, magnified ten diameters. It is oval in form, and shows twenty-two wedge-shaped masses of pseudo-vascular structure, radiat- ing from twenty-two different orifices, placed outside the central axis at regular distances, and parted by wedge-shaped masses of lax tissue, increasing and diminishing in opposite directions to what obtains in the pseudo-vascular tissue, as previously noticed in the larger specimen No. 1 (p. 20). The tubes composing the latter tissue gradually increase in size as they extend from the orifices towards the circumference ; whilst in the lax tissue the cells diminish as they are traced from the outside of the central axis to the exterior. The central axis has been destroyed, with the exception of two oval-shaped portions of tissue, which, however, are too imperfectly preserved to afford us much evidence of their original structure. Information as to this we shall receive from another and smaller specimen hereinafter described. Fig. 3 represents a longitudinal section of the pseudo-vascular part of the stem (magnified seven diameters), showing it to have been composed of quadrangular tubes, having their sides marked by oval openings, placed horizontally, like those previously described in No. 1 (p. 20). These tubes, when they approach the joints swell out ; but after they have passed those parts they assume their usual size. At the joints a diaphragm of coarse cellular tissue appears to divide the stem horizontally, not much unlike what is found in Dadozxylon; but, of course, less frequently, and at greater distances. Fig. 4 shows a tangential section of the stem, and affords evidence of the structure of three of the bundles (oval in section) of vessels which proceeded from the joints to the branches. These bundles, in their characters, both as to shape and structure, are like those seen traversing the internal woody cylinder of Sigil/aria vascularis, from the inside to the leaves. This section also shows other masses, of a lighter colour than the body of the specimen, and a more elongated oval shape, composed of three and four cells in breadth. These appear to be extensions of the wedge-shaped masses of lax tissue which divide the bundles of pseudo-vascular tissue near the orifices. One of these is shown magnified forty-five diameters in fig. 6, and will afford a good idea of their general appearance. CALAMODENDRON. 23 Fig. 5 exhibits a longitudinal section of the same stem, magnified fifty diameters. With this magnifying power the walls of the tubes give evidence of oval openings, placed horizontally ; but they are scarcely so well defined as those in the large specimen No. 1; and the walls appear to be slighter (even after allowing for the smaller size of the speci- men), than those in the specimen first described (p. 20). The two sections last mentioned, more especially the longitudinal one, are obscured in parts by patches of coaly matter ; and, as we should expect from the appearance of the transverse section of the stem, we find that a portion of the central axis has been destroyed, and no trace of its original structure left. The coaly matter is, no doubt, owing to the line of section being taken close to the bark, which is generally found converted into bright coal. § 3. Tae Specimen (Calamodendron commune) No. 4. (Plate ILI, fig. 7.) Plate III, fig. 7, shows a very small specimen of Calamodendron, magnified thirty-four diameters, with the central axis (or pith) in a complete state of preservation. The diameter at the broadest is only ;%th of an inch. The stem, no doubt, was originally cylindrical, like those of all other specimens of this genus of plants; but it has assumed an oval form from pressure when it was in a soft state. The central axis, or pith, is composed of large pentagonal utricles (some filled with a black colouring matter), arranged without order, except that the largest in size is found near the centre. The woody cylinder surrounding the central axis consists of nie wedge-shaped masses of pseudo- vascular structure, radiating from as many orifices, and similar in all respects to that of the two larger specimens, Nos. 1 and 3, previously described (pp. 20 and 22). Indeed it appears to be an individual of the same species, in a younger stage of growth. The form of the utricles composing the central axis somewhat resembles that found im small specimens of Stgil/aria vascularis. Yn all my longitudinal sections of this small stem, I have not been able to satisfy myself that they are marked on their sides by fine horizontal striz; but still they appear to be different from ordinary cellular tissue, and have more the form of utricles than of cells. This is a very material point to clear up, and probably the examination of other specimens may enable us to elucidate it more satisfactorily. § 4. Tue Specimen (Cone oF Calamodendron commune) No. 5. (Plate IV, fig. 1.) The fructification of Calamites was long ago supposed by Ettingshausen and others to be the same as Volkmannia; and the great number of small cones, evidently nearly 24 FOSSIL PLANTS. allied to the latter genus, found lying around my specimens, pointed out the probability of such being the case, when they first came before me; but it was only when the structure of their central axis was examined and found to be of the same character as the stem of Calamodendron that the connection of the one with the other was clearly established. Plate IV, fig. 1, represents a transverse section of Specimen No. 5, one of the cones, ‘th of an inch in diameter, magnified forty-five times. It shows the central axis of the column, composed of hexagonal and pentagonal tubes, which have been somewhat displaced from their original position. Around the part last described is a space where the structure is not shown; and then comes a six-sided girdle of rather larger tubes, giving rise to bundles of pseudo-vascular tubes (enveloped in cellular tissue), which constitute the thick portion of the disc-like scale, which divides the cone into receptacles, or cells, containing the sporangia, and from which the leaves proceed and go upwards. These sporangia (marked 7 7) are of an irregular oval shape, having the broader ends near the periphery and the narrower next the central axis. They were arranged in series of fours, around a stout spike, as described by Ludwig. In this transverse section, of course, only two are shown, and they appear to have been enveloped in a_bladder-shaped bag, traces of which are seen in the dark curved lines (marked 44) bounding the four best preserved sporangia, shown in the lowest part of the figure, and which, when in a perfect state, would in a transverse section have presented a cordate form. In this specimen distinct evidence of six of these bags, each containing twelve sporangia, is shown instead of the five containing ten sporangia in Ludwig’s figure. ‘The outer coating of the sporangium is composed of a single row of cells, and the sporangium itself is full of round bodies, like microscopic spores, some of which have an appearance of a triradiate ridge on their outsides, but the majority appear plain, as represented in the figure. The section seems to have been made across the cone midway betwixt the base of the scale forming the cell-partition, and the spike or spine supporting the sporangia, and therefore affords little evidence of the structure of either of those parts of the cone. For this we must resort to the longitudinal sections contained in Plate V, which will afford us the requisite information. § 5. Tue Spructmen (Conz or Calamodendron commune) No. 6. (Plate 4, fig. 2.) Plate IV, fig. 2, represents a transverse section of No. 6, (;,th of an inch in diameter, magnified 54 times), another cone, similar to that last described. This has the walls of its sporangia considerably disarranged, and only a few spore-like bodies are scattered about the section. ‘Ihe central axis is in a fair state of preservation, and is composed of amass of hexagonal and pentagonal tubes, smaller in size than those shown in fig. 1, but like them in other respects. A space without structure then intervenes between the CALAMODENDRON. 25 axis and an irregular zone of larger tubes, from which zone project six angular arms, that appear to me to be transverse sections of six of the processes which support the sporangia, and which it will be well hereafter to term ‘‘sporangium-bearers.” These seem to have been composed of tubes, having something of a pseudo-vascular character; and the lowest one in the figure appears to join with the outside covering of the bladder or bag (4 &) containing the sporangia; but of this we cannot be certain, as the parts of the specimen are much disarranged. These two transverse sections (Pl. IV, figs. 1 and 2), which are the best preserved amongst many in my collection, afford us evidence that the central axis or column of the cone was composed of hexagonal and pentagonal tubes, surrounded by a substance which has left no evidence of its structure, and outside this is the zone of hexagonal coarser tubes from which spring six heart-shaped bags or bladders containing the upper twelve of twenty-four sporangia, similar to those described by Ludwig, except that their number is two more than were seen in his specimen. § 6. Description oF Specimens (Conzs oF Calamodendron commune) Nos. 7—11. Plate V, figs. 1—5. Plate V, figs. 1, 2, 3, 4, 5, 5a, and 5d represent the specimens, Nos. 7, 8, 9, 10, and 11, and are longitudinal sections of small cones similar to those (Nos. 5 and 6) of which transverse sections are given in Plate IV, and described at pages 23 and 24. Fig. 1 (No. 7) represents a specimen one third of an inch in length, and magnified thirteen and a half diameters. It shows seven receptacles, or cells for sporangia; and more may have existed in the original specimen, as the extremities of the cone are probably wanting in the section. The bases or pedicles of the scales belonging to the second and third receptacles (from the top) are seen to proceed on each side from the central axis, and appear to consist of tubes or utricles, whilst the other divisions or floors, four in number, which are cut outside the central axis, are concavo-convex, and appear to be composed of a thick band of cells, whence spring the stout, fleshy, outside scales or leaves, which take a vertical direction, and enclose each receptacle until the bottom of the next scale or leaf is reached. In each receptacle is seen evidence of spo- rangium-bearers, one opposite the other, springing from the central axis. In the upper- most chamber is a tangential section of one; in the 2nd, a similar section of two; in the drd, a longitudinal section of two, which are seen to proceed from the central axis ; the Ath, a tangential section of two; the same in the 5th. In the 6th is a tangential section showing four sporangia grouped round a sporangium-bearer; and this affords us distinct evidence of the quadrate arrangement of the sporangia around the “bearer,” as first noticed by Ludwig. In the 7th and last receptacle, shown only on the outside of the central axis, no evidence of a longitudinal section of a sporangium-bearer is seen, but a part of one and larger portions of the terminal parts of four other scales or leaves are shown. A. 26 FOSSIL PLANTS. Most of the sporangia in this cone are disarranged, or their sections are such as not to give us a good idea of their original form; but they are all full of sporelike bodies. Fig. 2 (No. 8) is a cone rather less than one third of an inch in length, magnified fourteen diameters. It gives evidence of eight receptacles (all holding sporangia, more or less disarranged and containing a few spores), eight sets of scales or leaves, and spo- rangium-bearers, both in tangential and longitudinal sections, as follow :—in each of the first four receptacles, going downwards from the top, are two tangential sections: in the 5th, two longitudinal sections and one tangential: in the 6th, one longitudinal and one tangential section: in the 7th are two longitudinal sections, showing portions of the sporangium-bearers, connected with the central axis of the cone: in the Sth, there are no remains of the sporangium-bearers ; but a portion of the base of a scale and the whole of the terminal part of that organ are visible. In this figure, although the section only shows a portion of the column, the form of the scales and leaves forming the receptacles is well shown in the lower part. This specimen in all respects so exactly resembles that of fig. 1, that it requires no further description. Fig. 3 (No. 9), magnified nineteen diameters, is a tangential section of a single recep- tacle, and shows four sporangia, of a cordate form, two of them being full of dark- coloured sporelike bodies. The outsides of these sporangia appear to be thicker and darker in colour than the generality of the specimens. Probably these differences may arise from the sporangia being in a younger stage of growth. ‘The section is outside the central axis, and near to the ends of the sporangia. ‘These dark sporelike bodies are frequently met with, in a detached state, in the nodules ; but they are not often found in their position in the receptacle as seen in this case. ‘The sporelike bodies also appear to be larger in size than those commonly met with in ordinary sporangia. Hig. 4 (No. 10), magnified twenty-four diameters, represents a longitudinal section of little more than a single receptacle of a cone, but it clearly shows the marked difference of the scales, both in form and structure, to the sporangium-bearers. The thick fleshy bases or pedicles of the scales appear to have formed a disc- or cup-shaped division between the receptacles, and to have been composed of coarse cellular tissue on the outside, enveloping a bundle of pseudo-vascular tissue, which was prolonged into the apex of the scale or leaf. The sporangium-bearers are broad at their point of connection with the central axis, but they soon taper off, and form a spindle-shaped process like a thorn or spine, chiefly composed of pseudo-vascular structure. Some of the sporangia contain few sporelike bodies, and some are entirely empty. Fig. 5 (No. 11), magnified eighteen diameters, represents the most perfect longitudinal section of a cone that has yet come under my observation. In it we find the structure of the central axis to be of the same character as that of the pseudo-vascular bundles in Calamodendron (see p. 20 and 23), both being composed of tubes that have their walls perfo- rated with oval openings. A portion of this structure, magnified 130 diameters, is shown CALAMODENDRON. 27 in fig. 54, and will be at once recognised by all who have investigated the microscopical characters of coal, as frequently occurring in the charcoa! or “ mother-coal,” and which, as far as I know, has never been clearly traced to any particular fossil plant. This speci- men also affords evidence of the thick scales, or divisions of the receptacles, with their leaflike ends, and of the sporangium-bearers, all in their natural positions, connected with the central axis, and not separated and disarranged, as most generally met with. A highly enlarged sporangium (magnified forty-five diameters) full of sporelike bodies, taken from this specimen is given in fig. 5a. The structure of the central axis of the cone, as seen in fig. 5, and as previously shown in specimens Nos. 5 and 6, Pl. IV, figs. 1 and 2 (p. 238, &.), appears to have had its middle composed of tubes having a pseudo-vascular structure, surrounded by a zone of something which has not been preserved, and now shown by a blank space in the specimen. Next comes a zone of larger-sized tubes, of a hexagonal form, from which spring the receptacle divisions (or scales and leaves), and the sporangium-bearers. § 7. Description or Specimens Nos. 12, 13, 14, 15, and 16. [Plate VI.] In the strata near where the nodules containing the fossil wood showing structure occurred no specimens of Asferophyilites or of the fructification of that genus of plants were found except those previously described; but in my cabinet are three specimens, from the Carboniferous strata, of the fructification of plants most probably allied to Calamodendron, which are worth describing. In all their characters, but more especially as regards size and shape, they bear great resemblance to the specimens of fructification before mentioned. Unfortunately none of them afford any evidence of internal structure ; we get only an outside view of the sporangia, and see nothing of their internal parts or contents, as we see in my specimens from the coal-seams. Plate VI, fig. 1 (No. 12), represents a stout stem, having traces of ribs and furrows, and seven joints, at which knots appear. From these last-named parts, on each side of the stem, are seen to proceed seven cones, each about half an inch in length, springing outwards in a nearly horizontal direction in the specimen. These cones do not expose any trace of a central axis; but are composed of crown-shaped masses, most probably of sporangia, contained in receptacles, arranged around an axis. Eight or nine of these can be seen in one cone. Unfortunately the specimen being in soft shale, no evidence can be obtained of its internal structure, so as to ascertain if the sporangia con- tained any spores. If it is not the same as Dr. Goeppert’s Aphyllostachys Jugleriana,' it is very closely allied to that plant. 1 ‘Ueber Aphyllostachys, eine neue fossile Pflanzengattung aus der Gruppe der Calamarien, so wie 28 FOSSIL PLANTS. As previously stated, I am indebted to my friend Mr. John Aitken for the specimen which I believe was found by Mr. H. Stephenson, near the lowest Brooksbottom Coal, at Ewood Bridge, Lancashire, about fourteen yards below the position of the greater number of the specimens described in this memoir, and near the seam of coal marked * in the vertical section of the Lancashire Coal-field previously given (p. 12). The only point in which this specimen appears to differ from Ludwig’s (see above, p. 5), is that it only possesses eight to nine receptacles or cells, against his fifteen to sixteen. Another specimen of the fructification of a plant evidently allied to Asterophyllites and Calamodendron is given in Plate VI, fig. 2 (No. 13), magnified half as large again as the original. This consists of a stout stem, finely ribbed and furrowed, and affording evidence of four sets of fruit-cones, springing upwards at a high angle from four joints of the stem. Although only four cones are seen at each joint, more may be underneath, covered up in the matrix. In two of those sets which are more perfect than the rest we observe traces of four cones, and in the other only two. Each cone has a central axis or column, from which spring ten scales on a side, forming receptacles or sporangium cases, similar to those described in Pl. VI, fig. 1 (p. 27), except that there we only see eight on each side. In this specimen we do not find the terminal poit of the cone, so that there is no positive evidence of the number of scales it originally possessed. In size and characters, especially as to its stem, this specimen bears considerable resemblance to the Volkmannia sessilis, of Presl, figured by Dr. Goeppert,’ as well as to the fructification of Calamodendron ; and although probably the evidence may not be sufficient clearly to connect either this or the last-described specimen specifically with the Calamodendron commune figured by me, still they must be considered as nearly allied to it, as well as to Ludwig’s specimen (see above, p. 5), and they are valuable in showing the connection of the cones with the stems on which they grew. In the Upper Coal-measures of Ardwick, near Manchester, above the highest seam of coal there met with, is an abundance of Calamites and Asterophyllites, especially the species 4. longifolia, of Lindley and Hutton; and connected with the latter plant, and lying around both, are numerous fruit-cones, which, although showing no structure (beng embedded in a liver-coloured shale), give a clear idea of their external form, and the mode by which they were connected with the stem and leaves of Asterophyllites. Pl. VI, fig. 3 (No. 14), represents a specimen of Asterophyllites longifolia (from Ardwick) in my cabinet, magnified half as large again as its natural size. Springing from the stem, at each of the joints, twelve to fourteen verticillate leaves are shown in the part of the specimen exposed, and probably as many more may be concealed on the other side uber das Verhiltniss der fossilen Flora zu Darwin’s Transmutations-Theorie ;? von Dr. H. R. Goeppert, Dresden, 1864. 1 See Goeppert on ‘ Aphyllostachys,’ ante. CALAMODENDRON. 29 in the matrix. Some of these leaves measure nearly an inch and a half in length, and are marked with a keel. Altogether, in form and character (except being a little less in size), those leaves bear great resemblance to Lepidophyllum. In Pl. VI, fig. 4 (No. 15), is another specimen from Ardwick (magnified twice its natural size), having a stem of about two inches in length, and not so stout as the two last described, but more deeply furrowed and more sharply ribbed. At each of the joints of the stem (two of which are visible) are seen four fruit-cones, accom- panied by as many leaves of Asterophyllites longifolia, springing outwards ; and probably as many more may be concealed in the matrix underneath. The cones consist of a central axis, on either side of which, exactly opposite each other, are seen seven or eight pairs of cordate bodies, each having a division in its middle. ‘This organ somewhat resembles the bladder-shaped bag or envelope, containing sporangia, described by Ludwig (see above, p. 24). They are bounded by a scale or disc, coming at first from the axis nearly at right angles, but afterwards running almost parallel to it, and forming the receptacle or sporangium-cases. In three of these cones is the termination of the scales; and there are six cones seen, probably as many more lying underneath in the shale: thus it is probable that each receptacle had six bags, containing four sporangia each; or altogether twenty-four, as in the specimens showing structure previously described (p. 24). The thorn or spindle, around which the sporangia are fixed, is not well shown unless we take the dark line of division of the sporangium-case to be it. The whole of the specimen so far as the cone is concerned, in its external characters, resembles my other specimens of the fructification of Calamodendron commune rather than Ludwig’s specimens; and it is found, we must remember, not only with the leaves of Asterophyllites, connected with the stem on which it grew, but surrounded by an abundance of detached leaves and stems of that plant. Pl. VI, fig. 4 a (No. 16), represents the apex of a cone, from Ardwick, magnified three diameters ; and shows six leaves or scales on the side of the specimen which is exposed to view. 30 FOSSIL PLANTS. IV. Conciupinc REmMArRKs. After the description of the specimens of Ca/amodendron in this Monograph, it will not be out of place to notice the points in which this genus of plants differs from Styél/aria, especially from the plant which I have described as Sigillaria vascularis." The form of the roots of Stgi//aria varies from that of Calamodendron. It is true that the termination of Stigmaria is club-shaped, like that of Calamodendron, and the rootlets are arranged in quincuncial order, as was shown in the paper of mine previously alluded to (see above, p. 15); but in the latter plant we find no trace of the regular bifurcation of the main and lesser roots ; and the roots appear to have been small in size, with regard to the stem and branches of the plant, when compared with similar parts in Siglaria. From all the evidence which has been obtained by me, Calamodendron must have been a plant of small size when compared with Stgil/aria. The largest Calamodendron that has come under my notice is one from the Middle Coal-measures, and was found by the late Mr. John Atkinson, F.G.S., in the neighbourhood of Chesterfield ; it is now in my possession. ‘The cast of the central axis of this specimen is five and a half inches in ‘diameter; and, taking the proportions of smaller speci- mens, the woody cylinder, exclusive of the bark, would probably be about one foot in diameter; a small size, when compared with some stems of Sigl/aria, which have been found to measure seven feet in diameter at the base. The terminal branches of Calamodendron were also of small size, some of them not being more than ;%ths of an inch in diameter ; whereas the smallest specimens of Sigillaria vascularis, which have come under my notice have been about half an inch across; and the branches, like the roots before mentioned, in Calamodendron have none of the dicho- tomous characters so distinctly shown in Stgillaria vascularis. The organs of fructification did not reach more than one third of an inch in length, and were of a diminutive size even when compared with the small bulk of the plant. We cannot well compare them with similar parts of Sigi//aria, as at present we are unable to speak with absolute certainty as to the fructification of that plant, which was most pro- bably a cone much larger than that of Calamodendron. Indeed the small size of the organs of fructification in Calamodendron is one of the most singular characters of the plant. There is a difference also observed in the bundles of vessels passing from the centre to the circumference, dividing the wedge-shaped masses of pseudo-vascular tissue, and which M. Adolphe Brongniart and some other authors considered to be of the nature of 1 ¢ Quart. Journ. Geol. Soc.,’ vol. xviii, p. 106, pl. IV and V. CALAMODENDRON. 31 medullary rays, but which Mr. Carruthers, with apparently very good reasons, regards as not having that character in S. vascularis. At the joints, where the branches spring from the stem, the bundles of vessels there found much resemble those of &. vaseularis; but the tissue dividing the pseudo-vascular bundles in Calamodendron is very different ; and, when seen in the small openings towards the outer part of the woody cylinder, has more the appearance of a medullary bundle (being formed of cellular tissue) than those found in Sigillaria vascularis, which are barred on all their sides, like the tubes formiug the woody cylinder of that plant. Although some years since it occurred to me, as well as to others, that small Ca/a- modendron might possibly have been young S/gillaria, judging only from the rootlets and some other portions of the plants, still we have seen that in the external characters of the two genera there is no evidence to establish their identity. When also we come to compare the internal structure of the two plants, one differs from the other so much as to dispose of any outward resemblance. In Stgid/aria, whether we take Diploxylon cyca- doideum or Sigillaria vascularis, there are two woody cylinders, formed of radiating tissues, while in Calamodendron there is only one such woody cylinder. The tubes composing the internal radiating cylinders in the two first-named plants have their walls covered with fine horizontal lines or striae, either free or anastomising ; while in the last-named plant the walls of the tubes are much thicker, and are pierced by oval openings, having their major axes at right angles to the direction of the tubes. In addition, the central axis in Sigilaria has no horizontal diaphragms dividing it into separate portions, as is the case in Calamodendron, although the casts of the two central axes are each striated longitudinally. From the examination of the specimens described in this Monograph, S/gilaria and Calamodendron must be considered as two distinct plants, although they doubtless grew in similar positions, and in their habitats accompanied each other, in greater or less abundance, during the whole of the time in which the Carboniferous strata were in the course of formation, as is evident from the remains of both plants being so frequently found associated together in the “mother-coal” of this and other countries. In the upper seam of coal in the section of the strata previously given (p. 12), and there marked with a single asterisk, Ca/amodendron is by far the most common plant, and Sigilaria (in the form of Diploaylon cycadoideum) is but rarely found ; while in the seams marked with two and three asterisks, higher up in the series, Sigil/aria vascularis is by far the commonest form, and Calamodendron is rarely met with; Calamodendron being the smaller plant, forming the chief part of the small seam of coal, and Sigillaria the larger plant, forming the greater proportion of the thicker seam of coal. Something similar occurs in the Upper Coal-measures of Ardwick, whence the specimens Nos. 14 and 15 came. Calamites, then, is one of the most common plants met with, and it is found associated with Asferophyllites, Lepidodendron, Lepidostrobus, and Lepidophyllum, but with no traces of large-ribbed and furrowed Ségillaria, and only rare specimens of Stiyillaria 32 FOSSIL PLANTS. elegans. In this locality no seam of coal is found associated with the fossil plants; and a small bed, a few inches in thickness, occurs about fifty feet below them. My observations on this part of the subject have been made chiefly whilst searching for specimens affording evidence of structure, and not over any great thickness of the strata, and therefore further attention should be devoted to it in order to determine whether or no this condition occurs throughout the Coal-measures where Sigi/laria and Calamites are found associated. Many years since, when examining the thick seams of coal which are found in the middle division of the Lancashire Coal-field, I noticed the occurrence of large-ribbed and furrowed Sigillarie, and J stated in a paper on the Origin of Coal, read before the British Associa- tion in 1843, and printed in the Report of the Association’s first Meeting at Manchester, that where such large specimens occurred not only were the seams of coal thick, but they were open-burning coals, leaving a white ash. Many years’ observation on this subject has confirmed my first impression. Of course it is not contended that ribbed and fur- rowed Sigillarie are not to be met with from the lowest to the highest Carboniferous strata and their roots (Stigmari@) found in coal-floors throughout, but it is merely in- tended to state that during the formation of the thick seams of the Middle Coal-measures greater crops of these trees prevailed, and produced more mineral ingredients to form the white ash, as well as the thicker seams of coal. Nore.—The Author thinks it necessary to state that the Specimens described in this Memoir were discovered twelve years ago, and some of them immediately “mounted” and sent to Dr. Hooker, who had consented to join in publishing a description of them. With the specimens was written, November 29th, 1854, “ You will be delighted to find such beautiful sections of Volkmannia in the large slide I now send. “They are well worth the trouble and expense of the slide. These, doubtless, belong to Calamites.” Soon afterwards the Author had the opportunity of pointing out under the microscope the spore-like bodies in the sporangia to Dr. Hooker. At that time the whole of the structure of Calamodendron had been made out, with the exception of the centre of the stem and the connection of Volkmannia with it. The valuable paper of Dr. Ludwig on the Calamite fruit, and Dr. Goeppert’s on Aphyllostachys, afforded the writer much information ; and he obtained the evidence of Volkmannia being the fruit of Calamodendron from the similarity of the central axes. About two years since he cleared up both the above points to his satisfaction, but Dr. Hooker had returned the specimens, and was prevented by press of business from joining him in publishing. The author then commenced the present Memoir unaided, and the plates were put in the engraver’s hand. Owing to circumstances over which he had no control, the publication of the Memoir has been delayed longer than he expected. In the mean time, in Memoirs referred {o at page 10, Mr. Carruthers has described the Calamite and its fruit mainly from some of the Author’s specimens above mentioned, which Dr. Hooker had lent him with others. ‘he published observations are, of course, independent of each other; and, whatever may be their relative value, the Author wishes the dates and history of the specimens and investigations to be clearly stated.—E. W. B.—January 80th, 1868. PLATE. Calamodendron commune. Fig. 1 (No. 1). A decorticated specimen from the Upper Brooksbottom Seam of Coal, Lancashire. Natural size. Fig. 2 (No. 2). The furrowed and ribbed cast of the central axis of a large specimen from the same locality as No. 1. Natural size. In the following Plates the same parts in the specimens figured are indicated by the same letters, as follow : aa. The middle part, showing the central axis, or pith, composed of large hexagonal utricles. 66. The pseudo-vascular cylinder, composed of quadrangular tubes, having all their sides perforated by oval openings ; they are arranged in wedge-like bundles and in radiating series, originating from “orifices” near the thin end of the wedge next the central axis, increasing in size as they approach the circumference, and divided by medullary (?) bundles. ce. Wedge-shaped masses of lax tissue, composed of oblong cells, in radiating series, having the thick end of the wedge next the central axis. Both wedges and cells increase in size in the opposite direction to that seen in the tubes and masses of the pseudo-vascular cylinder, b 6. d, Diaphragm, apparently composed of cellular tissue, dividing the central axis or pith horizontally at the Joints. ee. Oval-shaped bundles of pseudo-vascular tissue, passing from the central axis, or pith, and com- municating with the Leaves. Jf. Bundles of cellular tissue, of an elongated oval form, near the circumference of the stem, dividing the pseudo-vascular cylinder, and having the appearance of medullary bundles, though only ex- tensions of the wedge-shaped masses, ¢ c. gg. The central axis or column of a Cone. hh. The Scales or Leaves forming the divisions of the Sporangium-receptacles or cells of a Cone. iz. The Sporangium-bearers of a Cone. jj. Sporangia full of Spore-like bodies. kk. Portions of the bag containing the Sporangia. Late 1 R Fitch imp hth Fitch. del. et JN is - er © . Nes - C = os - - « bl * * , ical 7 1 ‘ Ss I > ; te + ~ : + = a . . - 2 EY 7. - ry 3 , % i » ts Fi - 5 4.9 > * . cy . AP A wove wai ) - 7 ' yt ay, imisiav) tye: aaiivineae To ones uy svonm | =— tw gi airaege Sine wit Fire) w= Sovavrouy oi Ww ts TOR A ia Aya ' - ie tl a Wee UaeT MH BiL: A py sbourAp 10 eri bint) heqnias 7ethH) 1 Dai ) Ae pet i wt ei | Sevan A f UWA ATG! 77 ya a Dae F a Wy Huldroe. taltiagurd A co) of) dah Wellton Silugdee aise ast ty | a A Wal mariah él pit Astinde anitive & vy ay Peer sei eli Ue wire tre . g 7 — pa x PLATE II. Calamodendron commune. Fic. go. 1. Transverse section of Specimen No. 1. Natural size. Fig. 2. A portion of the transverse section of the same specimen, showing parts of two wedge-shaped bundles of pseudo-vascular tissue, originating at two “ orifices” on the outside of the central axis, and parts of the three wedge-shaped masses of coarse tissue, arranged in radiating series. Magnified 10 diameters. Fig. 3. A longitudinal section of the same specimen, showing the walls of the tubes forming the pseudo-vascular cylinder perforated by elongate-oval openings ; taken near to the central axis. Magnified 70 diameters. Fig. 4. A tangential section similar to the last, taken nearer to the outside of the specimen, showing the oval openings on the walls of the tubes. Magnified 70 diameters. Fig. 5. A tangential section of No. 1 Specimen, taken near the central axis, showing the pseudo-vascular bundles and the lax tissue dividing them. Magnified 15 diameters. Fig. 6. A section similar to the last, taken near the middle of the specimen. Magnified 40 diameters. FlatelH Fig. L INO Fig. 6. CG BW LV? 1 R. fitch. mp JN Fitch, delet hth PLATE III. Calamodendron commune. Fig. 1. Specimen No. 3, from the “ Hard Seam” of Coal at South Owram, near Halifax ; showing the outside of a decorticated plant, of small size, with ribs, furrows, and a joint. Magnified 2 diameters. Fig. 2. Transverse section of the same specimen, showing the wedge-shaped bundles of pseudo-vascular tissue, originating from “ orifices’ and parted by lax tissue. The central axis has for the most part disappeared. Magnified 10 diameters. Fig. 3. A longitudinal section of the same stem, showing one of the horizontal diaphragms. Magnified 7 diameters. Fig. 4. A tangential section of the same specimen, showing three oval-shaped bundles of vessels found at the joint, as well as some elongated oval masses of cellular tissue, not unlike medullary bundles. Magnified 12 diameters. Fig. 5. A longitudinal section of a portion of the same specimen, showing the oval openings on the walls of the tubes forming the pseudo-vascular cylinder. Magnified 50 diameters. Fig. 6. A tangential section of the same specimen, showing one of the medullary (?) bundles. Magnified 45 diameters. Fig. 7. Transverse section of a very young stem (Specimen No. 4), slightly compressed, showing the structure of the central axis, the wedge-shaped bundles of pseudo-vascular tissue originating from “ orifices,’ and parted -by coarse tissue. From the Upper Brooksbottom Seam of Coal. Magnified 34 diameters. al Late ill Zee ATO 3 Fug A Fig. 4. 4 en Imo REt Fitch, delet. hth JN 1 if iy 7 i 7 hy . i f ‘ i i 1 Ly 4 - ' j r ’ i y , ! ’ ; is . 1 . r y is 7 © 7 . . x, mn fj io i i 4 & ee é ® Me 5 sob As 4 x iF y ' we i bi re ’ 5 . a By he 8 P=# , . 7 al v ' . ei 7 ai ¢ s : : ‘2 j mh : “s 1 sae ‘ - i * ; A . i \ ; \ a Ao ) , i 4 i C “~ rs ! 7 i « z ; : ’ ¢ a re ; . J : * oS i d ‘ : x 2 £ ' y, < : F rt) ely AE L ule aa : ‘ : 9 P . y 7 it i ee . te A ' y : F ; 2 7] ~ a Z . i. © 5 Fe iT i] ‘ ae ' . y ’ a es A ® os al = % id ' i he “ ’ i ; 6 i - ‘ , ; ; ra , we sh, ee { P A ty : “ae - ‘ y : Ae 2 i ‘ : , si " , : . A . < : . r £ . vo j / :) iz PHA 1) PLATE IV. Calamodendron commune. The specimens in this Plate are from the Upper Brooksbottom Seam of Coal. Fig. 1 (No. 5). Transverse section of one of the Cones or organs of fructifica- tion, showing the central axis, composed of hexagonal and pentagonal vessels, and twelve Sporangia, full of Spore-like bodies; the four of the lower ones enclosed in two cordate bags. Magnified 45 diameters. Fig. 2 (No. 6). Transverse section of another Cone, showing the central axis and six processes or spines (Sporangium-bearers) radiating therefrom, and disarranged Spore-cases, formerly connected with them. Magnified 56 diameters. Flake, LV. esl igi Tt, LL de , . EZ ey de *e: Fy em roi “Xft Nx ae a < epee TOT pelos, 2 ae yet ign or , . as ca) aKa JN. Fitch del. et ith R Fitch amy ‘ t x 7 ' % ‘ i ° —_ ‘ oi ‘ ¥, y / ’ y PLATE V. Calamodendron commune. All the specimens in this Plate are from the Upper Brooksbottom Seam of Coal. Fig. 1 (No. 7). Longitudinal section of a Cone, showing seven Receptacles, Scales, and Sporangium-bearers, connected with the column or central axis, and Sporangia full of Spore-like bodies. Magnified 135 diameters. Fig. 2 (No. 8). Longitudinal section of a Cone similar to the last, showing eight Receptacles and Sporangium-bearers, connected with the column or central axis, and Sporangia containing Spore-like bodies. Magnified 14 diameters. Fig. 3 (No. 9). Longitudinal section of a single Receptacle with its two scales ; and a tangential section of four cordate Sporangia, two of them containing dark coloured Spore-like bodies. Magnified 19 diameters. Fig. 4 (No. 10). A longitudinal section of part of another Cone, showing the structure of the Scales and Sporangium-bearers, and the connection of both those parts with the column or central axis. Magnified 24 diameters. Fig. 5 (No. 11). A longitudinal section of part of another Cone, showing the Scales and Sporangium-bearers, connected with the central axis or column, and the Sporangia, in their natural position, full of Spore-like bodies. Magnified 18 diameters. Fig. 5a (No. 11). A longitudinal section of a single Sporangium full of Spore- like bodies, taken from No. 11 Specimen. Fig. 56 (No. 11). A longitudinal section of a portion of the central axis or column of No. 11, showing the oval openings in the walls of the tubes. Magnified 130 diameters. Flate V J.N Fitch, del, et hth R Fitch, imp PLATE VI. Fig. 1 (No. 12). Fruit-stalk, with Cones attached, resembling the Aphyllostachys Jugleriana of Goeppert; from the strata adjoining the Lower Brooksbottom Seam of Coal, at Ewood Bridge, Lancashire. Magnified half as large again as the original. Fig. 2 (No. 13). Fruit-stalk, with Cones attached, resembling Volkmannia sessilis of Presl; from the Mountain-limestone at Holywell, North Wales. Magnified — half as large again as the original. Fig. 3 (No. 14). Specimen of Asterophylhtes longifolia, from the Upper Coal- measures at Ardwick, near Manchester. Magnified half as large again as the original. Fig. 4 (No. 15). Fruit-stalk of a Plant resembling Calamodendron commune (?), with Cones and Leaves attached to it, from the Upper Coal-measures at Ardwick, near Manchester. Magnified twice the natural size. Fig. 4a (No. 16). Specimen showing six Scales, with their apices, attached to the central axis of a Cone, similar to those last described. From the Upper Coal- measures at Ardwick. Magnified 3 diameters. KE. W.-B: Hate V1 JN. Fitch, del. et hth R Fite hamp » fg, LOOTAOR as ry ~ A MONOGRAPH OF THE BRITISH FOSSIL CORALS. SECOND SERIES. BY P. MARTIN DUNCAN, M.B. LOND., F.G.S., SECRETARY TO THE GEOLOGICAL SOCIETY. Being a Supplement to the ‘Monograph of the British Fossit Corals, by MM. Mitne-Kpwarps and Juices Haine. PART IV, No. 2 CoRALS FROM THE ZONE OF AMMONITES ANGULATUS (continued). CorRALS FROM THE ZONE oF Ammonites BuckiaNnpdI, AMMONITES OBTUSUS, AND AMMONITES RARICOSTATUS, OF THE Lower Litas. Corats FrRoM THE Mippie Liss; FROM THE ZONES OF AMMONITES JAMESONI AND AMMONITES HENLEYI. ADDITIONAL Spectres oF CoRALS FROM THE ZONE OF AMMONITES PLANORBIS. CoRALS FROM THE AVICULA-CONTORTA ZONE. CoRALS FROM THE WHITE Litas. APPENDIX TO THE Liasstc CoRats. INDEX. Pages 45—73 ; Plates XII—XVII. LONDON: PRINTED FOR THE PALZONTOGRAPHICAL SOCIETY. 1868. = a “ sew « t a! { i) f *~ a => he = . i ™ ; +r : fb ) f 9% ’ ’ é ri j : \ Ag : : 4 ¢ j , e - iy ( i f Ab 1a Ba ii conta i Servove As . ‘ P ¢ ii u ,rysva» ' \ ’ j . is} i RNY i, ia® Ae NTL we pyle ‘ y wy i ' P yeu -eryyey a ’ 2 ~ a 4 ‘ t i bi ital i . — ” \ (vi CONTENTS. DESCRIPTION OF THE CoRALS FROM THE ZONE OF AMMONITES ANGULATUS (continued) DESCRIPTION OF THE SPECIES On THE CoRALs oF THE BritisH AND European Lower Liassic Deposits oF THE ZONES OF AMMONITES ANGULATUS, AMMONITES PLANORBIS, AND AVICULA CONTORTA List of SPECIES FROM THE CONTINENTAL ZONE OF AMMONITES ANGULATUS List oF SPECIES FROM THE CONTINENTAL AND British STRATA OF THE ZONE oF AMMONITES ANGULATUS DESCRIPTION OF SPECIES FROM THE ZONE OF AMMONITES BUCKLANDI (BISULCATUS) List OF THE SPECIES DESCRIPTION OF THE SPECIES FROM THE ZONE OF AMMONITES OBTUSUS DESCRIPTION OF THE BEDS WHICH CONTAIN THE MADREPORARIA OF THE ZONE oF AMMONITES RARICOSTATUS DESCRIPTION OF THE SPECIES F é List OF THE SPECIES : List oF THE SPECIES FROM THE ZONES OF THE Lownr LIAs ABOVE THE ZONE oF AMMONITES ANGULATUS : Corats FROM THE Mippin Lias: or THE ZONES OF AMMONITES JAMESONI AND AmMonitES HENLEYI : ENUMERATION OF THE British Liassic SPEcIEs . DEscrIPTION AND Notice oF SPECIES FROM THE ZONE OF AMMONITES PLANORBIS List oF SpECTES FROM THE ZONE OF AMMONITES PLANORBIS Noticrk OF THE INDETERMINABLE CoRALS OF THE AVICULA CONTORTA ZONE AND Waite Lias or THE Britisu IsLes (Ruatic or Moore) Note on THE AGE OF THE SuTTon STONE AND THE BrocastTLE, &c., Deposits . InpEx To Nos. 1 anp 2 oF Part IV PAGE nudes OA TP BAAR in) wel, were on wh Aver Ct, ex A MONOGRAPH BRITISH FOSSIL CORALS. (SECOND SERIES.) Part 1V.—No. 2. VIII. Corats From THE ZONE oF AMMONITES ANGULATUS. ( Continued.) Tuere are some Coralliferous deposits belonging to the Lower Lias at Inkbarrow, at Chadbury, in Worcestershire, and Fladbury, near Evesham, whose exact geological horizon has not been determined. They are low down in the Lower Lias, but their commonest Corals do not identify them with the Coralliferous beds of Brocastle. The genus /sastrea is dominant in these localities, and its species are unlike any which have been described. The Corals will not do more than associate these beds on one horizon. ‘There is a great probability, from the presence of small Gasteropoda, whose shells are left in the calices of the Corals, that careful search will yield a sufficient number of fossils to deter- smine whether these deposits are below the Zone of Ammonites Buckland. Our present knowledge does not justify the association of these /sastree with the Coral-fauna of the Zone of Ammonites angulatus. The Coralliferous deposits at Abbott’s Wood, Harbury, Aston Magna, and Down Hatherly may belong to more than one zone; but, from the association of Zhecosmilia Michelim, Thecosmitia Martini, and Septastrea Fromenteli, the presence of the Zone of Ammonites angulatus may be satisfactorily asserted. There is an Jsastrea found in the Lower Lias of Lyme Regis, which is said to belong to the Zone of Ammonites angulatus, but the mineralization of the specimen and its affini- ties are sufficiently distinct to associate it with the beds containing Ammonites Buckland. 7 46 BRITISH FOSSIL CORALS IX. Description OF THE SPECIES. Section—A4APOROSA. Faminty—ASTRAIDA. Division—LitTHOPHYLLACE SIMPLICES. Genus—MON'LIVALTIA. 1. Montirvattia Rupurti, Duncan. Pi. XII, figs. 3, 4, 5; Pl. XV, fig. 15. The corallum is turbinate ; it is truncated at the base, and is widest at the calice. The epitheca is strong, and is marked transversely with ridges, prominent lines, and constrictions ; the longitudinal markings are faint, but there is a tendency to split in their direction. The calice is moderately deep, and is circular in outline. The septa are crowded, unequal, long, and irregular ; the longest are thick internally, and reach so far inwards as to give the appearance of a false columella; all are slightly spined. There are five cycles of septa, in six systems, and those of the highest orders are small, whilst the primary and secondary are equal and very long. The wall is thick, and the epitheca does not project upwards as a ridge around the circular margin. The endotheca is abundant. The costa are small, and are rarely visible beneath the epitheca. Height of the corallum ths inch. Breadth of the calice Zths inch. Locality. Down Hatherly. In the Collection of R. Tomes, Esq. Division—ASTREACER. Genus—ISASTREA. 1. Isastr#a Tomersit, Dunean. Pl. XV, fig. 20. The corallum is massive, large, and irregular in shape. The upper surface is sub- gibbous. The calices are irregular in size, are separated by very thin walls, and are rather deep and polygonal, quadrangular, or more or less circular. FROM THE ZONE OF AMMONITES ANGULATUS. A7 The septa are very thin, and are faintly dentate ; they often curve and unite. They reach well into the axial space, and are united by dissepiments. They are subequal, but many rudimentary septa exist. There are not four complete cycles of septa. Diameter of calices 5 ths—ths inch. Locality. Long Coppice, near Binton, Warwickshire. In the Collection of R. Tomes, Esq. The delicacy and subequal character of the septa, their deficiency in decided den- tations, and the dissepiments between the septa, characterise this species. There is an immense /sastrea at Inkbarrow, with small calices and thick walls ; unfor- tunately it is not determinable specifically, but the honeycomb appearance and subgibbous upper surface, and the low septal number, may distinguish it. A specimen is in the col- lection of the Rev. P. B. Brodie, F.G.S. Lsastrea Murchisoni, Wright, is found attached to the Inkbarrow specimen, and thus this Scottish Coral has also an English habitat. X. On tHe Corats or THE British aND EvropraNn Lower Liassic Deposits or tHE ZonrES OF AMMONITES ANGULATUS, AMMONITES PLANORBIS, AND AVICULA CONTORTA. The strata of the Lower Lias evidently contain more than one Coral-fauna, and there is a strong distinction between the assemblage of species of the Zone of Ammonites Buck- landi and those of the zones below. ‘The Corals of the White Lias are few in number, and probably belong to the genus Aondlivaltia, but they cannot be distinguished specifically. The Avicula contorta series of France and England are uncoralliferous, but the Italian beds at Azzarola, which probably are on that horizon, contain a very remarkable Coral-fauna. The extent of the area of Coralliferous beds described by Stoppani as the Azzarola series is very considerable. The “ Madrepore-bed,” as it is termed by Stoppani, is seen above the Azzarola beds, with Cardium Rheticum, Myophoria inflata, Mytilus psilonoti, Avicula contorta, Terebratula gregaria, &c., wherever the succession of the rocks can be made out, either on the south-eastern slopes of the Alps, as on the Lake of Como, or on the north- western slopes to the south of the Lake of Geneva.’ The Madrepore-bed is described, more- over, as occurring below and in the midst of the Azzarola beds, and as forming a dense layer of eight to ten yards in thickness. The prevailing Coral is Rhabdophyllia Langobardica, Stop., and the genus is represented by three other species. The Rhabdophylie resemble in their habit of growth many Zhecosmilie, and form in the Azzarola beds great masses of tangle, like Thecosmilia Martini in the Coralliferous beds of the Cote d’Or and of Cowbridge in South Wales. Stoppani describes a Sfylima from some casts which ! Stoppani, ‘Monog. des Foss. de I’ Azzarola.’ 48 resemble those of Astroceenia gibbosa, nobis, from the Sutton Stone. BRITISH FOSSIL CORALS as distinguishable in the Azzarola deposits are— Rhabdophyllia Sella, Stopp. 5 Langobardica, Stopp. - Meneghini, Stopp. 5 De-Filippi, Stopp. Monthvaltia Gastaldi, Stopp. Stylina Savi, Stopp. Thamnastrea Batarre, Stopp. > Escheri, Stopp. ‘4 Meriani, Stopp. rf rectilamellosa, Wink. The species I select These Rhabdophylhe, Styline, and Thamnastree, are represented in the lowest zones the British Lower Lias by Zhecosmilie, Rhabdophyllia, and Astrocenia. The strata between the Trias and the Zone of Ammonites Bucklandi in Germany are very uncoralliferous, and the determinations of the species given by Quenstedt are not suf- ficiently exact. MM. D’Orbigny, Terquem et Piette, and De Fromentel, have noted and described the following species from the Lower Lias, below the Zone of Ammonites Buck- landi, Gryphea incurva, &c., in France and Luxembourg, and by omitting /sastrea basal- tiformis, De From., which belongs to the Zone of Ammonites planorbis, the following table will give all the species from the Continental Zone of Ammonites angulatus :— XI. List or Sprctes FROM THE CONTINENTAL ZONE OF AMMONITES ANGULATUS. 1. Montlivaltia Sinemuriensis, D’Orb. 10 dentata, De From. et Ferry. a Martini, De From. Rhodana, De From. et Ferry. discoidea, 'Terquem et Piette. Haimei, Chapuis et Dewalque. Guettardi, Chapuis et Dewalque. polymorpha, Terquem et Piette. 55 denticulata, De From. et Ferry. . Thecosmilia Martini, De From. (6 12. Michelini, Terquem et Piette. coronata, Terquem et Piette. 2) 2 13. Septastrea Fromenteli, Terquem et Piette. FROM THE ZONE OF AMMONITES ANGULATUS. 49 14 15 17 18. 19 . Septastrea excavata, De From. . Isastrea Condeana, Chapuis et Dewalque. 16. b Sinemuriensis, De From. . Stylastrea Sinemuriensis, De From. Martini, De From. 3? . Astrocenia Sinemuriensis, D’ Orb. 20. 5 clavellata, Terquem et Piette. Probably some of the species of Monélvaltia will have to be absorbed by others, but this list, when added to the Table of British Corals from the Zone of Ammonites angulatus, _ proves that, instead of the Lias being an uncoralliferous series, it was quite the contrary. The great development of Coral-life in the Azzarola series, the scanty remains of it in the western and north-western European Avicula contorta Zones and in the White Lias and in the Zone of Ammonites planorbis, and the luxuriance of the species in the Zone of Ammonites angulatus, ave very significant facts ; and the significance is not diminished when the paucity of the species of the Ammonites Bucklandi Zone and their distinctness from those of the Zone of Ammonites angulatus is considered. XII. List or Species or CoRALS FROM THE CONTINENTAL AND BritisH STRATA OF THE ZoNE OF AMMONITES ANGULATUS. 1. Oppelismilia gemmans, Duncan. Ireland. 2. Montlhvaltia Walle, Be South Wales. SOHMNATE w Murchisonia, ,, a Ruperti, s England. parasitica, 5, South Wales. simplex, PP » brevis, - % pedunculata, ,; » polymorpha, Verquem et Piette. South Wales ; East of France. Haimei, Chapuis et Dewalque. England ; Ireland ; Luxembourg ; France. Hibernica, Duncan. Ireland. papillata, e England ; Ireland. Sinemuriensis, D’Orb. France. dentata, De From. et Ferry. ,, denticulata, ¥. 5 Rhodana, 5 » Martini, Hs > Or Lae) BRITISH FOSSIL CORALS 18. Montlivaltia discoidea, Terquem et Piette. France. 10: — Guettard, Blainville. Luxembourg and England. 20. Thecosmilia Suttonensis, Duncan. South Wales. Pa — mirabilis, - % 22) — serials, Fr Ss 23. = erregularis, p 5 Q4. 7 Terquemi, gi fe 25. _— afinis, 5p is 26. — dentata, ‘ Dy OM fe — plana, a . QS. = Brodiei, _ * 29. _— Martini, EX. de From. South Wales ; England; France; Luxem- bourg. 30. — Michelini, Terquem et Piette. fy . x * ols — coronata, g France OQ: — rugosa, Laube. South Wales; St. Cassian. 33. Rhabdophyllia recondita, ,, 3 x 34. Astrocenia Sinemuriensis, D’Orb., sp. South Wales; France. oD. — clavellata, Terquem et Piette. France. 36. —— gibbosa, Duncan. South Wales ; Azzarola ? of. — plana, i a 38. — insignis, + » 39. — reptans, sf r AQ. —— parasitica, ,, oe 4). — pedunculata, ,, s 42, — costata, ae A3. — Javoidea, a = 4A. == superba, a 6 45. — dendroidea, ,, os AG. —- minuta, ee 55 47. Cyathocenia dendroidea, ,, is A8. == encrustans, ,, MY 49. — costata, Hs SS 50. Llysastrea Fischeri, Laube. * St. Cassian. bil: — Mooret, Duncan. 5 52. Septastrea excavata, BK. de From. 55 France. bd. — Fromenteli, Verquem. m England ; Ireland ; France. 54. Stylastrea Sinemuriensis, K. de From. ,, DY, —- Martini, = 29 56. Latimeandra denticulata, Duncan. = FROM THE ZONE OF AMMONITES BUCKLANDI. 51 57. Isastrea Sinemuriensis, E. de From. South Wales ; France. 58. — Condeana, Chapuis et Dewalque. Luxembourg ; France. 59. — globosa, Duncan. South Wales. 60. — WMurchisoni, Wright. Isle of Skye; Inkbarrow, England. 61. — TZomesii, Duncan. Worcestershire. Of these 61 species 50 are found in the British Isles. XIII. Corats From tHE Zone or AmMmonitEs BUCKLANDI (BISULCATUS). Corals are not numerous in this zone, and the commonest species of the Zone of Ammonites angulatus are not found in any of its strata. It is probable that Thecosmilia Martim, H. de From., which in France ranges from the beds containing Ammonites Moreanus into those in which Ammonites bisulcatus is found, has a corresponding vertical _ distribution in England. Zhecosmilia Michelini, 'erq. et Piette, appears to be present in the Zone of Ammonites Bucklandi, but only in the form of casts, which resemble those found at Abbott’s Wood, in the Zone of Ammonites angulatus. ‘These casts and some of Thecosmilia Martini have been assigned to the genus Cladophyllia, but without sufficient reason. TZhecosmilia is a large genus, and the species contain individuals of all sizes, so that to give to very small cylindroid Zhecosmile the term Cladophyllie is too artificial a distinction. The septa of Zhecosmilie are generally, but by no means universally, regularly toothed, granular, and slightly exsert ; and the septa of Cladophylli@ are said to be small, not numerous, and slightly dentate; moreover, the endotheca is scanty in Cladophylia, but abundant in Thecosmitia. ‘These are not generic distinctions, and it is very probable that one genus will absorb the other. . Srcrion—APOR OSA. Famity—ASTRAID J. Division—LiTHOPHYLLACE® SIMPLICES. Genus—MoNTLIVALTIA. 1. Montiivaittia Gurrrarvi, Blainville, 1830. Pl. XII, figs. 1O—14. The following is the specific diagnosis given by MM. Chapuis and Dewalque.’ Corallum simple and rather variable in shape; often conical, more or less depressed, rarely cylindro-conical ; the base is slightly pedicillate. 1 Chapuis et Dewalque, ‘ Descript. des Foss. des Terr. Second. du Luxembourg, 1854.’ 52 BRITISH FOSSIL CORALS Epitheca thick, ridged, and extending to the calicular border. Calice circular, ordinarily concave, shallow. Septa usually thin, granular; strongly toothed on their arched margins. Five cycles, the first three nearly equal. This Coral varies greatly in its height.and basal flatness. It may be sub-turbinate, or even discoidal ; and the specimen from Bottesford, in Lincolnshire, is flat below and very convex above, but it presents an axial depression. ‘The Continental specimens appear to be found in a lower horizon of the Lias than that in which the specimen figured in Pl. XII was found. | Locality. Bottesford, Lincolnshire. In the Collection of Rev. I. C. B. Chamberlin, F.G.S. There are specimens, which I believe are young forms, that were found at Fenny Compton and Aston Magna. PI. XII, figs. 6 and 7. There is a microscopic Coral at Willsbridge, in the Lima-series (Pl. XV, fig. 9), but the species is not distinguishable. It is figured, as perhaps a larger form may be discovered. Small and young Montlwvaltie are very common on Gryphee and on large Corals. Famitry—ASTRAIDA. Division—Faviacez. Genus—SEPTASTRAA. 1. Szprastr#a EvesHami: Duncan. PI. XIII, figs. 5—7. The corallum is large, tall, and flabelliform, and the surface is subgibbous. The base is small, and the corallites radiate and elongate rapidly. The calices are very irregular in shape and size, and many are twisted and irregular ; all are shallow, and those which are fissiparous are narrow. Some calices are polygonal, but fissiparity can be distinguished in most. The septa are small, dentate, and very irregular in size and arrangement. There are between thirty and forty septa in regular calices, but in the elongated there are many more. ‘The calicular wall is very thin, but where it has been worn a groove is noticed. The endotheca is rather scanty. Diameter of a polygonal calice 3ths inch, and of elongated calices from Ath to ths inch. Locality. Kvesham. In the Collection of the Rev. P. B. Brodie, F.G.S. FROM THE ZONE OF AMMONITES BUCKLANDI. 53 Division—ASTREACEA. Genus nov.—L&PIDOPHYLLIA. The corallum is compound, and the corallites are joined by their walls. The gemma- tion is calicular and gives an overlapping appearance both to the sides and the upper part of the corallum. The epitheca is distinct. There is no columella. The septa are dentate. The calicular gemmation and Astreacean characters distinguish the genus, There are two species ; one is found at Chadbury, and the other in the Island of Pabba, in the Middle Lias. 1. LepipopHytiia Srrickianpi, Duncan. PI. XII, fig. 15. The corallum is tall, and is composed of two sets of corallites. The calicular gemmation is very frequent and successive. he calicular margins are sharp and wavy; and they are free, except where the corallites join. The fossa is deep. The costz are distinct. The epitheca is scanty. Height of corallum 1 inch. Breadth of calice ths inch. Locality. Chadbury, Worcestershire. In the Collection of Mrs. Strickland. The specimens were collected by the late Hugh Strickland, F.G.S. Genus—iSASTRAA. 1. Isastrma ENDOTHECATA, Duncan. Pl. XII, figs. 17—21. The corallum is large, and either massive and flat, or tall and arising from a small base. The calices are very irregular in shape, size, and depth. The largest calices are very deep. The septa are small, and often wavy. ‘They are not exsert, but are very irregular. They are faintly dentate, wide apart, and project slightly from the calicular wall. 8 54 BRITISH FOSSIL CORALS The cyclical arrangement cannot be determined by the number of the septa; there are between four and five cycles. The largest septa reach the floor of the calice, where they join. The endotheca is greatly developed, and it forms small dissepiments, and others which stretch across the corallites almost like tabule. The marginal gemmation is frequent. Length of the largest calice } ch. Depth !— inch. Locality. Lyme Regis. In the Collection of R. Tomes, Esq. 2. Isastraa InsieNnis, Duncan. PI. XIII, figs. 10, 11. The corallum is massive and forms a flat mass. The corallites are very equal in size and regular in shape. The calices are placed very regularly in linear series; they are shallow, open, and are separated by a stout wall. ‘The calices are generally hexagonal, but many are square. The septa are small, project but slightly from the wall, are dentate and unequal. There are four cycles of septa in six systems in the largest calices. The primary and secondary septa are nearly equal; the tertiary are decidedly smaller, and the rest are the smallest. The endotheca is close. There is no columella. Diameter of largest calices ths inch, and of the usual size 3,ths inch. Locality. Lyme Regis. In the Collection of R. Tomes, Esq. This is a very well-marked species, and belongs to a section which comprises /sastrea Henocquei, Ed. and H., from the Lower Lias of Hettange, /sastrea polygonalis, Michelin, sp., of the Muschelkalk, and /sastrea Lonsdalei, Kd. and H., of the British Inferior Oolite. 3. Isastr@A Srrickianpi, Duncan. Pl. XIII, figs. 1—4. The corallum is very tall, has a small base, and is expanded superiorly. The corallites are unequal in size and length, and vary much in shape. The calices are very irregular in form and depth ; their walls are thick, and the septa stout and very dentate. ‘The dentations are blunted and are very large, and more so internally than near to the calicular margin. The septal number varies, and 32—40 appear to be the usual number. ‘The lamin are stout, and the primary and secondary septa reach downwards to the base of the fossa and are dentate. ‘Phe others, which are short, are also stout. — ss: FROM THE ZONE OF AMMONITES BUCKLANDI. 55 The endotheca is greatly developed, and consists of small vesicular dissepiments, and of larger masses which stretch across the corallites like tabulz and close in the calicular fossa. Height of corallum 6 inches. Breadth of largest calice 3,ths inch. Locality. Chadbury, Evesham. In Mrs. Strickland’s Collection. Genus—CYATHOC@NIA. 1. Cyarnocenta GLososa, Duncan, Pl. XIII, figs. 8, 9. The corallum is nearly spherical. The calices are numerous, small, and shallow. They are rarely circular, and are generally rather polygonal in outline, and they are separated by a small amount of ccenen- chyma. ‘There are no costz. The septa are stout at the wall and taper off inwardly ; they are subequal, distant, and form three more or less perfect cycles in six systems. Diameter of the calices {th inch. Locality. Fladbury, in Drift with Gryphea incurva. In the Collection of R. Tomes, sq. The shape of the corallum, the absence of costee, and the shallow calices, distinguish this species from Cyathocenia dendroidea, nobis, C. costata, nobis, and C. ncrustans, nobis, from Brocastle and the Sutton Stone. The following analysis of the genus will enable the diagnosis of the species to be determined readily. branching, having costz : ; p . C. dendroidea. encrusting, without costa, coenenchyma granular . CG. incrustans. flat, having large costee and a deep calice . C. costata. CyaTHOC@NI/” with the ae | globular, without coste, cconenchyma plain . . C. globosa. XIV. List or Specizs rrom tHe Zonet or AMMONITES BuckKLanDt. 1. Montlvaltia Guettardi, Blainville. 2. Septastrea Eveshami, Duncan. 3. Lepidophyllia Strickland, ,, 4. Lsastrea endothecata, - 5 » Msignis, - 6 » Strickland, a 7. Cyathocenia globosa, Y 56 BRITISH FOSSIL CORALS XV. CoraLs FROM THE ZoNE or AMMONITES OBTUSUS, Sow. Some worn and light-coloured simple Corals of the genus MZontliva/tia are found at Pebworth, five miles south-west of Stratford-on-Avon, in a bed with Ammonites Sauzeanus, D’Orb., and Ammonites semicostatus. One of the species (Montlivaltia mucronata, Duncan) will be described amongst those of the next zone, in which it is common. ‘The specimens are worn, the calices especially, and all the spines are broken off. The columellary space — is occasionally occupied by the prominent ends of the principal septa, the lamin having been worn away in their middle course. The longitudinal series of coste are rarely visible, and there are many examples of deformed corallites. Famity—ASTRAIDA. Division—LitHoPpHYLLACE® SIMPLICES. Genus—MONTLIVALTIA. 1. Monrnivantia PaTuLA, Duncan. Pl. XV, figs. 6, 7, 8. The corallum is turbinate, depressed, and slightly longer than broad. The calice is large, elliptical, very shallow, and open; its margin is sharp, and the wall shelves very gradually inwards, giving to the calice a very open appearance. The septa are unequal and numerous, and the largest are very long and dentate ; the tooth nearest the axial space points inwards and is rounded, and those of the longest septa form an irregular circle around the space. The smallest septa are very rudimentary, but the next in size have, in common with all the others, an internal oval tooth. All the septa are delicate, and they are not crowded. ‘There are five cycles of septa in six systems. The primary, secondary, and tertiary are nearly equal in length. ‘The septa are not exsert, but all are lower than the calicular margin. Length of the calice {ths inch. Breadth ¢ths inch. Locality. Walford Hill, Stratford-on-Avon, with Ammonites semicostatus and Ammonites Sauzeanus. In the Collection of R. Tomes, Esq. FROM THE ZONE OF AMMONITES RARICOSTATUS. 57 XVI. Corats From THE ZONE OF AMMONITES RARICOSTATUS, ZIET. The brick-fields in the vicinity of Cheltenham present dark-coloured clay beds, which have the following succession (see Wright, ‘ Fossil Oolitic Asteriade,’ p. 25). Marle Hill Section. NO. ED. IN. 1. Gryphea-bed; a hard, ferruginous clay, which broke into frag- ments, and contained Gryphea obliquata, Sow... . 38 to4 0 2. Coral-band; a thin seam of lightish-coloured unctuous clay, containing a great many small sessile Corals, Jontlivaltia rugosa, Wright and Duncan, most of which appeared to have been attached to the curved valves of the Gryphaw Lin. to 13 3. LHippopodium-bed . ; 10 0 4. Ammonite-bed , : : i In Warwickshire the railway-cutting at Honeybourne presented the same beds, and the Coral-band contained a considerable number of the Monthivaltia. A section on the line of railway at Fenny Compton, in Oxfordshire, near the station, presents the following beds in descending order; the bed No. 2 is highly coralliferous.? Fenny Compton Section. NO. FT. IN. 1. White clay, containing Gryphea obliquata (Maccullochii ?), G. incurva, Belemnites acutus, Hippopodium ponderosum, Pleu- rotomaria similis, &c. : ' 4 4 0 2. Blue clay, with included hard blue calcareous bands, containing Corals and the Mollusca mentioned in Bed No. 1 ; 2 0 3. Blue shale f : : : 10 0 Middle Lias clays and shales, with Ammonites Henleyi, are superimposed on Bed No. 1; and the blue shale (3) rests on a clay with calcareous masses, the ‘“ Cardinia- zone.” The Coral-bands at Marle Hill and Honeybourne are upon the same geological horizon as bed No. 2 of the Fenny Compton section. These beds contain some of the finest specimens of Montlivaltia ever discovered. ‘1 The Rev. P. B. Brodie, M.A., F.G.S., has given me great assistance, and has furnished me with this section. a8 BRITISH FOSSIL CORALS Famity—ASTRAID A. Division—LitHOPHYLLACEH SIMPLICES. Genus—MOoNTLIVALTIA. The Aonthvaltie from Fenny Compton, Honeybourne, and Cheltenham, belong to several species, and two of these are singularly polymorphic. Shape has not very much to do with the specific diagnosis of some recent simple Corals, and it is necessary to assert this m collecting under one fossil species Corals of very diverse external forms. Singularly enough, the Liassic J/ontlivaltie from the Zone of Ammonites raricostatus are common and are extraordinarily well preserved, although a few years ago a Liassic Coral was excessively rare. Even the ornamentation upon the dentations of the septa is pre- served, and the longitudinal striations of the epitheca also. ‘The Fenny Compton Coral-bed contains specimens of the species of all sizes, and this is the case with the deposits containing the so-called Zhecocyathus rugosus, Wright, at Honeybourne and Cheltenham. At Pebworth the Fenny Compton species are not found in a dark blue matrix, but in a white deposit ; moreover, the specimens are usually worn, and they appear to have grown under less favorable conditions than the others. Thecocyathus rugosus is referable to a group of forms specimens of which are very common ; it does not belong, however, to the same family that contains the Zhecocyathi. Dr. Wright gave the species a name in his MS., but the description and diagnosis have never been published. The Corals have been associated so long with the name of Dr. Wright, that it is just to append his name to the species. 1. Montiivattia Rucosa, Duncan and Wright. Pl. XIV, figs. 1, 2, 3; Pl. XV, figs. 14, 16, 17; Pl. XVI, figs. 5—15. THECOCYATHUS RUGOSUS, Wright, MS. ‘ The corallum is very variable in its shape ; it may be tall, conico-cylindrical, and curved, sub-turbinate and curved, short and cylindrical, short and turbinate and curved, or straight. It is pedunculate, and the scar by which it adhered to foreign substances, such as shells, is large and oval, or small and very irregular in shape. The epitheca is stout and identified with the wall; it is strongly ridged transversely and folded as well as grooved. It is rarely marked by longitudinal lines, and is usually deficient in ornamentation. When the epitheca is worn, the external ends of the septa are seen like costa, and the oblique external terminations of the endotheca are very apparent, sometimes having a herring-bone pattern. Young corallites are often adherent to the FROM THE ZONE OF AMMONITES RARICOSTATUS. 59 epitheca, they are therefore not buds, but accidentally attached Corals. When more than one Coral is attached to the same shell the bases appear to join. The calice is shallow, and its margin is formed by the epitheca, which often intrudes upon its periphery ; it is circular or slightly deformed, and it may be either contracted or very open. The septa are numerous and unequal; they are irregular in size and in their arrange- ment ; they are dentate, and the teeth are regular, rounded above, and ornamented with waving lines and are largest near the axial space. The worn septa show their bases in the form of oval swellings, and when these are of full size the appearance of a colu- mella and pali is simulated. ‘There are four perfect cycles of septa, and the fifth is very irregularly developed, the higher orders being often rudimentary. In some large calices the fifth cycle is complete. The endotheca is strong and well developed, and its dissepiments are numerous, oblique, and arched. Height of Coral 18ths inch, 2 inches, 1ths inch, sths inch. Breadth of calice ths inch, 2ths inch, {ths inch, {ths inch. Locality. ippopodium-bed and Coral-bed of Marle Hill, Honeybourne, and Fenny Compton. In the Collections of the Geological Society, British Museum, Dr. Wright, F.G.S., Charles Moore, Esq., F.G.S., R. Tomes, Esq., and Rev. P. B. Brodie, F.G:S. - The ornamentation of the teeth of the septa is very well seen in some specimens, but usually it is worn off and the teeth also. The Coral, although very polymorphic, is very easily distinguished from all others by its septa, epitheca, and base. 2. MontiivaLT1a MucRonATA, Duncan. Pl. XIV, figs. 4—11 and 14—16; Pl. XV, figs. 10—13. The corallum is very variable in shape; it has a small peduncle, and a small and more or less circular flat scar. ‘The corallum is turbinate and symmetrical, and widely open at the calice; or more or less compressed and subturbinate ; or cylindrical and com- pressed. When turbinate and with a circular calice, the calice is singularly shallow ; but when cylindrical and compressed, or in the young state, the calice is deeper. The epitheca is strong and rises up with the wall to produce a sharp margin to the calice. The transverse markings are very distinct, and there are constricting ridges and folds. The longitudinal markings are very distinct, ornamental, and symmetrical; they are in groups which are smallest at the base, where they are most distinct and rounded, but they are less distinct at the calice where they are flat. The groups dichotomize, so that there are usually 12 at the base and 24 at the calice; they are separated by well-marked grooves and consist of bundles of longitudinal epitheca swellings and cost. The calice is either very shallow and circular, or deep and circular, or deep and 62 BRITISH FOSSIL CORALS. 6. Monthvaltia radiata, Duncan. 7. Septastrea Hveshamz, 8. Lepidophylha Strickland, 9. Isastrea endothecata, > 9 10. — signis, dole — Strickland, 12. Cyathocenia globosa, 29 $9 rr) CORALS FROM THE MIDDLE LIAS. XIX. Corats rrom THE ZONE OF AMMONITES JAMESONI, Sow. Dr. Wright notices that this zone is well developed in the Island of Pabba, near Skye, in the Hebrides, and the remarkable Coral about to be described appears to form a bed there of some extent.’ Famity—ASTRAACEA. Genus—LEPIDOPHYLLIA. ]. LeprpopHyLuia Hesripensis, Duncun. Pl. XVI, figs. 1—4. The corallum is flat, and the corallites are short. The calices vary in size and number; they are open and shallow, and are crowded with delicate, unequal, and not prominent septa. ‘The septal arrangement is very irregular. ‘The lamine are dentate and narrow, and the largest approach the axial space. In calices of ordinary size there are four cycles of septa, and part of a fifth in some systems, whilst in the largest calices the fifth cycle is complete. The epitheca on the free wall of the corallites, where they overlap those below them in the general imbrication, is smooth. ‘lhe calicular gemmation occurs centrally, and also near the margin. Height of the corallum ;ths inch. Breadth of the calices +ths— ths inch. Locality. Pabba shale. In the Collection of the School of Mines, Jermyn Street. ' See note 1, page 41, Part IV, No. 1. FROM THE ZONE OF AMMONITES HENLEYI. 63 XX. Corans FROM THE ZONE or AMMoNItTES HeENLEYI, Sow. A great number of specimens of all sizes of a very polymorphic Monélivaltia have been found on the surface of the fields at Cherrington, near Skipton, and in a water-course or ditch section of the Middle Lias close by. Ammonites Henleyi, Ammonites Chiltensis, Car- diniu attenuata, and Cardinia elongata, were found with the Corals. Famity—ASTRAEID At. Division—LitHOPHYLLACE® SIMPLICES. Genus—MONTLIVALTIA. 1. Montiivattia Victorit#, Duncan. Pl. XVII, figs. 1—10. The corallum grows to a great size, and generally presents a scar where it was formerly attached. ‘The shape of the corallum is very variable, and it may be short, turbinate or sub-turbinate, or long and conical, or rudely cylindrical. The corallum is rarely straight, and generally there is a very decided curve in it and a twist also; more- over, there is frequently a constriction just below the calice, and at this point also there is generally a curve. The calice is either widely open or contracted and small; it is never very deep, but may be characterised either by exsert and rounded septa or by septa which dip at once into a concave fossa. ‘The outline of the calice is usually circular, and slightly compressed. The margin is sharp and is formed by the epitheca. The septa are numerous, crowded, long, and the principal often extend to and across the axial space, which is rather elongated. The lamine are not much thicker at the wall than elsewhere, and the dentation is more distinct close to the wall. There are six cycles of septa, in six systems, the highest orders being very small. The epitheca is very dense and is strongly marked with transverse elevations and de- pressions ; where it is worn away, the septal ends are seen, like coste with transverse dis- sepiments connecting them. ‘The wall is very thin, and appears to be identified with the epitheca. The endotheca is very abundant, thick, curved, and branching. Height of various specimens 5 inches, 3ths inches, 2%ths inches. Breadth of speci- mens 2 inches, 2ths inches, 2 inches. Locality. Cherrington, Skipton-on-Stour. J 64. BRITISH FOSSIL CORALS In the Collections of the Geological Society, British Museum, Rev. P. B. Brodie, F.G.S., R. Tomes, Esq., &c. This is the largest simple Coral of the British Fossil Coral-fauna, and is readily distin- guished. Its variability of shape almost equals that of Monthvaltia rugosa, Wright. ‘There are some fragmentary Corals in the Marlstone, but their genera are doubtful ; and the cast of a Montliva/tia was found by Mr. Charles Moore at Wells, but I cannot de- termine the species. The Corals from the Middle Lias are— 1. Lepidophyllia Hebridensis, Duncan. 2. Montlivaltia Victoria, : XXI. Toran NuMBER oF Sprctes OF MADREPORARIA WHICH CAN BE DISTINGUISHED IN THE LiAs oF THE Britisu IsLANps. In the zone of Ammonites planorbis . . . 2 00 ‘5 — angulatus . . . 50 » » — Bucklandi.. .. 4 - 5 a= obfusysy nicl ky otal ‘3 55 oa raricostatus 4 : 55 —— SAMCION si ctoikve ye DM 99 % — Flenleyt . l 66 From the Upper Lias described by MM. Milne-Edwards and Jules Haime®. . . 1 Motalle «if tov 67 Lower Lias. . . 64 Middle lias . . 2 Upper Lias . ] Total 7 " See page 65. Some are common to this and the next zone. 3 «Brit. Foss. Corals,’ Pal. Soc. FROM THE ZONE OF AMMONITES PLANORBIS. 65 XXII. Corats FRoM THE ZONE oF AMMONITES PLANORBIS.- Division—AstTREACE A. Genus—IsASTREA. 1. IsASTRHZA LATIMZANDROIDEA, Duncan. Pl. XV, figs. 18, 19. The corallum is massive, and has an angular and rather gibbous upper surface. The corallites are long, and their united walls are thick. The calices are very irregular in shape, and although some are small and polygonal, others are more like the serial calices of the genus Zatimeandra. The calices are deep, and gemmation takes place quite on the margin. The septa are numerous, very unequal, and there is a very small septum between the larger. The larger septa are very dentate, and the tooth near the axial space is very distinct, especially in the long calices. The larger septa are not very unequal, do not project much into the calice, and the axial space is left very free, but is closed by endotheca. ‘The existence of the small rudimentary septa makes the septal number very irregular, and the long serial calices contain very variable numbers of septa. ‘The endotheca is strongly developed, is vesicular, and closes in the corallites. Diameter of ordinary calices ths inch to ths inch. Diameter of serial calices %ths inch. Length of serial calices 3ths— XXIV. Coraus From THE ZonE oF AVICULA CONTORTA AND THE Wurtte LIAS. (The Rheetic series, JZoore.) It has been noticed that but one fossil which could be referred to a Coral has been discovered in the Zone of Avicula contorta in England. ‘The specimen is said to belong to the genus Monthvaltia, and to have a high septal number. The deposits containing FROM THE WHITE LIAS. 67 Avicula contorta mn England, Wales, and Ireland, are not of that character in which Corals would be usually found; but the Azzarola beds of Lombardy are, as has been already noticed, highly coralliferous. The Montlivaltia from the British Avicula contorta series is, however, of some importance as a species, for it is the oldest Secondary form, there being no Madreporaria between it and the Carboniferous fauna except the few species of the Permian. The White Lias, which was deposited under very different conditions to the Avicula contorta series, contains two genera of Corals, but the species are indeterminable, on account of the specimens being either m the form of casts or so altered by a destructive mineraliza- tion as only to present sections of their septa and part of the epithecal covering. The White Lias of Watchet contains Montlvaltie and stunted conico-cylindrical Thecosmlie. A species of this last genus has its wall and epitheca very well shown (fig. 3). Fig. 3. Thecosmilia, from the White Lias of Watchet. No Thecosmilia from the White Lias can be determined to belong to the species Michelini or Martini, but there is a cast of a Coral in the White Lias of Sparkfield which has some resemblance to casts of Thecosmilia Terquemi, Duncan. Fig. 4. Cast of a Thecosmilia, from the White Lias of Sparkfield. Several specimens, probably, of Montlivaltie, from the White Lias of W arwickshire, are only distinguishable by the radiating septal laminee (fig. 5). 68 BRITISH FOSSIL CORALS. Fig. 5. Montlivaltia, from the White Lias of Warwickshire. There is a great MJontlvaltia in the Leamington beds, which is elliptical and very large at the calice. It is only found in the form of casts, one of which is here figured. Fig. 6. Cast of a Montlivaltia from Leamington. A cast of a multiseptate discoidal Monthvaltia is found at Punt Hill, Warwickshire, and I believe it to belong either to Montlivaltia Haimei, Chapuis et Dewalque, or to one of its varieties which have been noticed in the part No. 1 of this description of the Corals of the Lias. It is figured below. Fig. 7. Montlivaltia, from Punt Hill, Warwickshire. The great vertical range of this Montlivaltia has already been noticed. When the local nature of the White Lias is appreciated, and it is acknowledged as “a passage-bed” between the Zone of Avicula contorta and the beds containing Ammonites planorbis, the discovery of these Corals, which in the East of France and in Luxembourg are found in the Ammonites planorbis series, and in that of Ammonites angulatus, will not be exceptional. FROM THE ZONE OF AMMONITES ANGULATUS. 69 XXV. APPENDIX. Notre ON THE AGE OF THE Sutton STONE AND Brocastuz, &c., Deposirs. A long and very elaborate essay, by Mr. Charles Moore, F.G.S., has been read before _ the Geological Society, and published in the ‘ Quarterly Journal’ of that society, with the title, “On Abnormal Conditions of Secondary Deposits when connected with the Somerset- shire and South Wales Coal-Basin, and on the Age of the Sutton and Southerndown Series,” which suggests that it is more or less controversial; but although this is the case, still it has great intrinsic merits. Mr. Bristow, F.R.S., read a paper before the Geological Society, which appeared in its ‘ Quarterly Journal,’ ‘On the Lower Lias or Lias-conglomerate of a part of Glamorgan- shire.’ Like Mr. Charles Moore’s communication, it is very valuable, besides being con- troversial. Lately also Mr. R. Tate, F.G.S., in his essay “On the Fossiliferous Develop- ment of the Zone of Ammonites angulatus, Schlot., in Great Britain,” has produced a paleeon- tological criticism which refers in one part to the “abnormal deposits” and “the Lias- conglomerate.” Hach of these essays refers to the characters and to the age of the Sutton Stone, whose Madreporaria have been described in this Part. Mr. Bristow considers the Southerndown series of Mr. T'awney' to be a portion of the Sutton Stone or ‘ Lias-conglomerate,’ and asserts that Mr. Tawney has made a great error in his section of the sea face of the deposit by giving it too great an elevation. Mr. Bristow also considers the Sutton Stone to be Lower Lias, and that the usual Gryphea incurva occurrmg in abundance renders his opinion incontrovertible. Mr. Moore, on the contrary, admits the correctness of Mr. Tawney’s section, but considers that insufficient altitude has been given. He considers that, as Ostrea Liassica (O. irregularis) occurs high up in the series as a Ammonites planorbis is wanting, the Sutton Stone is in the “Ostrea division” of the Ammonites planorbis Zone. Myr. Moore places the Brocastle deposit in the Ostrea series. He insists upon the presence of Gryphea incurva in the Sutton Stone and in the deposit at Brocastle “in abundance,” and localizes the deposits in the Lower Lias.” Mr. Tate proves what I had already demonstrated*—that Mr. ''awney placed the Sutton Stone too low down in the geological scale ; and, after a survey of the beds above the White Lias in Ireland and England, he considers that the Ammonites planorbis Zoue is * ce zone, and as ! Tawney, ‘Quart. Jour. Geol. Soc.,’ vol. xxii, p. 91. 2 A paleontological combination of the forms of the lower part of the Zone of Ammonites planorbis with Gryphea incurva would indeed be incredible. 3 P. Martin Duncan, ‘Quart. Jour. Geol. Soc.,’ Feb., 1867 ; and in the Ist No. of this Part. 10 70 BRITISH FOSSIL CORALS. so mixed up with that of the Ammonites angulatus that it had better disappear from British geology. Mr. Tate, however, supports indirectly the geological position I have given, from the study of the Madreporaria, to the Sutton Stone and Brocastle deposits. I agree with Mr. Bristow, or rather he agrees with me, as I was first in the field, that the Sutton Stone is what is usually called Lower Lias ;' but I dispute the possibility of — associating the Sutton Stone, Brocastle, and other equivalent deposits, including, of course, the Coral-bed of Cowbridge, with the strata composing the Ammonites Buckland Zone in the same division of a great formation. The word “ Infra-Lias,’” which refers to the deposits below the Ammonites Buckland series, does not assume separation from the Lias, and, although Low, Lower, and Lowest will apply to some places, it would rather confuse a geological series. To combine in one division of the Lias, under the term Lower, such zones as those of Ammonites raricostatus and Ammonites planorbis is to associate widely different faunee. There are many species which have a great range in this division of the Lias, but there is a clear paleontological distinction to be drawn in the British Isles, in France, Luxem- bourg, and in Germany, between the faune of the Zone of Ammonites Buckland and of those below. Ostrea irregularis (O. Liassica) is a shell so widely distributed, and has so great a vertical range, that it is of no value in fixing a geological horizon. It must be considered in relaton to the fauna associated with it; and the forms found in the Sutton Stone in company with this variable Oyster are not those which elsewhere characterise the Ostrea beds of the A. planorbis Zone. I have examined the Gryphee, and do not consider them typical zzewrve. The cha- racters of the Molluscan and Madreporarian fauna which I have already pointed out, and the affinities and grouping of the species, induce me to retain my opinion that the Sutton Stone, the Brocastle, Ewenny, and Cowbridge deposits are on one geological horizon, and still to assert that they are the equivalents of the French and Luxembourgian Zones of Ammonites angulatus. The deposits have a different Coral-fauna to the corresponding beds in the east of England, where simple ontlivaltie indicate different external conditions, but not a difference in time. CoraLs FROM THE Upper Lias. MM. Milne-Edwards and Jules Haime described Thecocyathus Moorei, Ed. and H., from the Upper Lias of Ilminster. Mr. Charles Moore has sent me specimens from Lans- down, near Bath. Thesame excellent collector has a fossil, probably a Sponge, with ‘ Sir Henry de la Beche was the first to pronounce the Sutton Stone to belong to the Lower Lias. FROM THE ZONE OF AMMONITES ANGULATUS. 71 markings upon it like those of a cast of the calice of a Coral; it is from Ilminster. Trochocyathus primus, Ed. and H., is too doubtful a species to be admitted into the Liassic Coral-fauna at present. I have to acknowledge with many thanks the great assistance I have received in completing this Monograph of the Corals of the Lias from Mr. H. Woodward, of the British Museum, Mr. R. Etheridge, of the School of Mines, and Mr. R. Tomes, besides those gentlemen whose collections have been placed at my service. (See Preface to Part IV, No. 1.) ERRATUM. In the Preface to Part IV, No. 1, “ Trochocyathus DMooret, Ed. and H.,”’ should be “ Thecocyathus Moorei, Ed. and H.”’ INDEX OF THE Astroccenia costata a dendroidea . ae favoidea an gibbosa 5 insignis e minuta op parasitica = pedunculata 3 plana ~) reptans i Sinemuriensis E. superba Astrocceniz, notes on the 7 scheme of the Cyathoccenia costata Be dendroidea 3 globosa a incrustans Elysastreea Fischeri s Moorei Isastreea Condeana As endothecata a globosa $5 insignis 5 latimeeandroidea 3 Murchisoni Sinemuriensis . - Stricklandi eS Tomesii SPECIES, ETC. DESCRIBED AND NOTICED IN Part IV, No. 1 anp No. 2, PAGES 21, 50. 22, 50. 21, 50. 18, 48, 50. 19, 50. 22, 50. 20, 50. 20, 50. 19, 50. 20, 50. 22, 33, 34, 49, 50. 21, 50. 26. 23. ee - bo oOo pw Ww @o or N io) DHA co i) . bo or i=) 29, 50. 30, 50. 49, 51. Seah BD, Oe 31, 48, 51. 04, 595, 62. 65, 66. 41, 47, 51. 30, 33, 34, 51. 54, 55, 62. 46, 51. CORALS FROM THE LIAS. Latimzeandra denticulata Lepidophyllia, genus of . A Hebridensis a Stricklandi Montlivaltia brevis % dentata 5 denticulata r discoidea . Gastaldi ‘ Guettardi ae Haimei te Hibernica . is Martini As mucronata . a5 Maurchisoniz ae nummiformis Re papillata ss parasitica ss patula : pedunculata 3 polymorpha 3 radiata eS Rhodana 5 rugosa c. Ruperti Pe simplex 55 Sinemuriensis i. Wallize A Victorize Oppelismilia gemmans PAGES 32, 43, 50. 41, 53. 62, 64. 53, 55, 62. 10, 49. 48, 49. 48, 49. 48, 50. 48. 48, 50, 51, 55, 61. 35, 40, 48, 49. 39, 40, 49. 48, 49. 59, Gl. 8, 49. 60, 61. 36, 38, 40, 49. 9, 49. 56, 61. 10,49. 8, 33, 48, 49. 61, 62. 48, 49. 57, 58, 61. 46, 49. 9, 49. 48, 49. 7, 49. 63, 64. 39, 40, 41, 49. -Rhabdophyllia De-Filippi . ae Langobardica # Menighini a recondita A Sellee "55 excavata |, Fromenteli . > Haimei ‘Stylastraea Martini - ,, Sinemuriensis ‘Stylina Savii ty Escheri ” Meriani ‘¢ rectilamellosa INDEX OF THE SPECIES. A7, 48. 48. 17, 42, 50. 48. 52, 55, 62. 32, 33, 49, 50. 37, 40, 48, 50. (Oe 49, 50. 49, 50. 48. 48. 48. 48. 48. Thecocyathus Moorei Ss rugosus Thecosmilia affinis > Brodiei oa coronata » - dentata es irregularis . - Martini 3 mirabilis y plana mE rugosa | 6 serialis 3 Suttonensis | 33 Terquemi | Trochocyathus primus os Michelini . 73 PAGES Tak 57, 58,68, 16, 42, 50. 13, 42, 50. A8, 50. 16, 42, 50. 15, 42, 50. 14, 33, 45, 48, 50, 51. 14, 33, 45, 48, 50, 51. 12, 42, 50. 17, 42, 50. 13, 50. 12,42, 50. 11, 42, 50. 16,42, 50, 65, 66. 7%. 11 Oe Bs oe rl ru ‘Pre te ie ee, ee eee eee co a 4 cs y pail, et gs eee ANN, - PLATE XII. CORALS FROM THE LOWER LIAS. Thecosmilia Martini, H. de Fromentel. (P. 45.) Part of its epitheca magnified. Monthvaltia Ruperti, Duncan. (P. 46.) Its calice. The calice, magnified. A young specimen of Montlivaltia Guettardi, Blainville. (P. 51.) The calice, magnified. ~ A full-grown specimen. (P. 51.) Its calice, magnified. The septa, magnified. The calicular surface of a young Thecosmilia Martini. The same, magnified. Lepidophyllia Strickland, Duncan. (P. 53.) A cast of Thecosmilia Michelini, Terquem. (P. 51.) Part of the corallum of /sastrea endothecata, Duncan. (P. 53.) A regular calice, magnified. A side view of a magnified calice, showing the endotheca in the calice. Oblique view of some calices, magnified. Endothecal dissepiments connecting the septa, magnified. ehh RS LPO P EI "€ ak i ‘Ap My Do ye | " f on : / ‘ ‘ i ‘ i ‘ = € ra ’ ‘ 4 ‘ ; y fe ‘ PLATE XV. CORALS FROM THE LOWER LIAS. . The upper calicular surface of Montlivaltia radiata, Duncan. (P. 61.) . The under surface. A side view of the corallum, showing the central depression of the base. . The calice, magnified, showing the quaternary arrangement of the septa. . The base, magnified, showing the pellucid epitheca and the coste. . The corallum of MWontlivaltia patula, Duncan. (P. 56). . The calice, magnified. . Part of the wall and one of the septa, magnified, showing the direction of the teeth. . A very young Montlivaltia, of an unknown species. The calice magnified (P. 52.) . A cornute variety of J/ondlivaltia mucronata, Duncan. (P. 60.) . One of its septa, magnified. . A portion of the external surface of the type of Monélivaltia mucronata, Duncan, showing the dichotomous longitudinal bundles of costa, magnified. . A conical variety of Montlivaltia mucronata, Duncan. . Asection, magnified, of Monthvaltia Ruperti, Duncan. . A large and unusual shape of Monthvaltia rugosa, Wright. . A side view of its dentate septa. . A dentate process, magnified, showing the ornamentation. . The corallum of Jsastrea latimeandroidea, Duncan. (P. 65.) . Its septa, magnified. . The corallum of Jsastrea Tomesii, Duncan. (P. 46.) iin Ax ID LU V/ = i r vad eet ‘ y dar 2 ¢ ‘ F he ye ae . i i" oe ¥ i y Yin Ay), ’ ‘ g : ‘ m4 ‘ . \e ‘ , . . \ fy = ; i Pa . : > ‘ r € . = j ‘ : ‘ ‘ . . . ] f > tly if 7 ‘ . 4 ‘ = > ‘ ' PLATE XVI. CORALS FROM THE LOWER AND MIDDLE LIAS. Fig. 1. Lepidophylha Hebridensis, Duncan. Natural size. (P. 62.) 2. Calices, magnified, showing calicular gemmation. 3. Side view of corallites, showing the epitheca. . Septa, magnified. Montlivaltia rugosa, Wright, sp. (P. 58.) Thecocyathus rugosus, Wright, MS. / Le: 13. 14. 8. Septal ends (external) and intermediate endotheca (magnified). 15. A section magnified, showing the strong and arched endotheca between the 3 4 D: 6. 7 ‘Common forms of \ 9 0 Unusual and young forms of the same species. septa. DewWilodle lith afd rr “s we PLATE XVII. CORALS FROM THE MIDDLE LIAS. . Montlivaltia Victorie, Duncan. Nat. size. >2) ap and slender; very un- of aw iiee aS esaD usual form spine. nda | | Total length of longest spine. Jen Almost cylindrical, uniformly but not slender; very rare. Slightly tapering ; rare. Pear-sbaped, longi- tudinal elliptical ; rather section rare. 4. ambitus | | | | if t | Diameter Diameter of spine at | at greatest neck. thickness. Length of neck before swelling commences. Form of apex. Diameter of test at ambitus. Hemisphe- rical. Hemisphe- rical. we Pear-shaped, longi- tudinal ovate ; section rather rare. Je Subacute. Uncertain, about 1:0 Uncertain. About °6 | | Subacute. Sub-hemisphe- rical. He Na) FROM THE WHITE CHALK. 73 “Table showing variation of form in the spines of Cidaris clavigera—continued. Total length of | Diameter | Diameter longest | of spine at | at greatest ambitus neck. thickness. spine. Length of neck before swelling commences. Ambitus Spine, characteristic of the general form of the whole of the spines attached to any individual test of C. clavigera, nat. size. Diameter of test RUD OOS G at ambitus, Club-shaped, stem slightly tapering; common. 6. Hemi- spherical. About 1-6 Club-shaped, stem cylindrical; com- mon. Hemi- spherical. a 1:0 2 3 6 7. Club-shaped, _ por- ti t th ion nearest the Beak: 13 ‘I “4 4 spherical. apex constricted ; common. 8. Club-shaped, _ por- tion nearest the apex constricted ; rather rare. 9. 2 sf Hemi- 8 ao . spherical. Constricted at about half the length ; very rare. 10. oi) 1 Be i | Acute. 6 10 74 CIDARIS “From the above Table it will be seen that this variation in form is independent of the size of the test, and is dependent rather upon some peculiar law in the formation of the spine, or some cause which has contributed to produce a greater development of calcareous matter in one part than in another. In flints which contain the spines of C. cla- vigera a fracture passing through the spine will often exhibit this growth very beauti- Fie. 1. fully ; thus, in the woodcuts (fig. 1), whilst an earlier form of e. the spine is clearly defined, the subsequent addition of material is also manifested by the change of tint. The same effect can also be observed in longitudinal sections of the ordinary spines, a difference of density and of hardness in the whole or parts of the enveloping layers being very apparent. Sections of body-spines of C. clavigera in flint. “Tn C. clavigera the difference between the spies of the peristome and of the ambitus is more marked than in most of the other species of the Ciparip# of the Upper Cretaceous group. The woodcuts (fig. 2) drawn from the spines of the tubercles adjacent to the mouth, and magnified four diameters, show that their apex is more acute, their ridges fewer, and their body more elongate-ovate than in the larger spines of the ambitus. The spines of the granules (fig. 3) are also dissimilar, being longitudinally striated, contracted at intervals, having almost parallel sides, and being in transverse section ovate. Fig. 3. Spines of C. clavigera from the tubercles adjacent to the peristome ; Spine of C. clavigera from the granules on the magnified four diameters. One spine has four serrated ridges, margin of the ambulacral areas; magnified the other six. eight diameters. «The spines in their original condition appear to have been tinted with parallel bands of colour, perpendicular to the axis. In several examples now in my cabinet the apex of the spine shows evidence of this peculiarity ; but in one specimen in particular (a test to which the spines are attached) that circumstance is so marked and is so persistent (the base and apex of the body of the spines being specially affected) that it can hardly be the result of accident. “The common longitudinal perforations in the outer layer of the spine alluded to at p- 50 seem to have been chiefly due to disease or to some difference in structure which caused those parts, now empty, to decay with greater facility in one direction than in FROM THE WHITE CHALK. 75 another. A transverse section of spines so affected proves that the canals are of neither uniform length nor dimensions, some being of greater extent and more open than others. “The central perforation not unusual at the apex of some spines, seems also due to ' disease or to parasitic borings, and will often be found to extend downwards as far as the acetabulum. ‘This is the case with the spine drawn in Pl. V, fig. 6. A portion of the surface of this spine having been carefully removed subsequent to the drawing being made on the plate, the cavity apparent at the apex was seen to extend through the whole length of the body, unaltered in size; just below this point it suddenly contracted in a circular curve (similar to the base of the perforations made by a Pho/as) as though to avoid break- ing through the walls of the neck ; at the lower point of the circular excavation the opening appeared again, only with a very much less diameter, and extended as far as the articular cavity, through which it passed. Another spine open at the apex, when cut lengthways, gave the same result (of a continuous tube, of two different diameters), except that the opening, which extended almost as far as the acetabulum, did not pierce it, but passed outwards in a transverse direction. “The test of C. clavigera varies in the proportions of its parts from youth to age; my smallest example, five tenths of an inch in width, differs consider- ably in appearance from my largest, which is one inch and seven tenths in width. Comparing these two it is seen that the tubercles in the former are relatively larger than in the latter; that the granules of the miliary zone are in the former almost as large as in the latter; that the number of plates are the same in both; that there is an oval rudimentary tubercle in the uppermost plate of the anal side in the largest specimen; that the areolas of the two superior tubercles of the anal side are in the smallest example separated by only three granules, including those of the scrobicular margin, whilst in the largest example there are fourteen. Both specimens have four rows of granules in the ambulacral areas at the ambitus; in the smallest they are of equal size and equally disposed, in the largest the two interior are much smaller than the two exterior, more numerous, and crowded together. These dif- ferences have a tendency to cause the two specimens, when placed with the anal side uppermost, to appear very dissimilar, particularly in the region of the miliary zone. Of these two specimens the smallest is much below and the largest much above the average size. “ Spines of C. clavigera are sometimes, but very rarely, found as far down as the middle of the flinty Chalk; the proper horizon of C. clavigera is above this part. “Figures 1 a, 1 6, 2,3 a, 3 6,3 ¢, 4a, 46, 5a, 5 6, Pl. XIII, are from the Upper Chalk of Bromley, in Kent. 76 CIDARIS “AppiTionaAL Nots on CIDARIS PERORNATA, Forbes. (See p. 62.) “This Cidaris is the largest of all the Cretaceous Crparipz ; portions of a full-grown specimen now before me, containing four complete columns of plates in contact, give the following dimensions for the test—height, two inches and two tenths; transverse diameter, two inches and one tenth. The spines, like the body, also exceed those of all other species. In a mass of spines of C. perornata from my cabinet, which are all one tenth of an inch in diameter, is one which, although deficient of a portion of its apex, measures in the remaining part of its length four inches and six tenths—this length is by no means a maximum. ‘The number of the plates and the form of the spines appear to have rendered perfect examples of the test with spines attached exceedingly rare. Separate plates and groups of broken spines are plentiful; complete columns of plates uncommon. Small Ostree are occasionally found affixed to the spines. “The test, when full-grown, has, in the ambulacral areas, eight rows of granules at the ambitus ; of which rows the two exterior are the largest and most evenly disposed, the six interior are more numerous, of less size, and not so regularly arranged ; at the mouth- opening there are six rows, at the anal four; the second discal plate has nineteen pairs of pores in the poriferous zone; the proximal discal plate in each column has a rudimentary tubercle and an elongate obsolete areola. The granules of the miliary zone are of two sizes, the smallest of which occupy the spaces between the largest. In specimens of the test of the usual size the first, second, and third of the plates, reckoning downwards from the anal opening, have the upper half of the boss crenulated. The spines belonging to the granules of the scrobicular margins are flat and somewhat fan-shaped; they are covered with minute striz, which converge from the circular base (in which there is an acetabulum) towards the smaller apex; length two tenths of an inch, greatest width one twentieth. The jaws of a full-grown specimen do not greatly differ in outline from those of other species ; they are half an inch in length. “ Cidaris perornata is tolerably common in the Upper Chalk ; it appears to commence (where it is rare) in the middle of the flinty Chalk. “ AppitionaL Notr on CIDARIS DIXON, Cotteau. (See p. 67.) “All the spines of this species hitherto found are of considerable size, and are clavi- form, and inflated; the apex is acute; the surface covered with numerous granules, which FROM THE GREY CHALK. ri are large and elongate on the upper half of the body, pointed at the apical region, circular on the lower half of the body, diminishing in area as they approach the neck, and ceasing at that part, arranged in rows gradually increasing in number from the apex to the greatest diameter, and afterwards more closely and less regularly deposited ; the neck smooth, short, and very much contracted ; the milled ring is slightly prominent, covered with fine longitudinal striz; the head smooth. “The spines of C. Divoni occur at the base of the Lower or Grey Chalk in the cliffs between Folkestone and Dover, in the band containing the spinous Ostrea carinata, Sow. (M.C., tab. 365, fig. 1), in company with C. Bowerbankw ; they are, however, very rare. The same species is found occasionally in the “ Coprolitic Bed” of Cambridge, a deposit containing rolled fossils from the Lower Chalk, Upper Greensand, and Gault formations. ‘The specimen figured in Pl. XII, fig. 6, and obtamed from the Coprolitic Bed of Cambridge, is identical in all respects with the Folkestone examples, except that the surface is more worn, and appears to have been subjected to much friction; the width of the Cambridge specimen is seven tenths of an inch, length of body one inch. The total length of spine (measured from a specimen in perfect condition in my cabinet), from Folkstone is one inch and four tenths; greatest diameter (midway between apex and edge of acetabulum) seven tenths; length of head and neck three twentieths ; diameter of neck three twentieths. “The test of this Cidarzs has not at present been discovered ; it would appear, however, judging from the analogy of its spines with those of C. Bowerbankii, that it must have much in common with the latter, except size; perhaps it may be an aged form of C. Bowerbankit. “ AppiTionaL Note on CIDARIS BOWERBANKIT, Forbes. (See p. 45.) “In this species, as in others of the Cidares, the form of the spine varies according to its position on the test; those at the peristome are tolerably cylindrical, with an acute apex; those at the ambitus are inversely conical, with the apex less acute, and those at the anal margin have the body inflated and the apex somewhat obtuse. The peri- stome spines have the surface covered with coarser granulations than is the case with those which occur on the opposite side. At Southeram Pit, near Lewes, Sussex (Lower Chalk), tests with the spines 2 sz¢w are occasionally found. In my cabinet is a specimen from Southeram Pit nearly perfect, in which almost the whole series of spines, from the anal to the oral region, are in position, and in which the variation of form in these spines, according to their situation, is well exhibited. From this specimen were drawn the figures shown in the woodcuts fig. 4 a—d, which are twice the size of the originals; @ is the spine 78 CIDARIS in connection with the tubercle adjacent to the anal margin; 6 that on the next tubercle, counting downwards ; ¢ that beneath 4 ; and d that below c, on the tubercle which is the third from the peristome : a is in length five tenths of an inch, in diameter three tenths ; d is in length two tenths of an inch, in diameter one twentieth. The test from which these spines are derived is five twentieths of an inch in height, and nine twentieths in transverse diameter. Fic. 4. Spines of Cidaris Bowerbankii; magnified two diameters. “ Several of the spines of this species from different localities are figured on Pl. XIII; figs. 9, 10, and 11] are from Folkestone, figs. 13 and 14 from Cambridge, 8 from near Arundel, from which last-mentioned locality also come the plates of C. dissimilis, figured Pl. XIII, figs. 6 a,66. Inall these a certain variation in general form is very perceptible. “ Cidaris Bowerbanku has great affinities im its test with C. clavigera, but is always much smaller in size. It is a very rare species. The horizon of C. Bowerbankii at Folkestone is just above the Upper Greensand. “At Folkestone, in company with the spines of C. Bowerbankii, occur globose spines with a short neck, and having the body covered with coarse spiny projections arranged longitudinally. They are drawn of the natural size in the woodcut fig. 5; they appear to differ from C. velifera, and are perfectly distinct from the spines of C. Bowerbankit. Vte. 5. c. Spines of a Cidaris from the Lower Chalk at Folkestone ; natural size. “ Appirionat, Nore on Tan CIDARES rrom tun Rep Cuarx. (See p- 44.) “Tn the thin red-coloured band met with at Hunstanton, in Norfolk, and in the lowest of the pink-coloured beds at Speeton, in Yorkshire, occasionally occur elongate, cylindrical FROM THE RED CHALK. 79 spines, which do not exactly agree with those previously referred to in this Monograph; four of these are drawn on Pl. XII, of which figs. 7, 8, and 9 are from Hunstanton, and fig. 10 from Speeton. They may be divided into three classes—(a) slender, having few (ten to sixteen) longitudinal ridges, with a prickly border, Pl. XII, figs. 7 and 9; (() thick, having numerous longitudinal ridges, with the prickles almost obliterated, Pl. XII, fig. 10; and (y) slender, without ridges, but with an occasional projecting prickle. “The drawing, Pl. XII, fig. 7, represents a magnified view (the natural size being depicted by a black line) of the expanded extremity of a spine with twelve ridges, not unlike in its general character that to be met with in some forms of the spines of Cidaris Gaultina, but differing from the latter in the valleys between the ridges being covered with very fine longitudinal lines, instead of being marked with fine granulations. _ Fig. 9, with ten ridges, is marked also by the fine longitudinal striz, and the absence of granulations in the valleys; the lines of spiny projections or prickles are thinner, more conspicuous, sharper, and less numerous than in the spines of C. Gaultina; it is very slightly tapering ; the fragment preserved measures an inch in length, and must when perfect have been at least two inches; in general aspect it bears a strong resemblance to C. subvesiculosa from the Upper Chalk. “The spine fig. 10 @ (natural size), and fig. 10 4 (a portion magnified) is found both at Speeton and Hunstanton; the specimen figured, which was from Speeton, and is not quite perfect, measures one inch and a half in length, and is two tenths at its greatest diameter ; the body of the spine increases very gently in diameter for a short distance from the acetabulum, and then as gently diminishes; the margin of the acetabulum is crenulated, a double milled ring surrounds the head, the neck is short and smooth, and the body is marked by about thirty longitudinal ridges, which are crowned by small and obtuse spiny projections. The valleys between the ridges are covered with fine longi- tudinal strie; the general aspect is that of a spine of C. dissimilis, but the latter generally has the spines much more slender. “ Fig. 8, from Hunstanton, is only a fragment, half an inch in length, and one tenth of an inch in diameter; it is uniformly cylindrical, with the surface quite smtooh and without strie; arising from the smooth surface are stout prickles, like thorns, which are repeated in longitudinal lines at about the distance of the tenth of an inch apart from each other. It is a very peculiar spine, totally distinct from all those of the Cretaceous species, and mostly resembles the spine of C. yerornata from the Upper Chalk; in the latter, however, the prickles arise from a small longitudinal ridge, and are not isolated and unconnected. ‘The same form of spine cccurs at Speeton. In the ratio of frequency, the form a is more common than that of 8; and the forms a and # are more common than that of y, which is very rare.” 80 CLASSIFICATION OF THE. Family 2.—Hemiciparipaz. (Not yet found in British Cretaceous strata.) Family 3.—DIADEMAD. This Family includes large and small Urchins having a thin, circular, pentagonal, and subpentagonal test, more or less depressed on the upper surface, and flat at the base. The ambulacral areas are wide and straight, with two rows of primary tubercles, often as large and numerous as those of the inter-ambulacral areas. The poriferous zones are narrow, almost always straight, and sometimes subflexuous ; the pores are unigeminal, bigeminal, and trigeminal in their arrangement in different genera. The inter-ambulacral areas are in general twice the width of the ambulacral, and occu- pied, at the equator, with two, four, six, or eight rows of primary tubercles, which diminish gradually in number near the poles. ‘The bosses of all the tubercles are small; their summits, in general, are crenulated, sometimes uncrenulated ; the tubercles are small, in general perforated, in Cyphosoma imperforate ; they are in general a little larger than those of the ambulacra; but are often of equal magnitude in both areas. The apical disc is small, and situated opposite to the mouth; it is composed of five ovarial and five ocular plates; the anterior pair of ovarial plates are a little larger than the posterior pair, and the right antero-lateral plate, with a small, spongy, madreporiform body on its upper surface, is the largest; the vent is round or oblong, and generally in the centre of the disc; the ocular plates are very small, and distinguished with difficulty. The mouth-opening is in general large and decagonal, and the peristome divided into ten lobes by deep notches ; the jaws in general are large and powerful. The spines in existing genera are long, slender, and tubular, sometimes three times as long as the diameter of the test." In the fossil extinct genera they rarely attain the length of the diameter of the test, and are short, stout, and solid, except in Hemipedina, which have long hair-like spines. The long tubular spines of living Diademas, and a rare form from the Cretaceous rocks, are encircled by spiral verticellate processes, or fringe- like scales, Pl. XIV, fig. 2, whilst the surface of the solid spines of Pseudodiademas is in general covered with fine longitudinal lines; neither prickles or asperities being developed on their stems. Lamarck divided the genus Cidaris of Klein into two sections, “ Les Zurbans” and “ Les Diadeémes ;” these were afterwards by Dr. Gray’ erected into genera; the Cidaris radiata, Leske, constituting a third type, formed his new genus Astropyga. The genus Cidarites of Lamarck was considered to form a natural family, including the genera Cidaris, Diadema, and Astropyga, which he constituted and characterised thus :— 1 Perens, ‘Ueber Gruppe der Diademen,’ p. 2, 101. Kénigl. Akademie der Wissenchaften Augt., 1853, Berlin. 2 «Annals of Philosophy,’ new series, vol. x, p. 426, 1825. ‘An attempt to divide the Echinide or Sea-Eggs into natural families.” tn 'DIADEMAD A). 81 1. Family—Ciwwariwez. Cidarites, Lamarck. Body with spines of two sizes; larger ones either club-shaped or very long ; spine- bearing tubercles perforated at the summit. Genus 1—Criparis, Klein, Lamarck. (Ls TurBANS.) Body depressed, spheroidal; ambulacra waved; small spines compressed, two-edged, two-rowed, covering the ambulacra, and surrounding the base of the larger spines. This genus may be divided according to the form of the larger spines: the extra- ambulacral beads have only two rows of spines. CrpaRis IMPERIALIS, Lamk. Klein., Nat. dispositio Echinodermatum, tab. vii, fig. a. Genus 2—Drapema, Gray. (Les Diapiss.) Body orbicular, rather depressed ; ambulacra straight ; spines often fistulous. EcHINOMETRA SETOSA, Rumph. Leske, Klein., Nat. disp. Echinid., tab. xxxvii, fig. 1, 2. EcHINUS DIADEMA, Linn. Syst. Nat., by Turton, vol. iv, p. 139. — CALAMARIA, Pallas. Spicil. Zool., tab. ii, fig. 4—8. Genus 3—AsrropyGa, (ray. Body orbicular, very much depressed ; ambulacra straight ; ovarial scales very long, lanceolate ; beads with several series of spines. Ciparis RADIATA, Leske, apud Klein, tab. xliv, fig. 1. The very meager characteristics by which Dr. Gray has defined the last two genera merely shows that a difference exists, and his description is insufficient for a correct dia- gnosis of either ; hence the various opinions extant regarding the character and limits of his genus Diadema ; only one of the species enumerated as types, Diadema setosa, Rumph., Pe 82 CLASSIFICATION OF 'THE is admitted to be a true Diadema. The valuable memoir of Herr W. Peters! has removed some of the difficulties that surrounded this subject, and his grouping of the — livmg Diademas makes an important step towards a natural classification of one section of this Family. Although the present state of our scientific knowledge of the Diademade may be considered as transitional rather than positive, still we possess enough to justify the separation of fossil Diademas from existing genera, as proposed by M. Desor.? The Diapemap2, in fact, appear to consist of two types; one of these, with a few rare exceptions, appertains to the present epoch, the other existed during the deposition of the Secondary and Tertiary rocks. ‘The living forms are in general large, depressed Urchins, with thin shells, having the tubercles and pores variously arranged in the different genera. They have, in general, very long, slender, twéu/ar spines, and the surface of the stem is covered with oblique annulations of small imbricated scales. ‘The fossil species, on the contrary, are smaller Urchins, with a thicker test; having the tubercles and pores variously disposed in different genera; the spines rarely attain the length of the diameter of the test; they are in general so/id, cylindrical, sometimes flattened or awl-shaped, and their surface is covered with fine longitudinal lines. I propose to include the following genera in this natural family. A Table showing the Classification of the Diademade. Section a. “Spines very long, slender, tubular, covered with oblique annulations of imbricated scales. Living in tropical seas. A few annulated tubular spines are found in the Upper Chalk and in the Coralline Crag. DIADEMADA Section B. Spines short, slender, solid; surface covered with fine longitudinal lines. Extinct; found in the Oolitic, Cre- taceous, and Tertiary Rocks. Diapema, Gray. Savienya, Desor. AsTropyGa, Gray. Ecuinoturix, Peters. PsEUDODIADEMA, Desor. CyPHosoMaA, Agassiz. Hemirepvina, Wright. PEDINA, Agassiz. Ecurnopsis, Agassiz. My learned friend M. Cotteau,* in his classical work on the Echinide of France, has lately proposed an extended classification of the family Diappmapa&, a résumé of which I 1 «Ueber die Gruppe der Diademen, Konig]. Akademie der Wissenschaften,’ Berlin Aug., 1853. ‘Synopsis des Wchinides Fossiles.’ 3 * Paléontologie Francaise, Terrain Crétacé,’ tom. vii, p. 371. - DIADEMADA. 83 herewith subjoin; the genera referred to this family are divided into four groups, based upon the structure of the tubercles, whether they are perforated or not perforated, and crenulated or not crenulated. In the Crparipa these characters have not much significance, and are present or absent in many species of congeneric forms ; in the Diapemapa”, however, they are more stable and persistent, and have served to form a great number of genera. If from an organic point of view this structure of the tubercles is only of secondary importance, in a paleontological sense it affords a character which is readily seen, and nearly always well preserved. The first group comprehends the Diaprmaps#' with tubercles perforated and crenu- lated: Hemicidaris, Agassiz; Acrocidaris, Agassiz; Pseudodiadema, Desor; Diadema, Gray ; Hibertia, Michelin; Microdiadema,. Cotteau; Heterodiadema, Cotteau; 4. Ambulacra straight, provided with tubercles in all their extent. x. Ambulacral and inter-ambulacral plates without angular impressions. z. Inter-ambulacral areas subgranular as they approach the summit. . Apical dise subpentagonal, peristome large. << R . Each of the ovarial plates of the apical dise carry a large tubercle nN z. Apical disc, without a large tubercle on its ovarial plates. 1. Poriferous plates unequal and irregular. @. Spines solid, aciculated, striated go. Spines tubular, verticillated 2. Poriferous plates straight, equal, regular. yy. Apical disc narrow, annular, peristome reentrant yyy. Apical disc elongated, prolonged into the middle of the single inter- ambulacral area; peristome narrow xa. Inter-ambulacra smooth near the summit, and presenting a stellate appearance . : : : . xx. Ambulacral and inter-ambulacral plates marked with angular impressions . B. Tubercles perforated and not crenulated. a. Ambulacra subflexuous, provided with tubercles only towards the ambitus and inferior surface . b. Ambulacra straight, provided with tubercles in all their extent. x. Pores simple near to the summit. x. Ambulacral plates unequal, irregular. y. Apical disc largely developed, peristome wide. z. Miliary zone extended ; principal inter-ambulacral tubercles very large, placed on the external border of the plates. . zz. Miliary zone narrower, tubercles tolerably large and placed in the middle of the plates . yy. Apical disc narrow, peristome slightly developed, tubercles very small. vz, Ambulacral plates, straight, regular, sutures very apparent . xx. Pores in double series at the superior surface and towards the ambitus I[EMICIDARIS. ACROCIDARIS. PsEUDODIADEMA. DIADEMA. HIBERTIA. MIcRODIADEMA. TIErERODIADEMA. ASTERIOCIDARIS. GLYPHOCYPHUS. CIDAROPSIS. DIADEMOPSIS. HEMIPEDINA. EcHINOPSITS. OrtHopsis. PRDINOPSIS. DIADEMAD“. 85 C. Tubercles not perforated and crenulated. a. Ambulacral and inter-ambulacral plates without angular impressions. x. Form depressed, tubercles rather large, peristome widely open . . Cypnosoma. xx. Form inflated, tubercles small, peristome narrow : : - Mrcropsts. 6. Ambulacral and inter-ambulacral plates marked with angular, and sutural impressions. x. Apical disc sub-circular, inter-ambulacral tubercles forming many rows towards the ambitus . : 7 : . TEMNOPLEURUS. xx. Apical disc pentagonal, inter-ambulacral tubercles forming two rows . Ecninocyrnus. D. Tubercles not perforated and not crenulated. a. Ambulacral and inter-ambulacral plates without angular and sutural impressions. x. Apical disc smooth, ovarial and ocular plates perforated below at their ex- . ternal angle . ; : P , . Gontopyeus. xx. Apical disc furnished with a large tubercle on each ovarial plate; ovarial and ocular plates perforated at some distance from the border . ACROPELTIS. xxx. Apical dise granular, deprived of tubercles; ovarial and ocular plates per- forated at some distance from the border. w. Tubercles rather large, forming regular vertical rows. y. ‘Two rows only of inter-ambulacral tubercles ; mammelon large and pro- minent : 2 : : . . LELosoma. yy- More than two rows of inter-ambulacral tubercles; mammelon small . Eciinocrparts. yyy. Inter-ambulacral tubercles not extending above the ambitus ; inter- ambulacral area forming, at the upper part, a depressed zone, perfectly circumscribed. 2. Four rows of inter-ambulacral tubercles towards the ambitus . C@LOPLEURUS. <<. Two rows of inter-ambulacral tubercles towards the ambitus; spines long, sub-tricarinated, and slightly bent : F - Karararnorvs. yyyy. Ambulacral and inter-ambulacral tubercles limited to the inferior sur- face, replaced above the ambitus by caducous granules . - Copropsis. wx. Tubercles small, forming very regular horizontal rows. y. Peristome small, pores simple towards the ambitus : . COTTALDia. yy. Peristome very wide, sub-pentagonal, pores forming double rows from the ambitus to the mouth ‘ : F . Macnosta. 86 PSEUDODIADEMA D. Tubercles not perforated and not crenulated—continued. yyy- Inter-ambulacral tubercles irregularly arranged above the ambitus, often lacerated : . : : . GutLyprTicus. ] d. Ambulacral and inter-ambulacral plates provided with angular and sutural impressions. x. Impressions angular : 3 : 3 . TEMNECHINUS. xx. Impressions sutural, and angular, and much more defined . OPECHINUS. The stratigraphical distribution of the Diademadz extends’ from the Trias to the modern epoch, where a few species now live in tropical seas. Of the thirty-one genera enumerated in the above table, seven are proper to the Oolitic period: Mccrodiadema, Asterocidaris, Cidaropsis, Hemipedina, Acropeltis, Glypticus. Seven to the Cretaceous period: /eterodiadema, Glyphocyphus, Orthopsis, Pedinopsis, Hchinocyphus, Leisoma, Codiopsis. Five are special to the Tertiary period: HHibertia, Hchinopsis, Celopleurus, Temnechinus, Opechinus. ‘Three to the Modern period: Diadema, Echinocidaris, and Keraiaphorus. One genus, Pseudodiadema, is common to the Oolitic, Cretaceous, and ‘Tertiary periods. ‘Three genera are found in the Oolitic and. Cretaceous periods ; Hemi- eidaris, which commenced in the 'T'rias, Acrocidaris and Magnosia, but neither extend above the Neocomian. Four genera are common to the Cretaceous and Tertiary periods: Goniopygus, Cottaldia, Cyphosoma, and Micropsis. 'The genus Temnopleurus appeared in the Tertiary period and exists in our present seas. Psrupop1apEMA, Desor. 1854. This genus is composed of small Urchins with a moderately thick test, which rarely attains two inches in diameter; the ambulacral areas in general are one third or even one half the width of the inter-ambulacral areas; the primary tubercles of both areas are perforated, and nearly all of the same size; the bosses are small, and have sharply crenu- lated summits. The ambulacral areas have two rows of tubercles; the inter-ambulacral areas two rows only, or two rows of primary and two or four short rows of smaller secondary tubercles, or they have four, or six rows of nearly equal-sized primary tubercles at the ambitus. The poriferous zones in general are narrow and straight; the pores in one section are unigeminal throughout, and in another they are bigeminal in the upper part of the zones. The apical disc is small; and the anterior ovarial plates are larger than the posterior pair. ‘he mouth-opening is large, the peristome deeply notched, and the oral lobes are nearly equal. FROM THE LOWER GREENSAND. 87 The spines rarely attain the length of the diameter of the test; in general they are much shorter, cylindrical, or needle-shaped, and have a prominent, milled ring near the articulating head ; the rim of the acetabulum is crenulated, and the socket perforated ; the surface of the stem is sculptured with delicate longitudinal lines. _ The Pseudodiademata are all extinct, and found in the Liassic, Oolitic, Cretaceous and Tertiary rocks. Pseudodiadema differs from Diadema in having solid spines, with a smooth surface, the sculpture, in most cases, consisting of microscopic, longitudinal lines; whilst in Diadema the spines are tubular, and have oblique annulations of scaly fringes on their surface. Pseudodiadema differs from Cyphosoma, a Cretaceous genus, in having the tubercles always perforated, those of Cyphosoma being imperforate. It differs from fHemipedina im having a small apical disc, and tubercles with crenulated bosses, those of Hemipedina bemg smooth; and from Pedina in having the pores unigeminal or bigeminal, those of Pedina being arranged in triple, oblique pairs. Pseudodiadema may be divided into two sections, from the different manner the pores are arranged in the zones. Jn one group the pairs of pores form a single file throughout; in another the pores are more numerous, and crowded together in the upper part of the zones. Professor M‘Coy has proposed the genus Diplopodia for the latter. It may be objected, however, that the crowding together of a greater number of pores in a zone is, at most, a sectional and not a generic character, inasmuch as the arrangement is subject to great variation in the diplopodous species themselves, and is, moreover, often only an adult development. A.—Species from the Lower Greensand. PsEUDODIADEMA ROTULARE, Agassiz. Pl. XIV, figs. 3 a, 4, ¢. DIADEMA ROTULARE, Agassiz. Mém. des Se. nat. de Neuchatel, vol. 1, p. 139, tab. xiv, figs. 10—12, 1856. _ — Des Moulins, Etudes sur les Echinides, p. 316, No. 25, 1837. — ORNATUM, Agassiz. Catal. Syst. Ectyp. foss. Musei Neoc., p. 8, 1840. — ROTULARE, Agassiz. Descript. des Kchin. foss. de la Suisse, part 2, p. 4, tab. xvi, fig. 1—5, 1840. — macrostoMa, Agassiz. Ibid., p. 10, tab. xvi, fig. 22—26, 1840. — ROTULARE, Agassiz et Desor. Catal. Raison. des Echinides, Ann. des Se. nat, 3e série, t. vi, p. 346, 1846. — MACkOSTOMA, Agassiz et Desor. Ibid., p. 347, 1846. _ — Bronn. Index Palzontologicus, p. 418, 1846. — CORONA, Gras. Oursins foss. de l’Isére, p. 33, pl. i, fig. 21—23, 1848. — ROTULARE, Marcou. Recherch. géol. sur le Jura Salinois, Mém. Soc. Géol. de France, Ire série, t. ili, p. 143, 1848. 88 PSEUDODIADEMA DIADEMA ROTULARE, D’ Orbigny. Prod. de Paléont. Strat., t. ii, p.89; Ht. 17, No. 489, 1850. — MACROSTOMA, D Orbigny. Ibid., No. 491, 1850. — ROTULARE, Cotteau. Cat. Ech. Néocom., Bull. Soc. de l’Yonne, t. vy, p. 285, 1851. DIADEMA DUBIUM, Sharpe. Sands and Gravels of Farringdon, Quart. Journ. Geol. Soc., vol. x, p. 194, 1853. = — Forbes. In Morris’s Catalogue of British Fossils, 2nd ed., p. 76, 1854. — ROTULARE, Cotteau. Paléontologie Frangaise, Ter. Cretact, vol. vii, p. 422, pl. 1097, figs. 11—13; pl. 1098 and 1099. PsEUDODIADEMA — Desor. Synopsis des Kchinides fossiles, p. 69, 1856. — macrostoMa, Desor. Ibid., p. 68. — ROTULARE, Cotteau, tudes sur les Hchinides de l’Yonne, t. ii, p. 24, pl. xlix, figs. 1—5, 1857. — PIETETI, Cotteau. Ibid., p. 31, pl.1, figs. 7—10, 1857. — TRISERIALE, Desor. Synop. des Echin. foss., p. 445 (Suppl.). 1858. — ROTULARE, Dujardin et Hupé. Hist. Nat. des Zoophytes, Echinoderm., p- 428, 1862. — PERIQUETI, Dujardin et Hupé. Ibid. — macrostoma, Duwardin et Hupé. Ibid. — TRISERIALE, Dujardin et Hupé. Ibid. Test small, circular, slightly pentagonal, moderately convex above, and flat below ; poriferous zones narrow, straight; pores in single file; ambulacral areas large, two rows of close-set marginal tubercles ; inter-ambulacral areas, four rows of tubercles at the ambitus, the outer rows disappearing on the upper surface; miliary zone wide, depressed near the disc, and covered with an abundance of well-formed granules. Mouth- opening large, decagonal ; peristome deeply notched ; lobes unequal. Dimensions.—Height four tenths of an inch ; transverse diameter, one inch. Description.—YThis is a very rare Urchin from the remarkable deposit of fossiliferous sands and gravels near Farringdon in Berkshire, about the age of which so many different opinions have been given; perhaps the Echinide found therein may assist to determine the problem whether these beds belong to the Lower Greensand, or to a ‘‘ more modern member of the Cretaceous Series than the Chalk,” as maintained by the late Mr. Daniel Sharpe, F.G.S." The Diadema now before us is a well-known and characteristic species, of the middle stage of the Neocomian formation, containing chinospatagus cordiformis ; and the extensive table of synonyms prefixed to this article shows how widely it is distributed in beds of the same age on the continent of Europe. The test is of medium size, circular or slightly pentagonal, moderately convex on the upper surface, and nearly flat beneath. 1 «On the Age of the Fossiliferous Sands and Gravels of Farringdon and its Neighbourhood,” ‘ Quart. Journ. of the Geological Society,’ vol. x, p. 176. 1853. EEE FROM THE LOWER GREENSAND. 89 ‘(he ambulacral areas are wide (fig. 3 a, 6) and have two rows of tubercles placed on the margin of the area; these are small, uniform in structure, set closely together, and eradually diminish from the equator to both poles; a band of granulations down the middle of the area divides the two series from each other: the poriferous zones are narrow and straight (fig. 3 4); the pores are round and simple, and arranged in a single file throughout the zones (fig. 3 ¢). The inter-ambulacral areas are occupied at the ambitus by four rows of tubercles ; the inner rows extend from the mouth to the disc, and the outer rows diminish in size on the upper surface and disappear before reaching the disc ; the tubercles forming the inner row are about the size of those in the ambulacra; those of the outer row are sensibly smaller (fig. 3 4). The miliary zone is large, and slightly depressed near the summit ; it is filled with numerous granules of unequal sizes, some of which are mammillated and perforated ; the granules are disposed in circles around the areole, and fill the entire area of the zones with a beautiful ornamentation ; the examples from Farringdon have lost much of this character from the process of fossilization in those gravel beds. The base of the test is flat, and presents a highly tubercular surface (fig. 3 @), the four rows of tubercles in the inter-ambulacral areas being all distinctly developed in this region. ‘The mouth-opening, one half the diameter of the test, is proportionally large ; the peristome is deeply notched into the lobes, the ambulacral portions being one half larger than those of the inter-ambulacral arches. In fig. 3 c, I have given a section of the base, magnified four diameters, showing the relation of all these parts to each other. Affinities and Differences—This Urchin presents many varieties of form, which have been described by different authors as so many distinct species, an error that has been now corrected, as shown in the table of synonyms. It resembles P. Bourgueti, Ag., found with it m the same Neocomian beds, but is distinguished from that species in having the primary tubercles less developed, more closely set together, and more homogeneous ; and in the secondary or outer series of tubercles being larger and more regularly arranged ; they are, however, nearly allied forms of one type of structure. Locality and Stratigraphical Position—This Urchin in England has hitherto been found only in the sands and gravels near Farringdon, where it is extremely rare. It was collected from these beds by the late Mr. D. Sharpe, and I obtaimed one specimen in the same locality. On the continent of Europe it is one of the most characteristic fossils of the “Terrain Néocomien,” and is found principally in the middle beds of that formation. M. Cotteau records the following localities in France where it has been collected :—Billecul, Miéges, et l’ermitage de Censeau, Nozeroy (Jura) ; Morteau, Hautepierre (Doubs) ; Ger- migney (Haute-Sadne); Vassy, Bettancourt, (Haute-Marne); Thieffrain, Vandceuvre, Marolles (Aube); Cheney, Flogny, Moneteau, Auxerre, Gy-I’Eveque, Leugny, Fontenoy, Saints, Pereuse (Yonne); in all these localities it is collected in abundance from the Middle Neocomian ; and at Le Rimet (Isére), Villefargeau, Perrigny (Yonne), it is very rare in the Upper Neocomian. In Switzerland it is found near Locle in the Lower 12 90 PSEUDODIADEMA Neocomian ; and at Landeron, Sainte-Croix, Hauterive in the Middle Neocomian, so that it forms a leading fossil of the Neocomian formations. History.—This Urchin was at first referred by Professor Forbes to the Diadema dubium of Albin Gras, but a careful comparison of specimens proved this to be an error. It appeared under that name in Mr. Sharpe’s list of Echinodermata from the sands and — gravels of Farringdon, and in the second edition of the ‘ Catalogue of British Fossils.’ Pszvpoprapema Firroni, Wright. Pl. XV, Figs. 1, a—g. DIaADEMA AUTISSIODORENSE, Wright. Ann. and Mag. of Nat. History, New Series, vol. x, p. 91, 1852. Test pentagonal, depressed ; inter-ambulacral areas with two rows of primary tubercles and two incomplete series of small secondary tubercles, which disappear on the upper surface ; ambulacral areas prominent, with two rows of primary tubercles much diminished in size at the upper surface; poriferous zones narrow, subflexuous. Pores bigeminal near the ovarial disc, and at the circumference of the mouth. Dimensions.—Height four tenths of an inch; transverse diameter nineteen twentieths of an inch. Description —In its general outline this beautiful Urchin resembles P. depressum of the Inferior Oolite ; in the details of structure, however, it is very distinct from that form. The circumference is pentagonal, from the convexity of the ambulacral areas, and the upper and under surfaces are much depressed (Pl. XV, fig. 1 a, 4, ¢, d). The inter-ambulacral areas are one third broader than the ambulacral ; two rows of primary tubercles occupy the centre of the plates; there are about ten pairs of tubercles in each area, which are of a moderate magnitude, and gradually diminish in size from the ambitus to the base and summit; the mammillary eminences are small, their summits sharply crenulated, and the tubercles, of proportional size, are deeply perforated (fig. 1 g) ; at the ambitus six rows of granules separate the tubercles from each other (fig. 1 e); towards the upper part of the miliary zone the four central rows are absent, leaving a naked space in the middle of the area ; three rows of granules in like manner separate the tubercles from the poriferous zones ; at the base of the area, and extending as far as the ambitus, there are incomplete rows of small secondary tubercles ; these gradually diminish in size, and disappear at the upper surface, which is occupied with an unequal close-set granulation about three rows deep (fig. 1 6); the ambulacral areas, one third narrower than the inter-ambulacral, are very prominent and convex, and occupied by two rows of primary tubercles about ten in a row; the lower six pairs of tubercles are nearly as large as the corresponding tubercles in the inter-ambulacral areas, but the upper FROM THE LOWER GREENSAND. 91 four pairs are much smaller, so that, whilst there is a great uniformity in the size and form of the tubercles at the base and ambitus of the test, there is a very marked difference between those of the ambulacra and inter-ambulacra in the vicinity of the apical disc (fig. 1 4) ; the inter-tubercular space is occupied by a zigzag band of granulation, which is narrow below where the tubercles are large, and broader above where they are small (fig. 1 e). The poriferous zones are narrow and subflexuous; and the pores arranged in single pairs; near the disc they are slightly bigeminal; the apical disc is absent in our specimen. ‘The mouth-opening is large and the peristome slightly decagonal (fig. 1 ¢). Afinities and Differences —Pseudodiadema Fittoni nearly resembles P. Bourgueti, Ag., but differs from it in the rudimentary condition of the upper tubercles of the ambulacra, and in having the intermediate granulation on the miliary zone less homogeneous. Locality and Stratigraphical Position.—I collected this Urchin from the Lower Green- sand at Atherfield, in bed No. 4 of the Cracker group, Dr. Fitton’s section ; it must be very rare, as none of the cabinets of Atherfield fossils hitherto examined by me contain a specimen. Listory.—I discovered this fossil m 1850, and in the first instance erroneously identified it with a specimen found in France, and then briefly described by M. Cotteau as Diadema Autissiodorense. ‘The fine figures and detailed description lately published by M. Cotteau in his additions to the “ Paléontologie Francaise” have enabled me to correct my error, and I now dedicate this species to the memory of my late friend Dr. Fitton, F'.R.S., whose admirable memoir on the Atherfield section and the strata below the Chalk will long remain models of patient research and accurate scientific investigation. Pszupop1apEMA Ma.sosi, Agassiz § Desor. Pl. XX, figs. 1, a—/f. Drapema Matzost, Agassiz and Desor. Catal. rais. des Echinides, Ann. Science. Nat., 3me sér., t. vi, p. 350, 1846. — — D’ Orbigny. Prodrome de Paléont. strat., t. ii, p. 201, 1850. DipLopopia — Desor. Synops. des Echinides fossiles, p. 78, pl. xii, figs: 12—14, 1856. —_— — Leymerie et Cotteau. Catal. des Echinid. Foss. des Pyrenées, Bullet. Soc. Géol. de France, 2° sér., t. xiii, p. 324, 1856. DiapeMa Mackeson1, Forbes. Woodward’s Notes on British fossil Diadems, Mem. Geol. Surv., Decade V, 1856. —- MAckIET, Woodward. Ibid. — Matsost, Pictet. Traité de Paleont., 2° éd., t.iv, p. 245, 1857. DieLorepia — D Archiac les Corbiéres. Mém. Soc. Géol. de France, 2° sér., t. vi, p. 384, 1859. 92 PSEUDODIADEMA’ DipLopopia Mavzost, Dujardin et Hupé. Hist. Nat. des Zooph. Echino- dermes, p. 501, 1862. PseupopiapeMA Maxsost, Cotteau. Echinid. Foss. des Pyrenées, p. 26, 1863. a — Cotteau. Paléontologie Frangaise, Ter. Crétacé, tom. vil, p- 448, pls. 1106 et 1107, 1865. Test large, subcircular, upper surface convex, slightly inflated, base rounded and flattened, ambulacral areas narrow, contracted at the upper part by the width of the pori- ferous zones, two rows of tubercles twenty to twenty-two in each row. Inter-ambulacral areas wide with four, six, or eight rows of tubercles at the equator, the two inner rows having eighteen to twenty tubercles in each, extend from the peristome to the disc, all the others disappear at different points on the sides. Small secondary tubercles scattered irregularly among the primary series in the inferior part of the areas. Poriferous zones narrow at the base and sides; pores in double file from the ambitus to the disc, where they mcrease in width, and on the upper third are largely bigeminal. Mouth- opening moderate in size; peristome nearly equal lobed; discal opening large and acutely pentagonal. . Dimensions.—Transverse diameter two inches; height thirteen twentieths of an inch. Description.—YThis is a very rare British Urchin, and as nearly all the tests have been either broken, crushed, or otherwise distorted, it is difficult to form a correct idea of its form. I have carefully examimed the original specimens collected by Mr. Mackeson, F.G.S., from the Lower Greensand at Hythe, and presented by him to the Royal School of Mines; these I have compared with a series collected by my friend the Rev. T. Wiltshire, from the Lower Greensand at Whales’ Chine, Isle of Wight, with which they agree, and both correspond with the figures and descrip- tion of Pseudodiadema Malbost given by M. Cotteau in the ‘ Paléontologie Frangaise,’ and with a good type specimen kindly presented to me by M. Bayle, of the Neole des Mines, Paris. I have no hesitation, therefore, in considering D. MJackesoni, Forb., identical with D. MWalbos?, Agass. Itis important likewise to note that both belong to the same geological horizon; the French specimens were collected from the Upper Neoco- mian, associated with Lchinospatagus Collegnii, Sism., and the British specimens from the Lower Greensand at Hythe, and the Crioceras-beds, Lower Greensand, at Whales Chine, Isle of Wight, the English equivalent of the Continental Neocomian formation. This Urchin attains a considerable size ; Mr. Wiltshire’s cabinet contains a specimen measuring two and a half inches diameter. The base of this fossil is nearly circular, and only slightly pentagonal. In some of the Hythe specimens in the Museum of the Royal School of Mines, the upper surface is convex and moderately inflated, and the base rounded and flattened. The ambulacral areas are narrow and contracted at their apices by the width of the poriferous zones above (fig. 1,y); they are slightly inflated, and furnished with two FROM THE LOWER GREENSAND. 93 rows of large tubercles, from twenty to twenty-five in each, according to the size of the Urchin, all deeply crenulated and perforated, and gradually diminishing from the equator to the apertures ; a single sinuous line of granules separates the tubercles, which are placed closely together in the area (fig. 1 d). The poriferous zones are narrow at the base and sides, where the pores are arranged in a single file (fig. 1 e); at the upper part they are bigeminal (fig. 1 @), the double rows encroaching on the width of the ambulacral area and diminishing the size of the tuber- cles therein. The poriferous plates are prolonged to the base of the tubercles in more or less apparent irregular sutures (fig. 1 ¢). The inter-ambulacral areas are widely developed, the large plates support tubercles closely resembling those of the ambulacra (fig. 1 gy). In the figured specimen there are six rows at the equator, and in larger specimens there are eight distinct rows. ‘The two internal rows have eighteen tubercles a little larger than the others, extending from the peristome to the disc ; the other rows have a more limited range, and disappear on the upper surface. It is only in the largest specimens that eight rows are found at the ambitus, the tubercles of the shorter rows being a little less than those of the two internal series (fig. 1 y); besides the primary tubercles a number of small secondary tubercles are crowded along each side of the median suture, between the peristome and the ambitus, and others occupy spaces by the side of the poriferous zones. ‘The miliary zone is wide, smooth, and depressed at the upper surface ; the granules are irregularly scattered on its lower half, and some of them are even developed into small mammillated tubercles on the upper surface; they form hexagonal circlets around the areas of the primary tubercles ; the median suture is very well defined, and lies in a smooth depression of the test (fig. 1 a, c). The mouth-opening (fig. 1 4) is large and pentagonal, and the peristome divided into lobes of unequal sizes; the arches that span the ambulacra are longer than those of the inter-ambulacra. The apical disc was very large; the opening is pentagonal and acutely angular, the angles extending far into the median suture of the inter-ambulacra (fig. 1 @ and 4g). The spines are slender, and circular; above the milled ring of the head, there is a short portion of the stem ornamented with fine longitudinal lines (fig. /), whilst the portion beyond is entirely smooth. I have represented this character in the fragment fig. 1 /. M. Cotteau has figured a large example of this species from the Upper Neocomian ; from this we learn that age produces important modifications in the structure of the test ; the poriferous zones are very wide, and bigeminal, not only on the upper surface, but as far down as the ambitus ; besides the eight rows of primary tubercles there are some rudiments of secondary tubercles ; the miliary zone is wide and depressed at the upper surface; the discal opening becomes more angular, and the ovarial plates penetrate 94. PSEUDODIADEMA further into the ambulacral areas ; the mouth-opening is circular, and the peristome nearly equally lobed. Afimties and Differences —Pseudodiadema Malbosi resembles some of the larger forms of P. Brongniarti, from the Grey Chalk of Folkestone, in the cabinet of my friend the Rev. T. Wiltshire, and figured in Pl. XX, fig. 2 a, 6, in Pl. XXI B, fig. 3, and Pl. XXIJ a, fig. 2. The tubercles in P. Brongniarti are not so numerous in each row; the poriferous zones are narrower, and the bigeminal arrangement of the pores, so well developed in P. Aalbosi, is less distinct in P. Brongniarti. These certainly are nearly allied species, and require a careful examination to detect the small differences existing between them. P. Malbosi resembles P. dubium, Gras, from the same horizon. I have only a mould im plaster of the latter, not sufficiently sharp for scientific accuracy. Locality and Stratigraphical Position—The specimens I have figured were collected from the Lower Greensand at Whales Chine, Isle of Wight, in the Crioceras-beds that pass across that chasm, associated with Ammonites Martini, D’Orb., Crioceras Bower- banku, Sow., Gryphea sinuata, Sow., &e. The specimens in the Museum of the Royal School of Mines were collected by Mr. H. B. Mackeson, from the Lower Greensand (Kentish Rag) of Hythe, and presented by him to that institution. Specimens are extremely rare in both the places quoted. The foreign localities, according to M. Cotteau, are La Classe (Aude), Opoul (Pyrénées Orientales), where it is abundant in the Upper Neocomian beds, associated with Achino- spatagus Collegnit, D’Orb. B.—Species from the Gault. Pseupop1aprMaA WiutsuirEet, Wright, nov. sp. Pl. XVI, figs. 1 a—F 2, 3. Test moderately large and equally depressed at both poles; ambulacral areas wide, with two rows of tubercles, large and approximated in the lower half of the area, small and detached in the upper; poriferous zones narrow, flexuous, pores in single file throughout ; inter-ambulacral areas narrow, two rows of primary tubercles, and a few irregular secondary tubercles at the base of the area, primaries large and approximated in the lower half, small and remote above; miliary zone wide and finely granulated above, narrow and with large granules below; spines long and slender, the stem ornamented with delicate longitudinal lines. Dimensions.—Height six tenths of an inch; transverse diameter an inch and a half. Description—We only possess a fragment of this beautiful form, still it has been enough to enable Mr. Bone to give a restoration of the test in Pl. XVI, fig. 14. The body is inflated at the sides, and nearly equally flattened on the upper and lower surface. The FROM THE GAULT. 95 ambulacral areas are wide and have two rows of tubercles; those on the lower portion of the area are large and closely set together, and those on the upper part are dispropor- tionately small and placed widely apart (fig. 1 4); some very fine granules divide the large basal tubercles, and a numerous granulation surrounds the smaller tubercles on the upper part (fig. 1 4). The inter-ambulacral areas possess only two rows of primary tubercles ; those near the base are about the same size as the corresponding tubercles in the ambulacra, on the upper part of the area, they are larger, and diminish more gradually in size, so that the difference in the tubercles on the upper surface readily distinguishes the ambulacral from the inter-ambulacral areas; an irregular row of four small secondary tubercles occupies the outer side of the base between the primaries and the poriferous zones, and a like central row extends through the middle of the lower part thereof (fig. 1 c). The miliary zone is wide and depressed in the upper part, and the plates are here covered with numerous small granules, that cluster chiefly around the bases of the small tubercles, the median sutural space being depressed and nude (fig. 1 4); the lower part of the zone is narrow, and the granules are much larger and more closely set together ; many of them are raised on small mammillons, with secondary tubercles interspersed among them (fig. 1 ¢). The large primary tubercles of both areas have very large areolas (fig. 1, c), with well defined margins. Each areola consists of two parts, an outer circle, consisting of a band covered with microscopic granules (fig. 1 d), and a smooth inner portion, from whence the boss arises (fig. 1 ¢). This kind of ornamention is very remarkable ; it is very well pre- served in the fragment before me, and correctly represented in figs. d and e. The summit of the boss is sharply crenulated, and the tubercle deeply perforated. The spines were long and slender, as seen by some imprints on the slab (figs. 1, 2, 3); the acetabulum of the small head is marked by coarse crenulations, the milled ring is prominent, and the whole surface of the stem covered with fine longitudinal lines. Afinities and Differences.—This species belongs to the group of which P. Normanie (Pl. XXI, fig. 3) may be regarded as the type. It differs from that species, however, in having smaller primary tubercles and fewer and smaller secondaries, in having narrower ambulacra and less flexuous poriferous zones. The miliary zone is likewise less distinctly marked ; the general contour of the test is different, for the upper and lower surfaces are more depressed and the sides less inflated. Locality and Stratigraphical Position —This unique specimen was found by the Rev. T. Wiltshire, F.G.S., in the Gault at Folkestone, in a bed near the base of that formation. I have very great pleasure in dedicating this species to my kind friend as an acknow- ledgment of the important assistance he has rendered me during the progress of this work, by the generous contribution of all his best specimens for figuring, his able notes on certain species of Cidaris, and other valuable aid frankly given on all occasions when required. 96 PSEUDODIADEMA c.—Species from the Upper Greensand. PsrUDODIADEMA Ruopant, Agassiz. Pl. XVIII, figs. 3 a—e. DrapEMA RuHopANI, Agassiz. Cat. Syst. Ectyp. foss., Mus. Neoc., Supplement, 1840. -—— Luca, Agassiz. Idem, Mus. Neoc., p. 8. — RHODANI, Agassiz. Desc. des Echinid. foss. de la Suisse, tom. ii, p- 9, pl. xvi, figs. 16—18, 1840. — Lvea, Agassiz. Idem, p. 8, pl. xvi, figs. 11—15, 1840. _ — Agassiz and Desor. Cat. Raison. des Echinid., Ann. des Science Nat., 3° sér., t. vi, p. 346, 1846. — fRnuopant, Agassiz and Desor. Idem. — Lvca, Bronn. Index Palzontologicus, p. 418, 1848. = RHOoDANI, Bronn. Idem, p. 419. — LUCA, Albin Gras. Oursin. foss. de l’Isére, p. 33, 1848. — —_ D’Orbigny. Prodrome de Paléontol. strat., t. ii, p. 142, Et. 19, 1850. — RuHopDANI, Renevier. Mém. Géol. sur la Perte du Rhone, p. 49, 1853. _ — Morris. Catalogue of British Fossils, 2nd ed., p. 70, 1854. — — McCoy. Mesozoic Radiata, p. 67, 1854. PseupopiapEMa Luca, —Desor. Synopsis des Echinides fossiles, p. 71, 1855. — Ruopant, Desor. Idem, p. 71. DIADEMA — Pictet. Traite de Paleontol., 2° ed., t. iv, p. 244, 1857. — Luca, Pictet. Idem. — Dzsort, Forbes. Notes by 8S. P. Woodward ; Memoirs of the Geol. Surv., Decade V, p. 8, 1856. — PUSTULATUM, Forbes. Idem, p. 8, 1856. PsrupopiapeMa Luca, Dujardin et Hupé. Hist. Nat. des Zoophytes, Hchino- dermes, p. 498, 1862. — Ruovant, Dujardin et Hupé. Idem. _— — Cotteau. Paléontol. Francaise, Terrain Crétacé, p. 460, pl. 1110, 1864. Diagnosis.-—Test circular, depressed, slightly convex above, very concave beneath, a little inflated at the angles; ambulacral areas with two complete rows of tubercles, fourteen to fifteen in each, and three incomplete rows of small secondary tubercles at the base, five or six in each; inter-ambulacral areas with two rows of primary tubercles, thirteen or fourteen in each, and four rows of small unequal secondary tubercles at the base ; primary tubercles large at the ambitus, suddenly diminishing in size in both areas on the FROM THE UPPER GREENSAND. 97 upper and under surface ; plates covered with a fine uniform granulation; mouth-opening situated in a concave depression. Dimensions.—Transverse diameter one inch and one tenth of an inch, height half an inch. Description —Although this Diadema exhibits a group of well-marked specific charac- ters, its history, nevertheless, is involved mm much confusion, from want of a careful examination of the anatomy of the shell. The prominent ambital tubercles in the inter-ambulacra, the sudden diminution in size on the upper surface, with the baldness of the test in that region, and the crowding of the base with small tubercles nearly uniform in size, form a group of persistent characters which distinguish Pseudodiadema Rhodani from all its congeners. The smaller forms of this species were figured and described by Professor Agassiz as Diadema Luce, and the large tests as Diadema Rhodani. A series of specimens, of different ages, has since shown that these two forms are identical. This initial error introduced the confusion that followed, and has rendered it a matter of some difficulty to understand the synonyms of this species; the careful study of a good type form sent by the late M. Semann from the Gault (Etage Albien, d’Orbigny) of Clars, near Escragnolle, department of the Var, has enabled me to compare our English examples with an undeniable specimen, and from this examination to determine that Diadema Desori, Forb., and D. pustulatum, Forb., are different forms of Pseudodiadema Rhodant. My late esteemed colleague Dr. S.P. Woodward adopted Professor Forbes’s materials in his “ Notes on British Fossil Diadems,”’ contributed to Decade V of the ‘ Memoirs of the Geological Survey ;’ and it is evident from these notes that he had his doubts as to the accuracy of our lamented friend’s determinations, as will appear in the description of the different species. There are two varieties of Pseudodiadema Rhodani—a large form, identical with the type, figured by Agassiz,’ and a smaller form, corresponding with P. Zuce. The former I have obtained from the Chloritic Marl of Chard; the latter from the Upper Greensand of Warminster, where it appears to be rare. The fine example figured in Pl. XVIII, fig. 3, a, 6, c, is of moderate size; the test is circular and depressed, slightly convex above, in- flated at the sides, and very concave below; the ambulacral areas are large, and a little expanded at the sides to give increased space to the ambital tubercles; from this point they taper regularly towards both poles. There are two rows of primary tubercles, from sixteen to seventeen in each, extending from the peristome to the disc; three of these in each row, at the ambitus, are large, and all those on the upper surface small, dimi- nishing to mere granules near the disc (fig. 3 a); the tubercles on the under surface are small, and have a uniform size to the peristome; in this region the area is filled in with several smaller secondary tubercles (fig. 34). The poriferous zones are slightly undulated at the 1 «Description des Echinodermes fossiles de la Suisse,’ tab. xvi, figs. 16—18, p. 9. 13 98 PSEUDODIADEMA sides and base ; they are composed of simple oval pores arranged in single file throughout (fig. 3d). The inter-ambulacral areas one half wider than the ambulacral, have two rows of primary tubercles, fourteen to fifteen in each; a little larger at the ambitus and upper surface than in the ambulacra; three pairs are much larger at the sides, those on the upper surface diminish rapidly in size between the ambitus and disc; and on the under surface they are small and nearly uniform in structure. Between the basal angle and the peristome there are short rows of secondary tubercles, about the size of the primaries in this region, with a few scattered secondaries between the lateral rows ; as all these small tubercles are nearly the same size, the under surface of the test has a highly ornamented appearance—the inter-ambulacra with four, and the ambulacra with two rows of small, uniform tubercles, and several secondary ones planted at every interval on the plates (fig. 3 4). On the upper surface the six upper tubercles are small, diminishing to mere granules around the discal opening (fig. 3 c). The large ambital tubercles are surrounded by shallow circular areolas (fig. 3d). In some specimens they are confluent, in others separated by one or two rows of minute granules. The small dorsal tubercles are surrounded by ring-like areolas, and the basal tubercles have a chain-like arrangement of granules encircling them, which adds to the ornamentation of this region. The miliary zone is very large ; from the sides to the discal aperture the entire surface of the plates of both areas, except those portions occupied by the areolas, is covered with small, numerous, close set of granules, which form divisional partitions on each side of the mesial sutures between the rows of the primary tubercles, and then expand into a regular corrugation on all the upper surface, the dwarfing of the tubercles being compensated by an increased development of granular ornamentation on this region of the test. The base is very concave, and the small, circular peristome, indented with well- marked entailles, is situated at the bottom of a deep depression ; the entire surface of the base is studded with small tubercles, surrounded with the circles of granules already described. ‘The disc is absent in all the specimens hitherto found; the opening is large and pentagonal, indicating a great development of this structure in the species. Affinities and Differences.—P. Rhodani is readily distinguished from its congeners by the subundulated poriferous zones, small dorsal, intermediate basal, and large primary ambital tubercles, by the shortness of the secondary rows limited to the base, by the smallness of the dorsal tubercles and the fine homogeneous granulation on the miliary zone; the concavity of the base, smallness of the peristome, and depth at which it lies, added to the highly ornamented character of the plates, form a group of characters that readily distinguish it from all others. It resembles most P. Normanie (Pl. XXI, fig. 3), from the Grey Chalk of Folkestone, in the varied development of the tubercles in each row; the ensemdle of the test in the latter form is sufficiently defined by good specific characters, and for the definition of these I must refer to the article on that species. Locality and Stratigraphical Position —The large example I have figured was found FROM THE UPPER GREENSAND. 99 in the Chloritic Marl, full of green specks of iron, at Chard, associated with Catopygus carinatus, Goldf., Discoidea subuculus, Leske, Pseudodiadema ornatum, Goldf., P. variolare, Brong., with Ammonites splendens, Sow., A. varians, Sow., and other forms characteristic of the Upper Greensand formation. The specimens from Warminster, in the Museum of the Royal School of Mines, and in the Collections of Mr. Soper and Mr. Cunnington, Devizes, were found in the Upper Greensand with P. Michelini, Agas., and P. Benettia, Forb., and other common Upper Greensand forms, as Catopygus carinatus, Goldf., Salenia petalifera, Agass., and Goniopygus peltatus, Agas., &c. &e. Foreign Distribution —Geraudot (Aube) ; Pérte du Rhone (Ain); Clars, Escragnolle (Var) ; very common in the Htage Albien (Cotteau). History.—First figured by Professor Agassiz in 1840, the large forms as Diadema Rhodani, the smaller as D. Zuce. After much confusion it was discovered that these forms are identical. Professor Forbes, from not possessing types of Pseudodiadema Rhodani, named the large form Diadema pustulatum, and the small ones from War- minster D. Desorz ; this nomenclature was adopted by Dr. S. P. Woodward, in 1856, in his “ Additional Notes on British Fossil Diadems,” published in Decade V of the ‘Memoirs of the Geological Survey.” M. Cotteau, in 1863, has given admirable figures, and a most correct description of the species, which my observations confirm in all their details. PsrupopiapemA Micue ini, Agassiz. Pl. XIX, figs. 2, a—/ DiapEMA MicHELINI, Agassiz. Catal.Syst. Ectyp. foss. Mus. Neoc., p. §, 1840. — — Agassiz et Desor. Catal. rais. des Hchinides, Ann. Se. Naturelles, 3° sér., t. vi, p. 347, 1846. = — Bronn. Index Paleont., p. 418, 1848. as — Sorignet. Ours. foss. de dép de Eure, p. 25, 1850. — — D' Orbigny. Prod. Paléont. strat., t. ii, p. 179, 1850. — Bowne, Forbes. In Morris’s Catalogue of British Fossils, 2nd ed., p. 76, 1854. == = Woodward. Mem. of Geol. Survey, Decade V, explan. of pl. ii, 1856. Psrupoprapema MicueLini, Desor. Synopsis des Hchinides fossiles, p. 72, 1856. Diaprema MicuHeE ini, Pictet. Traité de Paléont., 2° éd., t. iv, p. 245, 1857. PsEUDODIADEMA PULCHELLUM, Cotteau. Tichinides nouv. ou peu connus (Revue de Zoologie), p. 3, pl. i, fig. 7—9, 1857. Psrupop1apEeMA MicueELini, Dujardin et Hupé. Hist. Nat. des Zoophytes, Echino- derm., p. 499, 1862. _— — Cotteau. Paléontologie Frangaise, Terrain Crétacé, p. 476, pl. 1114, 1864. Diagnosis.—Test circular, or slightly pentagonal, depressed ; base flat, inflated at the margin, concave towards the mouth; ambulacral areas large, two rows of prominent 100 PSEUDODIADEMA primary tubercles, 12—14 in each row; interambulacral areas with two rows of primary tubercles, 12—14 in a row, and two external rows of small secondary tubercles, extend- ing from the peristome to the ambitus ;' mouth-opening small, in a concave depression ; tubercles of both areas nearly the same size. Dimensions.—Height six tenths of an inch; transverse diameter one inch and one fifth. y Description —This Urchin is in general of medium size, with a sub-circular or pen- tagonal test, convex above and flat below ; the ambulacral areas are large, slightly inflated, and provided with two rows of small primary tubercles, 12—14 in a row, rather less than those in the interambulacral areas, and separated by a double zigzag row of very small granules (fig. 2 d), gradually diminishing in size from the ambitus to both poles; the poriferous zones are subflexous and composed of pairs of small round holes placed in single file throughout, crowded together near the peristome, and spread out above ; the inter-ambu- lacral areas are twice the width of the ambulacral and furnished with two rows of primary tubercles rather larger than those of the ambulacral areas ; they are very uniform in size and gradually diminish from the ambitus to the poles; between these rows and the poriferous zones, and between the two rows themselves, a series of small tubercles, 6—8 in number, extends from the peristome to the ambitus, where they disappear ; these small secondary tubercles fill up the intertubercular spaces at the base, and give the under sur- face of the test a very ornamental appearance (fig. 2 4). There are fourteen plates in each column of the inter-ambulacra, the primary tubercle occupying the centre of each; the areolas are circular and superficial, the bosses prominent and sharply crenulated, and the mammillons large and deeply perforated (fig. 2d); the surface of the plates is sparsely covered with small granules which form circles around the areolas and are scattered without order over the interspaces; the internal borders of the four upper- most plates of both columns are nude (fig. 2 4); and the sutures distinctly visible throughout their entire course. The ambulacral areas have fourteen plates in each column, the tubercles are rather smaller than those in the inter-ambulacra, and the narrow areolas are separated by a zigzag line of single granules (fig. 2,/), which becomes double (fig. 2 d) near the ambitus. The miliary zone is large, nude, and depressed at the upper part, granular towards the equator, and narrow and sinuous as it approaches the peristome (fig. 2 c). The mouth-opening is small, and lodged in a deep depression, the basal portion of the test being inflated around the peristome, which is decagonal, with nearly equal lobes (fig. 2 4). The apical disc is absent in all the specimens I have examined ; the opening, however, is large, a little elongated, subpentagonal, and angular (fig. 2 a). Afinities and Differences—This Urchin is distinguished from its congeners by its in- flated base, depressed upper surface, simple pores,numerous primary tubercles nearly uniform in size and number in the columns of both areas, by its small unequal secondary tubercles, FROM THE UPPER GREENSAND. 101 extending at the base between the primaries and the poriferous zones, and by its narrow peristome sunk in a deep depression; a careful comparison of typical specimens of Diadema Michelini, Ag., from the Upper Greensand of Villers-sur-Mer, Calvados, kindly sent by M. Michelin, has satisfied me that Diadema Bonei, Forb., is identical with D. Wichelini; and that P. Benettie, Forb. (Pl. XV, fig. 2), both as regards its general form, the number, disposition, and character of its primary and secondary tubercles, the smallness of the mouth-opening, situated in a deep depression, and the equal lobes of the peristome, is identical with some forms of P. Afichelini of the same size. Locality and Stratigraphical Position—This species is very common in the Upper Greensand near Warminster ; the smaller forms are very closely allied to the Diadema Benettia, Forb., the larger to the D. Bonet, Forb.; it has likewise been found in the same formation at Durdle Cove, Dorset. Foreign Localities—It has been collected from the Htage Cénomanien of France (= Upper Greensand) at Villers-sur-Mer, Cauville, Vaches-Noires, Saint-Jouin (Calvados) ; Octeville (Manche) ; Fécamp, Orcher, Le Havre, Rouen (Mont-St.-Catherine), Seine-Infér. ; Vimoutiers, Gracé (Orne) ; Présagny (Eure). Pseupop1ApEMA Brnurriz, Forbes. Pl. XV, figs. 2, a—. DiaprEMa BENeErtTiAz, Forbes. Morris, Cat. Brit. Foss., 2nd ed., p. 76, 1854. — _— Woodward. Memoirs of the Geol. Surv., Decade V, Puzs 1856. PsruDop1aDEMA Brnert1m, Desor. Synopsis des Hchinides fossiles, p. 72, 1858. — — Dujardin et Hupé. Hist. Nat. des Zoophytes, Echino- dermes, p. 499, 1862. — MicuELinI, Cotteau (pars). Paléontologie Francaise, Terrain Crétacé, p. 476, 1864. Test circular, inflated, depressed ; ambulacral areas wide, with two rows of tubercles 15 in each, separated by a double row of minute granules; inter-ambulacral areas with two rows of primary tubercles, 14 in each; separated by a wide median space, unequally granulated ; miliary zone smooth above; a few small secondary tubercles at the base; primary tubercles of both areas nearly of the same size. Peristome very small, deeply sunk in a concave depression ; discal aperture large and pentagonal. Dimensions.—Transverse diameter one inch ; height four tenths of an inch. Description —This beautiful little Urchin was formerly considered to be the Diadema ornatum, Goldf., and recorded as such in the first edition of Morris’s ‘ Catalogue of British Fossils.’ It was subsequently considered by Professor Forbes to be a distinct species, and in the second edition of that work was dedicated by him to the late Miss 102 PSEUDODIADEMA E. Benett, of Norton House,’ Wilts. It was first accurately described by the late Dr. Woodward in his Notes on Fossil Diadems. The test is circular, depressed on the upper surface, concave below, and inflated at the sides. The ambulacral areas are proportionally wide, with two rows of tubercles separated by a double row of granules ; there are from 12—15 in each row, according to age; they in- crease gradually in size, from the peristome and disc towards the ambitus, where they are largest; the poriferous zones are narrow and slightly undulated; the poresarein oblique single file throughout, three pairs of holes being opposite each ambulacral plate (fig. 2 a), where they are magnified four diameters; the pores at the circumference have a small tubercle between each pair; the inter-ambulacral areas are about one fourth part wider than the ambulacral ; they have two rows of tubercles, from 12—14 in each, separated by a wide miliary zone, which is unequally granulated, and becomes smooth on the upper surface ; a similar sparsely granulated space separates the tubercles from the poriferous zones ; and at the base of the area a short row of small secondary tubercles extends from the peristome to the angle, between the large tubercles and the poriferous zones. The tubercles of both areas are nearly of the same size, those of the inter-ambulacral are the largest; they have all distinct oval areolas, which are sometimes radiated, and encircled by rows of very small granules. The base is inflated at the circumference, and concave in the middle; the peristome is small and deeply sunk, about one third the diameter of the test; its margin is divided by feeble entailles. The disc is absent in all our specimens; the aperture is wide and pentagonal, the angles pointing towards the median suture of the inter-ambulacral areas. Affinities and Differences.—This species so much resembles Pseudodiadema Michelini, Ag., that it has been considered by some authors to be the same. M. Cotteau says, “P. Michelini, such as we understand it, cannot be distinguished from P. Benettig, Forb., and we do not hesitate to unite the two species, which present in their form, in the disposition of their tubercles, in the structure of their peristome, in a word, in the ensemble of their characters, an identity almost complete.” Admitting, as I do, the great similarity of the tests of the two forms, still I think, by a careful study of both, we can detect characters which may justify us in retaining our lamented colleague’s species. P. Benettie¢ has the tubercles smaller, the sides more inflated, the circumference of the base more convex, and the mouth-opening smaller and lodged in a deeper depression. P. Michelini is a flatter Urchin, and the size and greater prominence of the tubercles impart to it a more spinous character; if these points of difference in the test are borne out by any corresponding difference in the spines, the species may be sustained ; if not, P. Bennetie may turn out to be a mere variety of P. Michelini. Locality and Stratigraphical Position —This is a very abundant species in the junction-beds of the Upper Greensand at Warminster, associated with P. Michelini, ' Authoress of ‘A Catalogue of Wiltshire Fossils,’ 1831. FROM THE GREY CHALK. 103 Catopygus carinatus, and the other species of this zone. It has been collected from the same beds at Durdle Cove, on the Coast of Dorset. Foreign Distribution—M. Desor states that P. Benettig is found in the Craie chloritée (Cénomanien) of Villers-sur-Mer, Vaches-noires, de Gacé, Orne, and is common im all French collections. p. Species from the Grey Chalk. PsEUDODIADEMA oRNATUM, Goldf., sp. Pl. XVI, figs. 4 a—e; Pl. XIX, figs. 1 a—c; Pl. XXI, figs. 1 a—d, 2; Pl. XXI 4, figs. 1 a, 6. CIDARITES ORNATUS, Goldfuss. Petref. Germaniz, tab. xl, fig. 10, p. 123, 1826. DIADEMA ORNATUM, Agassiz. Prod. d’une Monogr. des Radiares, Mém. Soc. des Sc. Nat. de Neufchatel, t. i, p. 118, 1836. Desmoulins. Fitudes sur les Kchinides, p. 314, No. 15, 1837. = = Dujardin. Yn Lamarck’s Anim. sans Vertébres, 2° éd., t. iii, p. 392, 1840. — — Roemer. Norddeutschen Kreidegebirges, p. 29, 1840. — —_ Agassiz et Desor. Catalogue raison. des Hchinides, Ann. Sc. Nat., 2° sér., t. vi, p. 347, 1846. — — Bronn. Index Paleont., p. 449, 1848. — — D’ Orbigny. Prod. de Paléont. strat., t. ii, p. 169, Et. 20, No. 560, 1850. — — Forbes. In Morris’s British Fossils, 2nd ed., p. 77, 1854. — — McCoy. Contributions to Brit. Palzeontology, p. 67, 1854. — —— Desor. Synopsis des Echinides fossiles, p. 72, 1856. — — Woodward. Mem. of Geol. Surv., Decade V, p. 7, 1856. — TUMIDUM, Woodward. Mem. Geol. Surv., Decade V, 1856. — CARTERI, Woodward. Ibid. PSEUDODIADEMA ORNATUM, Pictet. Traité de Paléontol., 2° éd., t. iv, p. 245, 1857. a — Dujardin et Hupé. Hist. Nat. des Zoophytes, Hchinod., p. 499, 1862. _— _ Cotteau. Paléontologie Frangaise, Terrain Crétacé, pl. 1115, tom. ii, p. 480, 1864. Test circular, inflated ; ambulacral areas straight; tubercles small, numerous, sub- equal, in. two rows, separated by a double series of granules; inter-ambulacral areas with two rows of tubercles, separated by four rows of granules at the ambitus, which disappear at the upper surface, and leave a smooth depressed space in the middle of the miliary zone, bordered by two rows of granules; between the tubercles and poriferous zones a row of secondary tubercles extends from the peristome to near the disc, consisting of small well-spaced-out tubercles, each surrounded bya circle of granules. Areolas cir- cular, many of them radiated ; tubercles of both areas nearly equal in size. Dimensions—Height half an inch ; transverse diameter one inch and two tenths. 104 PSEUDODIADEMA Description —This beautiful Urchin when full grown is moderately large; it has a circular, elevated, and inflated form, depressed on the upper surface, and concave beneath. The ambulacral areas are large, a little inflated, and furnished with two rows of moderately sized tubercles, 14—16 in a row; one of these is always longer than the other, and separated from its fellow by a double series of small compact granules. The poriferous zones are straight and narrow at the base of the test, and gently subun- dulated from the ambitus to the apical disc; they are composed of simple pores closely approximated to each other; the pairs are well spaced out in the infra-marginal region, closer together at the equator, and still closer placed near the disc. Pl. XXI, figs. 1 4, e. The inter-ambulacral areas are large, and provided with two rows of tubercles, nearly identical in size with those of the ambulacral areas at the base and ambitus of the same regions; in the upper part of the area, however, they are proportionally larger, so that we find only 13—15 tubercles in each row. The secondary tubercles form a distinct series of from seven to nine small tubercles, situated between the primary row and the poriferous zones; in general one small tubercle arises from each plate between the peristome and ambitus (Pl. XIX, fig. 1 d), and is absent from the three or four upper plates of the test (Pl. XVI, fig. 4 e); besides these, there are other smaller tubercles, of unequal size, scattered between the primary and secondary rows (Pl. XIX, figs. 1 3, ¢, d). The miliary zone is very wide; it is narrow near the peristome, enlarged at the ambitus, and nude and depressed on the upper surface (Pl. XVI, figs. 4c and e). The granules are numerous, unequal, sometimes mammillated, and distinguished with difficulty from the small secondary tubercles among which they are interspersed (Pl. XXI, figs. 1 a, ¢; Pl. XVI, figs. 4 ¢, d). The primary tubercles of both areas have well-defined circular areolas, prominent bosses, with sharply crenulated summits, and large mammillons with deeply drilled summits ; the inter-ambulacral are a little larger than those of the ambulacral areas. The areolas near the ambitus are mostly confluent above and below, whilst on the upper surface of the inter- ambulacra they are distinct, and surrounded by circles of small granules (Pl. X XI, fig. 1 ¢.). The under surface is convex at the circumference, with a deep depression in the middle, in which the mouth-opening is situated. The peristome is very small, and its margin feebly indented. The discal opening is large, elongated, and sub-pentagonal. Unfortunately, all the specimens hitherto found want the discal plates (PI. XIX, fig. 1 4). Afinities and Differences.—This Urchin resembles P. Michelini ; it is, however, distin- guished from that species by its much greater height, inflated sides, narrow base, slightly undulated poriferous zones, more numerous and better developed secondary tubercles, more deeply sunk peristome, and elongated discal opening; these diagnostic characters are not the result of age, as they are observed more or less in comparing young specimens of both species apparently of the same age with each other. Locality and Stratigraphical Position—The specimens figured, from the Grey Chalk, Folkestone, on Pls. XVI, XIX, and XXI 4, belong to the Rev. T. Wiltshire, the FROM THE GREY CHALK. 105 British Museum, and my cabinet. I have examined a specimen from the Red Chalk of Hunstanton Cliff, belonging to my friend C. B. Rose, Esq., F.G.S., who has most kindly communicated many of his Echinodermata for this work. The specimen figured in Pl. XXI, fig. 1, now in my cabinet, was collected from the remarkable bed of Chloritic Marl at Chard, which has yielded so many fine examples of Achinide. The specimen in the Cambridge Museum is recorded by Professor McCoy as having been collected from the Upper Greensand of Blackdown. Many of the specimens sent me as P. Carteri by my friend Mr. J. Walker, F.G.S., and collected by him from the coprolite beds of the Upper Greensand near Cambridge, are undeniable specimens of P. ornatum. Foreign Distribution.—France—Rouen, Mountain of St. Catherine, Seine-Inférieure , Vimoutiers (Orne), in the Ktage Cénomanien, where it is rare. Germany.—LEssen-on-the- Ruhr, Westphalia, whence Goldfuss’s type specimen was obtained. LMistory.—This Urchin was first figured and described in the ‘ Petrefacta Germania’ by Goldfuss, in 1826, under the name Cidarites ornatus. Professor Agassiz, in 1836, erroneously referred to this species a small Neocomian form, which was subsequently separated under the name Diadema Bourgueti. In the first edition of the ‘Catalogue of British Fossils,’ Diadema Benettie was recorded as D. ornatum on the authority of Dr. Woodward. Professor Forbes named the tumid varieties of this Urchin Diadema tumidum, and under this name they are described, from type specimens, in my cabinet (PI. XXI, fig. 1), by Dr. S. P. Woodward, in his “ Notes on British Fossil Diadems,” con- tributed to Decade V, ‘ Memoirs of the Geological Survey.’ Psnvpoprapema Norman, Cotteau, 1863. Pl. XXI, fig. 3, a, d. Pseupop1apEeMa Normania, Cotteau. Paléontologie Francaise, Terrain Crétacé, t. vii, p. 468, pl. 1112. Test of moderate size, subcircular, inflated at the sides, and convex above; base rounded at the margin, and very concave in the centre; ambulacral areas wide, with two rows of tubercles, large at the ambitus and small on the upper and infra-marginal regions, sepa- rated by several rows of minute unequal granules ; inter-ambulacral areas narrow, with two rows of primary tubercles, nearly similar in size and development to those of the ambulacra, and numerous small, unequal, secondary tubercles, forming in the infra-mar- ginal region six short series, two on each side, and two in the middle of the primary rows ; mouth-opening lodged in a deep depression. Dimensions.—Height six tenths of an inch ; transverse diameter one inch and four tenths. Description —This rare and beautiful Urchin is of medium size ; it has a subcircular form, with broad inflated sides, convex at the upper surface and very concave beneath. The ambulacral areas are wide in the middle, lanceolate in the upper part, and narrow in the infra-marginal region ; they have two rows of primary tubercles, of which two pairs at the 14 106 PSEUDODIADEMA ambitus are very largely developed, and occupy the greatest part of the area, having large areolas, prominent bosses, sharply crenulated summits, with projecting, deeply perforated mammillons (Pl. XXI, fig 3 d); on the upper surface they rapidly diminish in size ; four of them have small areolas and the others become mere granules. In the infra-marginal region they gradually diminish as they approach the peristome, where two short rows of secondary tubercles are regularly arranged. The primary tubercles are separated by two rows of very fine granules, which gradually increase in number, and fill the entire upper portion of the area (fig. 3 d). The poriferous zones are narrow and straight at the base, undulated on the sides, and straight again on the upper surface; the pores are simple, round, largely open, and arranged in a single file throughout (fig. 3 c and d). The inter-ambulacral areas are relatively small from the excessive width of the ambu- lacra; they have two rows of primary tubercles, of which three pairs at the ambitus, like those in the ambulacral areas, attain a great development; above they become suddenly smaller, and diminish to mere granules; and on the infra-marginal region they become gradually smaller as they approach the peristome. The secondary tubercles are very abun- dant, and limited to this region of the test ; at the base of each interambulacra there are six short rows, a long and a short row between the zones and the tubercles on each side, and two short rows between the tubercles themselves. ‘This great profusion of small secondary tubercles and diminished size of those in the primary series impart to the base of the test of this Urchin a remarkable ornamental appearance, which resembles, on a larger scale, the structure of the base in P. 2hodani (fig. 3 4). The miliary zone is largely developed; the two primary rows of tubercles at the ambitus have six rows of granules forming a band between them; in the upper part the granulation increases, and fills the whole zone, except the part occupied by the areolas of the small rudimentary tubercles, forming circles around them, and filling the whole space with a fine nearly uniform granulation. The base is very concave and crowded with small tubercles ; the mouth-opening is small, and lies at the bottom of a very deep depression; the peristome is narrow, and marked by feeble indentations. Affinities and Differences.—This remarkable species resembles P. 2hodani in having its base crowded with numerous small tubercles, and in having large tubercles at the ambitus, and rudimentary ones on the upper surface. Its form, however, is always inflated; the poriferous zones are straight at the base and upper surface, and undulated at the ambitus. The secondary tubercles are larger and more numerous, and distinguished with difficulty from those of the primary rows of the under surface. Locality and Stratigraphical Position.—The only specimen I have seen was obtained from -the Grey Chalk near Folkestone by the Rev. T. Wiltshire, to whose cabinet it belongs.—Foreign DistributionM. Cotteau records Vimoutiers (Orne), in the Htage Cénomanien, where it is very rare. FROM THE GREY CHALK. 107 PsEUDODIADEMA VARIOLARE, Brongniart, sp., 1822. figs. 1, 2. CIDARITES VARIOLARIS, DIADEMA VARIOLARE, CIDARITES VARIOLARIS, DIADEMA VARIOLARE, — SUBNUDUM, _ Rotssyt, TETRAGRAMMA VARIOLARE, DIADEMA — — SUBNUDUM, —_ RotssyI, TETAGRAMMA SUBNUDUM, CipaRIs VARIOLARIS, DIADEMA VARIOLARE, TETRAGRAMMA — DIADEMA SUBNUDUM, — VARIOLARE, DIPLOPODIA VARIOLARIS, — SUBNUDA, — Rotssyt1, DIADEMA VARIOLARE, — svUB-NUDUM, — Rorssyr, DIPLOPODIA VARIOLARIS, _ SUBNUDUM, PSEUDODIADEMA STRIATULUM, Cotteau et Triger. PsEUDODIADEMA Rotssy1, DIPLOPODIA VARIOLARIS, Pl. XVII, figs. 1—5 ; XVIII, Brongniart. Géog. phys. des env. de Paris, pl. v, fig. 9, 1822; Tableau des Terrains, p. 408, 1829; Desc. géol. des env. de Paris, 3° édit., pl. xvii, fig. 9, 1835. Agassiz. Prod. d'une Monogr. des Radiaires, p. 189, 1836. Roemer. Norddeutschen Kreidegebirges, p. 29, 1840. Morris. Catalogue of British Fossils, p. 51, 1843. Agassiz et Desor. Catalogue rais. des Hchinides, Ann. des Sc. Nat., 3° sér., t. vi, p. 350, 1846. Ibid. Ibid. Agassiz et Desor. Agassiz et Desor. Bronn. Index Paleontologicus, p. 1261, 1848. A, Gras. Oursins foss. de l’Isére, p. 33, pl. ii, fig. 16, 1848. D Orbigny. Prod. de Paléont. strat., t. ii, p. 179, 1850. D Orbigny. Ibid., t. ii, p. 201. Sorignet. Ours. foss. de l’Eure, p. 26, 1850. D’ Archiac. Hist. des progrés de la Geol., t. iv, p. 215 1851. Quenstedt. Handbuch der Petrefaktenkunde, p. 580, 1852. Giebel. Deutschlands Petrefacten, p. 319, 1852. Forbes. In Morris’s Catalogue of British Fossils, 2nd ed., p. 77, 1854. Forbes. Ibid. Desor. Synopsis des Echinides fossiles, p. 78, 1856. Desor. Ibid. Desor. Ibid. Woodward. Mem. Geol. Survey, Decade V, 1856. Woodward. ibid. Pictet. Traité de Paléont., 2° éd., t. iv, p. 245, 1857. Pictet. Ibid. Coquand. Synop. des Foss. Crétacés, Bulletin Soc. Géol. de France, 2° sér., t. xvi, p. 992, 1859. Coquand. Ibid. Echin. du départ. de Ja Sarthe, p. 144, pl. xxxvii, figs. 13—15, 1859. Cotteau et Triger. Ibid. : Coquand. Cat. rais des Foss. départ. Charente, p. 155, 1861. 108 PSEUDODIADEMA DIPLopopIA SUBNUDUM, Coquand. Ibid. — VARIOLARIS, Dujardin et Hupé. Hist. Nat. des Zooph. Echinodermes, p. 501, 1862. — SUBNUDA, Dujardin et Hupé. Ibid. — STRIATULUM, Dujardin et Hupé, bid. PSEUDODIADEMA VARIOLARE, Cotteau. Paldéontologie Francaise, Terrain Crétacé, t. vii, p- 488, pls. 1117, 1118, 1119, and 1120, figs. 1—3, 1864. Test large, subcircular, slightly subpentagonal, nearly equally depressed on the upper and under surfaces. Ambulacral areas narrow and contracted at the upper part by the poriferous zones ; two rows of tubercles, 15—17 in each series, separated by a single line of granules ; pores round, in oblique single pairs in the middle, widely bigeminal on the upper surface, and trigeminal near the peristome. Inter-ambulacral areas wide, with four rows of primary tubercles and two short rows of small secondary tubercles. The middle of the upper surface of the area nude and often depressed. Under surface convex, mouth-opening small; peristome with feeble entailles. Discal opening large, sharply angular, pentagonal. Dimensions.—a.—Height half an inch; transverse diameter one inch. B.—Height half an inch; transverse diameter one inch and a half. Description.—The identity of this species has long been uncertain from the impos- sibility of ascertaming the Urchin intended by the figure and description given by Brongniart, without reference to the type specimen; this comparison has now fortunately been made by M. Cotteau, who has given admirable figures and most ample descriptions of the various forms P. variolare exhibits under different conditions of age and habitat. I shall first describe the general characters of the species, and secondly point out the three chief varieties it assumes. The test is of medium size, subcircular, sometimes lightly pentagonal, and equally depressed at both poles. The ambulacral areas are narrow, and contracted at their upper part bythe excessive development of the poriferous zones in this region; they possess two rows of large tubercles, 15—-17 in each series, according to size and age, which gradually diminish in size from the equator to the peristome, and become small and rudimentary on the upper surface (Pl. XVII, fig. 3 a). The rows are closely approximated, there being only a single series of small granules, of unequal size, forming a zigzag line, between them; the tubercles have narrow areolas, prominent bosses, with sharply crenulated summits, and deeply perforated mammillons (fig. 4). The poriferous zones are narrow, the pores round, in single oblique pairs on the sides, in triple oblique pairs near the peristome, and they are widely bigeminal in all the upper fourth of the zones (fig. 5 a). The interambulacral areas are nearly four times as wide as the ambulacral in the spe- cimen figured in Pl. XVII, figs. 3 a, 4, c, rather more than one inch in diameter. There are four rows of large tubercles, and two secondary rows, at the ambitus; the two inner primary rows, with fourteen tubercles in each series, extend from the peristome to the apical FROM THE GREY CHALK. 109 disc, and the outer primary rows are absent from the three uppermost plates (fig. 3a). Ina large specimen from the Chalk-Marl of Dorset, B, one and a half inches in diameter, the ambital plates have six and eight rows of large tubercles, and two rows of small secondary tubercles. There are sixteen tubercles in each inner series which alone reach the disc, the second, third, and fourth rows disappear as the plates shorten on the upper surface. The small secondary tubercles, situated near the poriferous zones, form a short series between the peristome and equator; they are scarcely larger than granules, but are, nevertheless, mammillated and perforated, and their presence, position, and development, constitute one of the specific characters of this Urchin. The interambulacral tubercles are nearly identical in size with those of the ambulacral areas (fig. 4). They have narrow areolas, prominent bosses, with sharply crenulated summits, and large perforated mamil- lons. The miliary zone is narrow at the sides and infra-margin, with two rows of granuies of unequal sizes; at the upper surface it becomes nude and depressed (fig. 3 a) around the discal opening, a character which appears im excess in the var. sudnudum. A number of granules, of different sizes, form hexagonal circlets around the areolas (fig. 5 a). The under surface is convex, and the small mouth-opening occupies a slight depres- sion ; the peristome is circular, and its margin notched with feeble entailles (fig. 3 4). The opening for the apical disc was very large (fig. 3 a), widely pentagonal, and sharply angu- lar, extending into the nude portion of the nter-ambulacra. None of the specimens as yet found contain any of the discal plates. (See likewise Pl. XVIII, figs. 1 a, 4, and fig. 2). Authors have recognised three distinct forms of this species, which some have de- scribed as so many separate species, whilst others regard them as varieties of one. Ist. Var. a, variolare, identical with Brongniart’s type form, is found in the Upper Greensand of Wiltshire and the “ Chloritic Marl” of Chard (Pl. XVIII, fig. 2) and l’Etage Cénomanien of Villers-sur-Mer, Calvados, France, from which localities I have specimens. Its upper surface is more or less depressed, and its outline is circular or subpen- tagonal. The inter-ambulacra have four rows of primary and two rows of small secondary tubercles ; the under surface is convex, and the mouth-opening small. 2nd. The var. 6, swbnudum, has the upper surface remarkably nude, from the ab- sence of granules in the upper part of the milary zone; the test is higher, and my specimen from the “ Chloritic Marl” of Chard has a thicker structure than var. a. 3rd. The var. ec, Roissyi, is still higher and much larger than var. 6; it has a more tuberculous appearance, and from six to eight rows of tubercles in the inter-ambulacra. I have two specimens before me that agree very well with M. Desor’s diagnosis of this form, which he considers a good species, or at all events a large variety of P. sudnu- dum. After a critical study of all these forms, I can find no good structural character for separating them, and therefore consider them as varieties of P. variolare, depending on age or habitat for the differences they exhibit in the size, thickness, and number of tuber- cles in the inter-ambulacral areas. Affinities and Differences —Pseudodiadema variolare is one of the most perfect types of a 110 PSEUDODIADEMA tetragrammous Diadema with bigeminal pores. It very much resembles P. Brongniarti, Agas., from the Grey Chalk, but is distinguished from the latter by having its upper and under surfaces more depressed, the ambulacral areas narrower, their rows of tubercles shorter, and those on the upper surface more rudimentary; the poriferous zones are much wider on the upper fourth, and have the pores more largely bigeminal ; the base likewise is more convex, wider, and less contracted than in P. Brongniartt. Locality and Stratigraphical Position —P. variolare is found in the Upper Greensand of Warminster, and the “Chloritic Marl” near Chard; from the latter locality I have specimens that represent the var. 6., swbnudum, and var. c., Roissyi. The large specimen was kindly communicated by the Rev. C. W. Bingham, of Binghams Melcombe, near Dorchester; it was collected from the Upper Greensand of that neighbourhood—the precise locality is not recorded. Foreign Localities —M. Cotteau has given a wide range to the distribution of this species in France, and records—“ Villers-sur-Mer, Cauville, Vaches-Noires, Dives, Saint- Jouin (Calvados) ; Octevilie (Manche); Fécamp, Le Havre, Rouen (Seine-Inférieure) ; Vimoutiers, Gracé, La Perriére (Orne) ; Présagny pres Vernon (Eure); Berneuil (Oise) ; Grandpré (Ardennes) ; La Fauche prés le Villard-de-Lans (Isére) ; Le Mans, La Raglasse, Yvré-l’Hvéque (Sarthe) ; Corzé (Maine et Loire); Touvois (Loire-Inférieure) ; Angouléme (Charente) ; ile Madame, Saintes (Charente-Inférieure). Assez abondant. Htage Céno- manien, commun surtout dans la zone a Scaphites equalis.—Lillebonne (Seine-Inférieure). Rare. Htage turonien.”’ In the Hils conglomerate, near Essen, Hanover. Desor. History.—Figured for the first time by Brongniart in 1822, as Crdarites variolaris, in his ‘Géognosie Physique des Environs de Paris,’ from a specimen collected at Havre in the Upper Greensand. In the ‘ Catalogue raisonné des Hchinides,’ 1846, MM. Agassiz and Desor separated certain varieties of this species from the type which they found in the museums of France, under the names Diadema subnudum and D. Roissy ; those, however, I have endeavoured to demonstrate are merely varieties of P. variolare. FROM THE GREY CHALK. lll Psrupop1apemA Broneniarti, Agassiz, 1840. Pl. XX, fig. 2 a—c; XX a, figs. 2 a—f, 3,4; XXIB, figs. 1—3 a—e. TETRAGRAMMA BRONGNIARTI, Agassiz. Desc. des Echinides fossiles de la Suisse, t. ii, p- 25, pl. xiv, figs. 4—6, 1840. — — Agassiz et Desor. Catal. rais. des Echinides, Ann. Sc. Nat., 3e sér., t. vi, p. 350, 1846. a = Bronn. Index Paleontologicus, p. 1261, 1849. DIADEMA _ D’Orbigny. Prodrome, t. ii, p. 142, Et. 19, No. 328, 1850. — — Renevier. Mém. Géol. sur la Perte du Rhone, p. 32, 1853. aa — Forbes. In Morris’s Catalogue of British Fossils, 2nd ed., p. 76, 1854. PsEUDODIADEMA = — Desor. Synopsis des Hchinides fossiles, p. 74, 1856. DIADEMA — Woodward. Mem. of the Geol. Surv., Decade V, 1850. = — Pictet. Traité de Paléontol., 2e éd., t. iv, p. 244, 1857. PsEUDODIADEMA = — Dujardin et Hupé. Hist. Nat. des Zoophytes, Echino- dermes, p. 498, 1862. — —_— Cotteau. Paléontologie Francaise, Terrain Crétacé, t. vii, p. 456., pl. 1109, 1865. ‘Test large, subcircular, elevated; sides tumid, depressed at the upper surface, narrow, rounded, and contracted on the under surface ; ambulacral areas narrow, with two rows of tubercles, separated by a double series of small granules of unequal sizes ; poriferous zones narrow, straight ; pores round, in single pairs from the peristome to the ambitus, and bigeminal thence to the disc-opening; inter-ambulacral areas wide, with four rows of primary tubercles, nearly identical in size with those of the ambulacra, and two short rows of very small secondaries near the zones. Mouth-opening small, in a considerable depression ; peristome narrow, with feeble and nearly equal-sized entailles. Dimensions.—a.— Height six tenths of an inch; transverse diameter one inch and a half. ».—Transverse diameter two inches and a half. Description —The test of this species exhibits so close a resemblance in many of its anatomical details to that of Psewdodiadema variolare, that, were it not for some differences in the size, shape, and development of the shell, and in the structure of the ambulacra, I should hesitate to separate it from that form; after all, these differences may not be specific, but may have arisen from habitat and other physical conditions. A test of each species, with spines attached, for the purpose of comparison, is still with me a desideratum ; however, as this Urchin is considered by most authors to be distinct from P. variolare, I shall describe the fossils I have figured under the name 2. Brongniarti, Agas. These 112 PSEUDODIADEMA. specimens have been compared with typical examples from the Upper Greensand of the Perte du Rhone, and identified as the true forms of P. Brongniarti, Agas. The test is moderately large, sub-circular, and elevated ; the sides are tumid, and the upper surface is flat; the base is convex, contracted at the circumference, and having the peristome sunk in a considerable depression (Pl. X XI 4, fig. 2 ¢). The ambulacral areas are narrow, and sharply lanceolate ; they have two rows of primary tubercles, from sixteen to seventeen in each row, which diminish gradually from the ambitus to the peristome and the disc; they are placed closely together, and have narrow ring-like areolas ; the bosses are stout, with sharply crenulated summits, and the mammillons are large and perforated (fig. 2 e); a row of granules, of unequal sizes, sometimes mammillated, separates the tubercles at the middle and base of the area, whilst in the upper part branches of fine granules pass off horizontally, forming circlets around the tubercles (fig. 2 e). The poriferous zones are narrow ; the pores are in single pairs in the middle and infra-marginal region, near the peristome they lie in triple oblique pairs, above the ambitus they fall out of their regular ranks, and at the upper surface are distinctly bigeminal (Pl. XXTI 4, fig. 2 4). The inter-ambulacral areas are more than twice the width of the ambulacral; they have four rows of primary tubercles nearly identical with those in the ambulacra. The two inner rows are best developed, and extend from the peristome to the disc, whilst the external rows are absent from the three uppermost plates; in the large specimen there must have been sixteen to eighteen tubercles in each internal row (PI. XX, figs. 2 @ and 4); a series of small unequal secondary tubercles ascends from the peristome to above the ambitus, situated between the primary tubercles and the zones (Pl. XXJ 4, figs. 2 d, e); they are altogether absent from the upper surface, but constant in the region I have described. The miliary zone is narrow in the infra-marginal region, wider in the middle, and expands at the upper surface ; it is filled with four rows of small irregular granules, among which a number of small mammillated tubercles as large as the secondaries are placed (PI. XX, fig. 2 e) ; above the ambitus horizontal branches of granules extend from the median rows, separating the areolas, and forming hexagonal divisions between them (fig. 2 e) ; at the upper fourth of the area the granules disappear from the middle of the zone, and a triangular nude space is exposed, having its base at the disc and its apex at the fifth plate ; the circlets of granules are absent from the areolas of these plates. The upper surface is flat, and the opening for the apical disc large and pentagonal, the angles extending into the inter-ambulacra (fig. 2 6.) The base is convex and contracted at the side; the centre is concave, about one third the width of the shell, and the mouth-opening lies in a considerable depression ; the peristome is small, about one third the width of the shell; in a fine test showing the base one and a half inch in diameter ; that of the peristome is half an inch (fig. 2 ¢). Afinities and Differences—P. Brongniarti is distinguished from its congeners by its elevated test, with tumid sides, flat upper surface, and contracted, convex under surface. + tahea i a) me Fig. Fig. Fig. PLATE IX. Ciparis HiruDO, Sorignet, 1850. From the White Chalk. . Test and spines natural size in the Collection of Henry Willett, Esq., F.G.S. (P. 64.) . Lateral view of the same test, natural size. . Small spine from the ambulacral tubercles, magnified three diameters. . Upper surface of a test, natural size, belonging to the British Museum. . Interambulacral plate, ambulacral area and poriferous zones of the same, mag- nified three diameters. . Apical disc and anal plates of the same, natural size. . One ovarial and two ocular plates of the same, magnified twice. Under surface of another test, natural size, belonging to the British Museum. Primary spine, natural size, belonging to the British Museum. . Lower portion of the same, magnified three diameters. . Upper portion showing its stellate termination, magnified three diameters. Primary spine, magnified three times, in the collection of Rev. T. Wiltshire, EG: PIs.1X C.R.Bone, del et lith Printed by W. West = PLATE OX: Ciparis nixupo, Sorignet, 1850. From the White Chalk. Fig. 1 a. ‘Test the natural size, with spines attached, in the collection of Professor Tennant, F.G.S. (P. 64.) 1 G. Primary spine of the same, magnified twice. hig. 2. Lateral view of another specimen belonging to the British Museum. Fig. 3 a. Upper surface of a smal] specimen belonging to the British Museum. 3 6. Under surface of a small specimen belonging to the British Museum. 3 ¢. Lateral view of a small specimen belonging to the British Museum. 3d. Interambulacral plate, ambulacral area zones of the same, magnified three times. Fig. 4. Primary spine magnified twice, in the cabinet of the Rev. T. Wiltshire, F.G:S. Fig. 5. Primary spine magnified twice, in the cabinet of the Rev. T. Wiltshire, F.G.S. Fig. 6. Primary spine magnified twice, in the cabinet of the Rev. 'T. Wiltshire, F.G.S. a 7 W.West imp C.R Bone del et lith PRATE. Xd. From the White Chalk. Fig. 1a. Test and spines of Cidaris intermedia, Wiltsh., natural size. Collection of Rev. Thomas Wiltshire, F.G.S. (P. 69). 4. Spine belonging to the tubercle, the third from the peristome, magnified. Fig. 2. Spine of Crdaris hirudo (?), magnified. Fig. 4. Unusual form of spine, probably belonging to C. sceptrifera, magnified. Col- lection of Rev. Thomas Wiltshire, F.G.S. From the Lower Chalk. Vig. 3a. Spine of Cidaris dissimilis, Forb., natural size. Collection of Rev. Thomas Wiltshire, F.G.S. (P. 46) 6. Head and spine of same magnified. Fig. 5. Spine of Cidaris pleracantha, Agass., natural size. Collection of J. R. Capron, Esq., F.G.S. (P. 67.) From the Upper Greeensand. Fig. 6. Spine of C. Divoni, natural size. Collection of Rev. Thomas Wiltshire, F.GS. (P. 67) From the Red Chalk. Figs. 7,8,9. Spines of Cidaris, natural size, from Hunstanton. Cabinet of Rey. Thomas Wiltshire, F.G.S. (P. 79.) Fig. 10a. Spines of Cédaris, natural size, from Speeton. (P. 79.) Fig. 6. The same, magnified. IONS), ar. impr | ro i = E gS ig i 2 py = xX 2 Ss f ‘ « cur 4 ETN Py ier: Se Rrctictiot at Maint LAD LA Ethie . PP LLP PA I PDA AV IF IAS ene Cake roa) d 2 5 os ir Foe ‘g ¢ 6 Stepsoeematarceetoee 5 HH PMS, a 3 t iw) reer ee ey "een niane Sti inner cea PLATE XIII. From the White Chalk. Cipanris CLAVIGERA, Konig, 1822. Fig. 1 a. Test and spines, natural size. (P. 71.) 6. Spine of same, magnified. Fig. 2. Spine of same, natural size. Fig. 3a. Test and spines, natural size. (P. 71.) 6. Spine of same, natural size. c. Spine of same, magnified. Fig. 4a. Test and spines, natural size. (P. 71.) 4. Spine of same, magnified. Fig. 5 a. Spine of C. clavigera, of elongate form, natural size. (P. 71.) 46. ‘he same, magnified. From the Lower Chalk. Fig. 6a. Plates of Cidaris dissimiles, Forb. (BP. 46.) 6. One of the plates, magnified. All the above are from the Cabinet of the Rev. Thomas Wiltshire, F.G.S. Fig. 7 a. Spine of Cidaris pleracantha? natural size. Collection of J. R. Capron, Esq., F.G.S. b. The same, magnified. (P. 67.) Figs. 8,9, 10. Spines of Crdaris Bowerbankii, Forb. Collection of Rev. ‘I. Wiltshire, EG aGeeelia) Fig. 11. Do. do. Do. J. R. Capron, Hsq., F.G.S. Fig. 13 a. Do. natural size. Cabinet of Rev. Thomas Wiltshire, F.G.S. 6. Head and neck of same, magnified. Vig. 14. Spine of same, natural size. Cabinet of Rev. ‘Thomas Wiltshire, F.G.S. one es ad Wa lo: = ee nog, Pe on eee a kD ae Ce ete, . San ay 8 ACN Spy tt WD hw SAR ead 2B Seoes “4 CR Bone. delet lith Fig. Fig. PLATE XIV. From the Chloritic Marl. 1 a. Pepinorsis Wiest, Wright, upper surface of the test, natural size. In the collection of Mr. Wiest. 1G: ie a under surface, natural size. le. * ms lateral view, natural size. 1 d. Ambulacral area, poriferous zones, and interambulacra, magnified four diameters. 1 e. Base of an ambulacral area, showing the disposition of the pores, x four times. 2 a. Fistulous spme of a Diadema from the White Chalk, x six times. British Museum. 20, Do: do. do. Xx six times. Ao Do. do. do. x six times. Dad, é: Do. do. do. x six times. From the Lower Greensand. . 3 a. PSEUDODIADEMA ROTULARE, Agassiz, base magnified one half. In the Cabinet of Dr. Wright, F.R.S.E. (P. 87.) 3 6. 2 lateral view do. do. oC: A we segment of the base, do. four times. AV. PLAX ce SS — o Vila i ¥ | CR Bone del et lith rane ae erat wo ahs ees PLATE XV. From the Lower Greensand. > Vig. 1 a. Psrupoptapema Firronu, Wright, test natural size, cabinet of Dr. Wright, Pakesak, «(P:90:) | 4. Upper surface magnified one half. 1 c. Under surface do. do. 1 d. Lateral view do. do. | e. Ambulacra, poriferous zones, and interambulacra, magnified four times. 1 f. Portion of an ambulacra, magnified four times. 1 g. One tubercle and pores, magnified six times. From the Upper Greensand. Fig. 2. a. PseupopiapEMa Bryertim, Forbes, magnified one half, British Museum, and cabinet of Dr. Wright, F.R.S.E. (P. 101.) 2 6. Upper surface of the same test do. do. 2 c. Lateral view of do. do. do. 2d. Ambulacra, zones, and interambulacra, magnified four times. 2 e. Ambulacra seen in profile do. do. 2 f. Inter-ambulacral plate and tubercle, magnified six times. 1 eAD.QVE WWest imp CR Bone del et lith PLATE XVI. From the Gault. Fig. 1 a. Psnupoprapema Wirtsuirit, Wright, natural size, belonging to the Rev. T. Wiltshire, F.G.S. (P. 94.) 1 4. Lateral view of the same test, magnified one half. 1 c. Ambulacra, zones, and interambulacra, magnified four times. 1 d,e. Primary tubercle, magnified six times. 1/. Portion of a spine, magnified six times. Fig. 2. Spine magnified. Bic: 3) Do., natural size. From the Grey Chalk. Fig. 4 a. PskupopIaApEMA oRNaATUM, Goldfuss, sp., upper surface, magnified one half, belonging to the British Museum. (P. 103.) 4 6. Under surface of the same, magnified one half. 4c. Lateral view of the same, do. do. 4 d. Ambulacral area, interambulacral area, and pores, magnified four times. 4 e. Miliary zone and upper portion of an ambulacrum, do. do. PL. AVI | | ee a ae = Sa iene eet ne oti ee f .& - «G, ° oe ¢ att - : @» an a PES a ‘5 cise a4 2 J e\ Dr. Wright, F.R.S.E. (P. 103.) Same test restored and magnified half a diameter, showing the base. Lateral view of do. do., do. Portion of the ambulacra, inter-ambulacra, and zones, magnified four times. Inter-ambulacra, plate, and tubercle, magnified six times. From the Upper Greensand. . PsrupopiapemA MicweEwini, Agassiz. Upper surface, magnified one half. In the cabinet of Dr. Wright, F.R.S.E. (P. 99.) . Under surface of the same test, do. Do. . Lateral view of do., do. Do. . Portion of ambulacra, inter-ambulacra, and pores, magnified four times. Do. magnified four times. Base of the ambulacra, do. ap nhart PS M.&N Ha: . Bone del.et hth PLATE XX. From the Lower Greensand. Fig. 1 a. PseupoprapemA Matzsosi, Agassiz. Upper surface of the test, natural size. From the Collection of the Rev. T. Wiltshire, F.G.S. (P. 91.) 1 4. Upper surface of the same specimen. lc. Lateral view of do. 1 d. Inter-ambulacral plates, ambulacral area, and poriferous zones, magnified four times. 1 e. Basal portion of an ambulacral area, magnified four times. - 1 f. Portion of a primary spine, magnified five times. From the Grey Chalk. Fig. 2. a. PsrupopiapEMA Bronentarti, Agassiz. Fragment of a large test, upper sur- face, natural size, in the cabinet of Rev. ‘I’. Wiltshire, F.G.S. (P. 111.) 2 6. Under surface of the same, do. 2 c. Base of the ambulacral area, magnified four times. PL XX 26 Qu o) ae {ee WWest ump CR Bone del et ith F : ; ‘ ; ~ 7 ; : ' a. : ; » ‘ 7 j w ss ; i =n we iF a - 7 : n ' 7 7 ‘ . j r ? - 2 wi = J . - ¥ Fi i : 4 : . nl =) = : @ : ; } ; \ ’ PLATE XXI. From the Chloritic Marl. Fig. 1 a. PssupoDIADEMA ORNATUM, Goldfuss. Upper surface, magnified one half, in the Fig Vig cabinet of Dr. Wright, F.R.S.E. (P. 103.) 1 4. Lateral view of the same test, magnified one half. 1 c. Portion of the ambulacra, inter-ambulacra and pores, magnified four times. aa: Do. do. and spine, magnified four times. From the Grey Chalk. . 2a. Another specimen, with spines, in the cabinet of the Rev. 'T. Wiltshire, F.G.S. 2 6. Portion of the ambulacra, inter-ambulacra and zones, magnified four times. . 3 a. PseupopiapEMA Norman, Cofteau. Upper surface of the test, natural size, in the cabinet of the Rev. T’. Wiltshire, F.G.S. (P. 105.) 3 6. Lateral view of the same, natural size. 3c. Ambulacral and inter-ambulacral plates, magnified four times. 3d. Ambulacral area, magnified three times. PLXXI 16 WWest ump CR.Bone del. et lith PLATE XXI a. From the Grey Chalk. Fig. 1 a. PsrupoprapeMa ornatum, Goldfuss, from the cabinet of the Rev. 'I’. Wiltshire, F.G.S. Upper surface, natural size. (P. 103.) 1 4. Under surface, natural size. Fig. 2a. PsnupoprapEMa Bronentartl, Agassiz. British Museum, natural size. (P. 111.) 2 0. 2 c. Fig. 4. be) 2? by) 2? 2? 99 Upper surface, magnified one half diameter Under surface, do. do. Lateral view, do. do. Ambulacral area, poriferous zones, and one half of an inter-ambulacral area, mag- nified six times. Primary tubercle, magnified. Spines, natural size. Portion, magnified six times. ; Pl AAA TESAN Renae mortem ae Sea SCORE vere W West imp C.R Bone del etlith vs Ol eee eo i al A Le ae § oan ‘ 4 we -)) an] f ‘ ‘ ' i. | y 7 . i { ‘ A ‘ ‘ ‘ ' P . » ‘ . ’ , \ LL ae +* t . ‘ Bal , ‘ \ ’ PLATE XXI s. From the Red Chalk. Fig. 1 a. PsrupopiapEMA Broneniarti, Agassiz. Cabinet of the Rev. T. Wiltshire, HGS; (P21) 1 4. Lateral view of the same. Both natural size. Fig. 2. Another specimen, belonging to C. B. Rose, Esq., F.G.S. From the Grey Chalk. Fig. 3 a. PspupoprapEmMA Broneniarri, Agassiz. Cabinet of the Rev. T. Wiltshire, E.G:8. 3 6. The test restored from this fine large specimen. Upper surface. arc: Do. do. do. Under surface. 3 d. Do. do. do. Lateral view. 3 e. Portion of the ambulacra, inter-ambulacra and pores, magmified four times. PIAA Bs. CR Bane del et lith W West imp a PALAONTOGRAPHICAL SOCIETY. INSTITUTED MDCCCXLVII. VOLUME FOR 1867. LONDON: MDCCCLXVIII. ij rely i ry 5 7 ‘ - I t - e . i | ‘ P| : fj + ‘ 1 . 1 / i % ‘ 4 ; ‘ i i i ' 1 y . ' 1 " ; i ' - ‘ { ‘ ‘ ‘ js ' ‘ . ry A MONOGRAPH erie oS: ep Hs OLD RED SANDSTONE OF BRITAIN. BY JAMES POWRIE, F.G.S., H. RAY LANKESTER, JUNIOR STUDENT OF CHRIST CHURCH, OXFORD. PART I—THE CEPHALASPID. BY E. RAY LANKESTER. Pages 1—33; Plates I—V. LONDON: PRINTED FOR THE PALHONTOGRAPHICAL SOCIETY. 1868. Ny a CL PRINTED BY J. E. ADLARD, BARTHOLOMEW CLOSE, F.C. ‘ THE FISHES OF THE OLD RED SANDSTONE OF BRITAIN. Parr 1—THE CHEPHALASPIDA. INTRODUCTION. I wovutp here beg to claim indulgence for the deficiencies of the following pages relative to the Cepatasprp#. These Fish I have had to treat systematically for the first time since the work of Agassiz; a great deal of the material submitted to me has been quite new ; and, in addition to the absence of any previous essays on the Family (except- ing occasional papers of great value, but of limited scope), the very special and peculiar nature of the fossils has been a source of some difficulty. The generic and other divisions which I have adopted have been taken simply with the view of exhibiting breaks in the continuity of forms; and they may of course, at any time, be bridged over by further discovery. At the same time, I believe that the grouping here offered furnishes the most convenient method of arranging these fragmentary remains. I may mention, that I feel sure, from various indications, that many new species are to be found in the very localities which have furnished those now known. I take this opportunity of expressing my thanks to Professor Huxley for the use of a great number of specimens assembled by him for his own study, as well as of specimens in the Museum of the Survey ; to my colleague, Mr. Powrie, of Reswallie, Forfar, who is engaged on another portion of this Monograph, for the use of his unique and beautiful specimens, and other kind assistance; to Dr. Grindrod, of Malvern; Mr. Humphry Salwey, of Ludlow; the Rev. P. B. Brodie; the late Mr. Wyatt-Edgell; Mr. Lightbody, of Ludlow ; the Rev. J. Crouch; Mr. Salter; the Rev. Hugh Mitchell; Mr. Morton, of Liverpool; Dr. Harley, and others who have been kind enough to give their help in the communication of specimens and in other ways.—E. R. L. 2 OLD RED SANDSTONE FISHES. THE CHPHALASPIDA. We commence this work with the consideration of the Fishes belonging to the Family Crpuaaspip® of Professor Huxley,’ because they are the oldest Fishes the remains of which are known,—because these remains are of a special and peculiar nature, having very little in common with the remains of the other Fishes of the Old Red Sandstone,—and because a great deal of knowledge with regard to them has accumulated sce the last work, treating of them as a group, was published, viz., Agassiz’s ‘Recherches,’ dating from the year 1834. § I. History of the Cephalaspide.—tIn his great work, ‘ Recherches sur les Poissons Fossiles’ (1835), Professor Agassiz established the genus Cephalaspis, to include four species of Devonian Fishes obtained in Britain. These species were respectively named C. Lyell, C. rostratus, C. Lloydii, and C. Lewisii. The first was known to Prof. Agassiz by specimens both from the West of England and from Scotland; and two of these specimens showed, in addition to a large semicircular head-shield, the body covered with scales and provided with a well-marked caudal fin. ‘The other species were known to him only by oval discoid bodies corresponding to the large semicircular head-plate of the first; but he was led to conclude that they were allied forms partly by real, partly by fancied resemblances in construction, and by their occurring under the same conditions. He remarked, however, very strongly on the differences between the first and the last of his species of Cephalaspis, and pointed out that, in addition to the differ- ences in contour and shape, there were differences in minute structure and ornamen- tation, which would probably lead at some future time to a separation of the species into other genera. In 1847 Dr. Rudolph Kner published a memoir in Haidinger’s ‘ Naturwissenschaft- liche Abhandlungen,’ for the purpose of proving that Cephalaspis Lewisii and C. Lloydii were not the remains of Fish at all, but that they were the internal shells of a Cephalopod allied to Sepia, for which he proposed the generic title Pteraspis. Dr. Kner’s conclusions were based upon the examination of a fossil’ (evidently closely allied to C. Z/oydii) from the Upper Silurian (perhaps corresponding to our lowest Devonian) strata of Gallicia. The structure of the test in this fossil was observed by Dr. Kner to differ from that of any known osseous remains of Fishes; and he considered, from a somewhat superficial examination of both structures, that it agreed closely with the cuttle-bone of Sepza. (See also p. 12.) In 1856 Dr. Ferdinand Roemer, in Dunker and Von Meyer’s ‘ Paleeontographica,’ 1 The Family Cernanasprpss of Agassiz included the Genera Pterichthys, Coccosteus, &c. 2 Dr. Kner does not appear to have named this species, and I shall therefore hereafter speak of it a Scaphaspis Knerit. (See pages 19 and 20.) CEPHALASPID&. . 3 described a fossil from the Devonian strata of the Laacher See as Paloteuthis, referrmg it to the Sepiade. Professor Huxley has since shown this to be closely related to Agassiz’s Cephalaspis Lioydii. Roemer, in referring to Kner’s memoir, expressed the opinion that his Pteraspides were the remains of Crustacea. Whilst matters stood thus with regard to the last two of Agassiz’s species of Cepha- _ Jaspis, Sir Philip Egerton described, in the ‘ Quarterly Journal of the Geological Society,’ vol. xiii (1857), several new species by far more closely related to his Cephalaspis Lyellit, namely, C. Murchisoni, C. Salweyi, and C. ornatus; and further formed a new genus, Auchenaspis, for the reception of a very small form, which differed from Cephalaspis Lyellii in having the hinder portion of its shield separated as a distinct “neck-plate.” ‘ The genus Afenaspis was briefly described (without any figure) by Ewald, of Berlin, in 1848 ; and was stated to have affinities with Cephalaspis. His description, however, _ by no means warrants this conclusion. The fossil was found in beds of Permian age. a In Russia (1854), Hichwald described a new genus and species allied to Cephalaspis met, but differing in having no orbital apertures, as Zhyestes verrucosus (see fig. 7, _ p. 16), which Prof. Pander, in his ‘ Monographie der fossilen Fische des Silur-Systems’ a (1856), afterwards re-figured and described as Cephalaspis verrucosus, together with another _ species from the same locality (Rootsikiille), the latter beg termed Cephalaspis Schrenkit. _ Mr. Banks, of Kington, in Herefordshire, now discovered two species of Cephalaspide, 4 allied to Cephalaspis Lloydii, in the Downton Sandstone, and these were described by _ Professor Huxley and Mr. Salter, in the ‘ Quart. Journ. Geol. Soc., 1856,’ as species of - Kner’s genus Péeraspis. At the same time these authors withheld their opinion as to _ the piscine, molluscan, or crustacean nature of the genus. -. Shortly after this, in 1858, Professor Huxley published a most detailed and careful account of an inquiry into the intimate structure of the fossil shields forming Agassiz’s _ genus Cephalasyis, with a view to ascertain the correctness of Kner’s conclusions as to the nature of those species which he had separated as Pferaspis. This most valuable essay is contained in the ‘ Quart. Journ. Geol. Soc.,’ vol. xiv, and I shall have to refer to it very largely hereafter. For the present, it is sufficient to say that Professor Huxley conclusively demonstrated that there was no foundation in facts for the supposed resemblance between Péeraspis and Sepia, or between it and any Crustacean armature. Cephalaspis Lyellii was shown to have a bony structure, from which truly C. Lloydii and Kner’s species differed very widely, but not im such a manner as to render it probable that they were anything but the shields of closely allied Fishes. Kner’s genus Pteraspis was therefore definitely adopted for the three latter species of Agassiz’s genus Cephalaspis, and for the allied forms more recently discovered. Another species of true Cephalaspis was next made known by Dr. Harley (‘ Quart. Journ. Geol. Soc.,’ vol. xv, p. 503, 1859): and Mr. Salter in the ‘Ann. Nat. Hist.,’ July, 1859, described a Péeraspis occurring in the Upper and Lower Ludlow beds of the Silurian series, which is the oldest indication of a Vertebrate animal on record. 4 OLD RED SANDSTONE FISHES. In the ‘Quart. Journ. Geol. Soc.’ for 1860 Professor Huxley published a paper in which he showed the true nature of Roemer’s Palzoteuthis (see also fig. 10, p. 20), already alluded to, and at the same time gave a restored outline of the shield of the Cephalaspis (Pteraspis) rostratus of Agassiz. In 1863 I had the good fortune to obtain a specimen of Pferaspis, showing a few rhomboid scales attached to the shield, and thus the piscine nature of the fossils was definitely set at rest. (Plate V, fig. 3.) In ‘ Decade X’ of the Geological Survey published in 1861, Professor Huxley consti- tuted the family CnpHataspip to receive the genera Cephalaspis, Auchenaspis, Pteraspis, and Menaspis. This family he placed among the Chondrostean Ganoids, where he considered it should have a very distinct position. At the Meeting of the British Association at Bath, 1864, I proposed to divide the genus Pteraspis, as adopted by Professor Huxley, into three other genera, in accordance with three degrees of complexity in the structure of the head-shield observable in the known species : and in the following year, at Birmingham, I briefly reviewed the species of Cephalaspis then known, and exhibited specimens of these Fishes, of great interest, discovered by Mr. Powrie, in Forfarshire. Lastly, in the ‘Geological Magazine’ for April, 1867, I described the genus Didymaspis, allied to Cephalaspis and Auchenaspis. The following list contains references to all, I believe, that has been written on the Fishes of the Family Cephalaspide, the titles of the papers being given in chronological order : : Acassiz. ‘ Recherches sur les Poissons fossiles,’ vol. ii, p. 135, pls. la, 16, 1835. ‘Monogr. Poiss. Vieux Gres Rouge,’ p. 31, 1844. Kner (Cephalaspis Lloydii, C. Lewisii, and a new species). Haidinger’s ‘Naturw. Abhandlungen,’ vol. i, p. 109, 1847. Ercuwatp (Thyestes verrucosus). ‘ Bullet. Société Imp. Nat. Moscou,’ p. 108, 1854. Ewaup (Menaspis). Proceed. Acad. Berlin, p. 33, 1848. Panper (Ceph. verrucosus and C. Schrenkii). ‘Monographie der fossilen Fische des Silur-Systems,’ p. 47, 1856. Huxtry and Sauter (Pteraspis truncatus and Pt. Banksii). ‘Quart. Journ. Geol. Society,’ vol. xii, p. 100, 1856. Eerrton (Ceph. Murchisoni, C. Salweyi, and C. ornatus, and Auchenaspis). ‘ Quart. Journ. Geol. Soc.,’ vol. xiii, p. 282, 1857. Huxtey (Structure, &c. of Cephalaspis and Pteraspis). ‘Quart. Journ. Geol. Soc.,’ vol. xiv, p- 267, 1858. r (Observations on Pécraspis). ‘Brit. Assoc. Report,’ 28th Meeting, Trans. Sect., pp. 82, 83, 1858. Harwey (Ceph. asterolepis). ‘ Quart. Journ. Geol. Soc.,’ vol. xv, p. 503, 1859. Satter (Pteraspis from the Ludlow Beds). ‘ Annals and Mag. Nat. Hist.,’ 3rd ser., vol. iv, pp. 44 —48, 1859. Huxtny (Paleoteuthis and Restoration of Pteraspis). ‘Quart. Journ. Geol. Soc.,’ vol. xvii, p. 163, 1860. * (Zoological Position of Cephalaspide). ‘Memoirs of Geol. Survey, Decade X,’ p. 38, 1861. CEPHALASPIDA 5 LANKEsTER (Scales of Péeraspis). ‘Quart. Journ. Geol. Soc.,’ vol. xx, p. 194, 1863. és (On the Genera Péeraspis, Cyathaspis and Scaphaspis). ‘ Brit. Assoc. Report,’ Trans. Sect., p. 58, 1864. 265 (On the Genus Cephalaspis). ‘ Brit. Assoc. Report,’ Trans., Sect., p. 65, 1865. ? (On Didymaspis). ‘Geol. Magaz.,’ April, vol. iv, p. 152, 1867. _ Besides these papers, there are several Letters on the Cephalaspide in the pages of the ‘Geologist,’ 1858—64, by Messrs. Powrie and Marston, the Rev. Hugh Mitchell, and myself. § Il. Nature of the Evidence with regard to Cephalaspide now available.—Though the knowledge of these Fishes has been gradually increasing, as may be gathered from the brief review above given of their literary history, and though we have now far more perfect specimens to examine than Professor Agassiz originally had, yet, except in the case of one or two species allied to Cephalaspis Lyellii, nothing whatever is known of the form of the body, position of the fins, or arrangement of the scales.‘ In every case with the exception mentioned, our knowledge of the Fish rests solely, as did that of Professor Agassiz, on more or less imperfect specimens of the oval, discoid, or semicircular shields which covered the anterior portion of the animal. Owing to this fact, the modifications in the form and structure of this cephalic shield are necessarily made the basis of the arrangement and classification of these Fishes; and, though those few specimens which exhibit a scale-covered body, fins, &c., possess a very high interest, they cannot to any large extent be made use of for the purpose of drawing conclusions with regard to the other Cephalaspids. § III. Division of the Cephalaspide into two Sections.—The genera Cephalaspis and Pteraspis into which Professor Huxley divided Agassiz’s genus Cephalaspis, on account of remarkable differences in the histological character of the shields, each admits of subdivi- sions which perhaps might be viewed as sub-genera, but which, I think, it will be more convenient to regard as genera, adopting new names for the sections of the Family Cephal- aspide, indicated by Professor Huxley’s generic division. ‘Thus I propose to call the Pteraspidian forms Hnrrrosrract, in view of the special character of their test ; the Cephalaspidian forms, Ostxosrract, in reference to the occurrence of true bony structure in their cephalic shields. ach of these sections contains three or four genera of about equal value, all of which are very definitely distinguished, and nearly all embrace at least two or three species a-piece. | § IV. Nature of the Shields of Cephalaspide; and grounds for division into two Sections. 1 Professor Agassiz in his ‘Monogr. de Vieux Grés Rouge,’ published in 1844, remarked that the ten years that had passed since the publication of the ‘ Recherches’ had not brought avy fresh evidence to light as regarded the Cephalaspide. 6 OLD RED SANDSTONE FISHES. The grounds upon which the division into Hrrrrosrract and Osrrostract is based are derived chiefly from the intimate structure of the cephalic shields, though there are also other considerations in regard to the form of these shields which seem to warrant the arrangement. It is remarkable that, while in nearly all the Ostxosrraci (Cephadaspis, Auchenaspis, &c.) the orbits are large and placed near the centre of the shield, in the Hurerosrraci( Pleraspis Scaphaspis, &c.) there are either no orbits at all in the shield, or they are minute and placed marginally. ‘This distinction is not, however, persistent, for in Eichwald’s Zhyestes (Cephalaspis verrucosus and Ceph. Schrenkii of Pander) there are, according to the figures given, no orbits excavated in the shield, although its structure is similar to that of Cephalaspis. (a.) It will now be convenient to examine the question as to the structure of the shield in detail, in order to exhibit fully the reasons which have led to this division of the Cephalaspide ; and, in so doing, I shall have to: make most extensive use of Pro- fessor Huxley’s memoir on this subject. Professor Agassiz did not fail to perceive the differences in structure presented by his species of Cephalaspis. We thus describes the constitution of the cephalic shield in C. Lyellii :-—“ In C. Lyellit the head is covered with a pavement of polygonal plates altogether similar to that which covers the head of Ostracion. Hach plate is convex in the centre, and is marked by radiating grooves ending at the margin in denticulations, by which the scales interlock. These scales appear to be osseous, and to have their external surface enamelled. At the circumference of the disc they become confounded together, and the enamel presents wrinkles parallel to the edge.” Elsewhere these plates are said to be “true scales juxtaposed.” In the ‘ Recherches,’ M. Agassiz describes “ fibrous bones of the head” under “the scales,’ and he particularly mentions and figures the radiating direction of these “fibres.” ‘This view of the composition of the shield in C. Lyell, 1 need hardly say, has been shown to be erroneous. Professor Agassiz seems himself to have changed his opinion; for, in speaking of all four of his species together, he says—‘* It would appear from the condition of the specimens preserved, that all the cranial bones were only protecting plates which covered a cartilaginous cranium similar to that of the Sturgeons ; at least, I have never been able to discover any cranial bones deprived of that characteristic granulation which indicates that the plates were in direct relation with the integument. ‘Therefore, I think there can be no doubt that all these granular plates rested by their inner and smooth surface on a cranial cartilage, such as is found in Cartilaginous Fishes and in the embryos of Osseous Fishes.” —MJonog. Gres Rouge. He still seems to have entertained the idea that the shield was composed of a number of minute plates, though he abandons the notion of the existence of a subjacent osseous skull, and regards the shield as a tegumentary ossification. The shield of Cephalaspis Lloydii Professor Agassiz described as consisting of an external striated enamel, of a middle layer ‘‘ composed of granules similar to those of Chondropterygious Fishes,” and of an internal layer made up of superimposed lamelle. CEPHALASPID A cE Dr. Rudolph Kner describes the structure of his Pferaspis from Gallicia as follows :— “The form of the fossil is very similar to that of C. Lloydii; but it is larger, having a length of about four inches by a width of two. It consists of three layers. The innermost is shining, bluish-green, enamel-like, and presents four or five distinct lamella. This layer forms one continuous surface, marked in the centre by a longitudinal depression smaller at one end than the other, and by obscure radiating lines. The upper part of the conical depression is covered with minute pores or depressions, which are visible in the deeper as well as in the more superficial layers, but become evanescent in its lower part. Between the layer of enamel and the prismatic part which succeeds it, there lies a thin dull layer, in some places of a brownish colour. ‘This is followed by an excessively delicate lamina of enamel which lies upon the prisms. ‘The layer of prisms is one line thick, and in sec- tion presents a number of more or less hexagonal discs. ‘The enamel passes for a short distance between the prisms. Externally the prisms lie on a granular layer, to which the outermost very delicate epidermic lamina marked with parallel strize succeeds.”” Such was the structure which Dr. Kner believed he had seen in his Péeraspis, and compared with that of the Cuttle-bone, to which it really presents but a superficial resemblance. It was on account of this structure that Dr. Kner proposed to remove Cephalaspis Lloydii and —€. Lewisii from the Class of Fishes, and place them with his new formas Péeraspis among the Cephalopodous Mollusca. Professor Pander describes briefly the structure of the Russian Cephalaspide (Thyestes) in his work already quoted, from which it is evident that they have very little in common with Péeraspis. Sections of tubercles from the margin of the shield exhibited ‘“ a homoge- neous base, in which clear and dark cells of the most various forms (rounded, elongated, and angular, with fine radiating branches) lay scattered, and were frequently disposed in concentric layers, where a tubercle rose above the general surface. Although they have not the same general form as ordinary bone-lacune (such as occur in Pferichthys and Coccosteus), yet they can hardly be called by any other name.” Professor Huxley, from whose memoir the above account of his predecessors in_ this inquiry is chiefly quoted, has left very little to be added to his exhaustive description of the structure of the shield in Cephalaspis Lyellii and C. Lloydii. I have examined the struc- ture in Cephalaspis Murchisoni and in C. (Pleraspis) rostratus, and have nothing of importance to add, but am able to confirm his description from the examination of a very large number of specimens. Taking then, first, Cephalaspis Lyellii and its allies repre- senting the Osrnosrraci, we will return afterwards to Cephalaspis Lloydii, C. rostratus, ‘&e. (Kner’s Pteraspides), typical of the section Herrrosrract. A naked-eye examination of the shield of Cephalaspis Lyellii shows in most specimens that deceptive apparent division of the outer surface into polygonal scales described by Pro- fessor Agassiz. In some curiously preserved specimens of a species from Scotland I have observed the divisions really existing, the upper surface of the shield being split up into polygonal pieces like a tesselated pavement. In other species, however, from England 8 OLD RED SANDSTONE FISHES. (C.ornatus, C. Salweyt), the outer surface, if well preserved, and not broken away as it usually is in C. Lyellii, shows a continuous polished surface ornamented with tubercles and papillee variously disposed, without any indication of a tesselated structure. In all species of Cephalaspis, in Auchenaspis, and Thyestes, this tuberculate ornamentation on a continuous polished surface is an unmistakeable naked-eye character of the shield. As regards the internal surface in Cephalaspis Lyellii and the allied species, it never exhibits any trace, when properly preserved, of the apparent tesselated structure seen when the external surface has been to a certain extent removed. It appears smooth, polished, and enamel-like, except- _ ing where it is furrowed by numerous shallow depressions which radiate from the region of the orbits and occiput towards the margin, before reaching which they repeatedly sub- divide and anastomose. ‘There can be little doubt, Professor Huxley observes, that these are the impressions of the vessels which ramified under the disc during life. These radiating channels leave their impress in the convex coats of the interior of the shield, and it was their occurrence which led Professor Agassiz to speak of “fibrous bones.” In those forms allied to C. Lyelliz, but differing in having the shield divided antero-posteriorly into two plates, the anterior plate alone shows these radiating channels ; and, indeed, they are often deficient in the posterior part of the shield of true Cephalaspids. The inner surface of the shield in Zhyestes is not known. ‘This radiated channeling of the internal surface of the shield may therefore be taken as a constant character in Cephalaspis Lyellii and its congeners. When a vertical section is made through the shield in Cephalaspis Lyellit, it is seen to be exceedingly thin, scarcely anywhere exceeding the 3th of an inch in thickness, excepting at the margins and spine; it appears to the naked eye to consist of a compact white calcareous substance. The shield in C. Salweyi, C. asterolepis, and others, is not so thin, and the posterior plates of those species which possess them are also much thicker than the anterior orbit-bearing portions. If the section of C. Lyell be now prepared and placed beneath the microscope, the appearances thus described by Professor Huxley’ as visible in a section near the spine are observable :—The section “presents three regions or layers, distinguishable from one another partly by their minute structure, and partly by the different mode of distribution of the vascular canals by which the tissue is permeated in each. The innermost or deep layer is made up of superimposed Jamelle, not more than ,j,th of an inch thick, each of which sometimes appeared to be still more finely laminated. Interspersed among these, at greater or less distance, are numerous osseous lacunee, whose long axes are parallel with the planes of the lamin. The length of these lacunz varies greatly, but may be taken at 4th of an inch on the average; some, however, are twice or three times this length, while others are much less. ‘The transverse diameter is equally variable; but 1 «Quart. Jour. Geol. Soc.,’ vol. xiv, p. 271, &e. 2 The illustrations of the microscopic structure of the Osteostraci will appear in the next portion of this Monograph, which will treat of that Section. CEPHALASPID A. 9 none that I measured exceeded ;3,th of an inch in this direction. The form of the lacunze is very irregular, in consequence of the long branching and anastomosing canaliculi which are given off, not only from their ends, but from their sides. In some parts the innermost layer appears almost black when viewed by transmitted light, in consequence of the quantity of air retained in the multitudinous lacune and canaliculi. Large vascular canals, measuring from 4th to j,th of an inch in diameter, whose inner openings correspond with brown spots distinctly visible on the imner surface of the shield, traverse the innermost layer very obliquely in their course towards the middle layer. ‘Their branches are few, and for the most part run parallel with the main trunk ; but they give off a great number of minute canaliculi which anastomose with those of the nearest lacunze. Such of these canals as I have seen in section were oval, their long diameters being parallel with the planes of the lamella. ... . The middle layer is distin- guished from the inner by the rarity or entire absence of the lacunz, and by the indis- tinctness of the lamination as compared with that of the deep layer. Such striations of the nearly homogeneous base as seem to indicate lamination are, inthe middle and inner parts of the middle layer, so disposed as to be nearly perpendicular to those of the deep layer, appearing to follow the course of the vascular canals. ‘The latter are continuous with the large vascular canals of the deep layer, but they are smaller, and form a close network. ach of the large canals, on reaching the middle layer, gives off several branches, which run nearly parallel with the surface (and therefore greatly inclined to the course of the great canals), and anastomose with those around, above, and below them. In this particular part of the disc, in fact, a large canal gives off as many as three tiers of these lateral branches, separated from one another by not much more than their own diameter, and all ramifying and anastomosing with one another. These lateral vascular canals have at first a diameter of about ,)th of an inch; but many of their anastomotic branches are much smaller. Sooner or later all these branches appear to end in a close ‘superficial network,’ which lies in the boundary between the middle and the superficial layers. ‘The latter or third layer of the disc sometimes appears structureless, at others presents an obscure vertical striation, as if it were, like enamel, made up of minute fibres. The superficial vascular network sends into it a great number of minute short processes, which branch out abruptly at their ends, like a thorn-bush or standard rose-tree, and end in excessively fine tubuli, like those of dentine.t . . . . This substance, it will be observed, corresponds very closely in structure with the ‘cosmine’ of Professor Williamson. I have been unable to find any trace of a ‘ganoin’ layer external to it.” A structure in every essential respect similar to this is found in all the other regions of the cephalic shield, the three layers varying somewhat in relative thickness. When - flakes of the inner layer are detached and examined, the axes of the lacunze of each lamina are seen to be directed nearly at right angles to those of the laminze above and below; "In those species in which the tubercles are well marked, it is seen that a “bush” of fine tubuli forms the centre of each tubercle. b) 10 OLD RED SANDSTONE FISHES. so that under a low power the section appears to be cross-hatched by a series of dark lines. “In flakes of the disc similarly treated, but containing more of the middle and outer layers, it is obvious that the great canals divide into the branches of the middle layer, which have already been seen in the vertical section, chiefly, if not only, along lines corresponding with the apparent sutures of the so-called ‘polygonal scales.’ The canals of the middle layer are very singularly arranged, passing from their origin, across these sutural lines and nearly parallel with one another, towards the centre of the adjacent ‘scales.’ ”” This, then, is the cause of the appearance of a division of the external surface ito distinct scales; the minute canals, with their reddish infiltration of matrix, showing very distinctly against the lighter general substance, or even allowing the shield to crack along these lines. It is worth noting, in addition to this, that the more superficial parts of the hard calcareous matter in each of these apparently polygonal scales has a radiated fibrous arrangement, the fibres diverging from the centre. This fibrous structure is visible in many specimens of Cey/halaspis from the Cornstones of Herefordshire. (4.) We now turn to the other members of the family Cephalaspide, typified by — Agassiz’s C. rostratus, C. Lloydit, and C. Lewisii, embraced by Kner’s genus Péeraspis, and forming my Section HrrprostRact. A naked-eye examination of suitable specimens of any of these shields shows the exterior surface to have a very peculiar ornamentation: instead of enamel-like tubercles, minute ridges and grooves, running parallel to and concentrically with one another, everywhere mark the convex superficies. ‘These ridges are exceedingly small in some species, and vary in different forms from ;,th inch or larger to th of an inch in breadth: they are usually crenated at the margins, and give the notion, in some species, of a linear series of minute tubercles fused together. They may not unfittingly be compared to the markings of the epidermis on the palms of the hand and fingers in Man and other Primates. The imner surface of the shield is seen to be quite smooth and polished, and exhibits no vascular channels, as in Cephalaspis Lyellii, but a few irregular ridges of some size, diverging from the centre of the scute or running parallel to the margins, having the aspect of lines of growth. Scattered rounded apertures are to be detected on this smooth inner surface, not above ;{,th of an inch in diameter. Ifa vertical section is now made through the shield, or if it be judiciously broken, the structure of the interior becomes obvious. Firstly, the thickness of the ‘shell’ im most places is greater than in specimens of Cephadaspis of the same size, being frequently more than ;,th of an inch, and in terminal portions (such as the margins, spine, or rostrum) much thicker. The most striking feature, however, which comes to view is the separation of the material of the inner and outer surface of the shield by a stratum of polygonal cavities, sometimes filled up by the infiltration of carbonate of lime, and having the appearance of 1 Op. cit., p. 273. CEPHALASPID A. 1] prismatic bodies.’ The innermost of the layers is a solid laminated stratum, having somewhat the appearance of the nacre of molluscan shells, whence Professor Huxley has termed it the ‘ nacreous layer ;’ a few vascular canals pass through it, opening by the minute apertures mentioned as apparent on the inner surface of the shield. The middle layer. which consists of the polygonal cavities, is formed by vertical processes of the laminated substance of the innermost layer; these vertical plates are so disposed as to form a net- work enclosing polygonal (4-, 5-, 6-sided) cells, described by Professor Huxley in Pi. Banksii as having an average diameter of 3th of an inch. ‘The septa forming the vertical walls of the ‘cavities’ bend over to complete their enclosure above, and are thus connected with the outer layer—that which is produced into ridges. This general structure is common to all those species of Agassiz’s Cephalaspis which have been separated as Pteraspis. Professor Huxley minutely describes and illustrates the microscopic appearances of this structure, as seen in a section of the shield of P¢. Lloyd. The most important fact to be observed in this and Pteraspis rostratus, which I have examined, is the total absence of ‘bone-lacunee,’ or anything like them. The three layers—the outer or ‘striated,’ the middle or ‘cancellated,’ and the inner or ‘nacreous ’—consist of a continuous finely laminated material. The lamine in P¢. L/oydiz, Professor Huxley says, have a thickness of about ;4,th of an inch. The inner layer consists of nothing but a compact mass of these lamine arranged horizontally; it is totally devoid of vascular canals or tubules, excepting where here and there a canal of some size passes from its aperture upwards - into the walls of the next layer. The lamine are arranged concentrically round the cavities in the middle layer, and large canals, about ;,th of an inch in diameter, pass along these walls towards the upper layer. In some cases these canals appear to open from the polygonal cavities ; in other cases they come directly from their apertures on the inner surface of the shield. In the upper wall of the stratum of polygonal cavities the eanals take a horizontal direction, still very few in number and of large size, receiving here, undoubtedly, branches from the cavities. The calcareous laminz are here arranged horizontally. Passing onwards to the uppermost layer, if the section has been made transversely to the surface-ridges, these appear in section as so many papilliform pro- cesses; if the section has been made along one of the ridges, a continuous horizontal layer is exhibited. Into this layer, or into these papilla, the horizontal canals send short branches (one to each papilla), which give off minute tubules in every direction ; these arbo- rescent tufts correspond to the vascular bushes in the external layer of Cephalaspis.” The 1 Mr. Edward Fielding, the artist who has so well executed the plates of this part of the Monograph, has been obliging enough to make a chemical analysis of specimens of Pteraspis from the Cornstones of Herefordshire. He finds that the substance of the fossil itself is almost entirely phosphate of lime, whilst the material sometimes filling the polygonal cavities is carbonate of lime. 2 In the Devonian species of Heterostraci each enlargement of the doubly crenated ridge seems to correspond to a tubercle of Cephalaspis, having a vascular tuft for its axis. 12 OLD RED SANDSTONE FISHES. laminated material is arranged round these tufts concentrically, the finer branches traversing it, much in the same way as dentinal tubules traverse dentine. Indeed, each of the sections of the ridges recalls very strongly the structure of a tooth or of a dermal defence of a Placoid Fish. There is no trace of a ‘ ganoin’ layer beyond the laminated material forming the ridges; but it is noticeable that the outermost laminze in specimens from the Cornstones acquire a pinkish-brown colour distinct from the rest of the shield, and readily scale off. In some species this outermost surface is dull; in others it is brilliant and polished. The fact of the distinct colour of this outermost part of the outermost layer, together with its tendency to separate, might lead to the impression that we have here a ‘ganoin’ layer; no such layer, however, is to be detected with the microscope. The absence of bone-lacune, the paucity of vascular canals, and the excavation of the mid-layer of the shield into large vascular sinuses, cannot be too strongly insisted on as the characteristic structure of these shields. Professor Huxley obtained from Sir Philip Egerton specimens of Kner’s Pteraspis from Gallicia, and most completely reconciled the German paleontologist’s description of the fancied resemblances to the Cuttle-bone with the true structure of the shield as he had ascertained it from English specimens. Kner’s ‘innermost layer’ is obviously the ‘nacreous layer’ described by Huxley; the ‘prismatic layer’ is formed by the polygonal cavities of the middle layer, which have become infiltrated with probably carbonate of lime, as in many examples from Britain; the striated layer undoubtedly corresponds to the outer layer in our Fish, and the identity in structure is complete. The chief differences, then, between these two sets of shields in their minute structure consist—first, in the absence of osseous lacune in the Pteraspidian forms, their presence in the Cephalaspidian’ forms; secondly, in the different general character and arrange- ment of the vascular sinuses; thirdly, in the different mode of arrangement of the external layer, which is, it seems, invariably marked with fine long ridges in the former, with minute tubercles in the latter. (c.) The Pteraspidian forms present such marked differences among themselves in the form of their shields, that it would be inconvenient to retain them in one genus; the same is true of the Cephalaspidian forms: hence I have adopted the Section Hurmrostraci (éreooc, of another kind; dorgaxov, a shell or dermal bone) for those with the peculiar structure of Pteraspis ; and the Section OsrrosTRAct (ocréov, bone) for those with the bony structure of Cephalaspis Lyellii. ‘There are other characters which separate these two Sections ; as, for example, the disposition of the orbits mentioned above. It is interesting to observe that each of the Sections presents a genus in which the shield is composed of a single simple piece: in the Hernrosrract this shield presents no trace of orbits, but is a simple discoid body (Scaphaspis): in the OstxosrRactr the simple shield has large diverging appendages (cornua), and mainly consists of an orbital 1 [ have ascertained the presence of bone-lacunee in duchenaspis and Didymaspis, genera allied in external character to the type Cephalaspis Lyellii. CEPHALASPID/. 13 3 region, the eyes being placed centrally. In those genera of Hernrostraci which develop a more complex shield the evolution shows itself in the enlargement of the anterior region, into a rostrum, orbital pieces, &c. In those Ostnosrraci which exhibit a shield of more than one piece the complexity results from the diminution of the orbital region and cornua, and the development of a large posterior plate, which is represented by only a small portion of the simple shield. § V. Zoological Position —Whilst no one has ever doubted that Cephalaspis Lyellit and its allied forms are Fishes, in the face of the specimens showing the body and fins, the Hefe- rostraci have been sometimes regarded as Cephalopods and Crustacea as related above. After his examination of the structure of Pferaspis, Professor Huxley remarks'—“ No one can, I think, hesitate in placing Péeraspis among Fishes. So far from its structure having ‘no parallel among Fishes,’ it has absolutely no parallel in any other division of the Animal Kingdom. I have never seen any Molluscan or Crustacean structure with which it could be for a moment confounded.” He is then led to conclude that Pteraspis is a Fish from its close relation in structure to Cephalaspis. Above, the differences between the two have ‘been dwelt on; the resemblances have to be noticed as forming the original reason for ‘retaining Pteraspis among Fishes. In each the shield is excessively thin, and composed of three or four layers: 1st, an ‘internal,’ composed of lamellz parallel with the surface, and traversed more or less obliquely by vascular canals; 2nd, next to this a ‘ middle layer’ containing the network of wide canals, or polygonal sinuses, the upper floor of which forms a ‘reticular layer’ in Péeraspis; 3rd, the ‘external layer,’ consisting of a cosmine, like substance raised into ridges or tubercles. Later conclusions as to the form of the shield in some Hersrosrract, and the discovery of the scales of one species, have rendered the piscine nature of Pteraspis undeniable. With regard to the relations of the family Cephalaspide to other groups of Fishes, I cannot do better than quote the opinion of Professor Huxley, the only one current at the present time on the subject. It must be borne in mind in connection with this question, that no indication whatever of an internal skeleton belonging to these Fishes has been found ; and that specimens showing the body of Cephalaspis prove by negative evidence that such a structure did not exist. While assigning a distinct family to the genera Cephalaspis, Pteraspis, Auchenaspis, and Menaspis, he remarks :’ “ No one can overlook the curious points of resemblance between the Siluroids, Cadlichthys and Loricaria, on the one hand, and Cephalaspis on the other, while in other respects they may be still better understood by the help of the Chondrostean Ganoids. Compare, for example, Scapi- rhynchus with Cephalaspis, or the great snout of Pteraspis with that of Spatularia. Iam inclined to place the Cephalaspide provisionally among the Chondrostei, where they will form a very distinct family.” Ll. Opweit., ps 277. 2 «Memoirs of the Geol. Survey of Great Britain, Decade X,’ p. 3. 14 OLD RED SANDSTONE FISHES. § VI. Distribution of Cephalaspide.—There is not enough known with regard to these Fishes to warrant any conclusions as to their distribution in time or space. ‘The facts are these: first, as to time—Cephalaspids occur in the Lower Ludlow beds, in the Upper Ludlow beds, in the Downton Sandstone, in beds called Upper Silurian in Russia and Gallicia, in the Ludlow Bone-bed, in the Passage-beds, and in that lowest portion of the Old Red Sandstone, which some geologists think ought to be called Silurian. As to distribution im space: they have been found in the West of England, in the East of Scotland, in Russia at Rootsikiille, in Gallicia, and in the Eifel. There is reason to hope that they may be found in North America, and other localities where Silurio-Devonian beds occur. There is not satisfactory evidence that the same species of Cephalaspide occur in any two of the above-named localities. The species of the West of England appear to be distinct from those of Scotland; and even in the Cornstones of Herefordshire and Wor- cestershire the species appear to have the most limited range and definite horizons. § VII. Synopsis of the Divisions of the Cephalaspide. Familia vel Sub-ordo Curnaraspipsa, Huzley. Pisces sine ossibus internis; caput scuto calcareo magno simplice vel composito armatum: d¢rwncus parvus, caudiformis, lepidibus rhombicis magnitudine variantibus ornatus. Sectio AHETEROSTRACI. Scuti materia sine “lacunis”’ osseis, intime sinibus polygonalibus excavata, superficie striis vel liris ornata. Genus 1. Scaphaspis.—Scutum simplex ovale. Genus 2. Cyathaspis——Scutum in quatuor partes divisum, ovale. Genus 3. Péeraspis—Scutum in septem partes divisum, sagittiforme. Sectio B.—OstTEOSTRACTI. ? Scuti materia “lacunis”’ osseis et tubulis vascularibus numerosis instructa, superficie tuberculis ornata. a. Oculis in medio scuto positis. ? Genus 1. Cephalaspis.—Scutum simplex semicirculare, cornubus _ lateralibus instructum. Genus 2. Auchenaspis.—Scutum in duas partes, anteriorem et posteriorem divisum ; anterior major, cornubus lateralibus instructa. Genus 8. Didymaspis.—Scutum in duas partes, anteriorem et posteriorem divisum ; anterior minor, sine cornubus lateralibus. CEPHALASPID 2. 15 B. Oculis extra scutum positis (?)." Genus 4. Tkyestes.—Scutum simplex cornubus lateralibus parvis instructum. 6s. 1—7. Diagrams of the Shields of the several Genera of the HetERosrracous and OSTHOSTRACOUS CEPHALASPIDS. 7 R. Rostrum, or Rostral region. C. Cornua. D. Central disc, or Discal region. S. Posterior spine. 0. Orbits. Fie. 1. Fic. 2. Pre: 3. ScaPHASPIS. CYATHASPIs. PTERASPIS. CEPHALASPIS. AUCHENASPIS. DipyMASPIs. __ 1 The evidence as to the deficiency of the orbits in Thyestes is not very satisfactory. Should the Russian Cephalaspide prove to have had their orbits placed mesially, the central position of the eyes would become a common character of the Ostrostract. Thyestes might then be retained for a genus, allied to Cephalaspis, in which the posterior spine and cornua are rudimentary. Sir Philip Egerton’s Cephalaspis Murchisoni might then be joined with the Russian species in this genus. 16 OLD RED SANDSTONE FISHES. THYESTES. § VIII. Sectto A.—Hererostract. Before proceeding systematically to describe the genera and species of HreTsRostRAcous CrrHaLAsPiDs, illustrated in the Plates, it is necessary to say a few words with regard to the generic divisions given in the Synopsis above, and the resemblances and differences which the groups exhibit. It will be observed that the division into genera is based on the greater or less complexity of construction of the shield, it being stated to be simple in the first genus, composed of four pieces in the second, and of seven in the third. The nature of this separation into pieces is a question of some importance, which must be considered, as well as the limits of variation in ornamentation, and other matters. 1. Variations in the Intimate Structure —The variations in intimate structure in the shields of Heterostraci are unimportant. As far as they have been examined in this respect the differences are merely in size, in the thickness of the layers, dimensions of the polygonal cavities, &c., which have been fully described (pages 10—12). 2. Variations in the Surface-markings.—The highly characteristic striations of the shield in Heterostraci vary considerably in different species. In all they bear a definite relation in their disposition to the outline of the plate or piece of the shield which they mark, each addition to the size of the plate appearing to result from the addition of new ridges parallel to and representative of those which have preceded. In this way the oval scutes become marked by a concentric series of oval groovings of greater or less regularity, whilst the distinct portions of the scute in Cyathaspis and Pteraspis show a regular consecutive series of transverse or curved markings. ‘The grooves and their interjacent ridges vary in size in different species, from 50 to 200, or even more, going to the inch. The form of the ridges varies; some species presenting a flat-topped ridge, others a rounded one. In some species the inner sides of the ridges are crenated, in others smooth. whilst in one species there are ridges of large size interstriated by much smaller ones. CEPHALASPID A. 17 _ Other markings than these regular ridges are to be noticed, as common to these fish- shields. In some of those with an oval shield of simple construction, the frontal margin is a little thickened, and irregularly marked by coarse serpentine groovings. Besides this, an important piece of evidence with regard to the living condition of the scutes is furnished by the presence of small depressions or “ pits’ arranged in series on the striated surfaces diverging from the centre. These pits bear all the appearance of being the positions or sites of mucous glands, such as are abundant in all Fishes, and from them we may conclude that a secreting inembrane was closely attached to the striated calcareous material. 3. Variations in the Form and Construction of the Shield.—In all the Heterostract the shield is markedly concavo-convex, the lateral convexity being greatest. The margins are slightly inflected in all,’ and thickened, especially where there is a posterior spine developed, or an anterior rostrum, In the Osteostracous genus Cephalaspis the anterior margin of the shield is inflected to a larger extent and thickened into a well-marked and strong “rim.” In Pleraspis the rostrum is thick, strong, and solid, as also the posterior spine and lateral margins forming the “cornua”; they are striated on the under as well as the upper surface, and evidently projected to some extent beyond the softer portions of the animal covered by the curved shield. In the simplest genus, Scaphaspis, the shield is composed of a single piece, of an oval shape, usually terminating acutely or in a short spine. In the most complex it is formed by the coalition of an anterior rostrum, two orbital pieces, a large central disc, two lateral perforate cornua,’ and a posterior spine (see fig. 3, page 15). The genus Cyathaspis is intermediate between these two; it presents a short rostrum, two lateral cornua, and acentral disc. By the aid of this form of shield, it is perceived that the anterior transversely marked margin of the shield in Scaphaspis represents the “ rostrum” of the two other genera, which in Cyathaspis is much reduced ; at the same time the acute termination of the shield (or spinous tubercle) typifies the large posterior spine of Pteraspis.* It is interesting to note the gradual development from the simple to the complex form in these shields (the differentiation accompanied with integration), and it is much to be desired that renewed researches may bring to light additional intermediate forms to complete the connection. The pieces making up the complex shields are not actually separate. They form one continuous whole, but are distinctly emarginated by grooves and differing convexity. 1 The lateral cornua in the genus Pteraspis are greatly thickened and inflected at the margin, forming, indeed, hollow dilated rims to the shield, which are perforated on each side. It is very difficult to find any explanation of these open excavated structures, unless they be “spiracles.” See fig. 3, p. 15, and Plate VII, figs. 8, 9, 16, 17. * The separation of the orbital pieces from the lateral cornua in Pteraspis is not quite satisfactorily ascertained. 5 The simple ovoid scutes of Seaphaspis (Cephalaspis Lloydii, &c., of Agassiz) have been sometimes regarded by local collectors and others, as the “torsos’’ or central portions of the apparently more complete Pteraspis. This idea is, however, clearly shown by well-preserved specimens to be erroneous. 3 18 OLD RED SANDSTONE FISHES. Each piece has its own series of striations, and on the smooth concave surface of the shield exhibits an apparent line of juncture with its neighbouring piece, in the form of an irregular rounded ridge. ‘The pieces are in fact fused at their margins; they were, there is reason to believe, distinct when the Fish was young; and specimens of small central discs and rostra of Pteraspis occur unconnected and separate: when growth was com- pleted anchylosis took place, and the pieces formed a single united shield. The unmis- takable appearance of each piece having had a separate development and growth exhibited by the concentric markings and irregularities of form in complete shields of | Pteraspis, when considered together with the presence of the ridges formed by the fused margins of the different pieces, seems to point to the conclusion that they were once uncon- nected and separated. It may be said, that if these pieces originated by separate develop- ment they would be united, not by a fusion of their margins, marked by an elevated ridge, but by sutures. In reply to this, I must call to mind the peculiar nature and histological structure of the calcified shield ; besides which anchylosis occurs in the head-bones of living Ganoids (Polypterus), leaving a straight, elevated ridge as its indication. It does not appear that there is any reason to suppose that the union of separate pieces in Pteraspis would have occurred otherwise than in the manner indicated ; and hence, I believe, I am justified in regarding the demarcated portions of the shield in Cyathaspis and Ptleraspis as separate calcifications. ; 4. Scales of Heterostraci—All that is known as regards the scales of these Fishes is from a single specimen found in the Cornstones of Herefordshire, which shows a few nearly equilateral rhomboid scales, ranged behind the posterior portion of a head-shield. It is probable that all Heterostraci possessed scales of this form, and very possibly others larger as well. Fie. 8. Diagram of Preraspis, showing both Shield and Body. 5. Hypothetical Form of the Body ; Position of the Mouth, &c.—The Woodcut, fig. 8, gives a purely hypothetical view of a species of Péeraspis. It has been introduced to exhibit the probable position of the shield and the form of the body. We snow absolutely nothing of these matters; but it appears highly improbable (by the absence of such remains) that the mouth of this Fish was provided with hard teeth, or in fact was anything more than a suctorial organ; and there is reason to believe that what is known with regard to Cephalaspis as to the shape of the body is true to some extent for the parallel Heterostracous forms. CEPHALASPID/. 19 6. List of Species of Huterosrract, arranged in Order of their Occurrence. Lower Luptow Beps. Scaphaspis Ludensis, Salter. Upper Lupiow Beps. ¢ 5a f> Downton SanpsTong. Scaphaspis truncatus, Huxley and Salter; Oyathaspis Banksii, Huxley and i Salter. Luptow Bonz-BEeD. Scaphaspis Ludensis 7, Salter ; Scaphaspis truncatus, Salter. Upper Sizurtan (Gallicia). Scaphaspis Knerii, Lankester. Lower Devonian (of the Eifel). Scaphaspis Dunensis, Roemer. Cornstones (West of England). Scaphaspis Lloydii, Agass.; Scaphaspis rectus, Lankester: Cyath- aspis Symondsi, Lankester; Pteraspis rostratus, Agassiz ; Pteraspis Crouchii, Salter. Lower Oxp Rep Sanpstone (Forfarshire). Scaphaspis, sp., Pteraspis Mitchelli, Powrie. 7. Descriptions of Genera and Species. Genus 1.—Scaruasris, Lankester. British Assoc. Report, Trans. Sect., p. 58, 1864. CrpuHataspis (Lioypir et Lewisi1), Agassiz. Poissons fossiles, vol. ii, p. 149, 1835. PrerasPis (Luoypii et Lewisi1), Kner. Haidinger’s Naturw. Abhandl., vol. i, p. 159, 1847. Derivation.—sxagn, a boat; aozic, a shield. Characters—Scutum cephalicum simplex, ovale, elongatum ; postice aliquanto attenu- atum et fere brevispinosum ; superficie striis et liris longitudinalibus ornata antice trans- verse dispositis. General Remarks.—This genus includes all the Heterostraci with a simple oval or elongate shield consisting of but one piece. It is the earliest to appear, occurring in the Lower Ludlow beds of the Silurian system. Of its British species two are Silurian, two Devonian. ‘Iwo species have been described from Continental localities—one by Reemei from the Laacher See, in the Eifel; the other from Gallicia, by Kner. The beds in which these occur may correspond either to our Lowest Old Red or Upper Silurian deposits. These two foreign species exhibit the characters of the English species in an exaggerated form. Scaphaspis Dunensis of Roemer (p. 20), from the Eifel, groups with Sc. Lloydii and Sc. rectus, resembling them in its smooth ovoid form. Se. Knerii (p. 20), from the Silurian beds of Gallicia, on the other hand, resembles much more closely our own Upper-Silurian species, Sc. Zudensis and Sc. truncatus, m its angular lateral margins and median ridge. OLD RED SANDSTONE FISHES. HeEAD-SHIELD oF ScapHaspis KNERII. \ f HEAD-SHIELD OF ScAPHASPIS DUNENSIS. 1. Scapwaspis Luoypir. (Agass.). Pl. I; and Pl. VII, figs. 1, 6, 17, 18. CEPHALASPIs LiLoyntII, — Lrwisit, Preraspis Liuoypti, Kner. Haid. Nat. Abhandl., vol. i, p. 159, pl. 5, 1847. Scapnaspis — Lankester. Brit. Assoc. Rep. Trans. Sect., p. 58, 1864. ; Agassiz. Poiss.foss., vol. ii, p. 149, pl. 14, figs. 8—10, 1835. CEPHALASPID. 21 Derivation—Named after Mr. Lloyd, of Whitbach. Characters —Scutum ovatum, parte anteriore latiore quam parte posteriore ; postice sub-acute terminatum, antice margine lato, curvato; lateribus-aliquanto depressis et inflectis ; superficie externa striis regulariter ornata preter superficiem marginis anterioris striis paucis eccentricis notatam ; superficie interna levi, duobus colliculis obliquis antice, et rugis lateralibus notata. Stratigraphical Position —Lower Old Red Sandstone (Cornstones). History.—This is one of the species determined by Professor Agassiz. One of the type specimens figured by the distinguished Ichthyologist is a very good representative of the form (PI. I, fig. 5); the other (fig. 10) is a broken, distorted specimen: neither of them show the surface-markings or substance of the shield. The specimen drawn in fig. 9, which is a young and somewhat flattened example of this species, was described by Professor Agassiz as Cephalaspis Lewisii, the specific name being in honour of Mr. Lewis, of Ludlow. I have no doubt that the specimen figured belongs to the species Se. Lloydit. It is quite possible that Professor Agassiz may have meant to include some other form with that he figured under the name C. Zewisi7, but there is nothing in the text of his work which tends in any way to explain his views more clearly. ‘The specific name “Lewisii” is a synonym of Z/oydii, and has been applied by the local and other geologists to a distinct form of Scaphaspis, which must henceforth receive a new name. General Remarks.—Scaphaspis Lloydii is one of the commonest species in the Corn- stones of Herefordshire and Worcestershire, and is generally obtained in a better state of preservation as regards its form than others of the section. It is, however, rare to obtain a specimen showing the surface of the shield at all satisfactorily. The largest specimens T have seen measure from three to three inches and a half in length, and one and three quarters of an inch in breadth at the broadest part of the shield, which is about the middle. Specimens of from two to two inches and a half in length are more common, but are usually more crushed and imperfect than the larger ones. The outline of the scute presents a most striking similarity to that of a perfect ‘ Cuttle-bone ” of the common Sepia officinalis. The form is that of a true ovoid, with the broader curvature in front ; the egg-like outline is, however, slightly modified by a depression of the sides of the shield along the narrower portion of the ovoid, which appears as though flattened out and slightly produced along its posterior lateral margins. This, I believe, is caused by the growth of the shield, since it is not nearly so apparent in small as in large specimens. An examination of the internal casts, some beautiful specimens of which are figured in the Plate, shows a few well-marked, broad furrows, corresponding to fainter depressions on the surface of the scute, which mark out regions of differing convexity on that surface. hese lines appear to be caused by variations in the process of calcification of the shield by the thickening of its margin, and subsequent regular increase ; somewhat in the same way as ridges are produced in the shells of Brachiopoda and Lamellibranchiate Molluscs, if it is allowable to compare so dissimilar structures. They distinctly trace out 22 OLD RED SANDSTONE FISHES. the form of the shield when of smaller size, the portion within the oval furrows being more convex than that which is outside and nearer the margin. In addition to these large concentric impressions, there are two furrows to be observed on all casts of this species, also indicated by depressions of the surface of the test, which run from the slightly undulated anterior margin towards a point in the median line, varying from one sixth to one third of the length of the scute from that margm. ‘These lines indicate the ‘rostral region, since they mark off the anterior portion of the scute, which may be con- sidered homologous with the rostrum in P/eraspis. Similarly the elliptic furrows mark off the portion corresponding to the lateral coruna. In large specimens the oblique ridges are of much greater length than in the smaller ones, which is owing to the fact that the whole anterior portion of the shield grows more rapidly than the posterior: this is most obvious from an examination of such specimens as figs. 2 and 7, in which it is clearly indicated by the lines of growth that slight addition only is made to the posterior part of the shield, the whole tendency of its development being in an anterior direction. A small notch or inward curvature of the anterior margin, observable in some casts in this species, appears to be due to the thickening of the test, which is somewhat produced at this point. ) It is hardly possible to describe the arrangement of the concentric striations on the surface of the test, and indeed it is not necessary, as the artist has taken great pains to render them faithfully in the drawings of the specimens submitted to him. A series of small pits, arranged in pairs and diverging from the centre of the scute, mark the position of muciparous glands, :and are seen in figs. 4 and 8. The form of the minute ridges between each consecutive pair of striations appears to be characteristic of the species of Heterostraci. In Scaphaspis Lloydit the strie are jth of an inch apart near the centre of the shield; at the sides they appear as though com- pressed, and are only th of an inch apart; the-ridge left between has a flat, glittering surface, and is obscurely serrated within the groove. Fragments of this species may be readily distinguished from Péeraspis rostratus, with which it is often associated, since in the latter the intervening ridges are much narrower, rounded, and have a dull surface. Localities.—In addition to the numerous quarries in the Lower Old Red of Hereford- shire, Worcestershire, and Shropshire, which have furnished abundance of specimens of Scaphaspis Lloydii, My. Powrie has submitted to me specimens from Forfarshire (obtained by the Rev. Hugh Mitchell, of Craig, near Montrose), which, though very fragmentary, preserve sufficient character to warrant the conclusion that they belong to this or a closely allied species. Agassiz states that he obtained this species from all localities in England where Cephalaspis Lyellii occurred. CEPHALASPIDA. 23 2. Scapnaspis rectus, Lankester. Pl. II, figs. 5—8, 12, 13 (figs. 6 and 8 are drawn upside down) ; and PI. VII, fig. 2. Syn. PrEraspis Lewis, of Collectors, not of Agassiz. Derivation. —Rectus (straight), from the straight equal form of the shield. Characters—Scutum cephalicum oblongum antice nec postice paulatim attenuatum ; postice decussatum et in medio indentatum, marginibus lateralibus rectis, parallelis, superficie externa striis concentricis parallelis ornata, superficie interna colliculis obliquis marginalibus notata. Stratigraphical Position —Lower Old Red Sandstone (Cornstones). History —By some means or other the specific name Lewisz7 has been connected with this species, although it is not the Cephalaspis Lewisii of Prof. Agassiz. The reason of this is probably that geologists finding two forms of Scaphaspis, closely allied, and knowing that Agassiz had distinguished two such allied species, concluded that one of those found by them must be his C. L/oydii, the other his C. Lewisii. This conclusion, however, is wrong, and I have accordingly renamed this species. General Remarks.—This is a much smaller species than the last, and is much less widely distributed, though abundant in some quarries. The largest specimens measure about two inches and a half in length. Scaphaspis rectus is characterised by the great narrowness and convexity of the scute, and by the fact that its sides are im a great measure parallel, so that it does not present the broad ovoid which we have in Se. Lloydi, but a narrow, oblong, semicylindrical body. Anteriorly, the margins of the shield converge, not, however, in a sweeping curve as in the last species, but somewhat obliquely, leaving a small horizontal frontal margin. A more decided differentiating character (not always preserved) is the presence (in casts) of a very obvious notch or excavation in the median line at the posterior margin of the shield, this margin otherwise being more nearly horizontal than in Se. Lloydii, and presenting the appearance of being truncated, instead of being produced to a subacute termination. These characters are quite sufficient to distinguish Se. rectus from Sc. Lloydii, with which it is likely to be confounded. I have, however, often had great difficulty in determining fragmentary specimens. With regard to the markings on the inner and outer surface of the test there is little to be said. Those on the inner surface differ very little from the similar furrows in Sc. Lloydii, being, however, less marked. There is a slight excess of growth in an anterior direction, but far less exuberant than in the allied species. It is from this cause that the lateral margins’ of the shield remain parallel. The two con- verging furrows arising from the frontal margin are traceable in casts of this species, as 1 The furrows on these margins indicate ridges in the test itself. It has to be borne well in mind that the internal casts, by which most Cephalaspide are known, only partially indicate the characters of the 24 OLD RED SANDSTONE FISHES. well as a series of similar furrows arising from the lateral margins (Pl. II, fig. 12). The striations on the external surface follow the outline of the scute, as in Se, Lloyd. ‘hey are, on the average, ,t,th of an inch apart near the centre of the scute, and leave a flat, smooth-surfaced, crenated ridge between them. The striations do not converge poste- riorly, since the outline of this part of the shield is not subacute, as in Sc. Lloydii, but nearly horizontal; hence the striations following the outline do not meet at a smal] angle, but terminate at the posterior notch. Distribution —tThis species occurs frequently by itself, and is often absent where other species of Devonian Heterostraci abound. It is met with in Worcestershire (Hall’s Barn, Aightington), and in Herefordshire frequently. At Whitbach it occurs with Pteraspis Crouchi, 3. Scapuaspis truncatus (ualey and Salter). PI. Il, figs. 1—8. Syn. Pruraspts TRuNcATUS, Hualey and Salter. Quart. Journ. Geol. Soc., vol. xii, p. 100, pl. ii, fig. 1, 1856. Derwwvation.—Truncatus (cut off), from the abrupt termination of the front part of the scute. ! Characters —Scutum elongatum, ovatum, convexum, antice truncatum, postice attenu- atum et brevispinosum; superficie externa rugis undosis longitudinalibus, tenuissime interstriatis ornata. Stratigraphical Position —Downton Sandstone, and Ludlow Bone-bed. flistory.—This species was discovered by Mr. Banks, of Kington, and described by Messrs. Huxley and Salter in 1855. General Remarks.—Vhis and the following species of Scaphaspis form a pair much in the same way as the two Devonian species Se. Lloydii and Sc. rectus do. Though having the typical form of shield, the Silurian Scaphaspides differ from the Devonian very obviously in their surface-striations, and in the presence of a median ridge, which runs longitudinally at the posterior part of the shield. Se. truncatus is quite a small species rarely exceeding an inch and three quarters in length. A slightly elevated ridge on the median line, running from nearly the centre of the scute, terminates posteriorly in a very short spine or tubercle. ‘The general outline of the scute is that of a well marked ovoid, truncated, and slightly emarginated in front, and acutely terminated behind. The markings of the surface are very characteristic, and, at the same time, are retained in very few specimens that I have seen, most of which are indifferent examples of the internal cast. From twenty to thirty undulating ridges shield; and this, perhaps, may prove a source of error in some cases in the view taken of the forms of the various shields. CEPHALASPID 2. 25 extend in a longitudinal direction from near the frontal margin, and converge towards the spinous termination of the ovoid scute ; these are interstriated by the fine grooves charac- teristic of all Heferostraci. The space in front of the origins of the longitudinal ridges is transversely striated and somewhat thickened, indicating the rostral region, as in other Scaphaspids. Distribution —The only locality at present known for this species in the Downton Sandstone is Kington, Herefordshire. I have seen a fragment in a slab from the Ludlow Bone-bed. 4. Scapuaspis Lupxnsis (Salter). PI. II, figs. 4, 4 a. Syn. Preraspis Luprnsts, Salter. Annals and Mag. Nat. Hist., 3 ser., vol. iv, p. 45, woodcut, fig. 1, 1859 Derivation.—Ludensis, belonging to Ludlow. Characters—Scutum elongatum, oblongum, ovato-quadrangulare, antice et postice truncatum, postice spina brevissima armatum ; superficie externa colliculis longitudinalibus rectis antice transversis, non interstriatis ornata. Stratigraphical Position.—Vhis species has been found in both the Lower and the Upper Ludlow beds. History.—Mr. Salter, in 1859, described this fish-shield from specimens obtained by Mr. Lee and Mr. Robert Lightbody, of Ludlow. General Remarks.—This, the oldest vertebrate animal of which traces have been met with, is known by but a very few imperfect specimens. It was first obtained from the Upper Ludlow deposits, but was afterwards detected near Leintwardine, in Shropshire, by Mr. Lee, in the Lower series bearing that name ; this discovery added enormously to the age of the fossil. The form of the scute in Scaphaspis Ludensis is that of a much elongated ellipse truncated at both ends, the small spinous tubercle situated near the posterior margin of the shield hardly projecting beyond it. Like that of Sc. truncatus, the spine is the ter- mination of a lightly marked central ridge; and, as in that species, the inner surface of the shield exhibits indications of this ridge and of the coarse longitudinal furrowing of the outer surface. As Sc. rectus resembles Sc. Lloydii in many points, but differs in being much narrower, and in not terminating acutely behind, so does Sc. Ludensis resemble and differ from Se. ¢runcatus. The ornamentation on the outer surface of the shield is in the form of straight longitudinal ridges and furrows, coarser than in other Heterostraci. These ridges do not converge, but run parallel from one end of the shield to the other, a small frontal portion being left which is transversely marked. The length of the largest scute I have seen is about one inch and a half. Localities —The neighbourhood of Ludlow ; and Church Hill, Leintwardine. 4 26 OLD RED SANDSTONE FISHES. Genus 2.—Cyatuaspis, Lankester. Brit. Assoc. Rep., Trans. Sect., p. 58, 1864. Syn. Prerasprs (Banks11), Hualey and Salter. Quart. Journ. Geol. Soc., vol. xii, p. 100, 1856. 4 Derivation.—xvaboc, a spoon ; aczic, a shield. Characters.—Scutum cephalicum ovale, aliquanto elongatum, postice truncatum et brevi-spinosum; in quatuor partes divisum,—vostrum breve anterius,—cornua duo lateralia, margimibus scuti admodum depressis formata,—centralem discum. Superficie stris et liris longitudinalibus, in rostro transversis, ornata. General Remarks.—The evidence on which this genus is formed is more scanty than that on which the foregoing and the following genus rest. At the same time, it appears impossible to associate Pteraspis (Cyathaspis) Banks of the Downton Sandstone with either the simple oval-shielded Scaphaspis or the more elaborate Pteraspis. The specimens that I have examined are not very good or satisfactory examples, though they are, I believe, the best yet discovered; and from them it appears that in these shields the anterior portion is marked off as a distinct rostrum, projecting beyond the rest of the oval scute; the sides are much depressed, and also marked off from the rest of the scute, the demarcated pieces extending on each side to the rostrum, from which they are similarly separated. This ‘“ marking off” is effected by a break in the parallel striations of the upper surface, by a linear groove emarginating each portion, and by dissimilarity in convexity and curvature of the marked-off segments. These pieces had in all probability each a distinct development and growth, as in Pteraspis. The lateral pieces correspond to the lateral cornua of that genus; the spine and | orbital parts are not distinct pieces; the former being but a spinous tubercle on the disc” when it occurs, and the latter in all probability are connected with the rostrum. Some casts of Cyathaspis shields show a marginal tubercle on either side the rostrum, cor- responding with similar tubercles in Pteraspis casts, which are produced by the supposed orbital aperture. I have not had specimens which show the shield itself of Cyathaspis sufficiently well to demonstrate the nature of these tubercles in the casts, but it seems probable that they are connected with the orbits. The species belonging to this genus, concerning which more information would be most welcome, are two in number at present ; one is known by ascore or so of poor specimens from the Downton Sandstone of Kington, the other by a single fine internal cast from the Cornstones of Herefordshire. 1. Cyaraaspis Banxsu ({Hualey and Salter). PI. Il, figs. 9,10, 11; and Pl. IV, fig. 6. Syn. Preraspis Banksit, Hualey and Salter. Quart. Journ. Geol, Soc., vol. xii, p. 100, pl. ii, fig. 2, 1856. Derivation.—Named after its discoverer, Mr. Banks, of Kington. CEPHALASPID A. 27 Characters—Scutum ovale: discus centralis convexus postice truncatus et spina brevissima, striis longitudinalibus minutis externe ornatus: rostrum breve, rotundatum, striis transversis ornatum : cornwa cum disco usque ad truncationem juncta, externe striis longitudinalibus ornata. Stratigraphical Position.—This fossil has at present been found only in the Downton Sandstone. History.—Mr. Salter, in conjunction with Prof. Huxley, described this species as _Pteraspis Banksii, from specimens found by Mr. Banks at Kington in 1859. General Remarks—There is little to be said of this species beyond what has been said in the description of the genus. ‘The rostrum projects a little beyond the rest of the scute, and is rounded off in a gentle curve ; the lateral cornua embrace the central oblong convex disc throughout its length, terminating abruptly with it in the posterior region, which has the aspect of being sharply truncated. A little above the truncation of the scute in the median line is the origin of a short spine or projection. I have seen nothing in the specimens I have examined of the “costuli” and double series of ten “tubercles” mentioned in Messrs. Huxley and Salter’s description, which I cannot but think is erroneous in this respect. On either side of the rostrum in casts is a marginal tubercle, probably connected with the orbit, but specimens sufficiently well preserved to make this clear have not been found. ‘The surface is very finely striated, and the convexity of the central portion of the shield in uncrushed examples very great. The length of the scute of this species is from an inch and a half to an inch and three quarters ; its breadth from a little over an inch to an inch and a half. Locality—kKington, Herefordshire. 2. Cyatuasris (?) Symonpst, Lankester. Pl. VI, fig. 5. Cyatuaspis (?) Symonpst, Lankester. Brit. Assoc. Rep., Trans. Sect., p. 58, 1864. It is useless to give any definite character to this species at present, as it rests on a single specimen, which, though indicating a remarkable Fish belonging apparently to this genus, is too imperfect to warrant any further affirmation in regard to it. This specimen has been for some time in the Collection of the Geological Survey, and is labelled as coming from the Cornstones, Herefordshire; it is a well-pre- served smooth-surfaced cast of the interior of the scute, and measures three inches and a half in length, and two and a half in breadth, being one of the largest He- terostracit known. Its form is rather rectangular than ovoid. The central and median portion is convex and raised up above the rest of the scute; it projects backwards, some little way from the flatter portion, and has a rounded truncated posterior margin. In front the flat portion is produced into a small rostrum which has an incurved frontal 28 OLD RED SANDSTONE FISHES. margin, the flat portions at the sides form the lateral cornua, and terminated abruptly. with a rounded horizontal margin, leaving the central convex portion to pass backwards unembraced, for the length of about half an inch. A comparison of this cast with those of Cy. Banksii shows considerable similarity in the parts of the two scutes. I have named it after the Rev. W. S. Symonds, of Pendock, Worcestershire, who is one of the most energetic investigators of the Old Red strata of that district. Genus 3.—PTERASPIS. Syn. Prpraspis (Kner), Huzley. Quart. Journ. Geol. Soc., vol. xii, p. 100, 1856. _ Lankester. Brit. Assoc. Rep. Trans. Sect., p. 58, 1864. Derivation.—nrepov a wing, aomic a shield; either from the sculptured surface of the shield (?) or from its supposed wing-like cornua. Characters.—Scutum cephalicum sagittiforme; in septem partes divisum,—rostrum conicum elongatum anterius,—spizam longam posteriorem,—discum magnum centralem, quadrangularem vel ovatum rostro junctum,—dwuas partes orbitales, rostro et disco utrinque junctas,—duo cornua lateralia quicque fossa tubulari perforatum, utrinque disco et parti orbitali juncta; superficies externa in partibus variis diverse striis minutis parallelis ornata ; superficies interna levis, colliculis paucis et fossis notata. General Remarks.—In this genus the Heterostraci attain the most composite form of scute. Seven distinct portions are marked off as well by ridges on the lower surface of the shield as by furrows, and by the arrangement of the strize on the external surface. The form of the shield is no longer merely oval, or a modified ovoid, but the diverging appen- dages become so far developed before, behind, and laterally that the outline somewhat resembles what botanists term a sagittate leaf, though this does not express the form accurately. The head-shield of Pteraspis presents a recognisably piscine appearance, and the orbital apertures with the long snout and projecting perforated “ cornua” prevent the possibility of its true character being overlooked. The shield is composed of seven pieces instead of four as in Cyathaspis: a long conical rostrum ; a large more or less oval central disc, into which is fitted a large posterior spine; an orbital piece on each side attached to the rostrum and disc; and a lateral cornu on each side attached to the disc and orbital piece, from which it is scarcely distinct. In the median line, where the rostrum joins the disc, is a minute, round depression, with circular striations, perhaps forming an eighth, though very diminutive, piece. The distinctness of these pieces has been already spoken of (pages 17 and 18). ‘The facts bearing on the matter, as far as this genus is concerned, are the following: each of the seven parts has its own series of striations, arranged transversely on the rostrum, circularly around the orbital aperture CEPHALASPID A. 29 on the orbital portions, longitudinally on the lateral cornua, concentrically around a central point on the disc, longitudinally on the spine: none of these series of striations run into or form part of another series; each piece is marked off from its adjacent piece by a slight groove on the upper surface, and by a thickened ridge below : the striz are of the same size in young as in old specimens: in young specimens it is not unusual for the rostrum, spine, and orbital portions to be separate from the disc (particularly in the longirostrate species): the proportionate size of the various pieces of the shield is the same in large as in small specimens. From these facts the imferences are: Ist, that the growth of the scute was not simple,—that is, it did not proceed merely along the margin of the shield as from one centre ; 2nd, that the number of concentric or parallel striz must have been increased on each portion as the scute advanced in size ; 3rd, that this increase in size must have taken place around or along the margin ofeach piece. Hence, it appears that each portion had a separate growth; and when quite young the pieces were very possibly (so far as their calcareous substance is concerned) quite distinct ; they united as they advanced in size, and became firmly joined when fully developed by the thickening and fusing of their adjacent margins. The inner surface of the shield in Péeraspis is smooth and shining; for about half its length, however, on the under surface, the rostrum is striated as on its upper surface ; it is then suddenly excavated, and the hollow of the shield commences, with which the soft parts of the Fish were connected. The inflected thickened margin of the shield all round is striated, as also the under surface of the bulging cornua and the posterior spine. There are ridges on the inner surface caused by the anchylosis of the different pieces ; and there are a very few remarkable ridges radiating from the centre of the disc, varying in the different species. Corresponding to the minute circular piece between the rostrum and disc is a deep pit on the inner surface, which occurs in the species of Pteraspis ~ most markedly, and appears also to exist in Cyathaspis (PI. II, fig. 11). I can offer no explanation of this pit and the diverging ridges. The restoration of the shield of the genus Pferaspis was first attempted by Professor Huxley (‘ Quart. Journ. Geol. Soc.,’ vol. xvii, p. 165, 1861), who gave an outline of the scute in the commoner species Pferaspis rostratus. It has been a matter of very serious difficulty to obtain anything like a satisfactory notion of the form of the “ cornua”? in this genus, since the abundant casts only mislead, and specimens with the test preserved are hard to get. The figures 8 and 9 in Plate VII represent what I have been able to ascertain by breaking up many specimens. The “‘cornua” are thick solid bodies, which do not diverge from, but merely project above, the thickened margin of the disc. A large oval orifice opens on each side into the substance of the shield just below the projecting cornu. A short passage, partly perforating the cornu, runs obliquely forward from this orifice, opening widely again into the concavity of the shield, quite at its margin, and within the area of its posterior half. Perhaps this pair of holes were connected with the branchial apparatus. The Rev. Hugh Mitchell, of Craig, was the first to point out 30 OLD RED SANDSTONE FISHES. their existence. Professor Agassiz curiously misunderstood the nature of the scute in his Ceph. rostratus, for he was able in the specimen figured in Pl. IV, fig. 2, to find two large dorsally placed orbits similar to those of Cephalaspis Lyelhi ; it is obvious enough that these do not exist, while it is equally difficult to recognise the “ethmoid” spoken of by him, as well in rostratus as in his Lloydii and Lewisi. The genus Pteraspis includes three species, limited to the Lower Old Red Sandstone, two occurring in the Cornstones of the West of England, the other obtained in Forfarshire. I have had specimens from Herefordshire, which may indicate other species ; they are not sufficiently satisfactory to warrant further notice. 1. Preraspis Crovcnu, Salter. Pl. III; Pl. IV, fig. 5; Pl. VI, figs. 4, 7, 8; and Pl Vii igs; 458 ail: PrerasPis Croucutt, Salter, MS. = — Lankester. Brit. Assoc. Rep., Trans. Sect., p. 58, 1864. Derivation—Named after the Rev. J. F. Crouch, of Pembridge. Characters.—Scutum antice et postice attenuatum : rostrum longum et attenuatum ; discus elongato-cordiformis, apice utrinque oblique truncato: spina longa: cornua lateraha magna, crassa: superficies superior rostri parsque inferioris striis curvatis parallelis ornate, superficies externa disci et partium orbitalium striis concentricis ornata. Stratigraphical Position.—Lower Old Red Sandstone (Cornstones). History.—Specimens of this species in the Ludlow Museum have been named Pt. Crouchit by My. Salter ; and this name I have adopted. This species was not known to Prof. Agassiz. General Remarks.—This form of Pteraspis is distinguished by its long attenuated ros- trum, and by the casts of its disc being shaped like an elongated “ heart”’ truncated obliquely on each side of the apex, and having a broad, deeply undulate anterior margin. The rostra are frequently found detached from the rest of the shield, and I have seen no specimen in which they themselves were attached to it: many specimens, however, show the cast of the concave part of the rostrum, in conjunction with that of the orbital portions and disc (the substance of the rostrum itself having in all probability broken away in the stone, on account of the upward curve which it makes). One specimen belonging to Mr. Humphry Salwey, of Ludlow, shows the striations and contour of the elongated rostrum in connection with the cordiform shield; the specimen is the largest I have seen, and it is probably owing to its mature age that the connection between rostrum and disc was sufficiently strong to persist. It is from this specimen chiefly that the pointed long rostra have been shown to belong to the cordiform discs. The part of the rostrum, about half its whole length, which is striated both above and below, is thick and solid, and presents a considerable development of the layer of polygonal cavities within. Its CEPHALASPID #. 31 sides in this part are nearly parallel, converging slightly towards the anterior rounded termination. Where the rostrum becomes excavated, convex and smooth beneath, it broadens very much, and joins on either side the orbital pieces, and in the centre the disc and circular piece. Most casts show well the existence of a sharply marked pit corresponding to this circular bit. The posterior spine in this species is long and strong, its insertion extending into the disc for a distance of about one half of its own length, and one third of that of the disc. The test of the posterior part of the shield is very thick, and gives the perfect shield an outline somewhat different to that of the convex internal cast. The cornua appear to be thick processes, overlying the oblique penetration of the shield, as in those of the succeeding species. The markings of the internal surface of the disc in most cases afford means of determining otherwise doubtful specimens of P¢. Crouchii. The median line, after the insertion of the spine, is marked by a sharp depression, which becomes fainter towards the anterior portion. On either side of this diverges, from near the centre of the disc, an irregularly marked straight ridge, passing anteriorly in an oblique direction ; sometimes these are not perceptible as ridges. Below these straight ridges on either side are two curved ridges, the lower pair of which pass from the centre nearly right and left, the upper pair (nearest the straight ridges) curving obliquely towards the anterior margin. Tn the casts of the interior, which so frequently occur, these ridges are, of course, exhibited as furrows, and the median groove as aridge. Similar ridges exist in Pf. rostratus, but are not usually so well marked. The surface-striations of the shield are exhibited in the drawmgs. Those on the rostrum are semi-elliptical in direction, the open part of the ellipse bemg turned anteriorly on the upper, posteriorly on the under surface. The groovings are ,3,th of an inch apart (average), leaving a flat, bright, crenate ridge between them. Localities —Pteraspis Crouchii has been obtained from Whitbach and Ludlow. The greatest number of specimens now occur in that neighbourhood. ‘It is also met with at Abergavenny. I have never heard of specimens of this species occurring in immediate association with Pt. rostratus; and I believe it has a different horizon, Scaphaspis rectus occurs with it and the Cephalaspis asterolepis of Dr. Harley. _ WNote.—The scales of Péeraspis (Pl. V, figs. 3, 5, 8) which I obtained at Cradley, Herefordshire, perhaps belong to this species. 32 OLD RED SANDSTONE FISHES. 2. PreRasPis RosTRatTUS, Agassiz. PI. IV, figs. 1, 2, 3, 7, 8; Pl. V, fig. 4; Pl. VI, figs..1, 2,°3, 6, 10; and PL. VIL figs.) Sjyojuam 12) V3 iaG, Wo09: CEPHALASPIS ROSTRATUS, Agassiz. Poiss. foss., vol. ii, p. 148, pl. 14, figs. 5, 6, 1835. PTERASPIS ROSTRATUS, Hualey. Quart. Journ. Geol. Soc., vol. xvii, p. 165, wood- cut, 1860. — — Lankester. Brit. Assoc. Report, Trans. Sect., p. 58, 1864. Derivation.—Rostratus, bearing a rostrum. Characters.—Scutum grandius et minus attenuatum quam in Pf. Crouchii ; rostrum cuneiforme, discus elongatus, quadrangularis lateribus parallelis, spyima magna, cornua lateralia angularia; rostri superficies superior striis undosis, antice curvatis, parallelis ornata; superficies externa disci et partium orbitalium striis concentricis ornata. Stratigraphical Position—Lower Old Red Sandstone (Cornstones). History.—This species is more common than the last, and appears to be the one which was known to Prof. Agassiz. Prof. Huxley gave a restoration of its outline’ in 1860. Agassiz’s type specimen is drawn in PI. IV, fig. 2. General Remarks.—This is a larger species than Pt. Crouchii, sometimes attaining a length of four inches and a half from the tip of the rostrum to the termination of the disc. It differs from Pt. Crouchii in the shape of its rostrum, which is broader at its termination and shorter relatively than in that species,—in the shape of its disc, which is that of a parallelogram with the anterior angles rounded off,—in the shape of its cornua, which are thicker and stronger in Pt. Crouchii,—in the relative length of its spine, which is less deeply inserted in the disc. It is by no means uncommon to get specimens of this species with all the parts adherent. The rostrum does not readily break off as in Pt. Crouch ; and, not being proportionately so large nor upwardly curved, it is more frequently preserved, and knocked out of the stone with the rest of the shield; indeed, it is rare to see a specimen of Pt. rostratus in which all the parts are not attached, though still rarer to see one in which they are not to a large extent inextricably imbedded in the dense Cornstone. The markings of the inner surface of the disc consist of two straight ridges on each side passing obliquely forwards from the region of two small oblong depressions situated on either side of the median line, a little posterior to the centre of the shield, and of a broad anterior elevation in the median line, corre- sponding with a similarly broad depression of the outer surface. ‘Two curved lines, similar to those in Pt. Crouchii are noticeable in some specimens; but they are not so constant or so sharply marked as in that species. ‘The striations on the outer surface are 1 The very obscure cornua were not quite correct in this outline (nor in that which I gave in the ‘Quart. Journ. Geol. Soc.,’ vol. xx, p. 194, pl. xii, fig. 10); but in all other respects the restoration was most valuable. CEPHALASPID &. 30 disposed concentrically round a middle point, as in P¢. Crouchit, but this portion of the shield is apparently elevated into a slightly convex boss. The surface-striations of Pt. rostratus differ considerably from those of Pé. Crouchit. 'They are very fine, the groovings being only 4th of an inch apart near the centre of the disc, and leaving a rounded dull ridge with crenate margins between them. Localities —This species occurs abundantly at Cradley, near Malvern, Herefordshire, and in beds of the same horizon at Whitbach and in the neighbourhood. It is found associated with Scaphaspis Lloydit. o. Preraspis Mitcneiui, Powre. PI. V, figs. 1, 2, 6, 10, 11. Preraspis MitcHELii, Powrie. Geologist, vol. vii, pp. 170—!72, 1864. — — Lankester. Brit. Assoc. Report, Trans. Sect., p. 58, 1864. | have not had specimens of this species sufficiently perfect to enable me to charac- terise it properly. The specimen figured (Pl. V, figs. 6, 10) was briefly noted by Mr. Powrie in the ‘ Geologist,’ in 1864, and an outline-sketch was given. The lateral cornua are not seen, but the disc, rostrum, and spine are, to a certain extent. The disc appears to be intermediate in form between that of P¢. rostratus and Pt. Crouchii. The specimen and others even less well preserved were obtained by Mr. Powrie from quarries in Forfarshire, where Heterostraci had been discovered for the first time in Scotland by the Rev. Hugh Mitchell. A few rhomboidal scales also obtained by Mr. Powrie from this locality (Pl. V, fig. 1) probably belong to this or an allied species. ter ured, . vm cay ais da tat dea ate © ate ol ee eS: ; i } é 5 ny ‘ an rf : bh ia } ¢ q 4 7 Beare’ 0 ; . ik ae ‘ 7s ; i. ¥ F ; a ; 2 ! \ “ Fic. 4a. PLATE I. Scaphaspis Lloydii, showing a portion of the test. Cornstones, Herefordshire. In the collection of the late Mr. Wyatt-Edgell. A well-preserved internal cast, showing the lines of growth and anterior convergent ridges ; from Shropshire. In the collection of the late Mr. Wyatt-Hdgell. Internal cast ; from Cradley, Herefordshire. In the author’s cabinet. A specimen showing a large portion of the test; from Herefordshire. In the collection of the late Mr. Wyatt-Edgell. Anterior portion of the same, enlarged three diameters, to show the irregular striation of the anterior margin and mucous-gland-pits. 4 6. Ornamentation of the surface, to show the crenation of the ridges. 5. 6. 7. 10. ie A fine internal cast. Herefordshire. In the collection of the Rev. J. Crouch. An internal cast ; from Cradley, Herefordshire. In the author’s cabinet. A large internal cast; from Herefordshire. In the collection of the Rev. J. Crouch. 6 points to the lateral emargination, representing the cornua. a to the rostral region. A specimen showing nearly the entire surface of the test; from Herefordshire. In the collection of Mr. Lightbody, of Ludlow. A small example of Sc. Lloydii. Agassiz’s type specimen of Cephalaspis Lewisi, figured ‘ Poiss. foss.,’ Tab. 1 4, fig. 8 One of Agassiz’s type specimens of Cephalaspis Lloydii, figured ‘ Poiss. foss.,’ Tab. 1 4, fig. 10. A much crushed and indistinct cast of Sc. Lloydit. Cradley. In the author’s cabinet. calla M&N Hanhart omy F Fiel dling delt A \PI 1 n Wd ” Fic. 10. } da, 12. } 13. PLATE II. Scuphaspis truncatus, an internal cast. Kington, Herefordshire. In the Museum of Practical Geology. . Side view of the same specimen. Sc. truncatus, ternal cast. M. P. G. - 5s a beautifully preserved specimen, showing the sculpturing of the surface. M.P.G. Sc. Ludensis, from the Lower Ludlow beds of Church Hill. In the Museum of Practical Geology. . Se. Ludensis (concave surface) ; from the Upper Ludlow beds, near Ludlow. Mr. Lightbody. Se. rectus, a fragment, showing the surface markings. Shropshire. Mr. Wyatt- Edgell. A concave cast (figured upside down) of Se. rectus ; from the Cornstones, Worces- tershire. In the author’s cabinet. Observe the posterior notch. A very fine convex cast of Sc. rectus; from Worcestershire. In the British Museum. Sc. rectus ; from a sketch made for Professor Huxley. Cyathaspis Banksit, concave cast, showing the surface-striations; from the Downton Sandstones, Kington, Herefordshire. M.P.G. 4, Lateral cornua; c, rostrum ; d, central disc. Convex casts of C. Banksii; from Kington, Herefordshire. In Mr. Banks’ collection. a, Orbital ? tubercles. Convex casts of Sc. rectus; from Hall’s Barn, Worcestershire. In the author’s cabinet. HE Fieldings Del SCAPHASP U I é& CYVATEHAS'P er F Waller, imp 18, Hatton Garden a) Fia. 10. ih 2: 13. PLATE III. An internal cast of Pteraspis Crouchii, showing the form of the concave part of the anterior rostrum. Herefordshire. Rev. J. Crouch. A cast (laterally compressed), with the posterior spine attached. It shows well the radiating ridges. Herefordshire. A similar specimen, not compressed. Herefordshire. Cast of Pteraspis shield, probably P¢. Crouchii, in yellow sandstone. Shropshire. Mr. Wyatt-Edgell. Rostrum of Pt. Crouchii, showing anterior dorsal surface and cast of posterior ventral surface. Rev. J. Crouch. Rostrum of Pt. Crouchi, showing anterior under surface. From Mr. Lightbody’s cabinet. Portion of disc. Herefordshire. The only specimen showing the tapering rostrum attached to the cordiform disc, Acton Beachamp, Herefordshire. Mr. Humphrey Salwey. Portion of disc, with surface-striations preserved. Herefordshire. Mr. Wyatt- Edgell. An unusually well-marked cast of the disc. Herefordshire. M. P. G. Rostrum. Herefordshire. Dr. Grindrod. Part of rostrum in cast. View from below of a specimen of the rostrum, showing its excavated portion. Herefordshire. Mr. Lightbody. 13 a. Lateral outline of the same specimen, showing upward flexure. All the figures are of Pt. Crouchi. E Fielding del! Fic. PLATE IV. Pteraspis rostratus. A fine specimen, showing the concave surface of the shield. Observe the pit between rostrum and disc, the elevation into a ridge of the fused margins of these two parts, and the long ridges of the disc. Cradley, Herefordshire. In the author’s cabinet. Agassiz’s type specimen of C. rostratus. The hollow character of the cornua is well seen in the fractured cornu of the left side. Whitbach, Herefordshire. Sir R. I. Murchison. Poiss. Foss. ‘Tab. 1 a, fig. 6. Pt. rostratus. A specimen showing well the form of the rostrum, and its relation, to the anterior margin of the disc. Cradley. Author's cabinet. A portion of the disc of P¢. Crouchii, showing the truncated posterior margin and the deep insertion of the spine. One of the largest specimens (an internal cast) of Pf. rostratus yet found. Here- fordshire. M. P. G. Cyathaspis Banksit, a very well-preserved and complete shield. Kington. Mr. Lightbody. Disc of Pf. rostratus, with rostrum, cornua, and spine detached. Compare with Pl. III, figs. 8, 8, and 10. Abergavenny. Author’s cabinet. Pt. rostratus, showing, on the left side, the oval perforation of the cornua. M&NHanhart . imp E Fielding delt PITEPRASPIS. ROSTRATUS.Zrc 10. while PLATE V. Scales of Pteraspis sp., Perthshire. Mr. Powrie. Pteraspis Mitcheli. View of the concave surface ; on the right hand is seen the internal aperture of the cornual perforation. Perthshire. Dundee Museum. Scales of Pteraspis sp., attached to a portion of the striated head-shield (see figs. 7 and 8). Cradley, Herefordshire. Author's cabinet. Pt. rostratus. Cast of disc and rostrum, with cornua attached; on the right-hand side the matrix is seen filling the hollow passage of the cornu, and surrounded by bony matter. Cradley. Author’s cabinet. Enlarged view of some of the scales of fig. 3, showing surface markings. Pt. Mitchel, a specimen without the cornua and spine. Perthshire. Mr. Powrie. Magnified view of the striations of the part of the head-shield preserved in the specimen drawn in figs. 3 and 8. Side view of the specimen drawn in fig. 3. A huge Heterostracous head-plate, perhaps Scaphaspis Lloyd. Near Ludlow. Mr. Humphrey Salwey. The intaglio of fig. 6, showing the spine. Pt. Mitchelli? a very large but obscure specimen. Perthshire. Rev. Hugh Mitchell. oS ae 2 ens E Fielding del’ Pe) eee ns cee TO aad vane aman wt BA 1 os) 7 MONTOGRAPHICAL SOCIETY, INSTITUTED MDCCCXLVII. VOLUME FOR 1867. LONDON: > THE BRITISH PLEISTOCENE MAMMALIA. We BOYD DAWKINS, M.A. FES. GS., W. AYSHFORD SANFORD, F.G.S. PAKS DE BRITISH PLEISTOCENE FELIDA FELIS SPELAA, Goupruss. (Pages 29—124-. Prares VI—XIX.) LONDON: PRINTED FOR THE PALHONTOGRAPHICAL SOCIETY. 1868. PRINTED BY J. E, ADLARD, BARTHOLOMEW CLOSE, E.C. 2nd Phalanges. FELIS SPELAA. MEASUREMENTS. Digit 2. 14. Bl. B., Taunton. Felis tigris, Felis leo, Brit. Mus. 114 L. W. A. S. Inch. lnch. Inch. 1°76 1°26 1°38 2°00 1:45 1:42 0°78 0°68 0°60 0°60 0°43 0°50 0:78 0°64 0°60 0°91 0°50 0°54 22: Bl. B. | 1°80 1:58 1°47 | 1°87 Ica LZ 0°85 0°61 0°68 | 0°58 0:44 0°50 0°72 0°62 0°60 0°75 0°42 0°50 15. | Bl. B | 1°65 1°30 | 1:28 1°75 1°36 | 131 0°83 0°56 0°66 0°56 0°43 | 0°51 0°75 0°60 | 0°54 0°69 0°48 0°58 16. Bl. B. e577 2 1°20 We 75 McA oll 0°82 0°59 0°60 0°55 0°54 0°50 0:72 0°55 0°50 0°65 0°45 0°50 Sesamoids ; Pl. V, figs. 15, 16. The sesamoids figured are from Bleadon. They certainly belong to a large carnivore, and are precisely similar to those of a lion in form, though of course much larger in size ; and as they are from the Bleadon Cave, where Felis spelea abounds, and bear is extremely scarce, we do not doubt that we are right in ascribing them to the former animal. D 30 PLEISTOCENE MAMMALIA. CHAPTER VI. SKULL: Pls: Vi, VE Vill ie ox xa: CONTENTS. § 1. Introduction. Skulls of Felis spelea § 11. Sguamosal. Sound in Europe. § 12. Malar or Jugal. § 2. Basi-occipital bone. § 13. Lachrymal. § 3. Exoccipital ; Supra-occipital. § 14. Ethmoid. § 4. Basi-sphenoid. § 15. Wormian. § 5. Ali-sphenoid ; Pterygoid. § 16. Parietal. § 6. Pre-sphenoid ; Orbito-sphenoid. § 17. Frontal. § 7. Palatine. § 18. Nasal. § 8. Mazillary. § 19. Measurements. § 9. Intermazillary. § 20. Summary. § 10. Petrosal; Mastoid ; Tympanic; Malleus. | § 1. Introduction. Skulls of Felis spelaa discovered in Hurope—Perfect skulls of Felis spelea are so very rare that we have had the opportunity of studymg no more than three which at all approach completeness—that from Sundwig, in Westphalia, now in the British, Museum, of which Professor Owen has given excellent figures in his ‘Memoir on Thylacoleo,”! and the two which we figure from the Mendip Caves, now in the Taunton Museum. ‘The typical skull figured by Goldfuss, and copied by Cuvier,’ has altogether eluded our search. It was from the cavern of Gailenreuth from which Lord Enniskillen and Sir’Philip de Grey Malpas Egerton have obtained vertebrae and many other bones of Felis,“as well as large quantities of the remains of bears. There are, however, several figures of the skull. M. de Blainville® figures a fine and ' «Philosophical Transactions, 1859, pt. i, pls. xii, xv. 2 “Nov. Act. Nat. Cur.,’ tom. ix, p. 476, pl. Ixv; ‘Oss. Foss.,’ 1825, 4to, tom. iv, pl. xxxvi, fig. vi. 3 «Qstéographie, Felis,’ pl. xv. \ FELIS SPELAA. 31] apparently perfect specimen, which M. Pictet reproduces in his ‘ Paléontologie.” From the text of the former writer we gather that the figure was taken from a plaster cast in the possession of Count Munster, the original having been found in a cavern in Fran- conia. D’Alton? also figures a perfect or nearly perfect skull from Muggendorf. Other naturalists’ who have turned their attention to the Pleistocene Fauna describe and figure fragments only of the spelzean skull; for the species, though widely spread through central and western Europe, is nowhere abundant, nor are the remains generally found in ‘a perfect state. The largest English fragment hitherto figured is that found in Kent’s Hole Cavern by Mr. McEnery, and drawn on a slightly reduced scale by Mr. Scharf, and published by Mr. Vivian. It is also figured in a woodcut in Professor Owen’s ‘ British Fossil Mammalia.’ It represents only a portion of the right maxillary and intermaxillary, with the dentition of an animal of the average size. Of the two skulls from the caverns of Mendip which we figure, the more perfect and smaller lay for many years after its purchase from the Rev. D. Williams, broken and unseen in a box in the Taunton Museum. The fragments were put together by the pre- sent able curator, Mr. Bidgood, and the teeth were afterwards found among a quantity of those of hyzena; and thus a fair specimen of the skull of the British spelzean lion was obtained. Mr. Beard, the explorer of bone-caverns in Somersetshire, on his collection of bones being bought for the Somerset Archzeological and Natural History Society, told us of the presence of a skull of a lion from Hutton Cavern, among the bones some time before purchased from his old rival the Rev. D. Williams. As the skull agrees in its condition and colour with the remains from that cave, we have no hesitation in affirming that the original of Pls. VI, VII, VIII, and IX, is the specimen alluded to. When Mr. Beard’s collection was brought to Taunton, the small pair of nearly perfect lower jaws figured in Pl. VI was found to correspond exactly with the skull in respect to age, size, and colour. We know that it was the practice of Messrs. Williams and Beard to work at the same cave at the same time, and to share the contents. In this way very frequently a fine specimen was divided between them, even in the case of the long bones,—femora, humeri, and the like. These, now that both collections are in the hands of the Archeological and Natural History Society, are in many cases reunited, and form perfect bones. We have ' therefore every reason for believing that the lower jaws in Mr. Beard’s collection belong 1 «Paléontologie,’ 4to, 1853—7, pl. vi. 2 *Raubthiere,’ pl. vii, fig. a, 5, ¢, d. 3 Leibnitz, ‘Protogea,’ pl. xi; Buckland, ‘Reliquie Diluviane,’ pp. 17, 62, 261; Schmerling, ‘ Oss. foss. de Lidge,’ tom. ii, p. 14; Marcel de Serres, Dubreuil et Jean-Jean, ‘Oss. foss. de Lunel-Viel,’ pp. 101, 107, pl. vii, fig. 1; Rev. W. Vernon, ‘Phil. Mag.,’ 1829, p. 225; McEnery, ‘Cavern Researches,’ edit. G. E. Vivian, Esq., 1859; Owen, ‘Brit. Foss. Mam.,’ p. 161; ‘Rep. Brit. Assoc., 1842; Falconer, ‘Quart. Journ. Geol. Soc. Lond.,’ vol. xvi, p. 490; Blackmore, ‘Cat. of Fossils in Salisbury‘ Museum,’ p. 101; Boyd Dawkins, ‘Quart. Journ. Geol. Soc.,? xviii, p- 115; Ed. Lartet, ibid., vol. xvi, p- 475; Falconer, ibid., pp. 99, 104 ; Baron Anca, ibid., p. 460. The last two notices may refer to a dis- tinct species, as the animal is described merely as a large Felis. 32 PLEISTOCENE MAMMALIA. to the same animal as the skull in that of the Rev. D. Williams. They were, however, among the bones from Sandford Hill, and were therefore described as from that cave im our first chapter. A closer examination has shown us that bones from different caverns in the Mendip can be recognised with much probability by their condition and the colour of the matrix. In both these respects the lower jaws strongly resemble the remains from Hutton. They exactly fit the spelean skull from Hutton. We therefore suppose that they must have been accidentally misplaced either by Mr. Beard himself, or in the removal of his collection to Taunton, and that they really belong to the same animal that furnished the skull in the Hutton Cave to the Rev. D. Williams. The skull in question is that of an adult rather past the prime of life. The teeth are decidedly worn, and the alveolus of the right upper tubercular molar is partially removed by absorption, which proves the loss of the tooth during the lifetime of the animal. _ Its state of preservation is shown by the following list of its component bones. A minute portion of the right nasal is present in the angle of the frontal suture, also small portions of the palatines adjoming the maxillary suture, and also that with the pre-sphenoid. The maxillaries with their dentition are nearly perfect, the palatine process being excepted. From the (otherwise perfect) inter-maxillaries the incisors have gone. The left third — incisor was diseased, and probably lost durmg life. The right malar and squamosal are absolutely perfect, and the left nearly so. The posterior or cribriform plate of the ethmoid, and a part of its central plate, are present, so that the beautiful tracery with which it fills the anterior end of the cranial cavity may be seen by looking through the foramen magnum. The greater part, however, of the bone has disappeared. ‘The vomer is entirely wanting. ‘The pre-sphenoid and orbito-sphenoid are nearly perfect. The superior parts of both frontals are nearly perfect, but the orbital portions are much broken. The right tympanic bulla is perfect. The articular portion of the squamosal (‘“corsal” of Straus- Durkheim) is preserved on both sides, as also are the lower jaws, with the exception of the coronoid processes and a small portion of one of the condyles. The basi-sphenoids, ali-sphenoids, parietals, mastoids, basi-occipitals, exoccipitals, supra-occipitals, paro- occipitals, and Wormian, are nearly absolutely perfect. A small part of the lachrymals is attached to the frontals and maxillaries. The petrosals appear to be perfect, though of course they are but slightly visible. The second skull (Pl. X) is from Mr. Beard’s collection, and was found in Sandford Hill Cave. Along with it were found the lower jaws described in Chapter I and Pl. I of our work. ‘They were accidentally labelled as coming from Bleadon, and the mistake transferred to our pages was not discovered until the chapter had gone to press. Both skull and lower jaws belong to a young adult. Several bones of an animal exactly cor- responding im size and age were found along with them; and as those adjoiming each other in the skeleton exactly fit, we have reason to believe that we have a considerable portion of the same individual. Unfortunately it was the practice of Mr. Beard “to restore,” not very skilfully, the missing parts of crania and other fossils with hard plaster, FELIS SPELAA. 33 and in this case the result has been very great difficulty and risk in taking his work to pieces and articulating the skull for scientific purposes. After such rough usage the exactness of fit of the component parts and the symmetry have been to some extent lost. We have figured the skull as it stands now free from plaster, without attempting a restoration which very possibly might have been erroneous, and which certainly could have served no scientific end. The specimen retains a small piece of each nasal in siti, and a large part of the right palatine. The right maxillary is all but perfect, with a small portion of the left. ‘The right intermaxillary also is in part present. The only teeth remaining are the large premolars (four), and a portion of the right canine. Both premolars are nearly perfect, together with the left squamosal and a large part of the right, so that we can form an adequate idea of the size and form of the zygoma ticarch. The frontal bones are present, but their supra-orbital processes are much abraded. The left tympanic bulla is much broken, and the right is almost entirely gone. The basi-sphenoid is all but gone, and only the lower and posterior portions of the ali-sphenoid are left attached to the lower part of the ali-sphenoids and the squamosals. Both mastoids are imperfect. The basi-occipital is present, but the exoccipitals are abraded, and the supra- occipital is gone ; and of the par-occipitals, only the left fossa remains. In addition, we figure in Pl. XI the maxillaries and intermaxillaries of another specimen from Sandford Hill Cave, which is also from Mr. Beard’s collection. It is of very large size, and exhibits the perfect adult dentition, with the exception of the small tubercular upper true molars, the small premolars (two), and one first incisor. We also give a figure of the articular portion of the squamosal of a gigantic animal from Bleadon Cave (Pl. IX, figs. 2 and 3). Besides the skulls we figure, we have examined a large number of fragments, which are for the most part in the Taunton Museum, as well as the nearly perfect specimen from the Sundwig Cave, in Westphalia, now in the British Museum; from it are absent the greater part of the zygomatic arches, a portion of the left palatine, the pterygoid pro- cesses, the upper part of the supra-occipital, and two thirds of the nasals, the left orbito- sphenoid, together with the adjoining part of the frontal, the ethmoid, and the vomer. We have also examined the specimen figured by Professor Owen and Mr. Scharf from Kent’s Hole, and a similar fragment from Muggendorf, now in the British Museum, and another like it from Ravenscliff in Gower, in the possession of Colonel Wood. ‘These constitute the materials which we have at hand for writing this chapter on the skull of Felis spelea. We shall compare the spelzean skull bone by bone with that of the living species most closely allied to it, that is, lion and tiger, beginning with the basi- occipital. § 2. Basi-occipita. (Pls. VIII, IX. No. 1).—The basi-occipital forms the posterior portion of the base of the skull, and is regarded as the centrum of the occipital vertebra by all who hold the “vertebral theory.” From the slight bulging of the sides it 34. PLEISTOCENE MAMMALIA. is somewhat hexagonal in form. It is rather longer than wide, and is slightly longer than the basi-sphenoid, to which it is firmly attached by a straight transverse suture. It forms a strong plate of bone of nearly uniform thickness, articulated behind to the exoccipitals in the whelp by a suture, of which the lateral portions are transverse. In the median line, however, it sends back a square process, the free end of which is the lower and anterior border of the foramen magnum. ‘The lateral portions of this suture are interrupted by the “foramen condyloide,” which passes from the posterior edge of the “foramen lacerum posterius”’ (a) to the interior of the cranium near the anterior border of the foramen magnum. This transmits the large motor hypoglossal nerve. ‘The sides of the bone are in contact with rather than articulated to the tympanic, and above that to the petrosal, the junction between them being interrupted posteriorly by the large irregular “foramen lacerum posterius ” (a) or “foramen jugulare”’ for the passage of the eighth pair of nerves and a large vein connected with the internal jugular. ‘The medullary surface is somewhat concave, forming a lodgment for the “medulla oblongata.” The flatness of the lower surface is broken by two large rough depressions on each side close to the tympanics for the insertion of the “ recti antici majores”’ of the head, which take their rise in several roots on the pleurapophyses of the cervical vertebrae. Immediately behind them are the smoother but larger impressions of the “recti antici minores,” which have their origin in the “atlas.’’ These impressions are represented in PI. VIII, in front of the “foramen lacerum posterius.” In the median line we sometimes find in Lion the commencement of the tubercle for the attachment of the “ constrictor pharyngis superior,” which is, however, mainly attached to the basi-sphenoid. This does not occur, as far as we know, in Felis spelea. With the exception, perhaps, of a slight tendency to greater width in the speleean as compared with the leonine and tigrine basi-occipitals, there is no difference worthy of note. § 3. LHeoccipitals and Supra-occipitals. (Pls. VI, VII, VII, IX, X. Nos. 2,3, 4).— It is more convenient for purposes of description to treat these as one bone rather than in accordance with their centres of ossification, because they are never found separate except in the very youngest individuals, and because, firmly anchylosed together, they form the main surface of connection between the head and the trunk. ‘They compose a strong plate of bone, triangular in outline, firmly articulated to the basi-occipital (Pl. VIII, No. 1) at right angles, and with it circumscribing the foramen magnum (Pls. VIII, IX). On either side of the latter are two short thick pedicles of bone which point downwards and backwards, and support the condyles by which the head is articulated to the atlas. These point in their upper portion upwards, in the middle backwards, and in the lower downwards. Their edges project over the sides of the pedicle, forming a fossa, which is called the “condylian fossa” (4, Pls. VI and X). ‘These are the portions termed by Professor Owen the exoccipitals (No. 2). ‘The two inferior angles of the bone are composed of the paroccipitals, or paramastoids, FELIS SPELASA. 35 as they are sometimes called, which form on either side a wide and deep cavity on the outer surface, which may be called the paramastoid fossa (c, Pls. VIII and IX), receiving the projecting sides of the glenoid cavity of the atlas, and thus combining great firmness of articulation with freedom of lateral motion. It runs as far downwards as the origin of the massive bony pedicle which projects downwards, and ends in the paramastoid tubercle (Pls. VI, VIII, IX, ¢), which is homologous with the jugular tubercle in man. The par- occipital articulates with the mastoid in front, and inferiorly with the tympanic; if we hold to the “vertebral theory” of the skull, as propounded by Professor Owen, it is homo- logous with the parapophysis of the basi-occipital vertebra. The portion of the occipital which composes the apex of the triangle, and together with the exoccipitals complete the arch over the spinal cord, is the supra-occipital (Pls. VI, VII, VIII, IX, No. 3), which would be the neural spine of the vertebra. Its sides are decidedly convex, while the continuation on the paroccipitals is concave, so that the whole side of the triangle is distinctly sigmoid ; the interior and inner surface forms the back wall of the cranial cavity, and is entirely in contact with the cerebellum, for the convolutions of which it is deeply grooved and waved. The upper edges of the superior and outer surface are covered with high radiating ridges, of great sharpness and strength circumscribing depressions of various depth, which may be called the splenial fossze (Pl. IX, /), from their being the points of insertion for the tendon of the great splenius muscle. That descending from the apex, remarkable for its size, serves for the attachment of the cervical ligament (ligamentum nuchee). The articulation of the exoccipitals with the basi-occipital has already been described in our account of that bone. Each of the paroccipitals (paramastoids) covers by a broad overlap the posterior end of the tympanic bulla. Above this it is firmly articulated to the posterior border of the mastoid through the whole length of that bone. Above this, in some individuals among the larger Feles, the supra-occipital articulates with the parietal, but generally the descending process of the Wormian or interparietal passes down- wards so as to join the upper part of the mastoid and prevent the connection of the parietal with both the ex- and the supra-occipital. The suture with the Wormian is of great depth in the aged animals of the larger Feles, owing to the great height of the lambdoid or occipital crest. In advanced age the whole of the lambdoid suture is obliterated, and its position is marked by a very sharp and massive crest, the lambdoid or supra- occipital. Ligaments and Muscles of Occiput.—A very detailed account of the muscles and liga- ments of Felis are given in Straus-Durckheim’s great work,’ to which we would refer those who wish to become acquainted with the details of this part of the subject. We shall content ourselves with giving a list of the principal ligaments and muscles, with their 1 «Anatomie, descriptive et comparative, du chat, par Hercule Straus-Durckheim,’ 2 vols. 4to, plates folio, Paris, 1845. This work is perhaps the most perfect monograph on the comparative anatomy of a single animal that exists in any language. 36 PLEISTOCENE MAMMALIA. points of origin and insertion, so far as they are connected with the part of the skull under consideration. The ligaments which connect the head with the neck are the following:—The cervical ligament, or “ligamentum nuche,” which is comparatively small in the genus Felis, springs from the neural spine of the first dorsal vertebra, and passes among the muscles of the neck to its insertion on the summit of the occipital crest. The “atlo-cephalic capsular,” the representative of the capsular of the head in man, occupies the position expressed by its name. It connects the skull, not only with the atlas, but also with the odontoid process of the axis. The “anterior superficial atlo-cephalic” is the equivalent of the “ anterior cervical” inman. It springs from the anterior and upper border of the hypapophasis of the atlas, and is inserted into the posterior border of the basi-occipital. ‘The ‘ median superficial posterior atlo-cephalic,” the posterior superficial of the atlas in man, fills the space between the upper part of the foramen magnum and the corresponding part of the atlas. It also extends down the sides of the posterior portion of the paroccipital fossa, which it connects with the exterior of the glenoid cavity of the atlas. The “rectus posterior capitis” has its origin in this hgament. The “deeper” ligament of the same name as the last appears to be simply the fibrous envelope of the spinalcord. The “anterior lateral atlo-cephalic,” or hgament of the first vertebra in human anatomy, springs from the border of the glenoid cavity of the atlas, and is inserted into the basi-occipital on the inner border of the condyles. The “lateral atlo-cephalic,” a strong ligament not found in man, has its insertion on the internal border of the paroccipital fossa, whence it passes downwards and backwards, and is attached to the inferior border of the glenoid cavity of the atlas. It hinders the excessive rotation of the head on the atlas. Two other smaller ligaments, having the same function, are called the “ superficial,” and the “ deep transverse posterior,” “‘atlo-cephalic.” They have the same insertion as the last, but pass upwards and back- wards to their points of attachment on the upper border of the neurapophysis of the atlas. They are not found in man. The “lateral odonto-cephalic,” having the same name in man, closes the list; it passes from the end of the odontoid process to the inferior angle of each occipital condyle. In giving a list of the muscles attached to this part of the skull, we will begin with those that serve for the movement of the whole head. The great “complexus”’ of man is represented in Felis by two muscles, that called “biventer cervicis” by Eustachius and Albinus, the “intersectus” of Straus-Durkheim, and that to which the latter author confines the name of “complexus.” The insertion of the “intersectus” is on the inner portion of the occipital arch, over the foramen magnum. Thence it passes backwards, dividing into four principal tendinous roots, which are attached to the transverse processes of the seventh cervical and first three dorsal vertebra. It adheres to the cervical ligament through its whole length. Above this hes the “ com- 1 Op. cit., vol. ii, p. 241 et seq. FELIS SPELAA. 37 plexus” proper, which has the same insertion as the preceding, and passes backwards to the diapophyses of the cervical vertebra, and thence to the three anterior dorsals. The great, lesser; and middle recti posteriores of the head underlie these two muscles; their insertions extending nearly to the upper edge of the foramen magnum, and their origins being in the atlas and axis. ‘hese five muscles serve to lift up, and to a certain extent to rotate, the head. Their points of insertion are therefore necessarily of great strength in the larger and more powerful Feles. In Felis syelea they are not more massive propor- tionally to the size of the animal than in the living tigers and lions. The great splenius muscle springs from the “cervical ligament,’ and an aponcurosis which connects it with the last cervical and first five dorsals, and ends in a short strong tendon which is inserted into the occipital bone immediately behind the lambdoid suture. Its enormous size and strength in Felis spelea is seen by the large fossze for its insertion in Pl. IX, fig. 1. It takes part in the same movement as the preceding five muscles. The “trachelo-mastoid ” of Douglas rises from five tendinous roots attached severally to the last: four cervicals and the first dorsal, and passes in the form of a long thin band | along the side of the neck to its insertion on the paroccipital. Near it is inserted the “yectus lateralis,’ which has its origin on the ala of the atlas. The “ superior obliquus” of the head has nearly the same direction as the last, but passes within it to be inserted on the surface of the paroccipitals in the upper part of the condyloid fossa, at the point where they join the exoccipitals. These three muscles, together with the splenius, bend the head from side to side. The great size of the lateral ala of the atlas stands in direct relation to the development of these muscles, in animals that shake and worry their prey, such as lion, tiger, and Felis spelea. M. de Blainville' states that the occipital crest is prolonged further backwards in the tiger than in the lion, a point which we have remarked to be by no means of characteristic value, and that the condyles are more detached in the former than in the latter animal. We have frequently found the converse of this latter statement to be true. He also writes that in these two points the plaster cast of Count Miinster’s specimen of Felis spelea” agrees with tiger and differs from lion. We are unable to lay hold of any character in the occipital bone that would differentiate lion from tiger or from Fe/is spelea. § 4. Basisphenoid (Pl. VIII, No. 5).—The basisphenoid is articulated to the presphenoid by a transverse suture, which is clearly visible even in animals of considerable age. It is much wider though about the same length as this latter bone. The form of the inferior or guttural surface exposed in the perfect skull is roughly triangular, the apex of the triangle being cut off by the presphenoidal suture, and the base being formed by 1 € Ost, Felis,’ p. 28. 2 “Ost. Felis,’ p. 108. 6 38 PLEISTOCENE MAMMALIA. that with the basioccipital, which in the old animal is obliterated as completely as the frontal suture of human anatomy. The sides are covered to a great extent by the overlap of the anterior portion of the tympanic bulla, and in front by the guttural process of the alisphenoid. ‘The surface in the larger Feles is nearly flat or slightly concave. A small foramen on each side of the basisphenoid im the suture between it and the alisphenoid is the posterior opening of the canal which conducts the Vidian nerve and artery to the foramen sphenoidale : from this proceeds a well-marked groove backwards along the above- named suture, to the foramen lacerum medium ; it then passes along the suture between the tympanic and alisphenoid just outside the foramen caroticum, to the small foramen by which the nerve makes its exit from the petrosal proper, within which bone it branches from the facial nerve. When detached the bone in the lion is of the same truncated triangular shape, but it is somewhat wider than was before apparent in consequence of the overlap before described. Its vertical thickness is slightly greater anteriorly than posteriorly, owing to the com- mencement of the upward slope from the bottom of the “sella turcica” to the olivary process within the cranial cavity. The dorsum ephippii rises to a considerable height in all the larger Feles, and is turned much forwards, the inclination from the summit to the posterior edge of the bone being at an angle of 30° to 35° with the horizontal. The sides of the “dorsum” expand into lateral ale, somewhat like the wings of a moth, and homologous with the “ posterior clinoid processes” in man. ‘The arch formed by them, and the anterior clinoid processes is sometimes completed in Felis. We have not, however, met with an instance of this in the speleean skulls: on each side of the “dorsum” is a furrow for the “internal carotid artery.’ Sometimes there is a small median foramen on the guttural surface, and two minute foramina on the back of the “ dorsum ephippi.” Muscles.—The peristaphyline (Straus-Durckheim) muscle, the representative of the internal muscle of the same name in man, has its origin close to the alisphenoidal suture, at a point where it crosses a line joining the “foramina ovalia.” Its office is to lower the “velum palati.”” The superior constrictor of the pharynx is attached to the posterior por- tion of the bone close to the basi-occipital suture. The only point worthy of note in comparing this bone in Felis spelea with those of lion and tiger is that it has a tendency to be somewhat wider in the two former animals than in the latter. This width, however, is very variable, and cannot be considered characteristic. § 5. Alisphenoid; Pterygoid (Plate VIII, No. 6).—The alisphenoid, usually described as the ala of the sphenoid, and treated by Straus-Durckheim' as a mere process of that bone, lies immediately in front of the squamosal in the surface of the cranium. As, Op. cit., Vol. 1) P..o90- FELIS SPELAIA. 39 however, it is easily separated from the basi- and pre-sphenoid, while the suture between it and the pterygoid is obliterated at a very early age, we treat the alisphenoid and the pterygoid as one bone for descriptive purposes. Nearly the whole of the outer surface of the bone is visible in the perfect skull, as a vertical plate running upwards to form part of the walls of the cranium between the temporal and optic fossze. It also extends horizontally as far back as the petrosal, passing under the anterior part of the tympanic. Inferiorly, the pterygoids extend downwards and backwards on either side of the great guttural groove, ending in the thin, strong, hamular processes in lion and tiger, which in our spelaan skulls are unfortunately broken away. For purposes of description this compound bone may be divided into the horizontal or guttural portion, the supero-vertical or temporo-optic, and the infero-vertical or pterygoid portions. ‘The first of these is a narrow plate, transversely concave, covering the postero-lateral edges of the guttural surface of the presphenoid and the antero-lateral edges of the basisphenoid. In the basisphenoidal suture is the orifice of the Vidian canal, by which the nerve and artery of that name enter the alisphenoid, and pass forwards into the orbit at the external border of the “ foramen sphenoidale.” We have already described the groove connected with this canal on the surface of the basisphenoid in our account of that bone (§ 4). The infero-vertical or pterygoidal processes curve downwards from the horizontal portion, and articulate anteriorly with the palatine by a vertical suture. The hamular processes, in which they terminate, are the equivalents of the internal pterygoid plates of human anatomy, the externals being represented by a slight longitudinal ridge immediately in front of the foramen sphenoidale. M. de Blainville’ states that the hamular processes of the tiger are less delicate than those of the lion. The variations, however, in this respect, in both these species, do not enable us to confirm this observation. As might be expected, these parts have not occurred in a fossil state. The horizontal plate expands posteriorly behind the Vidian canal, and articulates with the squamosal just within the boundary of “the glenoid cavity.” At this point it joins the supero-vertical or temporo-orbital process, which is a long, thin, triangular plate, highly convex externally, articulating behind with, and passing under, the squamosal by a highly concave suture, above with the antero-inferior angle of the parietal, in front with the postero-inferior angle of the frontal, and the orbito-sphenoid. At the bottom of this suture is a notch, which, together with a corresponding surface of the latter bone forms the large “foramen sphenoidale,”’ to a certain extent the representative of the sphenoidal fissure or “foramen lacerum anterius”*® in man, and giving passage to the third and fourth nerves, to the first branch of the fifth pair, or the trigeminal, and to the sixth. Immediately behind this, and rather lower down, is the foramen rotundum, for the trans- 1 *Qst. Felis,’ p. 28. 2 Straus-Durckheim, op. cit., vol. i, p. 395. 8 Holden, ‘ Human Osteology.,’ 3rd edit., p. 78. 40 PLEISTOCENE MAMMALIA. mission of the maxillary portion of the trigeminal, and more widely separated and exactly opposite the glenoid cavity is the foramen ovale (y), which transmits the infra-maxillary, or mandibular branch of the fifth pair.’ The latter is called the carotid foramen by M. de Blainville? and some others, in the mistaken belief that it transmits the carotid artery. The foramen caroticum is in all the Feles extremely small; and the external orifice being entirely covered by the Eustachian process of the bulla, and generally, but not always, surrounded by the substance of that portion of the tympanic, it may be said to be within the foramen lacerum medium, immediately on the inner edge of the groove for the Vidian nerve. The internal orifice which is immediately in front of the apex of the petrosal proper leads directly to the groove for the artery described above as on the cerebral surface of the basisphenoid. ‘This is very distinct in the smaller cats, but is less so in the larger. The foramen itself is larger in the jaguar than in any other large Felis that we have examined, admitting a small wire. It is very small in the tiger, and still smaller in a panther, and even in very young lions; in some old animals of both these species it appears to be entirely closed. In Felis spelea it closely resembles the lion. The functions of the carotid artery, as it exists in most other mammalia, appear to be supplemented, or rather replaced, by those of the numerous vessels which accompany the nerves in their passage through the different foramina, and which in this part of the skull unite an external ‘“‘rete mirabile” to an internal, for the supply of blood tothe brain. We have observed in many skulls of Felis that the large foramina of the alisphenoid are accompanied by smaller, which appear to be appropriated to the transmission of vessels, though we have not ascertained this to be the case by the actual dissection of those spe- cimens. Muscles——The hamular processes of the pterygoidal portion of the alisphenoid being broken away, we can say nothing of the origin of the constrictor superior pharyngis, or of Folian muscle, in Felis syelea. ‘The origin of the circumflexus’ (Albinus) is under the foramen ovale (g), whence it passes round the hamular process to its insertion in the velum palati. § 6. Presphenoid and Orbito-sphenoid (Pl. VII, fig. 9).—All that hold the “ vertebral - theory” of the skull agree in assuming the homologies of the centrum of a vertebra for some part of the presphenoid, though there are differences of opinion as to the morphological value of the different parts. The bone is firmly anchylosed to the orbito-sphenoid while still foetal, and the sutures have all but disappeared at birth; for this reason we describe them as one bone. ' Straus-Durckheim, op. cit., vol. i, p. 295. Holden, ‘ Human Osteology,’ p. 395. 2 1% ‘ : a oh) 5 j ‘ p m arr a +5 ' 5 7 i ‘ . < ot \ . uy t > . ” dh —_ x : f ae 1 , , Hy j ‘- a . ' * Ap - » } ( aii Tie ' t Sih. j ; % * 7" - ~~ . - T ‘ ’ 7 f ag ’ ¢ + ; uy J & . j fh - " % t - O . ; an : ; £ 1 f 7 r hd e ¢ ¢ m i ‘ { - ‘ i ‘ h 4 ‘ oe ' - Al pe FIG. PLATE XIII. Felis spelea, Goldfuss. MILK DENTITION. (Natural size.) 1, 1’, 1. Outer, inner, and inferior aspects of left maxillary. The small premolar 2 inserted in fig. 1” is from that figured in fig. 2. Hutton Cave. Outer and inner aspects of anterior portion of left maxillary. Bleadon Cave. Outer and posterior aspects of right lower jaw, the crown of the canine and the incisor copied from other specimens, the lines of restoration being shown in the canine, and by the junction of the incisor with the jaw. Hutton Cave. Inner aspect of part of rmght ramus of young individual. From Sandford Hill Cave. Upper canine, inner aspect. Probably from Sandford Hill Cave. Upper milk molar 3, inner aspect. Sandford Hill Cave. Lower canine, inner aspect. Probably from Sandford Hill. This tooth furnished the restoration for that in fig. 3. Lower milk molar 3, inner aspect, of large size. Bleadon Cave. All the above are in the Taunton Museum, and were found by Messrs. Beard and Williams in the caverns of the Mendip Hills. PLATE XIll SPELAA FELIS m& W West time u W. Pid good 4s W.A Sanford del DENTITION MILK “sah 0, pone eae \ f ! te Mil , ay. Tete ys i aie Aet a h Aes } sth ih car tg et oi eae wey; : f ane ee se ; 2) F CRBNAEL -, ; tee | ASS OS ae . res tw Tether aria} ian ee hore suk bs’ ace ‘ait lites Swat Avaorin si a igile rat poe est eR Wh ‘ Jepeeny va ; Guinhs fies 2) era bike hae nal hile) Pe eateieowigee 2. 5 . ee pice pH) bitin Tae 4) oh ie mit 64/23 a: ag kitts ; M ere < nae 7 AN gE yay 0) oattgy stil, andi + ial oh eight Sails pitts. lainey getter tae yee Keg nf oe. es oe si nd bikart esol * an aatiatns. eat Wa vats Goan Mbinendh nihnns bain ie Fak x = ‘ 2 alt * Pa i ve hat) betchhayivy set Fey irk A ib: els 4th Bie we epital ett ais ae aeytynlabedey Bie Cera a prey ) . | rcceeimedas th itis heal ‘ pci Oey ; ae a ad Hage , ay stag igen: aU a tall eae ae some Dupree ‘ ea Che 7 lye Tdi 2s Ri AGS A ro Gat Puy Agcy wii Re 4 : . Pu ns leh SRY nh a ite haley oe 1 | ay ee | Par Haine teres 4 , ' eee ty ge Be oy ; Pe ahah wy AT 48: unary a ey Y Dae a tae er iene gsr Bore peer BEL aah Rt Nay 7 > f. ’ . PLATE XIV. Felis spelea, Goldfuss. VERTEBRA. (Natural size.) Fig. Brad es, Proximal, dorsal, and distal aspects of atlas. Sandford Hill Cave. The restoration of the transverse process is slightly enlarged from the cast of a specimen from Gailenreuth, in the possession of Sir Philip Egerton. It is to be seen in the British Museum and the Museum of the College of the Surgeons. 2, 2’, 2”, 2. Dorsal, lateral, proximal, and distal aspects of sixth cervical vertebra. Sandford Hill Cave. 3, 3, 3”, 3”. Proximal, dorsal, lateral, and distal aspects of seventh caudal vertebra. Bleadon Cave. All these specimens were found by Mr. Beard, and are now in the Taunton Museum. The following letters are used for the different parts of the vertebra in Pls. XIV, XV, XVI: ce, centrum. d, diapophysis. ae, anterior epiphysis. a, anapophysis. pe, posterior epiphysis. m, metapophysis. n, neurapophysis. az, pre-zygapophysis. ms, neural spine. p2, post-zygapophysis. pa, parapophysis. ac, neural canal. pl, pleurapophysis. v, canal for vertebral artery. Ay, hypapophysis. aa W A.oantord del W PLATE. XIV. W.West imy ae PLATE XV. Felis spelea, Goldfuss. VERTEBRA. (Natural size.) Fig. 1, 1’, 1”. Proximal, lateral, and distal aspects of a perfect second dorsal vertebra. Sand- ford Hill Cave. Mr. Beard’s Collection, Taunton Museum. ; W West ump SECOND DORSAL VERT Bir: W.A.Sanford del. J.Dinkel. lith Ae dep tlaney Lats 4 o if i i i ‘ 1 t ama | k er =: mee ay AN eae wee): CC a eee go Hi ade Pane a Cay ee CMe ued! my L: : . am - ‘ AY ny Ars rer ese Pasha nuke ait eit ae a erie abo beat se S80 ae tee = 5a ae bl twit - | ty “Rk any ee ye ec) ohsiery: : "3 me | oa) ege irethby: aed shoes inion) das io Se piel a SoBe aig tog Lilia Hints : ; y ant y ; , Agri igen deh: ie Seniesa “Son, abla tagy te hg Ae val na | Poa ee : 7 ‘ ee et Jt te nA } a) 4 ~ De, * aeghal liad atitihe'n tetany , pe te esis bie me sdk’ bapa Ani eter, ak davig> jean: seaddion 4 ee oe) f ie cia) v9 me pa * lenge nies gar cat ‘Ge Frat, ae 4 osha heh sm bi Page ’ sab Pee Te ih aw Sue iat sino coped aw FA By bt SPE ay ane HE ars kh c L i Pa ues Sri et caw | M, atte a he % HK, is mbar mn} heresy be Dh | 5 bah ay me Spas - Esa use nee tak Moyne ee piv | TOA ah nA cad i? hs re iy ie Bd pee bor coud RP ‘ nice Nema PLATE XVI. Felis spelea, Goldfuss. VERTEBRA ; STERNUM. (Natural size.) Fic. ae Seventh dorsal vertebra, distal and lateral aspects. 2, 2’, 2”, 2”. Eleventh dorsal ; lateral, distal, proximal, and dorsal aspects. 3, 3°, 3”, 3. Second lumbar vertebra, proximal, lateral, dorsal, and distal aspects. 4, A’. Fourth caudal vertebra, ventral and distal aspects. 5. Ninth caudal vertebra, dorsal aspect. 6. Tenth caudal vertebra, ventral aspect. Talat Eleventh caudal vertebra, dorsal, proximal, and distal aspects; very large specimen. 8. '‘l'welfth caudal vertebra, lateral aspect. On 9%, Fourteenth caudal vertebra, dorsal, proximal, and distal aspects. 10;2'07; Third sterneber, lateral and ventral or superior aspect. The above are in the Taunton Museum, and were in Mr. Beard’s Collection. All were derived from Bleadon Cave, except No. 3, which was from Sandford Hill. Nos. 5, 6, 8, 9, and perhaps 4, have the appearance of having belonged to one animal. \ FE LIS# i i W* Bidgood del et.lith May1867 V F, DQRTFroeP ae ae Os eee FE LALA | PUA OVE | , STE RNALS. PLATE XVII. Hels spelea, Goldfuss. SCAPULA. (Natural size.) Fie. 1. Glenoid cavity and distal surfaces of the right scapula. From Sandford Hill Cave. Found by Mr. Beard ; now in the Taunton Museum. 2. Outer or superior surface of the same bone. Rah ae is ie W” Bidg ood del.et lith PRAT Ex Var. LEA. W.Westimp hs A. . PLATE XVIII. Felis spelea, Goldfuss. HUMERUS. FEMUR. (Natural size.) FIG, 1. Composite figure of the posterior or palmar aspect of the left humerus. The ground- work in pale tint is taken from Dr. Schmerling’s great work on the caverns of Liége (‘Oss. Foss. de Liege,’ vol. ui, pl. xv, fig. 2). The distal end is from a humerus found in Bleadon. ‘The small compressed shaft was found in Oreston, and is now preserved in the Bristol Museum. ‘The imperfect proximal and distal articular portions are from Bleadon Cave. All the specimens from this latter cave are in the Taunton Museum. 2. Anterior aspect of distal end of left humerus, obtained by the Rev. H. H. Winwood, F.G.S., from the gravel of Larkhall, near Bath. 3. Distal articulation of very large humerus, left side. Sandford Hill Cave. ‘Taunton Museum. 4. Composite figure of left femur. A cast of a specimen obtained by Sir Philip Egerton, from Gailenreuth Cave, supplied the groundwork in light tint. The proximal and distal ends and the fragment of shaft were obtained from Bleadon Cave by Mr. Beard, and are now preserved in the Taunton Museum. . Distal articulation of femur. Bleadon Cave. ‘Taunton Museum. or 5 5 ae HUME ~ PLATE: XS Felis spelea, Goldfuss. TIBIA. FIBULA. PATELLA. METACARPAL. METATARSAL. (Half natural size.) Left tibia of young adult, anterior aspect ; perfect, with the exception of the proximal articulation. Sandford Hill Cave. ‘Taunton Museum. Proximal end of shaft. _ Distal articulation. Anterior aspect of proximal end of left tibia. Sandford Hill Cave. ‘l'aunton Museum. Proximal articulation of the same bone. Posterior aspect of right fibula, which probably belonged to the same animal as fig. 2. Sandford Hill Cave. Taunton Museum. External aspect of distal end of left fibula. Bleadon Cave. ‘Taunton Museum. Anterior aspect of patella. Sandford Hill Cave. Taunton Museum. Posterior aspect of ditto. (Natural size.) Second left metacarpal of gigantic size. Lower Brickearths, Crayford. Dr. Spurrell’s Collection. Second left metatarsal of gigantic size. Lower Brickearths, Crayford. Dr. Spurrell’s Collection. These are referred to in p. 22 as figured in Pl. VIII. W. West imp PLAT owe PEAR A ~ nw PE IS BONES OF EEVess J Dinkel del ethth, CAD OF SCIENCES CALIF A fey) ~ N N LIBRARY l 3 1853 10007 |