Py) GETS Phe N8 oe ' alee beet a): tnt A EE Be ae . ent ba Rene Rte Tyee teat { ia iteyaisanettle Wt) a) unites Nica ttiadbts perecinh th CHS Hinata isl tian fi hi i ee ay tite fund Site iyi en ike ee hip iJ re telege By hy atl i? 1) suit whyeme wal Lye eels i thi hates Tires bed) ah WATS ag a pik ae eae pa Se A et beal tt AL mitt Nutt ale nee ye Gab ate ey ty, Paha tan ¥ Mi ; castes NP ate oy v4 tet itt Porrrver PMTNCIDE Mt td Ln th i PE OM aaah gL } ore Wh Te a'ih Heaney ly co With r Henan ang ai! Wis Vaden tier bA Pectin Et | ‘f he yaad i ati a A te vy seit Huon My Ne Wait hee OPO: * ay ee Pht “ AMEY iit AH Eat pale een Wit Chan eh ‘iain ba} hie, Obit hay iA vif VOCE: et ts Heri, hy ila oA Pat § Abi iat Hh nih ent Re eb ee abhi arse SEED ANY Ati Bees de ts Be ius Nala Gh WP yee 4 1 MT i iit eed ane eee hed i ta tet aM ey han; ihe b bits cy, shih) 7 ne MA Hiptcene pate teen } Mem) rT Ay het ify hy sit Lam el ai gts Ye TV ERE hath Why urine {i ry’ “wl n) ran 40h He bbiaat tay vias it \ a 4 tptbie sts AF) ait? ae Wee yeh *} Sila | vy] RY Le ut ite yan co Ava sibs i Aig \ Ash at et SGN reg thay ay APA BL EN iy sah saa | wae mittee 4) ane + poetry tiaale 90 thy hi a4 Ny) vent a yen erry brn hit a i he a ivr sy yey i hby! watts wh eens als Hitt rads he “e Matusies ttt vib leh " aaee uM shore Hy aN fala " LA Yat 48 ein ite ns ay rylt y Ti “Abe it teat Bit Hai bel , sFEr Hy ANY wey 7. ee Wahl Leva \ \ Vanent Wye th bh rk ab as beh ah ralare ay ST) dyselta ey aa Hilut ttt He eH iat peste vy Wie a} sites “af Hi hau i BF hat ay yi if cotati! + pith f tf ead \ i) Peay feta coat i! bids * it baht | ay i Nay bat sui th det}: Hattie puts ' Ne le i ty AME HM rat lease vat Uae nan yy AA la i pe es VAAIAIA ARIA AAAR ' rv a WAAR A) AA A fa Y Wewey FN aA AAA ~\ VV AV Be YVAN AY \ RaaWAAAA AAAAAR RAAAAAAAAAL.. Val AAA AAAA AANA | | a/AN\ oN, \i/ SA AAA|E WN, \ \ \ i \ i } aA Eiaiaial \i \ } | - | \} } 7 if | r \ | AAP 1/ \ AY —\ y W/o Wi V/ \ AAAAAAAAAAA A AA\ AN |e) | y IAN | . Wa | \/ } \] \ | | | AA [A aN; | =. \ AA I \Von\ | AA | ae) i! \ \ ry | A AEN VE Cl (almlalale NAIA } ip Baaea.| || | a § LW 7. aanr IN ANY ig a An § ENA NT Vn YY AAAz Aa A a NM R a eC CCC Ca CCC Cu RC GaGa Gada ae iterceste soccer as ig sees estat: ass EO QO CO Gan € 2 KO Cm = ca Gr aa c_ <& C1. Le ag ac Gi @ ES eae ete ei tea Ce CLC C ga Ea Ca CG ( <" CCE HE ace (Gr @ac aq CE ECE aad a aa (i .¢ (Ce eq < ue Ca a ; = \( ( (KC CEE | re C C «a oc Ce @ ( Ce« CCC Ee Cee CC Ce QC @€ Qc C AG CEC (CE OE ‘. OC aK . Ce ec < G CC@ CEG Ce («| CE COEF cee (EG COC a KK ( G CC Ci Ce COE CE ay a COE CEC Ct CC CCE @ co CCC CC CGC CEE i . COE CCG COG Ce ee © QCOE€E COCR ce « (GG ( © ( a a C ¢« i C@ CC eseeattis sieactarsat: AC CCR Ce (a CCam Qn c Keres tC Ger wears << Can auves ee OE aC| Cccee_ © ( C|€ CGE CC Qe Cee €cKcC «KC CC CEC OC KC CCOE@ ¢ ( (COE CG KC LCE CECE & ee |q Cccce € & ( CG (COE « acc € ( Ke CCG fer cae Tat Es < ca CC ( C oe C CGE cS a Ta CCC £G ; ee, GE Cae € BGS Neg Ce ce OSE OLE eI as saw ysibug u: ajoo@ UU 1Iquia sy) ane S"THONVETL wia791A0p4 inoew GNAD AT NOIDHU © AVOTOAD I. INTRODUCTION AND ACKNOWLEDGMENTS The Paradoxides beds of the famous Cambrian section at Manuels, Newfoundland, whose stratigraphical and faunal succession are described in this paper, are of unusual interest because a large proportion of the many species of fossils which they contain are found also in the contempo- raneous beds of northwestern Europe. The description of the beds here presented is based on field work done at Manuels by members of four geological expeditions from Princeton University, and on laboratory studies pursued by the writer at that university and other institutions in eastern North America. The writer makes grateful acknowledgment of the en- couragement and financial support that have been extended to him in this research by the Department of Geology and the Graduate School of Princeton University, and tenders his sincere thanks to the many friends who have placed information or collections at his disposal or have assisted him in the field. He is indebted to Dr. A. O. Hayes, Profes- sor N. C. Dale, of Hamilton College, and Professor A. F. Buddington, of Princeton, for data collected by them while members of the Princeton field parties in Newfoundland, to Professor A. H. Phillips, of Princeton, for numerous chemi- cal analyses and mineralogical determinations, and to Dr. E. C. Cairnes, lately Fellow in Geology in the same insti- tution, for drawing and lettering his maps. Professor Gil- bert D. Harris, of Cornell University, loaned him the Hartt Collection from the Paradoxides beds of New Brunswick; Professor Percy E. Raymond, of Harvard University, and Professors Charles W. Brown and Richard M. Field, of Brown University, afforded facilities for the examination of IO BULLETIN 43 10 collections in their care and loaned fossils from Newfound- land and New Brunswick and Professor Raymond con- ducted him to the famous Paradoxides harlani quarry at Braintree, Massachusetts, and assisted in obtaining a col- lection from there. Mr. James P. Howley, the late director of the Geological Survey of Newfoundland, opened the New- foundland Survey’s Cambrian collections for his examina- tion; Professor W. A. Parks allowed him to study the Mat- thew Collection of New Brunswick and Newfoundland types in the Royal Ontario Museum of Paleontology; and Dr. Charles D. Walcott gave permission to examine the won- derful series of specimens from the Paradoxides beds of many countries that he has gathered together at Washing- ton. The late Dr. G. F. Matthew and Mr. William McIntosh, of St. John, guided him to a number of the Cambrian locali- ties in southern New Brunswick, and presented a valuable collection of fossils. Professor George H. Perkins supplied information about the Cambrian of Vermont; Mr. L. D. Burling furnished a list of the species from the Paradoxides beds of Newfoundland and eastern Canada that are repre- sented in the collections of the Canadian Geological Survey; and Dr. Charles E. Resser, of the United States National Museum, provided much valuable information, especially about Cambrian bibliography. Most of all, the writer desires to express his apprecia- tion to his wife, for her constant encouragement, helpful criticism, and efficient aid in the field and laboratory; and to Professor Gilbert vanIngen, of Princeton, for the use of his large library, assistance in the preparation of the photo- graphic illustrations, and a great deal if invaluable guidance and advice. II PARADOXIDES SECTION LI Hl. PREVIOUS WORK AND LITERATURE J. B. Jukes, who made the earliest official geological survey of Newfoundland and was probably the first man to study and classify the rocks of the Conception Bay region, noted the presence of shales at Manuels, and referred them to the ‘‘Belle Isle’ division of his ‘Upper Slate Formation.”’* (See column I of Table IJ, p. 12.) He found no fossils in these shales, and therefore could not determine their age. He judged from their appearance that they were very old, although he knew from their stratigraphic position that they were younger than most of the other rocks of the dis- Emcin) (dukes, 1342, vol. 11, pp. 245, 249, 275, 329° 1843, pple 55, oF, 135, and map.) Alexander Murray, who was the first director of the second official Newfoundland survey, appears to have been the next geologist to work at Manuels. The first published result of his investigations was probably incorporated in Logan’s “Geological Map of Canada,” which appeared in 1865, and in which the beds at Manuels were mapped as belonging in the “Calciferous” division of his “Lower Silurian” “Quebec group” (Logan, 1865, p. 15, pl. 1). The next reference to the beds at that locality was in Murray’s report of progress for 1866 (Murray and Howley, 1881la, p. 75) in which Murray stated that a ‘‘very good section of the more recent formation” (Jukes’ “Upper Slate Forma- tion’) was exposed “on Manuels’”+ Brook, at Topsail Head, and at Kelly’s Island.” He wrote that “the obscurity or absence of organic remains” rendered it “‘unadvisable to express too decided an opinion” as to the age of this “forma- * All terms and statements included in quotation marks in the present paper are quoted verbatim from other authors, and are used in the sense in which those authors used them. + Various spellings of Manuels occur in geological literature. The spelling employed in the present paper is the one that was used by Mr. Howley on the 1907 official geological map of Newfoundland. (Howley, 1907.) 12 BULLETIN 43 12 tion’; but added, in a foot-note, that a fossil, which appeared to be of earliest “Silurian” age, had been dis- covered in its upper beds. In his report for 1868 he pre- sented a hypothetical section of the ‘Lower Silurian (Pots- dam)” strata of the Conception Bay basin, showing the beds arranged in what he thought was probably nearly their true order (Murray and Howley, 1881la, pp. 156, 157, and plate facing p. 160) ; and in his report for 1870, he included -a generalized composite section of all of the ‘Primordial Silurian” strata of southeastern Newfoundland that he and his then assistant, Mr. J. P. Howley, had discovered up to that time (Murray and Howley, 1881la, pp. 237-239). Sum- maries of these sections are given in columns 2 and 3 of Table I of the present paper. Murray included the beds at Manuels in both his 1868 and his 1870 “Silurian” sections, but he had never succeeded in finding any fossils in them, and was therefore not certain that his estimate of their age was correct. In 1874 he arranged to have T. C. Weston, then collector for the Canadian Geological Survey, visit Manuels and make a thorough search for the needed paleontological evidence. Weston’s search was successful. He found fossils in the Paradoxides beds in the valley of Manuels Brook, and identi- fied the first one he discovered as a species of “Microdiscus”’ that was common in the Paradoxides beds of New Bruns- wick (Weston, 1896, p. 153). The fossils that he collected were apparently sent to Ottawa, for four years later J. F. Whiteaves, then paleontologist of the Canadian Survey, referred most of them to species that had previously been described from New Brunswick, and correlated the beds containing them with the “St. John’s Group” of that prov- ince (Whiteaves, 1878). In 1884 Dr. C. D. Walcott correlated the Paradoxides beds at Manuels with those at St. John, New Brunswick, and with “the lower part of the Menevian, or possibly with portions of the Harlech and Longmynd groups” of Great lassi fication Used in the Present Paper. \f most of the subdivisions, see Table III, Del) s Ay Paleozoic beds that are known to occur ott Cove beds about Conception and Trinity 'yograptus™ occurs in lower beds. rusia lenticularis and other fossils. sane bods :. :Agnostus pisiformis obesus, "Olenus™, : and other fossils. : : named beds ; Agnostus pisiformis. named beds :Small fauna of unidentified forms. :Possi bly a "Paradoxides forchhammeri :fauns", Tligrew 2 ook forma- : on. ds 93=125 of the present paper aradoxides ; See ppe )e vidis zone Pon fort tion. : Beds 36~92 of the present paper . aradoxides : (See pp. )e icksi zone }; ‘amber lin's: rook forma-: ions : aradoxides : Beds 1-35 of the present paper. onne tti : (See pp. 5, ne io named : mation . : Manpaniferous beds. named beds: atadoxides:Beds with Catadoxides magnificus zon eds ian other fossil otolenus eds ith Protolenus and other YDA : fossils. illavia :beds with Callavia brdgeeri and me, and :other Vepgeitse Some of the beds of ecball :this kp may be of pre-Callavian a avia:age; they Bee to contain no :trilobites, ut hold a "Coleoloides". Suara eas Cn eee eae eee Rd eR eRe te a eT M nd Howley Mat thow Walcott : Walcott =: LO > Marcou : Mateott Mat thew : Matthew : Walcott : Murray and How : 5 : Renenteiee 1868. : Report for 1870. cover, VanInegwn : 889 : 1690 : 189 1896 : 899 : : 1 and 1843 Sion eipote teeGa,. (Weresets : (Waleote, ii (ueeonaa 0, i ,,(faleott, , |(Matther,1996, : (Matthow,1899a) ; (Walcott,1900g,: (For details, see (Sikes 1b42; 1249) AL teers B er eeyeent paver: (Matthew, 1660) ME es fossb)” : 1889a,p.383) : p, 326) : 1891e, p.dds) | pp. 195-194) : : “pps 304) 31 Y ; : 188la,pp.156, oo: 80 3 : : Classification Used in tho Present Papor. A For furthor details of most of the subdivisions, soe Table Thee pp: SOL; 316)’: “Table Tit, s, Op Martine dee EN : Be) oD 5 ry : Rs ae: i a H : : ; : : Beds above :Lower Tower : All tho early Paloogoio bada that aro known to ocour : Bes ‘Gaactnes ah : : g i 0 A ° gq :0rdo- Paaeagten : above ee zone Cera beds plone voneap tion and Trinity : tones : 4 e 3 : g : ss s :vician ;}e! anadian) : Bays. "Bryograptua'' occurs in lower boda. : pandetone Great : ene Tee of Bell Serauna of the : Belle Isle,: Belle Isle, : : Belle Isle, ane Olenus sone: i ian) 3 Rey Jogray ( : Bell Isle. si island. 2 8 : °. : + On . wants t 3 : Upper : : Upper : a : 6 ! Little Bell Isle : : Sandstones and : Olenug Rags : PP : : ; : : Te : _ 8 : us t sand : °: ghales of Little ; ae : Cambrian, : Cambrian, : : Cambrian, > Olenian SS] : : Possibly : Beda with Orusia lontioularia and othor fossils. re) : sandstone : : Belle Island and ; division. pe: : : : :@2% Olenus : Upper: Blliott Osarkian : 2 Dice io: adjggont part of { (lems) : (lems) stone | to : eG ‘Upper :Uimamd bods igioa tie pint ToraTe SbSsuS-T OIE 8 *: Kelly's Island :.,: Conception Bay. : ; d : : tap zono Cam- ; ove == aa nee * Unnamid bods ue be pisif ea ’ » o . . 2 . . ’ ml Py Py :Cambrian : nad other fossils, i>: sandstones and ¢ ,,: 8 $ : : : : brian: Sori in a ro- ; AS ta: shales. Pars Rett yeeue rend : : : ; iS : brian ae oe : aentted Vonamod bods”; Agnoatus pisiformia. 7 evan pelt : : : : , Cambrian ; SS Nee ~ Ft rs] Fi Bo 8 : &: shales. : s ‘ : ks igenans :Uinnamod bods ;SialT Tauna of unidentitied tormes ° BS : Brown, black, cle : : ; 3 7 t : S :knowne :Possibly a "Paradoxides forchhammori 2: Dy jt eens Doe See art oe ee Si eS ae Ee oe fr ae Sub-zone o : eee ate : in tho ; a — faunal, is aipee: :™: Red, green, and :Horizon of : : : : Paradoxides —. 9 OCU Be ay LOE j Newfound: Kel ignew Stone ener) ’ : g : : : = 5 idis. 3 : : : ; : d"boda: Brook forna= : 2 a BOE red and green i ie ie no aie reredoaidee Newfoundland, : Avalon, ; Avalon, iBohomian, or : Avalon, Davidis : zone. :Cam- : Series seuerey hen ta Wau: fornia ps OSS398 oF the present Tante $A : +: black shales,and : 9 git ao 3 A 6 : : Middle Sub-zone of * Paradoxidian wanes iparts of :(Paradoxides; (Soo pp. : 4 @ :7: shaleg,con- : : eeliiavat onan: Horieonvos Lower Cambrian,: Middle : Middle —; Paradoxides : Paradoxides : (Avalon, —: brian ‘uth if Adie gopod; u B EI Dat : : 3 ‘ : : : : : 3 : ‘ slong Pon t ° Gg 4 ee heceinien with : Cambrian, : Cambrian, :zone, late ©; Cambrian, aeubttel). : Acadian, iat ~ipatton, : Pade 96-92 of the prosont paper as ini :: containing 3 S © : : P ae H des : : , : land. :(Paradoxidos ; (Seo pp. . 5 i 5 2 eed Bo 8 Paradoxides ‘Horizon of Paradoxides. —: (Parailox- :(Paradoxides) baile Taconic, (ere Sub-zone of : Middle hn which :Roprosonts Hokat gona): ae a Mens E Sete Brat :Paradoxides : ides) : : 2 HERO SIG: 5 : Cambrian). : Concopt ion:Brook forme: o :,: Paradoxides Tee ae nett tal ispinosus (?). : : ; : Rteminicus. : : Hate taymu BR CTnRatRe th eee : a Bop ieaeeee a : A f£ the : : ; : ; enue ls" Paradoxides ; Beds 1-35 of tho aie papor. Bol. . Sibel; :Horizon of. : : Q : : Paradoxi= :bonngtti : (Soo pp. . :H: bennetti Sete :Conocoryphinge. . : ‘ : : Aire AB Ch CDI eS 2 ae gta : : : : 5 is uso ‘dpa -bada HIGHT TA : oUF oe g o . 0 5 5 : : re ;: :_ known. formation . +: D boda ee s a :1 in upper Eee : ae ame ee a eee eae ees : 2 : :(Diaconformi ty) iv aia “anna etn ae = _ ann a : H eee. : : * P : : i aril . :(Catadoxidos:Bods with Catadox magnificus aes o : celta | SasSttithaa, | Potten ff ator Ae ranalalie i Ott (ote 123 ae : : : : i s ; > magnificus beds. ‘ : : rotolonus Bode with Protolonua and othor m8 part. Bit 3 ‘Terra Nova, : Terra Nova, poenveuiadls epee tecentia; : on ae : zone ? Lameneniamcine :TaROF ago iFrotolonus :Bode vi th Protolanua an : tot ‘ 3 B o g : r > Protolenian R : : ‘ees : 2 gp 0 : Lower : Lower : tome : x :Cam- + :Towr ‘avian ~:Bods with OalTavia brvpgort and 2 Bip P : : : : é 5 F or : Cambrian, : : : : ; ‘Cambrian ;s0n0, and other fossils. Some of the bods of ® tu ipopsail limestono.: o:Topsail limostone.: : Cambrian, : Cambrian ee atee zone : : : :brian :(Disconformity ) :inaro~ :posaibl See GTS ASTD Le Ta ras H Org te: G : : . *. | * A ‘ ‘ io ipre-Callavia:agm; thay appear to conta : Naeger Pe oe j(Otsaaliue (Clete dae cen f(a) ; Olonelius ——_:Btghominian fa0n80s ibedan ate tHobL Eon but hotdcaNestoctotdes", pire eae an 8 : g + Taco: ° : 9 inian D 5 : ries ie ae 7 ‘ A, s ie :Conglomorate and 3 ,, as room nal mo : : } : } Btchemini eoua ; : Serie ; : A, :, :black and brown :Conelomerate seen: : : : : : (Placentia : i { : 5 :shales (soon at : on anuels). : : g : : : : Georgian,” : : a :Manuel1s ). : Red and green : : B : 2 : Li b os *, © :shales and lime~ : : : : b igtones. : 1 hime a aS. ~~ ee ae oe ow fo) ave oclassifie : ar P lon Peninsula, southeastern Newfoundland, howd th ways in which a rent suthors have classi d the early Paleozoic rocks of|the Ava . Table I. Table 8 ng ) Affe pes ee BOAOK ae ae erm : ‘ eceaiaras meneame % | hae I ecsk ay 5 % ai yaa Wine tade :% ; soe , eel NR ee ies ay i pensar * a BG iad tee 2 i és a * j ‘ Oe Bt 4 4 . ae } eae ji ie tiNEE A ise: Of) + emeteecil pews hn aaa ia} aOR AN et i * ‘oa, i . * i ey : EASES elo eereerenr ido ng PARADOXIDES SECTION 13 Britain (Walcott, 1884, p. 18), and in the following year Dr. Matthew correlated them with the “Acadian” beds of St. John, New Brunswick and the Solva of Great Britain (Matthew, 1885, pp. 121 and footnote, and 122). In 1886 Dr. G. F. Matthew tentatively divided the “Paradoxides” beds of Newfoundland into five “horizons,” as follows: 5. “Horizon of Paradoxides Davidis.”’ 4. “Horizon of Paradoxides Tessini.”’ 3. “Horizon of Paradoxides spinosus (rg) si 2. “Horizon of the Conocoryphinz.” 1. “Horizon of Agraulos strenuus.”’ He stated that the “Horizon of the Conocoryphine” occurred at Manuels, and that the others were found at other locali- ties on the shores of Conception, Trinity, and St. Mary’s Bays (see map, Fig. I). He listed the species that had been recorded from Manuels by Whiteaves in 1878, and added a number of other forms which he had found in mate- rial from the same locality that had been sent to him by Mr. Howley (Matthew, 1886b, 1887). His “Horizon of Agraulos strenuus” has since proven to be of pre-Paradoxi- dian age. Two years later he correlated the “Shales of Manuel R.” with “Division” 1c of the “St. "John Group” of Canada, the upper part of the ‘Upper Sparagmite forma- tion—Etage 1b and c” of Norway, the upper part of the “Tower Paradoxides Beds” of Sweden, and doubtfully with the upper part of the “Solva group” of Great Britain (Mat- thew, 1888a, p. 25). At this time he considered that the “horizon of Conocoryphe at Manuel Brook” was older than the “limestone beds of Topsail and Brigus, in Conception Bay,” because Murray, unable to find any fossils at Manuels, had placed the beds of that locality below the limestone of Topsail and Brigus in his stratigraphic section (Matthew, 1888d, p. 74). In the May, 1888, issue of the American Journal of Science, Dr. Walcott applied the term “Newfoundland” to the Paradoxides beds ‘“‘of the St. John’s area of Newfound- 14 BULLETIN 43 14 land” (Walcott, 1888a, p. 399). He apparently intended his new term to apply to all the known Paradoxides beds of southeastern Newfoundland. This name appears to be the earliest one applied to these beds aione; and, if it does not prove to have been used previously in any other sense, should be used for these beds in the future. In the summer of 1888, Dr. Walcott found an “‘Olenellus fauna” in beds beneath the Paradoxides beds at Manuels. He announced his discovery a few weeks later in London, at the meeting of the fourth International Geological Con- gress (Walcott, 1888b, 1891c). As most American geolo- gists had believed, up until that time, that ‘“Olenellus” belonged stratigraphically above Paradoxides, this discovery aroused a great deal of interest, and Manuels quickly became one of the famous Cambrian localities of the world. Dr. Walcott published a brief description of the section at Manuels, and gave lists of the fossils that he had found there. From the Paradoxides beds, which he divided into three zones, he recorded thirty species, some of which he recognized as forms that were characteristic of the Paradoxides beds of New Brunswick or Wales (Wal- cott, 1889a, pp. 378-381). He referred the beds above the “Olenellus zone” in the Manuels Brook section to a new “terrane,” the “Avalon* terrane” (Walcott, 1888b; 1889a, p. 383; 189la, p. 548; 1891b, pp. 66, 306; 1891c) ; but he later (Walcott, 1899, p. 219) ceased using the term “Ava- lon” for these rocks, and applied it to the great group of pre-Cambrian sediments that underlies the Cambrian in southeastern Newfoundland, in which sense it is now gen- erally employed. (See Van Hise and Leith, 1909, pp. 43, 99-100, 518-529; Willis, 1912, pp. 14, 16; Buddington, 1919, 10s 41511.) In 1889 Dr. Walcott (1889b, p. 445) described a tiny trilobite, which he named “Karlia minor,” from the beds containing Paradoxides davidis at Manuels. He later * Spelled “Avalan” when first used (1888) ; but this was probably due to a typographical error, as the name was taken from the Avalon Peninsula and was spelled “Avalon” by Dr. Walcott in all his sub- sequent papers. PARADOXIDES SECTION 15 _ On referred this species to the genus Corynexochus (Walcott, 96a, p. 224; 1916b, p. 319). In 1890, Jules Marcou correlated the Paradoxides beds of Newfoundland with those of New Brunswick and with “the lower part only of the Paradoxides zone at Braintree,” Massachusetts (Marcou, 1890, p. 226). In the same year, Dr. Matthew (1890, p. 137) correlated the uppermost of Dr. Waicott’s three Paradoxides zones at Manuels with the Menevian of Wales, “Etage 1d” of Norway, and the ‘Upper Paradoxides Beds” of Sweden; and the two lower zones with the “Lower Paradoxides Beds” of Sweden, a part of the “Upper Sparagmite—Etage 1c” of Norway, and a part of the Solva of Great Britain. In 1891 he (Matthew, 1891) correlated the Newfoundland faunas with similar ones else- where, in the manner shown in Table II of the present paper. In the same year Dr. Walcott published the horizontal section of the beds at Manuels that is repro- duced in Figure 2a of the present paper (p. 16), and gave a resume of the work that had been done up to that time on the Paradoxides beds and faunas of Newfoundland (Wal- cott, 1891a, pp. 528,548, 554, 555, 565, 582, 583, and figs. 51 and 52; 1891b, pp. 50-55, 78-80, 118, 257-262, and 374; 1891d, pp. 533, 548, fig. 75, and pl. 42). TABLE II. Dr. G. F. Matthew’s 1891. correlation of the Paradoxides faunas of southeastern Newfoundland with those of other regions. Copied from Matthew (1891, p. 265) a b c d e it ey F nM ae: E 5 — 5 | 2 a ns feU SRS Miner) veces 95.50 s 5 ‘A B B 5 oO an o oS ° R © = = A Ey = p Ay ai AY A A Sardinia (italy); 222 === 7) ae Montagne Noire (France) ___ % BODEN ope te os Se zs a VIER aise se 2 A epee its 2 cs me Sweden and Norway______-- % * F - - Newtoundlan dyes oe cs * i * * Acadia (N. Brunswick) ___-- Massachusetts #222525 -— x“ *yoorg sjonuey jo Aopyea oy} ur doroyno yey} sso} are UMOYS Spoq oY], ‘salzred ploy UojeoulIg ey} JO yrOM oY} uodN peseq UOTJIEgG "py “SIT SSSSE [2Aa] POSY SS SOSA ap SS) SS vv 2404S RAD s1avjnzijuay BIS04O EAeee oleae oe sistof id oR SIpiaep DUDE 18YD14 eae apes pedi : " 0 Eg Hihs Sa} SoU GOPIKOPRARY GoOpinopLae, . x Ga Gla a Ges Sageoublyso ou SAY SOPMEPRARL SPMPYEE aa uUEG SAP|KOPTAL A Ree wiaaye “(9TE “4 “6 “SU ‘2006T) 3OOBM Worz potdoy ~“UOTZD9S ONGT 8.47092 M “AC OS “OIA 43 BULLETIN 5706 Aye vis Ne “(7aG “a “TG “SY ae ‘eT EST) WOM Worz pordog “UoryDeS TERT S4IO9TEM AC “BS “SAL ‘UDIYILY = : —— OSI ETI TOT = —SSSSSSSSS ———S== = SSS SSS ep OF Cg Ae TAL ws Pp SSS SSS re SY oe an > = == SSS <—s Ae hl “RA c —SS SSS SSS NA EI 3) Ys ——= d SSS ea ee SSE CONAN ~, = SS SS Wau ——_—— SS Ser te ‘aUu0Z sapIimop Used PUOZ SA2)VIUI [() je (0 eS wuld. = gpd? 16 Ly PARADOXIDES SECTION 17 Figs. 2a-d. Sections of the Cambrian beds at Manuels, by Dr. G. F. Matthew, Dr. C. D. Walcott, and the writer, to illustrate the increases in our knowledge of the limits of the “Olenellus,” ‘“‘Pro- tolenus,” ‘“‘Paradoxides,” and “Olenus” zones at that locality. Scale of all the sections, about 1,500 feet to 1 inch In 1896 Dr. Matthew subdivided the Paradoxides beds of Newfoundland as follows (Matthew, 1896) : 3. “Sub-zone of P. Davidis.” 2. “Sub-zone of P: Abenacus” (doubtfully identi- fied in Newfoundland). 1. “Sub-zone of P. Eteminicus.”’ He recorded from the “sub-zone of P. Davidis” at Manuels several species not previously known from that locality, one of which he described as a new form, naming it ““Plumulites manuelensis.” In 1898 Dr. Walcott described a new species of brachio- pod, “Obolus (Lingulella) fragilis,’ from the ‘‘Middle Cam- brian” shales “on Manuels Brook” (Walcott, 1898, p. 404). In the same year Dr. Matthew visited Manuels, and soon afterward published the horizontal section that is repro- duced in Figure 2b of the present paper. He stated that the dividing line between the ‘“Olenian” and “Para- doxidian” parts of his section was drawn arbitrarily (Mat- thew, 1899a, pp. 50-52, fig. 4). He recorded an Erinnys “from the Paradoxides Davidis sub-fauna at Manuels Brook,” and “Atops trilineatus” from an unknown horizon, which he thought might be the “‘Davidis zone,” at the same locality (Matthew, 1899c, pp. 89-95). In 1899 Dr. Walcott visited Manuels a second time. His account of this visit, published in 1900, included a description of the beds which seemed to him to mark the lower and upper limits of the ‘‘Paradoxides zone” in the brook valley, and contained, as one of its illustrations, the horizontal section that is reproduced here as Figure 2c (Walcott, 1900a, pp. 313-317, 329-331, and figs. 9 and 10.) In 1902 Dr. Walcott recorded a brachiopod, ‘“Acro- treta misera Billings” from the beds of the “Paradoxides 18 BULLETIN 43 18 zone” at this locality (Walcott, 1902, pp. 590, 591). In 1905 he described a new species of brachiopod, ‘“‘Plectorthis papias,” from the same beds (Walcott, 1905, p. 268). He afterward referred the latter species to the genus Eoorthis (Walcott, 1912, p. 785). In 1910 Professor Charles Schuchert wrote of the Para- doxides beds as follows: “In southern Newfoundland, on Manuels brook, the Acadic begins with a conglomerate havy- ing pebbles holding fossils of the Georgic strata below (Walcott, 1900, p. 315). Above this layer, which is 18 inches thick, follow argillaceous shales having a depth of 170 feet. These are succeeded by a thin limestone zone marked by the presence of Paradoxides, the latter extending 66 feet higher in a series of shales interbedded with lime- stone. Here occurs Paradoxides davidis and P. bennettti. Apparently a stratigraphic hiatus exists above these beds, followed by the Olenus fauna.” (Schuchert, 1911, p. 522.) In 1912 Dr. Walcott, in his monumental monograph on the Cambrian Brachiopoda, described and figured all the brachiopods known to occur in the Paradoxides beds at Manuels and listed the other fossils that he had recorded from those beds in 1889 (Walcott, 1912, pp. 140, 141, 161, 162, 170, 392, 393, 496-500, 647-649, 653, 654, 695, 785, and plates XXIII, XXIX, LXXII, XCI). In the same year Dr. B N. Peach, referring to Dr. Walcott’s description of the Manuels Brook section, wrote “In Southern Newfoundland Walcott showed that the base of the Middle Cambrian divi- sion is marked in Manuel’s Brook by a conglomerate con- taining fossils of the lower or Georgian terrane, thus indi- cating elevation and erosion of the Lower Cambrian rocks. Higher up the strata yielded Paradoxides davidis and P. bennetti” (Peach, 1912, p. 455). During the summers of 1912, 1913, and 1914, Professor Gilbert vanIngen, Dr. A. O. Hayes, Dr. N. C. Dale, Dr. A. F. Buddington, and the writer studied and mapped the rocks at Manuels and collected many fossils there, including more than 3,C00 from the Paradoxides beds. In 1914, Professor 19 PARADOXIDES SECTION 19 vanIngen (1914b) issued the “Table of the Geological Formations of the Cambrian and Ordovician Systems about Conception and Trinity Bays’ that is reproduced in Table III (p. 20) of the present paper, and Dr. Dale gave a brief description of the manganiferous beds which underlie the lowest known Paradoxides beds of that region (Dale, 1914). The complete results of Dr. Dale’s investigation of this “manganese zone,” which were published soon afterward, included a detailed description of the “zone” at Manuels (Dale, 1915, pp. 371-409, and figs. 1-29). In 1919 the writer examined foot by foot the Para- doxides beds in the valley of Manuels Brook, and, with Mrs. Howell’s assistance, collected some 7,000 fossils from them. Summaries of two preliminary papers based largely on the work of this and the previous Princeton expeditions were published in 1920 and 1922 (Howell, 1920a;-1920b, 1922). In 1920 Dr. Walcott recorded Protospongia fenestrata Salter from ‘‘the black shales of the Paradoxides hicksi zone” at Manuels (Walcott, 1920, pp. 305, 306). Ill. GENERAL GEOLOGY OF THE REGION ABOUT MANUELS The major features, and some of the minor details, of the geology of the Avalon Peninsula, on which Manuels is situated, have been described and mapped by Jukes (1839, pp. 1—4, 6-16, 27, 28; 1840a, pp. 104-108; 1840b, p. 1; 1842, pp. 219-226, 245, 249-254, 256-276, 321-334; 1843, pp. 25-32, 51, 55-60, 62-82, 127-140, map, and sections 1-8), Murray and Howley (1881la, pp. 188-179, 199-205, 232-249, 279-297, 478-483, 532-536; 1881b; 1918, pp. 4-32, and maps), Walcott (1889a, pp. 378-381, 388; 1891a; 1891b, pp. Hae S21 —20L, 21a, 000-568, 565, pl; Ti; 189ic; 13899, pp. 201, 218-221, 230-231; 1900a; 1900b), Chamberlin (1895), Matthew (1899a, pp. 45-52, figs. 3 and 4), van Hise and Leith (1909, pp. 438, 99, 100, 518-520), vanIngen (1914a, 1914b), Hayes (1914; 1915), Dale (1914; 1915), 20 BULLETIN 43 20 and Buddington (1914, 1916, 1919). Most of the informa- tion contained in this chapter has been taken from the works of these authors. The peninsula is composed of a complex cf Pre-Cam- brian igneous and sedimentary rocks, with scattered rem- nants of a once wide-spread blanket of sediments of Cam- brian, Lower Ordovican (and perhaps also Ozaikian) age clinging to it in the places where they have been protected from erosion. (See map, fig. 1, p. 8.) The Pre-Cambrian rocks include interbedded rhyolite and basalt flows, with corresponding breccias and tuffs, and volcanic dust beds, shales, sandstones, and conglomerates (the Harbour Main volcanics *; total thickness unknown) ; thin-bedded, green- ish-gray, dense slates, feldspathic sandstones, and con- glomerates (named by Dr. Walcott the Conception slates; estimated to be about 3,000 feet thick) ; green and reddish slates (Dr. Walcott’s “Torbay slates’; estimated thickness, 3,300 feet) ; dark brown and blackish slates (Dr. Walcott’s “Momable slates” ; estimated thickness, 2,000 feet) ; reddish- brown and green feldspathic sandstones and conglomerates, with intercalated shale beds (the “Signal Hill sandstones” of Jukes; estimated by Dr. Buddington to reach a thickness of 10,000 feet) ; reddish, greenish, and white sandstones and quartzites (Dr. Walcott’s “Random terrane” ; supposed to be 1,000 feet thick) ; and intrusives of gabbro, granodio- rite, cranite, granophyre, quartz syenite, aplite, rhyolite porphyry, and diabase. It is probable that the Harbour Main voleanics are the oldest members of this group; and that they are succeeded in age, from oldest to youngest, by the Conception, Torbay, and Momable slates, and the Signal * Dr. Buddington named these rocks the “Avondale volcanics” in 1916 (Buddington, 1916, map, p. 181). “Avondale” was, however, used by Frazier as a formation name in a different sense many years earlier (Frazer, 1883, p. 307). It becomes necessary, therefore, to assign a new name to the Newfoundland volcanics; and, at Dr. Bud- dington’s suggestion, “Harbour Main” is here proposed. The rocks are well developed near Harbour Main, Conception Bay. ‘(qbr6r) uesuy ueA mory pardoy = *pa}7}W10 Weed sey 9[q¥} OH} jo ped uelqme)-e1g oY, + sjuayeamby uvedomyy - Uio}saA\ pur UvoHeNIy - Utoysvoq VION Tey} PUP ‘pue[punoyMON ‘sAvg Ayur1y, pus uoydaosu0D ynoqe stajshS UvIOIAOpPIQ pues uURTIquIR) 9} JO SUOT EUIO [eorsoTo35) ay} fo a[qey,, S,Uasuy UeA IOssajfo1d Jo syied UeIoTAOpIO pue URIAquIeD 9} Jo uoljupoidsy “€ HIAV];, FES LE CED CR RT mn eh sajnpou auojsauty : TeyTOED 939 “SOPIOfOF19D | yy sayeys sets pur poy I @ BySIAPUOg aI BNUdIyS B[JINUsI}S au0js = urmMWayory 1195901 BIARL[eD | -dtuly sejnpou YIM sojeys poy 6 @ sn311g oF ees | Se aS Ap[uod JaMoT ‘O79 ‘sayqjoAPy ABYys pot YIA\ 9UOJsoLUT] PAY € @ quIog YuWwWS IE} = AjyIWIO;UOISIG 2 oO “a sasuods pue eliejoipey ayoydsoyg x a ALONG SdUOJsaLUI] SHOJOJIUeSUeLU Pesaaed 2 2 Aayuog sadd snjeydasosdiia : f S S me) mu snuajo}o1g GSA eae ei) a ueIplojuBpy ce 2 — = = Z Ajyrus10yuodsig sapixopeieg 1amo7 apsoydsoyg SNOSIPOAIIFY snjsousy uayyjeq sjanueyy ueIpeoy snaesqsory | -4Jey puke sauojspues ulyy dnory uyof 35 BAIOS aydXso20u0D |} ‘sayeys aaljo pue ‘uMosg ‘Lg pue uvrAouayy sopixopeied POUIsNI1aF Bpapnsury] sauojspues pappaq uIYI ssepq epndury SHE[Nd4Ua] BISNIGO) | pue suoa12U0d auod-UI-au0Dd 9A0D HON snuag[Q | yuM sayeys yoejq puke Adin : ueiuueyof snjydeisosig sajeys Aors) LACE Ae ease ayoonyseyos eS Se TIpseunys sajeys UMOIg es | asdy “AOU ‘RYAILIeY eUuljOgeie | s[eys “AOU ‘BOIAOURIIO} LIUOIWIULI ee i i ‘(qy161) uaZuy ues wio01y patdoy 9 ~payztulo useq sey 3[qr} aN} Jo qed ueIqMie)-s1g ayy *, Syuajeainby uvedomy - wis3saj\ PUL ULOLIaWIY - Wis}seeqIION 4133 pue ‘Pur[pUno}MAN ‘skeg Ayu1y, pus uoijdaou0D ynode siiszshG UPIDIAOpIQ pue ueiiqmED 933 Jo SUOTEMIOT [eIIZojoag ay} Jo aIGe],, S,ussuy ueA s0ssajoig jo syivd uvIDIAOpiO pu uetiquieD am jo “uononpoiday -€ atayy | = = Pat as | so[Npow sucysawity | Wye 19 *S9P19199199) | yim sayeys use2Z pue pry Ia EISIAEUOg } te enuad1s E{fNUsaS | auois ye: 1492804q PIARI[ED | -dusy sejNpou YIM sayeys pry 68 sngug = | UrrutMysIy Aapuor MOT “Dy ‘saytjoryy | A[eYs pot YA JUOIsoLUN, PI oo | qiog ys , = = Hl Ajltusoyuoasiq || = i] = SS ea saduods pur eiurjoipey | ausoydsoyg lk A) | a = |i A = | AIAN sauoisaiut, siosaj1ueSuew DPioyuery 5 Aauioy saddg Ue YUM ajeys pet pue uasaicy @.9 Cy I s 9 [010. | = || UEIpsojuBTT Aqyitumsoyuorsiq 5 | en eee | 4 sapIxopesed 19M07] aywioydsoyg Ia } = 1 | Ee | snosiposoryy = Ee | ?) s uayeq : = J ot SNDBLISOIT | -4[e4H pur sauoispues iy) Sionvelt ee tl] SEIDEN = | = PAlOS A4090U0D | ‘sajeys AAt[o pue ‘umosq *yDeIq za a i = pure urtaauayy sopixopeseg a‘ ———— ee ! POUIsNssay vL[a[NsUIT Suo|spues pappaq ulya ow a sdepy endury SHE[NI4Ua] BISN4IC) | Pue suoNd1I9U0d 9u09-UI-au0D one ueIuuRYO Suri | WIM sajeys ydeIq pue fo15 a nD HOM | J = as 3 os me aot ra _ Parr mm PIRES = 2 a sinjdessoxigy sales Sass) tla Pedy uMoig ayoonyseyss PIpPsePUINYS soyeys UuMOIG (A a Fanaa = = fasdy fa : 1S VIDLILY eULoquarg sepa ea = AOU “EIAOURIIA} PIUOWIUL 2 doprwary z *AOU ‘I[AMOY aqOLN seqeys ya WWaWupleyy = “AOU “IWOpuEs UoYdo.d||aq auojsauty pues £019) Goa 4Oo1g ssapry areys 94 | Aytwaoyuoosiq 5 _———— ] hi neyy ‘MopMoy vjnSurq | sa[eys pur’ sauoyspues 19 purysy Ayjaxy IsSurpiq eppaynSury saulospues 69 puelsy [eq eT deg sa[eys Puke sauo|spues @ 9 aAOD aouey] AWIBLUIAY 1TJOO YAY : © auoyspues snoulsn4s9.j = fh) pag Mergoyy | sajeys pur sauoyspues ¢ 9 yorag 0 Bs A é Pyaq eppaysury fo) ei Ispunyiq epansury | sayeys aos pue z D £ ssuypig ‘tein engurq |b aPWIWIY IWIWjOO YIM a S “Aou ‘snjetiquiy snjoqo1aeyds auoyspues snoursn4tef 9 9 prop wsayseq, 2 = Susy MOT “AOU ‘ISMOLING sHOgG = 2 | sa[eys Pue s}I5 Dh ) | auO}spues ay \\ 8 9 puowpay ] PMAOURIID) BOIIEZIT Pe Pais slulye shyoqoyngurq eee ee ero SRUES 6 9 PAMEY EMSUIT aneuay INI]0 uroowly avS]y Jayj0 pue Surog c ; nnd ee i glia ag | Ajyiwioyuoosig | ‘u ds ‘vurgea viuasseurayy | ansoydsoyg I H ayeys “ois { yeys Adis snjdeisowpig | € Fi R Sa]Npou aiid g peg ais, In] d @ oH | qd d | Z pag ity daaq sesa.0yuC, Isdaquizys snjyouojewoyu(s | ajeys S219 €@ Hu Suasy apprypy supytu snqideaSow j "4 ie Ai) ome ia i wy Pgh \ a lew UMN senile PilewNO 43: BULL. AMER. PAL., Wisi, Ib,’ iE. Fic. 1. View down the gorge of Manuels Brook, showing exposures of beds of the Paradoxides bennetti zone (beds 1-20) and the underlying manganiferous beds. Fic. 2. View of beds 1-15 and the underlying manganiferous beds, as they are exposed in the left wall of the gorge of Manuels Brook. The cliff here pictured is the one shown in the extreme lower left corner of figure 1. Pre ea eNO: oR. BuLL. AMER. PAL., Nor 438 Pre Fic. 1. View down the gorge of Manuels Brook, down the stream from the part of the gorge shown in plate I,, figure 1, showing exposures of beds of the Paradoxides bennetti zone, the P. hicksi zone, and the P. davidis zone, and the shales containing Agnostus pisiformis and A. pisiformis obesus and Olenus. Fic. 2. View of beds of the Paradoxides bennetti zone (beds 20-35), the P. hicksi zone (beds 36-92), the P. davidis zone (beds 93-125), and some of the overlying beds of undetermined age, as they are exposed in the left wall of the gorge of Manuels Brook. The exposures here pictured are those shown at the extreme left side of figure 1. 33 PARADOXIDES SECTION 33 time. The fossils that have been found in this basal 3 feet are so fragmentary that they can not be identified, but they appear to be most probably referable to a Paradoxides fauna, and a description of the beds is therefore included in the present chapter. The 10 feet of manganiferous shales beneath these beds, and the black shales overlying the highest known. Paradoxides horizon, may also be of Para- doxidian age; but they appear to show no evidence of being so, and therefore will not be considered in the present chap- ter, where only the 299 feet that were surely, and the under- lying 3 feet that were probably, formed during Paradox- idian time will be discussed. A section and table, showing the stratigraphical and faunal succession of the known Paradoxides beds that are exposed in the valley of the brook, are exhibited in figure 2d and Table IV (pp. 16 and 56), and the outcrops and exact positions of the various beds and faunal horizons are indicated in the figures on plates 1 and 2. No exposure showing a continuous section of these beds is to be found on either side of the brook valley; but a practically complete composite section can be obtained by combining the parts outcropping on the two sides. The section shown in figure 2d (p. 16) and described in the present chapter, was gotten in this way. Some mistakes may have been made in joining up the parts of this section that were exposed in unconnected outcrops, but it is believed that none of these errors can have been large or important. The lithological characters and the faunas of the individual] beds are described in detail in the succeeding pages of this chapter, and are discussed on pages 57 to 72. Beds 21 to 125 are described as they occur on the western side of the brook valley. They appear to be of essentially the same character in the exposures on the eastern side. The growth of our knowledge of the relationships existing between the Paradoxides faunas of Newfoundland and those of northwestern Europe and northeastern North America will probably necessitate the chaneine cf a num- ber of the specific and varietal names used in this paper, 34 BULLETIN 43 34 for the writer has attempted to identify his specimens from Manuels with previously described American and European species and varieties without entering into any discussion of the exact relationships or possible equivalences of those previously described forms. Detailed comparisons of the European and American forms will be included in a mono- graph on the Paradoxides faunas of southeastern New- foundland, which is now in preparation. The new species and varieties mentioned in the faunal lists of the present paper are described farther on. All the species of agnostid trilobites in these faunal lists are placed in the single “genus,” Agnostus. The writer believes, with Corda (Hall and Corda, 1847), Tullberg (1880, pp. 11-15), Jeekel (1909), and Raymond (1913, pp. 2 and 3), that these species should be referred to more than one genus; but, as he has been unable to assign some of them to the genera of any classification of the group that has yet been proposed, he has followed the old practice of srouping them all in “Agnostus.” This procedure has at least the advantage of making the names of many of the snecies more intelligible to the majority of his readers than the names would have been had he referred the species to what are really their proper genera; for to the minds of most paleontologists ‘““Agnostus” undoubtedly conveys more meaning than do “Condylopyge,” ‘‘Pleuroctenium,” ‘Lejo- pyge,” “Peronopsis,” or “Phalacroma.” The best generic classification of the agnostids that has yet been proposed is that advocated by Professor Raymond (Raymond, 1913). The writer believes that, in most respects, that classification is satisfactory as far as it goes, but that it is not complete; and Dr. C. E. Resser and he are now making a detailed study of all the agnostids, in the hope of obtaining a fuller knowledge of the true lines of descent and the proper classi- fication of that very interesting group of little trilobites. 35 PARADOXIDES SECTION iss) on - STRATIGRAPHICAL AND FAUNAL SUCCESSION OF THE PARA- DOXIDES BEDS EXPOSED IN THE VALLEY OF MANUELS BROOK The letters in parentheses indicate the relative abundance or rarity of the species, thus: r—rare, c—common, a—abundant, va—very abundant. The ° and / mark the earliest and latest known occurrences of the species in the section, thus: °“—earliest known occurrence; /— latest known occurrence. THICKNESS BED No. LITHOLOGICAL CHARACTER Fret. INCHES. 125. Dark gray slightly micaceous, phosphatic, carbonaceous shale, containing so many small lumps of black, car- bon2ceous, phosphatic material that it looks almost like a conglomerate. These lumps, which appear to be either pebbles or concretions, vary from a small fraction of an inch to 3 inches in diameter, and often zontain fossils. The top and bottom of the bed are uneven, but appear to show no clear evidence of the beds having been eroded after its consolidation______ Op 2 Paradoxides sp. undet. (probably P. davidis Salter) (r) Hyolithid gen. and sp. undet. (possibly Hyolithes tenuistriatus Linnarsson) (r). This is the highest known Paradoxides bed in the section, the top of the “Paradoxides davidis zone” of this paper (see p. 59). 124. Hard greenish gray shale. The top is uneven against the uneven bottom of bed 125 and the change in lithological character between the two beds is abrupt, but there does not appear to be any clear evidence of a sedimetnary break or of the erosion of bed 124 ichelus consolidations == ee ee @ oil Paradoxides sp. undet. (probably P. davidis Salter) (r). 123. Grayish white clay-like material, apparently resulting from the weathering of a gray shale, small pieces of Whichwoccun ine thernclay ==) = 2 8 ee ee Ol! No fossils found. 122. Hard, heavy-bedded dark greenish gray and blackish shales, containing numerous black, carbonaceous, phosphatic “pebbles,” 1 inch or less in diameter, Similarito, those vnebed) 1 2p. a ee OW) & /Paradoxides davidis Salter (c). ieweyriticerous. black (shales ie 2 fee ee ye Paradoxides davidis Salter (r). 36 120. 118. IL 116. 115. BULLETIN 43 Ryritiferous black shale eee Paradoxides davidis Salter (r). /Agnostus punctuosus Angelin (r). /Agnostus levigatus ciceroides Matthew (r). 9. Black shale containing a few limestone nodules, some of which are a foot or more in diameter, and many small nodules of pyrite and lumps or pebbles of phos- DMAtic) Matera, eS ae ee ee ee ere ewer Paradoxides davidis Salter (va). /Corynexochus cf. minor (Walcott) (1). Agnostus punctuosus Angelin (c). Agnostus levigatus ciceroides Matthew (r). Brachiopod gen. and sp. undet. (r). Soft pyritiferous bluish black shale, intermediate in character between 119 and 117____________________ Paradoxides davidis Salter (r). Agnostus punctuosus Angelin (r). Agnostus levigatus ciceroides Matthew (r). Soft, thin-bedded, dark gray shale containing many Sill ComencsiiGms @it jae Paradoxides davidis Salter (a). Centropleura sp. undet. (1). Agnostus punctuosus Angelin (r). Agnostus levigatus ciceroides Matthew (c). /°Agnostus cf. incertus Brégger (r). PmeemiCeons: Cad —amehy Gale Paradoxides davidis Salter (a). /Paradoxides rugulosus Corda (r). Corynexochus cf. minor (Walcott) (1). Conocoryphe sp. undet. (1). Agnostus punctuosus Angelin (r). Agnostus levigatus ciceroides Matthew (c). /Agnostus cf. fallax Linnarsson (r). /°Protospongia fenestrata Salter (1). Pyritiferous dark gray shales, slightly harder and heavier-bedded than 116, containing occasional flat limestone nodules, 1 inch to 6 inches thick, some of which hold fossils. A 2-inch bed of shale, full of small phosphatic “pebbles” like those in bed 125, Qeounas A 1A, Blows oie lWAge——_—_ Paradoxides davidis Salter (va). °Paradoxides rugulosus Corda (1). /°Centropleura henrici (Salter) (x). /°Solenopleura variolaris (Salter) (1). /°Solenopleura communis Billings (1). /°Holocephalina primordialis Salter (c). /Eodiscus punctatus Salter (1). Agnostus punctuosus Angelin (r). Agnostus levigatus ciceroides Matthew (c). /Agnostus levigatus terranovicus Matthew (c). 36 o/ 114. 113. 112. 111. 110. PARADOXIDES SECTION /° Agnostus levigatus mamilla Matthew (r). Agnostus cf. fallax Linnarsson (r). /Agnostus ef. acadicus declivis Matthew (r). /Obolus fragilis (Walcott) (x). /°Stenotheca cf. cornucopia Salter (r). /°Hyolithes cf. tenuistriatus Linnarsson (1). Hard, heavy-bedded, dark gray limy shale, some parts of which, more limy than the rest, are full of frag- mentsvoL Laradoxides! davidism === Paradoxides davidis Salter (va). Centropleura sp. undet. (1). Agnostus levigatus terranovicus Matthew (r). Agnostus cf. fallax Linnarsson (r). Hard, dark gray shale with flat nodules of limestone LNMIUSHUp Peep anew een = ee ore See Centropleura sp. undet. (1). /Bailiella venulosa (Salter) (1). /Hartshillia inflata (Hicks) (1). Agnostus levigatus ciceroides Matthew (c). Agnostus cf. acadicus declivis Matthew (c). Brachiopod gen. and sp. undet. (r). Hard black and dark gray shales, some layers of which are covered with fragments of trilobites___________ Centropleura sp. undet. (r). Hartshillia inflata (Hicks) (c). /Solenopleura cf. applanata (Salter) (c). Kodiscus punctatus (Salter) (r). Agnostus levigatus terranovicus Matthew (c). Agnostsus cf. fallax Linnarsson (r). /Agnostus rex (Barande) (r). Hyolithid gen. and sp. undet. (r). Pyritiferous black and gray shales, not quite so hard PENS). GALT eS ee Ry ea EE 0 Be Te On ee ey Paradoxides davidis Salter (r). Solenopleura cf. applanata (Salter) (r). °Hartshillia inflata (Hicks) (r). Agnostus punctuosus Angelin (va). ° Agnostus levigatus terranovicus Matthew (a). ~ Agnostus cf. fallax Linnarsson (c). Pyritiferous black shale with occasional thin gray loninels =” INGE CUMS Go Inehyolerspalilils Paradoxides sp. undet. (r). Solenopleura cf. applanata (Salter) (r). Agnostus punctuosus Angelin (va). Agnostus levigatus ciceroides Matthew (r). Agnostus levigatus terranovicus Matthew (a). Agnostus cf. fallax Linnarsson (c). /Agnostus cf. nudus (Beyrich) (r). /Agnostus cf. granulatus (Barrande) (r). /Agnostus sulcatus Illing (r). ie) “I 38 BULLETIN 43 /°Agnostus cf. pusillus Tullberg (r). Hyolithid gen. and sp. undet. (1). 109. Pyritiferous brown-weathering gray and black shales, in alternating thin beds. Slightly softer than 1106. Contains occasional small nodules of pyrite, and a Sreat “many fossils ses Se ae a ree aed ae fie Paradoxides sp. undet. (r). Bailiella venulosa (Salter) (1). Solenopleura cf. applanta (Salter) (a). °Corynexochus minor (Walcott) (r). Eodiscuc punctatus (Salter) (va). Agnostus granulatus (Barrande) (r). Agnostus cf. nudus (Beyrich) (r). Agnostus sulcatus Illing (r). /°Agnostus longifrons parvulus n. var. (r). /Agnostus cf. exaratus tenuis Illing (c). Agnostus punctuosus Angelin (r). Agnostus levigatus ciceroides Matthew (r). Agnostus ievigatus terranovicus Matthew (a). Agnostus cf. fallax Linnarsson (a). Agnostus cf. acadicus declivis Matthew (r). Agnostus rex (Barrande) (r). /Agnostus gracilis Illing (r). /Agnostus cf. kjerulfi Brogger (r). °Obolus fragilis (Walcott) (r). /Acrotreta misera (Billings) (1). Hyolithid gen. and sp. undet. (r). 108. Soft slightly pyritiferous black shale________________ Centropleura sp. undet. (r). Solenopleura cf. applanata (Salter) (1). Eodiscus punctatus (Salter) (a). Agnostus punctuosus Angelin (a). Agnostus levigatus terranovicus Matthew (a). Agnostus cf. fallax Linnarsson (a). Agnostus granulatus (Barrande) (c). Acrotreta misera (Billings) (r). 107. Soft pyritiferous black shale________-_______________ Solenopleura cf. applanata (Salter) (r). Kodiscus punctatus (Salter) (r). Agnostus punctuosus Angelin (a). Agnostus levigatus terranovicus Matthew (a). 5 Agnostus cf. acadicus declivis Matthew (r). Agnostus cf. nudus (Beyrich) (r). Agnostus granulatus (Barrande) (c). Agnostus cf. kjerulfi Brogger (r). /Agnostus cf. parvifrons mammillatus Brog- ger (r). /Agnostus bibullatus (Barrande) (r). Acrotreta misera (Billings) (r). 106: Soft, pyritiferous black sshalese me se Paradoxides sp. undet. (r). 38 PARADOXIDES SECTION Eodiscus punctatus (Salter) (c). Agnostus punctuosus Angelin (c). Agnostus levigatus terranovicus Matthew (va). Agnostus cf. fallax Linnarsson (c). Agnostus granulatus (Barrande) (c). °Agnostus bibullatus (Barrande) (r). Acrotreta misera. (Billings) (c). Hyolithid gen. and sp. undet. (r). 1065. Soft black shale with some very thin brownish layers__ Solenopleura cf. applanata (Salter) (r). Eodiscus punctatus (Salter) (c). Agnostus punctuosus Angelin (r). Agnostus levigatus terranovicus Matthew (c). Agnostus cf. fallax Linnarsson (c). Agnostus granulatus (Barrande) (r). Agnostus cf. kjerulfi Brogger (r). °Agnostus cf. parvifrons mammillatus Brog- ger (r). Acrotreta misera (Billings) (c). 104. Thin-bedded soft gray and black shales______________ Agnostus levigatus terranovicus Matthew (r). Agnostus cf. fallax Linnarsson (r). Acrotreta misera (Billings) (1). 103. Thin-bedded pyritiferous soft black and gray shales___ Eodiscus punctatus (Salter) (c). Agnostus levigatus terranovicus Matthew (c). Agnostus sulcatus Illing (r). Acrotreta misera (Billings) (c). 102. Thin-bedded pyritiferous soft black shale with a few thin gray and brown layers. In its upper 5 inches it is much like bed 103. Contains occasional small MOOBICS AOL. IOV EIte ss. essed Deed ESE eh SU ee Paradoxides sp. undet. (r). Solenopleura cf. applanata (Salter) (r). Eodiscus punctatus (Salter) (va). Agnostus levigatus ciceroides Matthew (c). Agnostus levigatus terranovicus Matthew (va). Agnostus cf. fallax Linnarsson (a). Agnostus granulatus (Barrande) (r). Agnostus sulcatus LIlling (r). °Agnostus cf. gracilis Illing (r). Agnostus cf. kjerulfi Brogger (r). /Agnostus vaningeni n. sp. (r). Acrotreta misera (Billings) (r). 101. Thin-bedded pyritiferous soft black shale with a few TH auie Fee WES ee Aa aa eS Se Se es eee Peet Paradoxides sp. undet. (possibly P. davidis Salter) (r). /Paradoxides cf. hicksi Salter (r). Eodiscus punctatus (Salter) (a). 40 BULLETIN 43 Agnostus cf. fallax Linnarsson (a). Agnostus ef. nudus (Beyrich) (c). Agnostus granulatus (Barrande) (r). Agnostus sulcatus Illing (r). Agnostus cf. kjerulfi Brogger (r). /°Agnostus parvifrons punctifer n. var. (r). Acrotreta misera (Billings) (r). /Lingulella ferruginea Salter (r). 100. Thin-bedded pyritiferous soft black shale____________ Paradoxides sp. undet. (possibly P. davidis Salter) (7). Agnostus levigatus terranovicus Matthew (r). Agnostus cf. fallax Linnarsson (c). Agnostus cf. nudus (Beyrich) (r). Agnostus granulatus (Barrande) (r). Agnostus sulcatus Illing (r). Agnostus cf. kjerulfi Brégger (1). Agnostus vaningeni n. sp. (r). Acrotreta misera (Billings) (r). Stenotheca sp. undet. (r). 99. Thin-bedded pyritiferous soft black shale____________ /°Centropleura pugnax Illing (r). Solenopleura cf. applanata (Salter) (r). Eodiscus punctatus (Salter) (1). Agnostus punctuosus Angelin (c). Agnostus levigatus ciceroides Matthew (r). Agnostus levigatus terranovicus Matthew (r). Agnostus cf. fallax Linnarsson (va). Agnostus cf. acadicus declivis Matthew (r). Agnostus cf. nudus (Beyrich) (r). Agnostus granulatus (Barrande) (r). °Agnostus cf. kjerulfi Broégger (1). Agnostus sulcatus Illing (c). /Agnostus cf. parvifrons Linnarsson (r). Agnostus cf. exaratus tenuis Illing (r). /°Agnostus cf. fissus perrugatus Gronwall (va). °Agnostus vaningeni n. sp. (r). Acrotreta misera (Billings) (c). Hyolithid gen. and sp. undet. (r). Stenotheca sp. undet. (r). 98. Soft. but compact, black shale, in layers 1/4 to 1/3 of an “rel: thick. 22322 2 a ee ele eres No fossils found. The absence of fossils from this black shale is remarkable, as the very similar black shales above and below it are richly fossiliferous. 97. Soft pyritiferous black shale with occasional flat lenses of limestone. Contains several beds of white clay- like material, 1/16 of an inch thick. and has a similar bed, 42 of an inch thick at the topmee 40 41 PARADOXIDES SECTION Agnostus punctuosus Angelin (r). Agnostus cf. fallax Linnarsson (r). Agnostus sulcatus Illing (1). 96. Soft pyritiferous black shale, similar to bed 95, but with Agnostus punctuosus much the most abundant mMemMperROhseNe et aUM ae ee St ee Paradoxides davidis Salter (c). Agnostus punctuosus Angelin (va). Agnostus cf. fallax Linnarsson (r). Agnostus sulcatus Illing (r). Sie Sort pyeltinerous black shales === a6 ee Paradoxides davidis Salter (c). °Agnostus punctuosus Angelin (c). Agnostus cf. fallax Linnarsson (a). Agnostus cf. nudus (Beyrich) (r). Agnostus granulatus (Barrande) (c). Agnostus sulcatus Illing (a). Hyolithid gen. and sp. undet. (r). 94. Very soft pyritiferous black shale, similar to 95, but softer and less fossiliferous. Breaks into little hexa- gonal pieces, about 4 of an inch in diameter, and almost as thin as paper. Pyrite nodules, % of an inch or less in diameter, are common_______________ °Paradoxides davidis Salter (c). Agnostus levigatus terranovicus Matthew (c). Agnostus cf. fallax Linnarsson (r). Agnostus granulatus (Barrande) (r). Agnostus sulcatus Illing (a). Agnostus cf. acadicus declivis Matthew (r). 93. Thin-bedded soft pyritiferous black shale_____:_-____ Paradoxides sp. undet. (possibly P. davidis Salter) (r). Agnostus evigatus ciceroides Matthew (c). °Agnostus levigatus terranovicus Matthew (a). Agnostus cf. fallax Linnarsson (c). Agnostus cf. granulatus (Barrande) (1). Agnostus cf. sulcatus Illing (a). /Agnostus cf. umbo Matthew (r). Stenotheca sp. undet. (c). This is the lowest bed of the “Paradoxides davidis zone” of this paper. ear orsy JiMestone==—— = io. es Tk 8 oe Agnostus cf. acadicus declivis Matthew (r). Acrotreta misera (Billings) (r). This is the top bed of the “Paradoxides hicksi zone” of this paper. 91. Soft pyritiferous black shale with limy lenses and NOGULESSIny ALS upper parus= ss ee ee a Solenopleura cf. applanata (Salter) (r). () 41 42 BULLETIN 43 ) /Liostracus globiceps jaculator n. var. (1). Agnostus levigatus ciceroides Matthew (c). Agnostus cf. fallax Linnarsson (c). Agnostus cf. sulcatus Illing (c). Agnostus granulatus (Barrande) (r). Agnostus cf. umbo Matthew (r). /Agnostus cf. parvifrons Linnarsson (r). /Agnostus fissus Lundgren MS (r). Acrotreta misera (Billings) (1). Stenotheca sp. undet. (r). v9, Soft pyritiferous black shale, slightly greenish in spots. Contains nodules of pyrite, some of which are as aol A IE mavelay ihn Chiginneiere Paradoxides hicksi Salter (1). Agnostus levigatus ciceroides Matthew (r). Agnostus cf. fallax Linnarsson (c). Agnostus cf. nudus (Beyrich) (1). Agnostus granulatus (Barrande) (r). °Agnostus cf. sulcatus Illing (a). °Agnostus cf. umbo Matthew (r). °Agnostus cf. parvifrons Linnarsson (r). Agnostus fissus Lundgren MS (r). /Agnostus cf. gibbus Linnarsson (r). Acrotreta misera (Billings) (r). 89. Soft pyritiferous black shale with layers of brownish Slee tin Tes; llonyeste tient Paradoxides hicksi Salter (r). Agnostus levigatus ciceroides Matthew (r). Agnostus cf. fallax Linnarsson (va). Agnostus fissus Lundgren MS (r). 88. Pyritiferous black shale, harder than bed 89__________ Paradoxzides hicksi Salter (c). Solenopleura cf. applanata (Salter) (1). °Bailiella venulosa (Salter) (c). Agnostus levigatus ciceroides Matthew (r). Agnostus ef. fallax Linnarsson( va). Agnostus cf. acadicus declivis Matthew (r). Agnostus rex (Barrande) (r). Agnostus granulatus (Barrande) (r). Agnostus fissus Lundgren MS (r). °Agnostus cf. gibbus Linnarsson (r). Acrotreta misera (Billings) (1). Me Soule | Onseenyennone lille sla. 2 Paradoxides hicksi Salter (c). Centropleura sp. undet. (r). /Conocorpyhe zequalis Linnarsson (1). /Agyraulos socialis Billings (1). Agnostus levigatus ciceroides Matthew (c). Agnostus cf. fallax Linnaysson (r). Agnostus fissus Lundgren MS (r). We) 42 43 PARADOXIDES SECTION 86. Pyritiferous black shale, slightly brownish in some layers and greenish blue in the bottom inch__ Paradoxides hicksi Salter (1). /°Centropleura venusta (Billings) (1). Solenopleura cf. applanata (Salter) (a). *Liostracus globiceps jaculator n. var. (1). Conocoryphe ezqualis Linnarsson (r). Agraulos socialis (Billings) (1). Eodiscus punctatus (Salter) (c). Agnostus cf. fallax Linnarsson (r). Hyolithid gen. and sp. undet. (1). 85. Pyritiferous black shale, harder than bed 86. TwitheacOncho1lal srachUres= = — = ee Paradoxides hicksi Salter (1). Solenopleura cf. applanata (Salter) (r). Agnostus fissus Lundgren MS (r). Breaks 84. Alternating beds of dark gray limestone and thin- bedded, soft, pyritiferous gray and bluish gray shale Paradoxides hieksi Salter (c). Centropleura sy. undet: (r). Solenopleura cf. applanata (Salter) (a). Conocoryphe zequalis Linnarsson (a). Agraulos socialis Billings (c). Eodiscus punctatus (Salter) (a). Agnostus levigatus ciceroides Matthew (c). Agnostus cf. fallax Linnarsson (c). Agnostus cf. acadicus declivis Matthew (c). Agnostus fissus Lundgren MS (r). Hyolithid gen. and sp. undet. (r). Somes Olbmolive, Shalete.. 2.2 Sl Res Sa Paadoxides hicksi Salter (c). Solenopleura cf. applanata (Salter) (1). Conocoryphe zqualis Linnarsson (c). Agraulos socialis Billings (c). Eodiscus punctatus (Salter) (c). 82. Soft pyritiferous bluish black shale. Some of the pyrite occurs as small nodules. The fossils are BoorlyasnreSerVvied se 5 a ee Paradoxides hicksi Salter (c). — Agraulos socialis Billings (c). Hyolithid gen. and sp. undet. (c). 81. Thin-bedded dark gray shale, containing a few small flat phosphatic “pebbles’”__________ aa Paradoxides hicksi Salter (c). Agyaulos socialis Billings (a). EKodiseus punctatus (Salter) (c). °Agnostus levigatus ciceroides Matthew (r). Agnostus cf. fallax Linnarsson (r). Agnostus cf. acadicus declivis Matthew (r). /Agnostus barrandei Salter (r). 1 43 4 44 BULLETIN 43 Agnostus fissus Lundgren MS (c). /Acrothele cf. matthewi (Hartt) (1). 80. Gray shale, greenish in spots. Occasional thin lenses of limestone or limy shale occur in its upper part, and 4 inches above the base there is a %-inch bed of hard gray limestone or limy shale with uneven Upper) And lowers JSUT el Ce ee eee ee ee Paradoxides hicksi Salter (c). Solenopleura cf. applanata (Salter) (r). Agraulos socialis Billings (c). Agnostus cf. acadicus declivis Matthew (r). Agnostus granulatus (Barrande) (r). Agnostus barrandei Salter (r). /Agnostus cf. parvifrons tessella Matthew (r). Hyolithid gen. and sp. undet. (r). 79. Hard dark gray shale, with a 1-inch bed of dark gray limestone at the top. The bedding surfaces of the shale’ are tough 22) eee Ow : : : 5 se ce Tee oe oe oe celee ae oe oe ee lee oe oe soles ce ee ce an lee ae ne om ae lee ae a oe calen oe ae ce aelee eee oe eles te oe oe celee 8 ae ot cel ee oe ce oe aelee oe ae 8 celee oe 5 : 5 ut 5 : : i) : os : : - se slew ae ae se tetee ce ae oe ce lee oe oe salen ee oe ne cles oe on se ealee an oe oe celee oe oe ae cores oe ue Oe sults oe oe oe eeget Oe ae oe etree on a8 we aelee oe ae ae welee ae i iS) . is o > Wh : His : 3 ° ¢ 5 5 soles ae te oe on tee oe oe sores ce ae ce eee oe on a0 cares oe 0 on caret on 08 oe aeree on oe on cocee oe 08 oe oanee 00 08 os sores 48 ae a oe ee oe ce oe ae,ee oe : 5 : 5 Z i ‘ 5 5 a: , é & 5 : hot 5 Dike some : b : 5 5 u soles en ee ae selon oe oe soles ae ce ae eelee oe ce ae celeb ae ae ae seven oe oe oe cares a oe ae ee ee ae ne oe cles ae Oe 4s cel ee oe ae 48 tees oe ae oe orien oe ; ea, eae : : u: ee a , os. ee MEE Pie ee : se setae ae ae ae ceTee ae oe ne setae ae oe soles oe oe oe celee gisthsse doles oe on ae coven on oe 48 serene ae 00 00 ce lee ce oe oe ceren oe oe oe ounce on 08 08 wales 08 a0 oe wer ee oe : Q a : , co ra , a= . a BS Fae ~~ 3 ie e eh ee 9) Sie * . ‘ Ms a eed ater e a = x cathe. ec se cetedeb ee Shae eee Pes Ap ae ae. “le: 12m pee io Recs a ~ . he ° — > sh ni ht- 92 . ey tatk ¥ 5 a r < a ? + —- ‘ ply* Ae otis - , Nit Ss £ (" ae id “- ft + Ee eS * | Bois § 4 fo" Bee 4 : & if Few co Ne 5 $ = ese : : ; ¢ bo @ es a . | apate é " . as * Ns dase “Ty ¥ aati aa Dus * ¢ - ae > i * . ie. 4 4 ft theres x ear: ‘ ¥ 4 ; ie | : | : 2 " ; aitgues jee a, « * * . Sip a Pe ee ee te ae abate : : r : . bs eo. ; é ;, : ekg rapier b ‘ : : ee . : 7 : (Male 7,23 5 oe PSOE LIC CL ERE LT Whee i" ; . eck = ek ; ; r! 2 us Fi . ; - ; 2 abs ; Te - Ss tes : , eee rere eels ok tre Fs ae ae " , H : ; ‘ ; ae ie tee ‘ s $ Be ce te Saat t : Sed BM : Bg s L 3 5 PaO PR Vie tere a rye fe Wire rete Fy oe a 7 4 i : ae Be ? 2 rt 3 ae > r t 3 ; Ee ee eee Se eh oe pes * * ’ e * a) as re ve ed - in vs xh > La 5 : + 4 a ’ Y 7 oR ee i : ’ ee + ‘ } C4 a NY Ge Pareeee res eee ra | te ; f ‘ies wer ‘Se @ ‘ ahs igen : ie t eer + Fav ot Yea dey ae : : ops wads oak rele y | n ° é s* > ‘ : sear Pe ay ‘em ; piled padre pe Patt ig + ee BOG for 'y epee 7 ¥ sage : tae : ee Se . t ; Pre ; £° Bia ta ; : f ee Ae ; ; ; $ 2 ek a he eae Te ‘ (Se Pee Pe ee Cee YS Ai ; + ee a ae es ; + tne te oer ta es x ep ie ee ¥ ; ; i . 4 ine eae } ~ ‘ — psp tee ee ee ee a “se : - Gg - a ed Raa ange tineien Lx ihe ve . + 2 é ete ts Pi Ee! T+ be eee @ sabre miei ene > => we ce e+ a3 oe « / > , ‘6 ta ee te p= She = = aes < =e win +e © Pl eee <> Se -& ae a eS es ee ~~ we ee -isinesenaad) ak de atpiiis +e (atest i Se Lee wiges DORE TONAL Bite tire alla cite eke ee | - he ite. dep inieopseed Sn Siig Rican Bors | crepe OME NE Salen nay mc RIS ar a bital, 3 ex 57 PARADOXIDES SECTION 57 ‘Lingulella, Micromitra (Iphidella), and Obolus) are brachi- pods, and the remaining 16 (Agnostus, Agraulos, Bailiella, Centropleura, Conocoryphe, Corynexochus, Ctenocephalus, Eodiscus, Goniodiscus, Hartshillia, Harttella, Holocephalina, Liostracus, Paradoxides, Ptychoparia, and Solenopleura) are trilobites. Algz are also probably represented. DISCUSSION OF THE STRATIGRAPHY The classification of the subdivisions of the Paradox- ides beds at Manuels that is used in this paper is shown in Table I, p. 12. The beds are there referred to the Acadian epoch of the Cambrian period. In 1874 Murray wrote of the shales exposed in the gorge of Manuels Brook as “my Manuel’s River rocks” (Weston, 1898, p. 153), and in 1889 Mr. Howley mentioned ‘‘the Manual Creek shales just above the Topsail Head limestone” (Howley, 1889, p. 123) ; but it is doubtful whether either of these authors intended to use their terms strictly in the sense of formation names; at any rate, they did not define them as such. In 1888 Dr. Walcott spoke of the Paradoxides beds of southeastern Newfound- land as the “Newfoundland” beds (Walcott, 1888a, p. 399). In 1914 Professor vanIngen (1914b) described the Para- doxides beds about Conception and Trinity Bays as the “Manuels series,” composed of a single formation, which he called the “Manuels formation” (see Table III, p. 20). The known Paradoxides beds at Manuels and elsewhere in southeastern Newfoundland appear, however, to be divisible into at least three distinct formational units, and the writer therefore suggests, with Professor vanIngen’s approval, that the term ‘““Manuels” be abandoned as a formation name and retained in its larger sense alone. In the present paper Dr. Walcott’s term ‘‘Newfoundland” is used as a general series name for all the Paradoxides beds of southeastern Newfoundland (some of which differ lithologically from the Paradoxides beds at Manuels), and Professor vanIngen’s series name “Manuels” is restricted to the Paradoxides beds 58 BULLETIN 43 | 58 of the Conception Bay region, where the lithological char- acter of the beds appears to be everyhere approximately the © same as it is at the type locality, Manuels. It is possible that the Paradoxides beds of Newfoundland do not repre- sent a whole “series,” in the sense in which that term was defined by the International Geological Congress of 1900; but the course suggested above would seem to be the best one to pursue at the present time. (This question is dis- cussed further in Chapter VII. The three major units into which the beds known to contain Paradoxides at Manuels seem to be most naturally divisible, both lithologically and faunally, are here named the Chamberlin’s Brook formation, the Long Pond forma- tion, and the Kelligrew Brook formation. The Chamber- lin’s Brook formation, which is the oldest of the three, out- crops in the banks of Chamberlin’s Brook, a mile or two northeast of Manuels. It contains a fauna whose most char- acteristic species is the big trilobite, Paradoxides bennetti. (This species is perhaps identical with the earlier described one, Paradoxides harlani Green, of Massachusetts, and it is possible that the zone ought to be called the ‘harlani zone’; but until the Newfoundland and Massachusetts fossils can be proved to be referable to a single species, it seems best to continue to use the name bennetti. See pages 60, 61). The Long Pond formation, whose type locality is on the shores of Long Pond, a couple of miles southwest of Man- uels, is characterized by a smaller Paradoxides, P. hickst. The Kellizrew Brook formation is exposed on the northeast side of the valley of Kelligrew Brook, about five miles south- west of Manuels. It is distinguished by the presence of another large trilobite, Paradoxides davidis. It is not known whether these divisions should be con- sidered as “stages” or as “zones” when measured by the stratigraphic units of the International Geological Con- gress. In the present paper they will be spoken of as “formations” or “zones.” They are probably recognizable as lithologic units (formations) throughout the Conception 59 PARADOXIDES SECTION on No) Bay region, and as faunal units (zones) throughout south- eastern Newfoundland. As no subdivisions of Acadian time that are applicable to the whole world have yet been defined by students of Cambrian history, it is not possible to refer these zones to any smaller universal time units. The Chamberlin’s Brook, Long Pond, and Kelligrew Brook formations grade gradually into each other. The same is true of the bennetti, hicksi, and davidis faunas. It has therefore been found necessary to draw the exact verti- cal boundaries between the zones more or less arbitrarily. The zones may be characterized briefly as follows, begin- ning with the highest: III. Paradoxides davidis zone (Kelligrew Brook formation). Beds 93 to 125; 31 feet thick. Soft black and gray shales, the latter often containing flat nodules and lenses of dark gray limestone, which are sometimes as much as 6 feet in diameter and 1 foot in thick- ness. Small nodules of pyrite are common in the upper 414 feet. A bed, 1 to 2 inches thick and full of small black phosphatic nodules, occurs 24 feet above the hase, and there is a similar, but slightly thicker, bed at the summit. Fossils are common in almost all of the beds and are very abundant in many of them, especially in the softer shales, which are often full of agnostids and other trilobites. II. Paradoxides hicksi zone (Long Pond formation). Beds 36 to 92; 37 feet thick. The lower 25 feet are composed of heavy-bedded, brown-weathering and black-weathering, dark gray shales (with a few thin beds of soft black shale and some thin limy layers and lenses in their lower part) which grade upward into 10 feet of softer gray and black shales containing small lenses and nodules of limestone. Small concretions of pyrite occur at various horizons. The top is a 4-inch bed of dark 60 BULLETIN 43 60 gray limestone; the bottom is a 114-inch layer of unctuous white clay. Fossils are common in most of the beds, and agnostids are very abundant in many of the soft black shales. J. Paradoxides bennetti zone (Chamberlin’s Brook forma- tion). Beds 1 to 35; 234 feet thick. Hard olive gray, and somewhat softer bluish gray, shales, with several beds of impure nodular limestone. Two of the limestone beds are at the base, about two feet apart, a third is 134 feet above the base, and the others are scat- tered through the upper 1314 feet. A few of the shale beds in the upper 1314 feet are full of small limestone nodules. Small black phosphatic nodules, usually an inch or less in diameter, occur commonly in the lower 2 feet and 11 inches, and bodies of similar size and appearance, some or all of which are probably phosphatic, are found occasionally in the shales of the rest of the zone. lLens-shaped cavities, varying from a fraction of an inch to 6 inches or more in diameter and wholly or partly filled with a yellow earthy material that appears to have resulted from the weathering of pyrite concretions, occur at several horizons, especially in the 40 feet of olive gray shale that overlie the lower 28 feet. The shales break with a conchoidal fracture. The two low- est limestone beds (beds 1 and 4) are manganiferous. The shales are mostly barren except in the upper 13! feet, where they contain many fossils and appear to be limier than below. The limestones are fossiliferous. The beds of these three divisions appear to occur in an unbroken stratigraphic succession from the heavy-bedded, hard, olive gray and bluish gray shales of the bennetti zone, through the somewhat thinner-bedded, softer and darker shales of the hicksi zone, up to the predominately thin- bedded soft black shales of the davidis zone. The shales of the bennetti zone usually weather to a very dark olive brown color, those of the hicksi zone to brown or brownish black, 61 PARADOXIDES SECTION 61 and those of the davidis zone to black (except where stained brown by the weathering of pyrite). The limestones of the bennetti zone are generally light gray or pinkish in color, and occur in thin beds or in roughly oval or spheroidal nodules that appear to be of the same general type as the limestone nodules commonly found in the pre-Paradoxides beds of the Conception Bay region. The limestones of the hicksi zone are generally thinner than those of the bennetti zone, have smoother bedding planes, and occur usually as discontinuous sheets or thin lenses, rather than as continu- ous beds, and are dark gray, instead of light gray or pink, in color. In many ways they resemble the limestones of the overlying davidis zone, in which zone the limestones are not found as continuous beds, but as lenses and nodules, some of which are as much as a foot in thickness and six feet in diameter. Two beds filled with small masses of phos- phatic material are present in the davidis zone—one at the top of the zone and the other 7 feet below it—and phos- phatic bodies occur more or less abundantly in many of the shales of each of the three zones and in the two lowest lime- stones (beds 1 and 4) of the bennetti zone. The phosphatic bodies in the bed at the top of the davidis zone, and in the similar bed 7 feet below, sometimes contain fossils. Pyrite is common in all the zones, but is most abundant in the upper two. It usually occurs in small concretions and lenses, up to an inch or more in diameter, or in thin sheets, a small fraction of an inch thick. Some of the pyrite is associated with fossils. In a few instances hyolithid shells seem to ‘have been filled with it. Barite is present in the dolomitic limestone (bed 1), at the bottom of the bennitti zone, and in the hard black shlaes of beds 38 and 40, near the base of the hicksi zone. In the latter two beds it occurs as little crys- tals, about 14 of an inch long, scattered irregularly through the black shales. Thin beds of unctuous white clay, appar- ently derived from the decomposition of some sort of a shale, occur at several horizons in the hicksi and davidis zones. 62 BULLETIN 43 62 The writer has discerned no good evidence of sub-zrial erosion in any part of the Paradoxides section at Manuels. The varied character of the sediments involved proves that the conditions of deposition changed from time to time. Some of the beds, such as the phosphatic ones at the base and summit, may possibly indicate some sort of strati- oraphical break, for some or all of their phosphatic bodies may really be pebbles that have been rolled into place; but no good evidence of the erosion of any bed after its con- solidation has been found, nor have any sun cracks or indu- bitable ripple marks been discovered. More field work will probably have to be done before these questions can be finally answered. The upper and lower surfaces of bed 1 are uneven. Dr. Dale (1915, p. 406) has stated that the upper surface is ripple marked, and when looked at in cross section, it does appear to be so; but in the late summer, | when there is comparatively little water in the brook, sev- eral square yards of it are exposed, presenting evidence that the uneveness should probably be interpreted in a dif- ferent way. The bed is divided by joint planes into roughly rectangular pieces with approximately square upper faces, so that, when viewed from above, it looks like a pavement of blocks. The upper face of each block has an area of about one square foot, and is slightly higher in the center than around the edges. Blocks whose upper portions have been removed by erosion show a black central core, which shows evidence of being more resistant than the rest of the mass, and seems to indicate that each block is essentially a nodule and that the unevenness of the upper and lower surfaces of the bed is due to these nodules rather than to ripple mark- ing or erosion. Bed 4 shows the same features, although in a less pronounced degree, and beds 21 and 24 have similar, but less regular, nodular structures and uneven upper and lower surfaces. Bed 27 seems to be much like bed 24, except that its individual nodules have not originated near enough to each other, or have not grown large enough, to form a continuous bed of limestone, so that a shale full of limestone 63 PARADOXIDES SECTION 63 nodules has resulted instead. Bed 125 seems to have been interpreted by Dr. Walcot: as a true basal conglomerate, for in notes on the Manuels Brook Cambrian section, published in 1900 (Walcott, 1900, p. 317), he says: “Six feet above the thickest band of lime- stone, in which Paradoxides davidis occurs, there is a thi» layer of calcareous conglomerate, varying from 2 to 6 inches in thickness. It contains many dark argillaceous concre- tions, also pebbles of a reddish siliceous rock. This narrow band of conglomerate is found on both sides of the 11 and is taken as the base of the Upper Cambrian. Of the fauna occurring below it one species of Agnostus and one brachiopod, Obolus (Lingulella) ferruginea, pass up into the Olenus fauna.” The “thickest bed of limestone’? men- tioned in the first sentence of this quotation is probably a shale bed containing one or more of the large limestone lenses characteristic of the upper 7 feet of the davidis zone. A number of these lenses, lying end to end, might easily be mistaken for a part of a continuous stratum. The writer has not seen in bed 125 any of the “pebbles of a reddish siliceous rock” mentioned by Dr. Walcott. This bed shou!d probably not be considered to be a conglomerate, in the usual sense of that term, because many of the pebbles, con- cretions, or whatever they may be, that occur in it appear to show by their shapes and positions that they could not have been rolled into place. It is true that the bed marks the line between the fossiliferous shales of the Paradoxides davidis zone and the superjacent barren measures of un- known age; but, that a deposit of this peculiar character does not necessarily prove the presence of an important sedimentary break, is suggested by the fact that the faunas of the shales that overlie and underlie the similar phosphatic “pebble” layer in bed 115 appear not to differ from each other in any appreciable way. However, as the only fossils that have been discovered in bed 125 occur in the masses of phosphatic material (none having been found in the sur- rounding matrix), it is possible that the phosphatic masses 64 BULLETIN 43 64 of this bed are really fragments of older deposits. It seems more probable, however, that they are either concretions formed in the bed in which they now occur, or that they are pieces of unconsolidated contemporary or slightly older sedi- ments that were being broken up at the time when bed 125 was being formed. It is true that the Agnostus pisiformis and Agnostus pisiformis obesus (‘“‘Olenus’) faunas, which occur not far above the Paradoxides beds in the brook section (see p. 27), appear to be fundamentally different from the davidis fauna, and that there is, therefore, probably a stratigraphic break somewhere in the 30 or more feet of shales that lie between the highest beds known to contain Paradoxides and the lowest beds in which Agnostus pisiformis has been found; but this break may well occur somewhere above bed 125. At the base of the Paradoxides section, the fossils of the bennetti zone seem to differ considerably from those of the underlying ‘‘Catadoxides magnificus” beds (see pp. 16 and 26), and some sort of a sedimentary break may very possibly exist in the 10 feet and 10 inches of unfossiliferous manganiferous shales that intervene between the lowest known Paradoxides bed and the highest known Catadoxides horizon. These question will be taken up again in a future paper, in which the microscopical character, chemical composition, and mode of deposition of the Paradoxides beds of the Con- ception Bay region will be discussed in detail. - DISCUSSION OF THE CHARACTER AND OCCURRENCE OF THE FAUNAS The general composition of the Paradoxides faunas at Manuels is shown in the faunal lists on pages 35 to 56 and in Table IV and the Manuels lists of Chapter VII. An examination of these lists and table will show that in each of the three zones the trilobites far exceed in both numbers 65 PARADOXIDES SECTION 65 and variety all the other types of life. The number of individuals of each group (trilobites, brachiopods, etc.) in each zone is probably roughly proportional to the number of species present. It will be noted that almost all of the Newfoundland species are represented by identical or nearly identical forms in New Brunswick, Massachusetts, or Europe. The European and continental American types that have been recognized at Manuels are indicated in Table IV and discussed on pages 100-132. Wherever Paradoxides faunas have been carefully studied, they appear to be roughly assignable to one or more of four large groups. Three of these four groups cor- respond to the Paradoxides bennetti, Paradoxides hicksi, and Paradoxides davidis faunas of Newfoundland; the fourth group, which is the youngest, is characterized by the presence of Paradoxides forchhammeri, and is known to be well developed only in Scandinavia. Possibly future work will demonstrate that this classification is not the most natural one, but it is a convenient one to use when discuss- ing the faunas of southeastern Newfoundland and correlat- ing them with their contemporaries in northwestern Europe and continental North America. At Manuels the bennetti, hicksi, and davidis faunas can each be divided into sub- faunas; but, as the composition of some of these subfaunas may be due to very local facial conditions, it seems best to defer a discussion of them until more is known about the Paradoxides faunas of other Newfoundland localities. The bennetti, hicksi, and davidis faunas are not sharply separated from each other in the Manuels Brook section, for their species frequently range from one fauna into the next. There seems to be no evidence of the sudden dis- appearance of a fauna except at the top of the davidis zone. There is good evidence of the more or less sudden arrival of new groups of species, such as the group characterized by Paradoxides davidis and Agnostus punctuosus, which appears abruptly in the bottom of the davidis zone; but, in 66 BULLETIN 43 66 all known instances, a number of the species previously established in the region continued to exist there for some time after the arrival of the new forms. As a result of these conditions no bedding planes forming natural fron- tiers between the different zones have been discovered, and the stratigraphical limits here set to the zones are therefore of necessity chosen more or less arbitrarily, and may prove on further investigation not to be the best ones. Of the 25 genera and 79 species and varieties identified by the writer in the Paradoxides beds at Manuels, 13 genera and 54 species and varieties have been found in one zone only, 9 genera and 8 species and varieties have been found in both the bennetti and hicksi zones, 8 genera and 16 spe- cies and varieties have been found in both the hicksi ard davidis zones, and 6 genera and 4 species have been found in all three zones. No member of any of the Paradoxides faunas has been recognized in beds below the bennetti zone at Man- uels; but in the upper part of that zone there has been found a, tiny brachiopod which appears to be referable to Acrotreta gemmula Matthew, a species occurring in the pre-Paradoxides beds of New Brunswick and Cape Breton (Walcott, 1912, p. 686). Dr. Walcott (1900, p. 317) has stated that two species (Lingulella ferruginea and an Agnostus) range up from the Paradoxides beds into the Olenus beds at Manuels (see p. 63). No comprehensive discussion of the probable habitats and facial pecularities of the Paradoxides faunas at Manuels will be attempted in the present paper, as it is believed that a full consideration of these subjects should be deferred until after further examinations have been made of the other exposures of Paradoxides beds in the Conception Bay region. A few observation on the distribution of the fossils in the different kinds of beds represented in the section can, however, properly be recorded at this time, and are there- fore presented below. 67 PARADOXIDES SECTION 67 DISTBIBUTION OF FOSSILS ACCORDING TO LITHOLOGICAL CHARACTER OF BEDS Fossils are rare in the hard olive gray and bluish gray shales of the lower 220 feet of the bennetti zone, are com- mon in the nodular limestones and the upper 14 feet of gray shales of that zone, and are usually common, and often abundant, in the gray shales of the hicksi and davidis zones. It is in the thin-bedded soft black and gray shales of the hicksi and davidis zones, however, that they occur in the greatest profusion, some of these beds being literally full of agnostids and other trilobites. They are rare in the upper 5 feet of heavy-bedded black and dark gray shales of the davidis zone and in the thin phosphatic bed (bed 125) at the top of that zone. Trilobites are the most abundant types in most of the beds, but they are outnumbered by the brachiopod, Lingulella ferruginea, at a few horizons in the heavy-bedded dark gray shales of the upper part of the bennetti, and the lower part of the hicksi, zones (beds 35, 37, 38, and 39). The limestone nodules and beds appear to hold the same species as do the shale beds in, or between which, they lie. It is probable that the limestone nodules were formed in the muds after the muds had been laid down and before they had become consolidated into shales; for the fossils in the limestones retain their convexity, where those in the surrounding shales are flattened, and the shales are often slickensided where they have settled down around the resistant nodules. The two limestones of bed 19 were per- haps deposited originally as impure lime muds. They con- tain many fossils, whereas the shales which precede and succeed them are almost entirely barren. The fauna which they contain is, however, not greatly different in its general composition from the one which occurs in the shale beds, 23 and 26, higher up in the bennetti zone. The general rarity of fossils and the scarcity of lime- stones in most of the gray shales of the lower 220 feet of 68 BULLETIN 43 ~ 63 the bennetti zone may both be due partly to the same cause, whatever that cause may be. The smaller fossils that are preserved in these shales frequently retain some or all of their convexity, but the larger ones, such as Paradoxides bennetti, are usually flattened and crumpled, as though the limy material that must have been originally present in their shells had been removed before the consolidation of the surrounding sediments. The question of chemical com- position and method of preservation of the tests and shells of Cambrian organisms has been discussed in an interesting paper by Hicks (Hicks, 1875). Considering the different types of fossils separately, we find that the larger trilobites, such as Paradoxides, Cono- coryphe, Bailiella, Harttella, and Centropleura, are mainly confined to the harder, heavier-bedded, shales and the lime- stone beds and nodules. The trilobites of medium size, such as Solenopleura, Liostracus, Agraulos, and Holocephalina, are likewise found in the harder and heavier beds, but also occur commonly in, the soft, thinly-bedded shales. The agnos- tids are most abundant in the soft black shales, from which most other trilobites are usually absent; but they are some- times present in large numbers in the soft gray shales and the limestone nodules of the davidis zone. They are rare in the bennetti zone, where only 6 different forms have been discovered; they are common in the hicksi zone, 14 kinds having been found there, some of them in considerable num- bers; and they are the most abundant and characteristic fossils of the davidis zone, which contains countless thou- sands of individuals, belonging to no less than 26 species and varieties. The diversity of species of agnostids in the davidis zone may be more or less independent of the lithological char- acter of the containing rocks, but the great abundance of individuals in some of the beds of that division, particularly in some of the black shales, surely must have some facial significance. Bed 48 and its fauna are especially interesting 69 PARADOXIDES SECTION 69 in this connection. This bed, which occurs below the mid- dle of the hicksi zone, is a soft black shale of the sort com- monly found filled with agnostids in the lower part of the davidis zone, but very rare in the lower half of the hicksi zone, where most of the beds are somewhat heavily bedded gray shales containing relatively few of these little trilo- bites. Bed 48 is, however, as full of agnostids as a typical davidis zone black shale. Most of the agnostids present in the bed are found at other horizons in the hicksi zone, but the most abundant form, Agnostus cf. gibbus hybrida Brég- ger, has not been found at any other horizon at Manuels, and its presence in bed 48 is probably directly due to the black mud facial conditions obtaining at the time that that bed was deposited. Agnostids are conspicuously absent from the olive and dark gray shales and the nodular limestones of the upper- most part of the bennetti zone (beds 32 to 35), and the hard, heavy-bedded, dark gray shales of the lower part of the hicksi zone (beds 36 to 39). Trilobites are comparatively rare in beds 36 to 39, and the absence of agnostids from those horizons is therefore not surprising. It is more diffi- cult to imagine why they should be absent from beds 32 to 35, which contain many other trilobites. The relative abundance of agnostids in the soft black and gray shales does not appear to be due to a better pre- servation of their tests in those kinds of deposits, for the individual specimens are no more perfect than the ones that are enclosed in the harder, heavier-bedded, gray shales. There does seem, however, to be a greater proportion of com- plete individuals in the softer shales, particularly in the black shales. This may indicate that the softer shales were deposited in quieter waters than the harder ones, although the harder, heavier-bedded, shales usually appear to be just as fine-grained as the softer ones. Remains of the larger trilobites, such as Paradoxides and Centropleura, are com- paratively rare in the softer, thinner-bedded, shales, but when they do occur they are usually fairly well preserved, 70 “- “BOLLETIN 43 70 so that their comparative scarcity in such beds would seem to be due, in some instances at least, to other causes than lack of conditions favorable to good preservation. For ex- ample, although specimens of Paradoxides are common in the soft black agnostid shales of beds 94, 95, and 96, they are usually absent from the soft black and gray shales of the lower part of the davidis zone, and have not been found at all in the black shale, bed 48, although occurring in large numbers in most of the other beds of that part of the section. Of the brachiopods, Lingulella ferruginea and Acro- thele matthewi are found most frequently in the hard, heavy-bedded, greenish gray shales of the bennetti zone and the equally hard and- heavily bedded dark gray shales of the lower part of the hicksi zone, and-Acrotreta misera is com- monest in the soft black agnostid shales of the lower half of the davidis zone and in the similar shale of bed 48 of the hicksi zone. Although there is this general relation between certain types of fossils and particular kinds of beds, yet beds of the same lithological character that are separated from each other by a bed, or beds, of a different sort seldom contain exactly the same faunas. For instance, three soft black shales in the lower half of the davidis zone may be, to all appearances, identical in lithological character, and may contain typical agnostid faunas, but each of the three faunas will differ slightly from the other two, and, when any spe- cies occurs in two of the beds, it is likely to be rare in one and abundant in the other. Many species seem to have lived at Manuels only inter- mittently, for they appear in, and disappear from, the faunas there a number of times. In some cases the presence or absence of these forms seems to be due to the presence or absence of a particular type of sediment, but in other cases no such cause is apparent. For example, Corynexo- chus minor is found in beds 109 (a soft gray shale), 116 (a hard gray shale), and 119 (a black shale of medium hard- ness), but has not been discovered in any of the intervening gt PARADOXIDES SECTION vA! gray and black shales. The individuals occurring in beds 116 and 119 may possibly differ slightly from those in bed 109 and may perhaps belong to a variety of the species, but, if any differences exist, they must be small. Similarly, Aegnostus punctuosus appears to be absent from beds 100 to 105, although it occurs (often in great abundance) in the overlying and underlying strata. Agnostus granulatus, with a much longer time range, is even more irregular in its occurrence. It has been recognized only in beds 23, 28, 29, A3, 46, 51, 74, 80, 88, 90, 91, 98, 94, 95, 99,100, 101, 102, 105, 106, 107, 108, 109, and 110, but these beds include examples of the hard olive shales of the bennetti zone, the hard gray and the soft gray and black shales of the hickci zone, and the soft black and gray shales of the davidis zone. The species was a long-lived one, with a wide geographical range (it has been reported from Bohemia, England, and Newfoundland), yet it seldom occurs in large numbers in any one of the beds of the Manuels Brook section. Eodiscus punctatus also has a remarkable vertical distribution, being present in beds 40, 44, 51, 52, 54, 67, 69, 75, 81, 83, 84, 86, 99, 101, 102, 103, 105, 106, 107, 108, 109, 112, and 115, but seemingly absent from the intervening beds. It shows a preference for gray, rather than black, shales, but some- times occurs in the latter, and is frequently absent from the former. It is probable, however, that future work will disclose the presence of these four species, as well as many others, in beds which have so far yielded no traces of them. In contrast to these species of intermittent range are others which seem to have lived continuously at Manuels, and to have been much less susceptible to the effects of changing environments. Such species sometimes range through a succession of beds of different lithological char- acter. Agyraulos socialis, one of the most characteristic members of the hicksi fauna, is one of these persistent types; Paradoxides hicksi is another. On the whole, it seems likely that, although certain of 72 BULLETIN 43 72 the members of the Paradoxides faunas were largely influ- enced by the changes that took place in the local environ- mental conditions at Manuels, others were very little affected by them; and the invasion of the district by new forms from some unknown outside region was probably the most important factor in bringing about the faunal changes that took place from time to time. Only a few of the spe- cies and varieties present in the hicksi and davidis faunas seem to have evolved from forms previously existing in the district; the majority appear to have been invaders from without. This matter is discussed further on pages 1o1-3. IMPORTANCE OF AGNOSTIDS IN THE PARADOXIDES FAUNAS Although the various species of Paradoxides are the most characteristic members of the Paradoxides faunas of the world, they are far surpassed in numbers and variety by the agnostids. Paradoxides supplies the best index fossils for the Paradoxides beds in general, and for the main faunal subdivisions in particular, because it is found all through the Paradoxides beds, is confined entirely to them, and is represented by different and widespread forms at different horizons. The agnostids are, however, usually the most useful group for distinguishing and identifying the smaller faunal divisions because of their great number and variety, their apparent ability to spread rapidly over large areas, and the ease with which their small shields are pre- served entire, so that they can be specifically identified. Moreover, although they are most abundant in the Para- doxides beds (particularly in the hicksi, davidis, and forch- hammeri zones), they are also found commonly in other Cambrian strata; and will probably prove ultimately to be one of the greatest aids in correlating not only the Para- doxides faunas, but also some of the later faunas of differ- ent parts of the world. One of the most striking differences between the Para- doxides faunas and those which underlie them is that, while 73 PARADOXIDES SECTION 78 the Paradoxides faunas contain agnostids, the others do not. Obviously the ancestors of the Paradoxidian agnostids must have lived in pre-Paradoxidian times, and their absence from the known pre-Paradoxides beds in all the regions where Paradoxides is known to occur indicates that the waters in which those older beds were deposited were either cut off by some barrier from the seas which contained agnostids at those times, or (what is less likely—for the later pre-Paradoxides beds, at least) that agnostids had not yet been evolved when those beds were laid down. Agnos- tids occur in Siberia, however, in faunas which may be of Catadoxidian (pre-Paradoxidian) age. (See vonToll, 1899.) Agnostids occur in the beds which overlie the Para- doxides beds, but most of them are very different from those in the Paradoxides beds, and there are not nearly so many different species as there are in the Paradoxides faunas. VI. DESCRIPTION OF NEW SPECIES FROM THE PARADOXIDES BEDS AT MANUELS The following nine forms, believed to be new, have been discovered in the Paradoxides beds at Manuels in the course of the present investigation: Agnostus clare. Agnostus barlowi definitus. ie Agnostus parvifrons punctifer. Agnostus vaningeni. Agnostus longifrons parvulus. Agnostus rectangularis. Conocoryphe bullata. Paradoxides parvoculus. Liostracus globiceps jaculator. These forms are described below and figured on Plate 3. In describing the agnostids the writer employs the terminology of parts that has been in common use since Barrande’s day. Professor Raymond has called attention to 74 BULLETIN 43 © 74 the fact that paleontologists have perhaps been confusing the pygidia and cranidia of the agnostids, calling the pygid- ium the cranidium and vice versa (Raymond, 1917; 1920). The writer believes that this is possibly true, and that the question of which end of an Agnostus is which is still an open one. He feels, however, that such evidence as we have at present is more in favor of the long-accepted view than of Professor Raymond’s newly-suggested one. Agnostus clarz, new species Pl. 3, Fig. 1 Only the pygidium is known. Pygidium, subquadrate, with a thin, tapering, sharply pointed, cusp or spine at each of the two posterior angles. Lateral lobes, wide, tapering and depressed behind the axis, but confluent, separating the rear end of the axis from the marginal furrow. Axis cylindrical, 7/10 as long and 1/3 as wide as the whole pygidium, nearly parallel-sided, but with a very slight constriction and faint indications of a trans- verse furrow at a point about 1/3 of the way back from its forward end, evenly rounded posteriorly. Margin of mode- rate width all around. The marginal furrow and the fur- row surrounding the axis are of moderate width and depth. In the single specimen known, which is slightly flattened, the axis is very convex, with some indications of a tubercle or keel on the anterior half. The side lobes are only slightly convex. The pygidium, as a whole, seems to have been a little higher at the front than at the rear. The ends of the spines at the posterior angles are broken off, but there is no evidence to indicate that they were unusually long. The surface of the test is smooth. Length, 5mm. Breadth, 5 mm. at the rear, 6 mm. at the front. The greater breadth in front may be due to expan- sion on flattening, if the front was more convex than the rear. Horizon, bed 23, in the upper part of the Paradoxides 75 PARADOXIDES SECTION 75 bennetti zone. Very rare. The specific name is given in honor of Mrs. Howell. This species is based on a single pygidium, which is specimen 8000 in the paleontological collection of Princeton University. The most striking features of this pygidium are its quadrate form and its short, blunt, very convex axis. Of the agnostids that have been described from the Para- doxides beds of Europe and North America, Agnostus fallax vir Matthew (1886, p. 69; 1896, p. 215), of New Bruns- wick, and A. fallax ferox Tullberg (1880, p. 32), of Scandi- navia, appear to approach it more closely. The tails of vir and ferox differ from that of clare in having more taper- ing and slightly longer axes, and in being less quadrate in shape. The cranidium of clarz is unknown, but will prob- ably prove to be more or less like that of fallax vir. Cranidia of that general type have been found in bed 23, but some or all of them probably belong to a form like Agnostus fallax trilobatus Matthew (1886, p. 68), pygidia of which occur commonly in that bed. Two cranidia (Nos. 8003 and 8025) of a similar form, but with the glabella only very faintly defined, also found in bed 23, may belong to clare. It seems more probable, kowever, that they belong to another form, Agnostus barlowi definitus, and they are therefore described and discussed in the account of that species, which follows the present description. Until the eranidium of clare is discovered, it will be impossible to assign the species definitely to any one of the numerous subdivisions into which different authors have proposed to split up the old genus, Agnostus. If it proves to have a fallax-like head, it will belong to Tullberg’s subdivision, “Fallaces,” of his group, “Limbati” (Tullberg, 1880, p. 11), and to Corda’s genus, Peronopsis (Hawle and Corda, 1847, p. 115; Raymond, 1913, p. 2; Zittel-Eastman, 1913, p. 710). 76 BULLETIN 43 | 76. Agnostus barlowi definitus, new variety Pl. 3, Figs. 2,3 Only the pygidium is definitely known. Pygidium semi-elliptical, about 1 longer than wide, with a broad, shallow, mariginal furrow and a somewhat convex margin. The margin is of medium width in front, but becomes very wide behind, where it bears two short, broad, sharply pointed cusps or spines. Lateral lobes of medium width, confluent behind the axis, where they are almost as broad as at the sides. Axis broad and very bluntly pointed; slightly constricted in its forward third, where it bears an elongated tubercle; widening rapidly toward the front where it reaches its greatest breadth. It is about 3/7 as wide and 5/7 as long as the whole pygidium, and is com- pletely surrounded by a narrow, shallow furrow, which is often very weakly impressed around the rear half of the axis. The whole pygidium is of medium convexity, highest in the front half and sloping gradually toward the sides and rear. . Length, 314 to 4144 mm. Breadth, 314 to 4144 mm. Horizons, beds 23, 26 and 28, in the upper part of the Paradoxides bennetti zone. Rare. The holotype is specimen No. 8001 in the paleontologi- cal collections of Princeton University. It is from bed 23. The varietal name refers to the fact that the axis is completely defined by a furrow. This pygidium is very similar to that of Agnostus bar- lowi spinatus Illing (1916, p. 413). It differs from that form in having the axis completely defined by a furrow, instead of merely indicated by two short grooves near the front of the pygidium, in having the marginal border broader at the rear; and in having the cusps or spines farther back. Barlowi spinatus has been found only in the hicksi zone of England, while barlowi definitus is known to occur only in the upper part of the bennetti zone of New- — foundland. Only a few specimens of either form have been 77 PARADOXIDES SECTION ver discovered, and it is possible that a large series of specimens from either England or Newfoundland would show that the two varieties are one. However, as they occur at different horizons so that their varietal pecularities may have some phylogenetic significance, it has seemed best to describe the Newfoundland form under a separate name. The author has followed Illing’s example in calling definitus a variety of barlowi. Whether spinatus and defini- tus are really varieties of Agnostus barlowi Belt (1868, p. 11) must remain an open question until the crinidium of either spinatus or definitus is discovered. There are two cranidia from bed 23 in the Princeton collections which may possibly belong to definitus (Nos. 8003 and 8025 in the paleontological collections of Princeton University; see plate 3, figure 3). They agree with Lake’s (1906, p. 17) description of the head of barlowi except that they have a marginal fold of medium width, instead of the very nar- row one of that form, and that a two-lobed glabella is faintly outlined, instead of having only the rear end of it suggested by short furrows, as in Belt’s species. Lake’s figure of bar- lowi seems to show a faintly outlined glabella, with no sign of a cross-furrow separating it into two lobes. It is barely possible that these two Newfoundland heads belong to Agnostus clare (see description above), the head of which is unknown and may very well be of this type. They are referred to barlowi definitus mainly because definitus seems, from the number of pygidia found, to have been more com- mon than clare in the beds in which the heads were found. If the heads do belong to barlowi definitus, we would appear to have in the phylogeny of the barlowi line a progressive loss of furrows from both cranidium and pygidium, as well as a loss of spines from the pygidium— indicating, perhaps, a progressive loss of vitality in the race, resulting ultimately in extinction. Belt stated that the type specimens of bar- lowi came from the Tremadoc. Lake supposed that since Agnostus cicer Tullberg (1880, p. 26), which appears to be identical with barlowi, was recorded by Tullberge from the 78 BULLETIN 43 78 Conocoryphe equalis zone and Agnostus intermedius zone (both of which zones are well up in the hicksi zone), Belt must have been mistaken when he recorded barlowi from the Tremadoc. The finding of ‘‘cicer”’ in the Paradoxides davidis beds by Groénwall (1902, p. 167), of spinatus in the lower hicksi beds by Illing, and of definitus in the upper bennetti beds, as here recorded, indicates that the race may have ben a lony-lived one and may really have ranged up into the Tremadoc, after all. Gronwall (1902, pp. 59, 60, 210) has described a va- riety of barlowi from the davidis zone of Bornholm Island, Denmark, which he has named “cicer forfex,” the tail of which differs from cicer, as that of a definitus differs from that of a spinatus, in having the axes completely outlined. Forfex’s tail differs from that of definitus in having a more bluntly pointed axis, no spines, and a narrower marginal — fold. Barlowi, barlowi spinatus, and barlowi definitus have unusually wide marginal folds, however. Belt notes how the wide fold of barlowi makes the species resemble Agnos- tus nudus (Beyrich), (Beyrich, 1845, p. 46; Barrande, 1852, p. 903), and this resemblance is even more striking in defini- tus, with its still wider border, widest at the rear. Agnostus parvifrons punctifer, new variety Pl. 3, Figs. 4, 5 Only the pygidium is definitely known. Pygidium semi-elliptical, about as wide as long, with a slightly convex margin, which is of medium width in front but becomes very broad at the rear, reaching its greatest breadth back of the rear end of the axis. Lateral lobes of medium width, tapering at the rear, not confluent. Axis of medium width, very slightly constricted near its anterior end, tapering backward to a fine point which separates the lateral lobes from each other and reaches the marginal fur- row ; very convex, reaching its greatest height in the middle, where it is prolonged upward and backward into a sharp- pointed, somewhat elongated, peak or super-tubercle, which = 79 PARADOXIDES SECTION 79 slopes toward the front of the pygidium at an angle of about 45°, slopes somewhat more steeply toward the sides, and has a very steep, concave slope toward the rear, the concavity being due to the fact that the peak bends backward so that it slightly overhangs the posterior slope. The lateral lobes and marginal fold are only very slightly convex, and the axis rises very prominently above them. The lateral lobes are separated from the axis and the marginal fold by fur- rows which are very wide and fairly deep in the type speci- men, which is exfoliated, but which are probably verv much less prominent—perhaps even indistinct—in unexfoliated specimens. Length, 3 mm. Width, 3 mm. Height of the peak of the axis above the lateral lobes, 1 mm. Horizon, bed 101, in the lower part of the davidis zone. Rare. : The holotype is specimen 8004 in the paleontological collections of Princeton University. The natural mold is specimen 8005. The varietal name refers to the elongated tubercle borne on the axis. The pygidium of this remarkable little aghostid is very similar to that of “Agnostus parvifrons mammillata” Broég- ger (1878, p. 56), appearing to differ from that form only in havine the peculiar elevation of the middle of the axis more pronounced and slightly bent backward instead of being symmetrical. An immature pygidium 114 x 114 mm. in size (specimen 8006 in the paleontological collections of Princeton University) found in the same bed as the type of parvifrons punctifer and probably belonging to that form, appears to differ in no essential respect from imma- ture pygidia of “parvifrons mammillata” from Scandinavia contained in the Princeton collections (specimens 8009 and 8010). The thorax and cranidium of parvifrons punctifer are not definitely known. The cranidium is probably like that of “narvifrons mammillata.” An immature cranidium about So BULLETIN 43 fe) 144, x 114, mm. in size (8008 in the Princeton collections), from the same bed as the holotype pygidium of punctifer, appears to differ in no way from an immature cranidium of “mammillata” on specimen 8010. It is similar to the adult head of “mammillata” figured by Brogger (1878, pl. V, fie. 3a), being semi-elliptical, rather convex, and surrounded by a margin of medium width, which is separated from the rest of the cranidium by a furrow that is of medium width and depth in this particular specimen, which is exfoliated, but would probably be much less distinct on an unexfoliated specimen. The glabelia is very convex, slightly longer than wide, and is surrounded by a furrow of medium depth and width, which would probably be shallower on an unexfoli- ated specimen. The basal lobes are of medium size, and are confluent behind the glabella. It is possible that the single pygidium upon which this hew variety is based is a deformed or aberrant example of “parvifrons mammillata,” as pygidia nearer to the normal “mammillata” type occur only a few feet higher in the section (in bed 74). Aenostus parvifrons and its varieties appear to have been an extremely variable group, however, and it is quite possible that punctifer is a valid variety. Whether puncti- fer and mammillata should really be classed as varieties of Aenostus parvifrons Linnarsson (1869, p. 82) is perhaps an open question. It seems best to follow Brogger for the present, however, and consider them.as such. If future research proves that the two are not varieties of parvifrons, then punctifer will become a variety of mammillata. Jekel (1909) has proposed the genus, Hypagnostus, for Agnostus parvifrons and its allies; and if that genus is accepted as defined by Jeekel, punctifer will belong to it. \ Pl. 3, Fig. 6 Agnostus vaningeni, new species Pygidium semi-elliptical, moderately and fairly evenly convex, surrounded by a margin of medium width and con- 81 PARADOXIDES SECTION 8I vexity, which is separated from the rest of the tail by a shal- low furrow of medium width. Axis large, 114 times as wide as the lateral lobes; long; evenly tapering, except at a point just back of the front end, where it is very slightly con- stricted ; ending in a blunt point a short distance in front of the marginal furrow; separated from the lateral lobes by a narrow, shallow furrow; bearing a somewhat elongated tubercle about one-third of the way back from its forward end. Lateral lobes of medium width in front, tapering regu- larly to rather blunt points at the rear, where they are separated by a narrow, shallow furrow, which connects the furrow surrounding the axis with the marginal furrow. The whole pygidium is evenly and very moderately convex. ‘ Length of the holotype pygidium, 314 mm.; breadth, 3144 mm. Other specimens identified as belonging to this species (specimens 8013 and 8030) are 1 mm. or more larger. The axis of the holotype is 3 mm. long, and 134, mm. wide. Horizon of the holotype, bed 99. Other specimens pos- sibly belonging to this species( specimens 8031 and 8032) were found in beds 100 and 102. All of these beds are in the lower part of the davidis zone. Rare. The holotype is specimen 8011 in the paleontological collections of Princeton University. The natural mold of the holotype is specimen 8012. The specific name is given in honor of Professor Gilbert vanIngen. The pygidium of this species resembles those of Agnos- tus levigatus terranovicus Matthew (1896, p. 233), Agnostus parvifrons tessella Matthew (1886a, p. 71), and Agnostus rotundus Grénwall (1902, p. 78). From levigatus terranovicus it is distinguished by its longer axis and the separation of its lateral lobes at the rear. It differs from rotundus in having a longer and wider axis, and tapering side lobes, completely separated at the rear. It is not greatly different from parvifrons tessella in general shape, but has a slightly wider and longer axis, the marginal furrow and 82 BULLETIN 43 ~ 382 the furrow surrounding the axis are less distinct, and there are no distinct furrows crossing the axis, such as there are. in that form. Although the crandium of Agnostus vaningeni is un- known, it is probably more similar to that of levigatus terranovicus and rotundus than to that of parvifrons tes- sella, as certain cranidia of the terranovicus-rotundus type occur in association with the vaningeni pygidia. Possibly vaningeni represents an intermediate stage of evolution between parvifrons tessella and levigatus terranovicus. It is interesting to note, in this connection, that the unexfoli- ated pygidia of vaningeni look very much like those of — levigatus terranovicus, while exfoliated tails much resemble those of parvifrons tessella. _Vaningeni is probably nearest to terranovicus, as its crandium is almost certainly of the terranovicus type. It may prove, however, to be even nearer to the British form which Lake (1906, p. 14) has referred to rotundus, but which, to judge from his figures and description, is perhaps nearer to levigatus terranovicus than to rotundus. Perhaps parvifrons tessella, vaningeni, levigatus terranovicus, and levigatus Dalman (1828, p. 136) represent a single line of development, to a branch of which rotundus and the closely related Newfoundland form, levigatus mamilla Matthew (Matthew, 1896, p. 234) may belong. Agnostus lens Grénwall (1902, p. 65) appears also to be very similar to levigatus terranovicus, and, as lens is found at a horizon in Scandinavia that appears to be lower than the one at which vaningeni occurs in Newfoundland, it is possible that terranovicus is descended from lens, and not from vaningeni. Vaningeni, like levigatus terranovicus, levigatus mamilla, lens, rotundus, and the somewhat similar Paradoxidian form, Agnostus levigatus ciceroides Mat- thew (1896, p. 234), and the very similar or identical form, Agnostus altus Grénwall (1902, p. 58), together with Agnostus levigatus forfex Brégger (1878, p. 58), Agnostus barrandei Salter (Hicks, 1872, p. 176), Agnostus lens fron- tosa Groénwall (1902, p. 66), Agnostus barlowi Belt (1868, 83 PARADOXIDES SECTION 83 p. 11), Agnostus barlowi forfex Grénwall (1902, p. 59), Agnostus barlowi spinatus Illing (1916, p. 413), and Ag.bar- lowi definitus (described above, see p. 76), does not appear to fit into any of the genera of Jekel’s or Professor Ray- mond’s classifications. Some of them, such as levigatus terranovicus and levigatus forfex, are probably so closely allied to levigatus, the type of Corda’s genus, Lejopyge (Hawle and Corda, 1847, p. 51), that they should be classed in that genus, but the others should probably be assigned to one or more new genera. If vaningeni is really derived from parvifrons tessella, it is probable that teseslla belongs to a different group from parvifrons Linnarsson (1869, p. 82), and should not be called a variety of that species. Agnostus longifrons parvulus, new species Pl. 3, Figs. 7, 8 Only the cranidium is definitely known. Cranidium small, semi-elliptical, evenly and moderately convex, surrounded by a narrow, convex margin, which is separated from the rest of the cranidium by a narrow mar- ginal furrow of medium depth. Glabella bilobed, parallel- sided, the front lobes separated by a straight furrow of me- dium width and depth. The glabella is surrounded by a deep, narrow furrow, which is continued forward in front of the front lobe, so that it joins with the marginal furrow and separates the two cheeks. The front lobe of the glabella is roughly triangular in shape, the two sides of the triangle being convex and the apex an obtuse angle. It is of moderate convexity, approximately the same as the cheeks. The rear lobe is about twice as long as wide; the front third is moder- ately convex, like the front lobe, but the posterior two-thirds is evenly and highly convex, and is crowned with a tubercle. The basal lobes are small and triangular. The cheeks are somewhat wider than the glabella, are moderately convex, of approximately equal width throughout, and are separated by a furrow in front of the glabella. Length of the holotype cranidium, 2 mm.; breadth, 13, 84 BULLETIN 43 . 84 mm. This is an average sized adult cranidium. Horizon, bed 109, the middle part of the davidis zone. Rare. The holotype is specimen 8014 in the paleontological collections of Princeton University. The varietal name refers to the small size of the variety. The holotype of this variety is a cranidium. No well preserved and certainly identifiable complete individual has ~ been discovered to prove beyond question just what the pygidium is like; but a poorly preserved specimen showing both cranidium and pygidium (specimen 8015 in the Prince- ton collections), which is almost certainly of this variety, and numerous little pygidia found associated with longifrons parvulus cranidia, enable us to be reasonably certain about the matter. The description of the pygidium given below is based upon an impression on a piece of shale (Princeton specimen 8016) broken from the piece holding the holotype cranidium, and upon pygidia found on Princeton specimens 8017, 8018, 8033, and 8037. Specimen 8033 and its counter- part 8037 hold a cranidium that is believed to belong to this species. Pygidium small, semi-elliptical, surrounded by a margin of medium convexity and width, and by a shallow, narrow, marginal furrow. Axis convex, narrow, slightly constricted in its anterior third, tapering to a rounded point at the rear. It is divided into three segments. The anterior lobe is short. The middle segment is about the same length as the anterior one, but is expanded in the middle so that it juts forward and backward until it is nearly twice as long there as it is at the sides. It bears a prominent median tubercle or keel. The posterior segment is about twice as long as the two anterior segments combined, but does not reach the marginal furrow behind. The lateral lobes are convex, but not so much so as the axis; they are wider than the axis, and are somewhat narrowed behind, where they are separated by a narrow, shallow, furrow, which joins the marginal furrow to the narrow, but rather deep, furrow 85 PARADOXIDES SECTION CO Zp | that surrounds the axis. The pygidia have about the same general dimensions as the cranidia. This form is very closely related to Agnostus longifrons Nicholas (1916, p.453). It differs from longifrons in being smaller, in having the anterior lobe of the glabella shorter, and in having the posterior lobe of the glabella longer, with the greatest convexity 1/3 of the length of the lobe forward from the posterior margin, instead of just in front of it. The pygidia of the two forms appear to be identical, except for size, and the thorax of longifrons parvulus will pro- bably prove, when it is discovered, to be like that of long- ifrons. Agnostus longifrons, which is a Welsh species, appar- ently occurs in the hicksi zone, while longifrons parvulus is found well up in the davidis zone. Mr. Nicholas has pointed out, in his description of longifrons (1916, p. 454), how this type of agnostid resembles Agnostus gibbus Linnarsson (1869, p. 81). Agnostus longifrons parvulus is one of the smallest of all known agnostids, and therefore one of the smallest of known trilobites. It is perhaps the smallest trilobite yet discovered. It is almost surely descendéd from the Agnostus longifrons line. It belongs in the genus Agnostus, sens. strict., of Jeekel’s and Raymond’s classifications. Agnostus rectangularis, new species Pl. 3, Fig. 9 Only the cranidium is known. Cranidium roughly rectangular, slightly longer than broad, with rounded corners. Margin broad and nearly flat. Marginal furrow broad and shallow. Cheeks of me- dium width, approximately equal in width throughout, con- fluent in front of the glabella. Glabella of medium length and width, about as wide as the cheeks, and two-thirds as long as the cranidium, not bilobed. It narrows slightly to- ward the front, where it is evenly rounded, and it is separat- ed from the cheeks by a shallow furow. The basal lobes are 86 BULLETIN 43 86 large, triangular in shape, and rather indistinctly marked off from the rest of the glabella. There is a very indistinct tubercle on the glabella about 1/3 of the way forward from the rear margin. The whole cranidium is rather flat. It is highest at the rear. Length of the holotype, 31% mm., breadth, 314, mm. The only other specimen known is 4 mm. long and 314 mm. broad. Horizon, somewhere in the upper half of the davidis zone. Exact horizon unknown, but probably bed 115 or a nearby bed. The holotype is specimen 8019 in the paleontological collections of Princeton University. The natural mould of the holotype is specimen 8039. The other specimen is num- bered 8020. The specific name refers to the rectangular appearance of the cranidium. At first glance, the cranidium of this species reminds one of that of Agnostus fallax Linnarsson (1869, p. 81), because of its quadrate form and wide margin. Fallax, however, has a, bilobed glabella. The cranidium of Agnostus obtusus Belt (Belt, 1868, p. 10; Lake, 1906. p. 28) is similar to that of rectangularis in general shape, but it has, like fallax, a bilobed glabella. Agnostus tardus Barrande (1852, p. 913), and the closely related forms, A. glabratus Angelin (1878, p. 6) and A. lentiformis Angelin (1878, p. 7), have cranidia somewhat similar to that of rectangularis, as do A. sidenbladhi Linnarsson (1869, p. 82), A. callavei Raw MS., (Lake, 1906, p. 25), and A. sallesi Munier-Chalmas and Bergeron (Bergeron, 1889, p. 337). Tardus, glabratus, len- tiformis, sidenbladhi and callavei are of much younger age than rectangularis, but sallesi is from the Paradoxides beds of southern France. The figure and description of sallesi given by Munier-Chalmas and Bergeron are not clear and detailed enough to show just how similar their species is to rectangularis, but they are sufficient to suggest that it is very close, so that the Newfoundland form may yet prove Sz. PARADOXIDES SECTION 87 to be a variety of the French one, or even identical with it. Rectangularis is probably referable to the genus Arthro- rachis Corda of Raymond’s classification. The cranidium of rectangularis exhibited on specimen 8020 shows faint traces of two peculiar diagonal furrows on the front of the glabella in just the same position as those described and figured by Raw and Lake on the glabella of callavei (Lake, 1906, p. 26, pl. II, fig. 20). Lake states that these furrows may be adventitious cracks in callaveli, and the same may be the case with the one cranidium of rectangularis. The holotype cranidium of rectangularis does not show any clear evidence of such furrows, although there appear to be slight indentations present, which may be traces of similar grooves. These furrows, if they really are such, the general character of the cranidium, and the pres- ence in beds of the same, or nearly the same, horizon of pygidia somewhat resembling the pygidium of Agnostus quadratus Tullbere (1880, p. 34), suggest the possibility that rectangularis may prove to be related to quadratus, rather than to the other species mentioned above. To judge from Tullbere’s figure, quadratus has a more pointed gla- bella than rectangularis, and shows more definite indica- tions of furrows. The quadratus-like pygidia from Manuels also differ from the pygidium of quadratus in having much shorter axes. Quadratus occurs in the Paradoxides forch- hammeri beds of Sweden, above the Paradoxides davidis beds, and perhaps rectangularis is an ancestor of quadratus. The exact horizon of the quadratus-like pygidia found in loose limestone nodules at Manuels is not known, but it is somewhere in the davidis zone, and, to judge from the fossil evidence, from the same, or about the same, horizon as the rectangularis cranidia. A typical example of this sort of pygidium is shown on Princeton specimen 8038. Conocoryphe bullata, new species ~ 1215 8}, ies al) ala Only the cranidium is known. 88 BULLETIN 43 ~ 8s Cranidium roughly semicircular, twice as wide as long, with broad, upturned anterior and posterior margins and broad deep anterior and posterior marginal folds, and a nearly straight facial suture. Glabella conical; about 3/5 as long as the whole cephalon, with probably three furrows (the only known specimens are imperfect and show only the forward part of the glabella). The glabella is surrounded by a furrow of medium depth and width. The cheeks are confluent in front of the glabella, where they are slightly convex. The main part of each fixed cheek is remarkably convex, reaching its greatest height opposite the middle of the glabella. The whole surface of the cephalon is covered with small tubercles, about 4 in each square millimeter. A small keel-shaped prominence on the outer front edge of the swollen portion of the cheek may be the rudiment of an ocular ride. Length of the holotype cranidium 1 cm., breadth, 2 cm. Height of elevated portion of fixed cheek above bottom of posterior marginal furrow just back of highest part of the fixed cheek, about 5 mm. Horizon, bed 19, in the middle of the Paradoxides ben- netti zone. Rare. The holotype is specimen 8021 in the paleontological collections of Princeton University. The natural mould of the holotype is specimen 8040. A smaller head, believed to be referrable to this species (fig. 11), is numbered 8026. The specific name refers to the bubble-like elevation of the fixed cheeks. } This Conocoryphe resembles Conocoryphe dalmani Angelin (1878, p. 63) in its general shape, but differs markedly from that species in its greatly elevated fixed cheek. This elevation of the fixed cheek is characteristic of Ctenocephalus, rather than Conocoryphe, as exemplified in Ctenocephalus tumida Gronwall (1902, p. 99), and its presence in this very early Conocoryphe may perhaps indi- cate a convergence of the Conocoryphe and Ctenocephalus stems. 89 PARADOXIDES SECTION 89 Paradoxides parvoculus, new species Pl. 3, Figs. 12,13 Only the cranidium is known. Cranidium of the ordinary Paradoxidian form. Gla- bella large, evenly rounded in front, and narrowed at the rear so that it is only 34, as wide at the neck ring as it is at the widest part in front. The widest part of the glabella is 1/3 of the way back from the front. At this widest point the glabella is a little more than half as wide as the cephalon is long. There are four glabellar furrows in addition to the neck furow. The neck furrow is transverse and rather lightly impressed across the axis of the glabella, but is curved forward at an angle of 45 degrees and is deeply impressed at the sides. The second glabellar furrow is nearly transverse at the sides, but curves backward to form a shallow, backward-pointing arc over the axis. The third furrow is not continuous across the glabella, but is repre- sented by two short furrows, extending in from each side about half the way to the median axis of the glabella, and about at right angles to that axis. The anterior, or fourth pair of glabellar furrows, which are situated opposite the anterior ends of the palpebral lobes and at thé widest part of the glabella, are slightly shorter than the third pair and are deflected slightly forward from the outside toward the inside of the glabella. The neck ring of medium width, with a small tubercle. Fixed cheeks subtriangular, about 1/6 longer than wide. Brim flat or perhaps with a slightly elevated rim on its front edge. The brim is narrow in front of the glabella, but widens toward each side. In the holotype, which is 22 mm. long, the brim is 1 mm. wide in front of the median axis of the glabella, and 4 mm. wide at the sides. The glabella is separated from the brim and from the fixed cheeks by a continuous furrow of medium width and depth. The posterior marginal furrow, running trans- versely across the rear ends of the fixed cheeks, is also of medium width and depth. The eyes are very short and 90 BULLETIN 43 Bee go narrow. They are curved, and extend obliquely backward from a point close to the glabella at the elabella’s widest part. They are only about half as lone as the free cheeks, reaching back only to a point opposite the middle of the sec- ond glabellar furrow. They are very narrow, bcing slichtly less than 1 mm. wide on the holotype cranidium, which is 22 mm. long. The anterior branch of the facial suture is at an angle of about 45° with the median axis of the crani- dium; the posterior branch is straicht, and is directed at an angle of about 25° from the median axis of the cephalon. The surface of the fixed cheeks is covered with many anas- tomosine raised lines, which form a fine and somewhat irregular wrinkled network. The rest cf the cranidium appears to be smooth or ornamented only with very fine concentric or parallel lines, and perhaps very fine tubercles. Length of the holotype head, 22 mm.; width at rear, 16 mm.; length of eye lobe, 5 mm.; width of eye lobe, 114 mm.; length of posterior branch cf facial suture (be- hind the palpebral lobe), 6 mm. Horizon, beds 23 and 26, in the upper part of the ben- netti zone. Rare. The holotype is specimen 8027 in the paleontological collections of Princeton University. It is from bed 23 A smaller head, whose exact horizon is unknown but which was found either in bed 23 or an adjacent bed, is number 8022. Both it and the holotype are figured. Another small head from bed 23, which seems to be referrable to this spe- cies, is numbered 8028 (natural mould, 8041). The specific name refers to the size of the eye lobes, which are unusually small for a Paradoxides. Parvoculus has perhaps the smallest eye lobe of any known species of Paradoxides. It appears to be most closely related to Paradoxides hicksi, and may prove to be an ances- tral variety of that species; but it seems best not to call it such until more is known of the ontogeny and phylogeny of hicksi. Hicksi appears to be a variable species, but some of the British forms assigned to it should perhaps be con- gI PARADOXIDES SECTION gI sidered as varieties of the typical form. Parvoculus differs from hicksi as described by Salter (1869, p. 55) in having narrower eye lobes, a shorter glabella that is more evenly rounded at the front, with the two rear glabella furrows extending all the way across the glabella in adult species, and a flat or nearly flat brim. It closely resembles one of the heads figured by Illing (1916, pl. XXXVI, fig. 3) as P. hicksi, differing only in the fact that, while the second glabellar furrow of Illing’s specimen arches forward, that of parvoculus bends backward. Illing’s specimen came from a much higher horizon than parvoculus, however (the upper part of the hicksi zone). In the general shape of the elabella and the thinness of the eye lobe, parvoculus resem- bles young specimens of hicksi occurring in the hicksi zone at Manuels, and it seems probable, therefore, that it is the ancestor of hicksi, and that hicksi may have evolved in America. The flat or nearly flat brim of parvoculus is also characteristic of early types of Paradoxides, as pointed out by Professor Raymond (1914, p. 235). Liostracus globiceps jaculator, new variety Pl. 3, Fig. 14 Only the cranidium is known. Cranidium moderately convex (in specimens preserved in shale), the elabella only a little more convex than the rest of the cephalon. The front of the cephalon is an are of very slight curvature, and the facial suture runs so nearly straight back from this front edge to the palpebral lobe that the front half of the cranidium has a rectangular appear- ance. Back of the palpebral lobe the facial suture runs diagonally backward and meets the posterior marginal fur- row at an angle of about 45°. The brim is flat and broad, and is separated from the rest of the cephalon by a broad, shallow furrow. The fixed cheeks, which are moderately convex, join broadly in front of the glabella, the distance between the front of the glabella and the anterior marginal furrow being slightly more than half the distance between g2 BULLETIN 43 g2 the side of the glabella and the palpebral lobe. The fixed cheek is almost twice as wide at its posterior end as it is opposite the palpebral lobe, so that the cephalon is only 2/3 as wide at the front as it is at the rear. The posterior mar- ginal furrows are wide and shallow, as is the furrow that Separates the glabella from the fixed cheeks. A facial ridge runs from near the front of the glabella outward and slizhtly backward, in a gentle curve, to the anterior end of the nalnebral lobe The neck ring is of medium size, and is prolonged backward into a spine, which is so long that the spine and neck ring combined are almost as long as the elabella. The glabella tapers slightly toward the front, where it is evenly rounded. It has four pairs of short, shal- low, indistinct furrows which run obliquely backward at an anele of about 20°, from the outer edge of the glabella 14 to 2/3 of the distance toward the middle line of the glabella. The anterior pair of furrows are the shortest, the posterior ‘the longest, the other two being intermediate in leneth. The anterior furrow starts from a point opposite the inner end of the facial ridge. The surface of the cranidium is covered with closely set tiny pits, not visible to the naked eye, and there are faint indications of parallel lines running across the part of the free cheeks in front of the facial ridve in a direction perpendicular to the facial ridge, such as are fre- quently found in other species of Liostracus and allied genera. Length of largest cranidium referred to this form, 10 mm.; width at front of cephalon, 10 mm.; width at rear of cranidium, 15 mm.; length of glabella, 6 mm.; width of olabella at rear, 6 mm. Horizon, beds 86 and 91, in the upper part of the hicksi zone. Rare. The holotype is specimen 8023 in the paleontological collections of Princeton University. It is a small cranidium, being only 7 mm. in length, including the spine. It is chosen as the holotype because it is the only specimen now 93 PARADOXIDES SECTION 93 in hand in which the spine is well shown. It is from bed 8&6. Other specimens referred to this form are numbered 8034, 8035, and 8036. The measurements given above for the “largest cranidium” were taken from specimen 8035. The specific name refers to the spine, which distin- guishes the variety from Liostracus globiceps. This form appears to be certainly a variety of Liostra- cus globiceps Grénwall (1902, pp. 145, 146, 218, pl. IV, figs. 12a, 12b).. It seems to differ from that form only in the possession of a large neck spine. Gronwall describes and figures only a little spine on the neck ring of globiceps. Grénwall’s cephala seem to have been more convex than the specimens from Manuels, but, as the Manuels specimens are in shale, and probably somewhat flattened, it is not likely that the two forms differed in convexity in life. Gronwa!! records globiceps from Bornholm Island, Denmark, prob- ably from the davidis zone. 94 BULLETIN 43 : 94 EXPLANATION OF PLATE 3 NEW TRILOBITES FROM THE CAMBRIAN PARADOXIDES’ BEDS AT MANUELS, NEWFOUNDLAND PAGE Fic. 1. Agnostus clarx, n. sp. Holotype pygidium. 4/1. Bed 23, in the upper part of the Paradoxides bennetti zone. Princeton catalog number 8000 .. . 74 Fic. 2. Agnostus barlowi Belt, var. definitus, n. var. Halaeane pygidium. 4/1. Bed 23, in the upper part of the Para- doxides bennetti zone. Princeton catalog number 8001 76 Fic. 3. Agnostus barlowi Belt, var. definitus, n. var. Cranidium referred to this form. 4/1. Bed 23, in the upper part of the Paradoxides bennetti zone. Princeton cate Pe enumbers S00 3s seee ae 76 Fic. 4. Agnostus parvifrons Limmnaessorn., var. =, jannacie n. var. Holotype pygidium, top view. 4/1. Bed 101, in the lower part of the Paradoxides davidis zone. Prince- ton catalog number 8004... . 78 Fig. 5. Agnostus parvifrons Linnarsson, var. punctifer, n. var. Side view of the pyei dun whose top view is shown in Fig. 4. 4/1 Oe 78 Fic. 6. Agnostus vaningeni, n. sp. Einloae: ayeitien. 4/1. Bed 99, in the lower part of the Paradoxides davidis zone. Princeton catalog number 8011 .. . 80 Fic. 7. Agnostus longifrons Nicholas, var. parvulus, n. var. Holotype cranidium. 4/1. Bed 109, in the middle part of the Paradoxides davidis zone. Princeton catalog number SO14 oo ee Salis neg ete eh oe Fic. 8. Agnostus longifrons Nicholas, var. parvulus, n. var. Pygidium referred to this form. 4/1. Middle part of the Paradoxides davidis zone. Princeton catalog number 8017 .. .. 83 Fic. 9. Agnostus rectangularis, n. Sp. Holom pe each 4/1. Upper part of the Paradoxides davidis zone. Prince- ton cataloc number S019 > 3 5). eu Fic. 10. Conocoryphe bullata, n. sp. Holotype eranidium. 3/2. Bed 19, in the middle part of the Paradoxides bennetti zone. Princeton catalog number 8021 ... . 87 Fic. 11. Conocoryphe bullata, n. sp. Cranidium referred to thi species. 2/1. Bed 19, in the middle part of the Para- doxides bennetti zone. Princeton catalog number 8026 87 Fic. 12. Paradoxides parvoculus, n. sp. Holotype cranidium. 1/1. Bed 23, in the upper part of the Paradoxides bennetti zone. Princeton eatalog number 8027 .. . 89 Fic. 13. Paradoxides parvoculus, n. sp. Cranidium. 1/1. Wipeer part of the Paradoxides bennetti zone. Princeton cat- alog number 8041 .. . rs SG SS Fic. 14. Liostracus globiceps Ciaveill, var. jaculaton n. var. Holotype cranidium. 2/1. Bed 86, in the upper part of the Paradoxides hicksi zone. Princeton catalog number 8023 = 4.) 6.9 2 eS Norms iRings BuLL. AMER. PAL., ee3) VOL lr, at te, bak BS! Peels ty: 95 PARADOXIDES SECTION 95 VII. CORRELATION Very few of the species of the Paradoxides faunas occur in any of the other known Cambrian faunas of the world; and none of those species are sufficiently diagnostic to enable us to use them with any confidence in close corre- lation. Therefore, although the Paradoxides beds and their fossils have been long known and much studied by geolo- gists and paleontologists, and have proved so easy to corre- late among themselves that it has been possible to make close comparisons of beds on opposite sides of the Atlantic, yet their exact age relationships with the other known faunas that are supposed to have existed in the world at about the same time are not well understood. For many years after their discovery the Paradoxides faunas were considered to be the oldest well developed faunas known. They were spoken of as the “first fauna” or “primordial fauna.” (Barrande, 1852; Murchison, 1872; Logan, 1866, in Murray and Howley, 1881, p. 49; Dawson, 1868, p. 638; etc.). Later when the term “Cambrian” came into general use, they were referred to as “Lower Cam- brian” (Walcott, 1884). Finally, when in 1888, it was dis- covered that the ‘“‘Olenellus” faunas really belonged below and not above them, the Paradoxides faunas were assigned to their present position in the Middle Cambrian (Walcott, 1888b). In 1914 Professor vanInven correlated te Paradoxides beds of Conception and Trinity bays, Newfoundland, includ- ing those at Manuels, with the “Acadian” division of the “St. John Group” of New Brunswick and the “Menevian and Solva” of Great Britain (vanIngen, 1914b). In the present unsettled condition of Cambiian nomenclature it seems wisest to refer to the beds of the Manuels series merely as “Cambrian Paradoxides beds,” without using the qualifying “Middle.”’ That the Paradoxides beds belong roughly in the middle, or what is now generally called the “Acadian,” portion of what we now know as the Cambrian 96 BULLETIN 43 - 96 is possible, but they cannot be correlated definitely with any part of the Cambrian of the southern, central, or western parts of North America, or of Asia or Australia, and the base and, more especially, the summit of the Cambrian sys- tem are at present so undefined, and may yet be so shifted, that it seems quite possible that the Paradoxides beds may at some time in the future be called something quite differ- ent from “Middle Cambrian.” Indeed, Professor Charles Schuchert has proposed raising the “Middle Cambrian, Acadian, or Paradoxides epoch” to the rank of a period, the “Acadic Period” (Schuchert, 1910, pp. 520-522). In his great work, ‘Revision of the Paleozoic Systems,” Dr. E. O. Ulrich accepts the Middle Cambrian age of the Paradoxides beds, but objects to the assignment of the over- lying “Olenus” beds to the Upper Cambrian, stating reasons for his belief that they were formed much later, probably in “Canadian” time (Ulrich, 1911, pp. 621, 678-680). Whether the Paradoxides beds should be considered to include all of the Middle Cambrian of the Manuels Brook section is another question. The lowest of the overlying Agnostus pisiformis beds are certainly not Middle Cam- brian in the sense in which that term is usually used today, and the same can be said of the underlying beds containing Callavia broggeri. The age of the black shales which inter- vene between the highest known Paradoxides beds and the lowest known Agnostus pisiformis horizon must remain in doubt until more is known about their fossils. The same is true of the beds lying between the lowest known Para- doxides beds and the highest beds in which Catadoxides magnificus has been found (see discussion of this question in Chapter IV, pp. 26, 27 ). The beds between the highest known Callavia horizon and the uppermost stratum in which Catadoxides magnificus is known to occur have been assigned to the Middle Cambrian by Dr. Walcott (1900a, pp. 315, 316; 1912, p. 140), and to the Lower Cambrian by Professor vanIngen (1914b). The Paradoxides faunas are so distinctive and char- : Massachu-: 2 : Zones :setts and: New : Cape : Manuels :? Vermont:Brunswick: Breton : Great :Scandi- :Britain :;navia 2 ee os oe Paradox- : ides : forch-. hammeri zone Paradox- : : nese” {: : : davidis : zone : ee ry Paradox- ides : hicksi : zone ; Paradox- : GCE bennetti zone TABLE V. Correlation of the Paradoxides faunas found at Manuels with those found in northwestern Europe and northeastern North America 97 PARADOXIDES SECTION 97 acteristic that they are generally easy to recognize and cor- relate with each other. Salter (1859) and Jackson (1859a) compared the first specimens of Paradoxides discovered in Newfoundland (examples of Paradoxides bennetti) with Paradoxides spinosus of Bohemia and Paradoxides harlani of Massachusetts. Billings, who studied the collections made by the Geological Survey of Newfoundland, correlated the beds containing Paradoxides faunas there with “the Lower Lingula flags of Great Britain, or the Menevian Group of Salter and Hicks” (Billings, 1872, p. 470; 1874, p. 69) ; and Dawson (1873, p. 5) correlated them with the “Acadian Group” of New Brunswick, “the gold-bearing rocks of Nova Scotia,” “the slates of Braintree in Massachusetts,” and the “Menevian of Salter, Etage D of Barrande.” In his report of the progress of the Newfoundland Survey for 1868, Murray referred the Paradoxides beds at Manuels to the “Lower Silurian,” although he had found no fossils in them. Weston, who in 1874 made the first discovery of fossils at Manuels, correlated the beds containing them with the Para- doxides beds of New Brunswick on the evidence of the first specimen he found (Weston, 1898, p. 153). The New Brunswick beds had been correlated in 1865 by C. F. Hartt and Dr. G. F. Matthew with the Paradoxides beds of Massa- chusetts, the Paradoxides bennetti beds of St. Mary’s Bay, Newfoundland, the Lingula flags of Great Britain, the “alum-schists” of Scandinavia, and “Etage C of Barrande in Bohemia” (Matt., 1865, p. 427). In 1878 Whiteaves also correlated the Manuels Paradoxides beds with the “St. John’s group” of New Brunswick (Whiteaves, 1878). In 1881 Etheridge correlated the Menevian of Wales with “the St. John’s group in Newfoundland” (Etheridge, 1881, p. 68). In 1884 Dr. C. D. Walcott (1884, p. 13) correlated these beds with the New Brunswick beds and with “the lower part of the Menevian, or possibly with portions of the Harlech and Longmynd group” of Great Britain. He pointed out the fact that the species then known to occur in the Paradoxides harlani beds of Braintree, Massachu- 98 BULLETIN 43 -_ @8 setts, all appeared to be represented by very similar forms in Newfoundland (1884, p. 44). None of the Newfound- land species that he mentioned as comparable with the Braintree forms were then known to occur at Manuels, but most, or all, of them have since been found there. In 1885 Dr. Matthew stated that the fossils then known from the Paradoxides beds of Manuels Brook appeared to be of early Paradoxidian age, like those of the “Acadian” of St. John, New Brunswick, and the Solva of Great Britain (Matthew, 1885, pp. 121 and footnote, and 122). In the following year he referred the Manuels fossils to the “Hori- zon of the Conocoryphine,” the lowest Paradoxides horizon then known in Newfoundland (Matthew, 1886, p. 149) ; and two years later he correlated the “Shales of Manuel R.” with “Division” 1c of the ‘St. John Group” of Canada, the upper part of the “Upper Sparagmite formation—Etage 1b and c’” of Norway, the upper part of the “Lower Paradoxi- des Beds” of Sweden, and doubtfully with the upper part of the “Solva group” of Great Britain (Matthew, 1888a, 10. 2/5) 6 In 1888 and 1891 Dr. Walcott (1888b, p. 551; 1889a, pp. 383, 386; 1891a, pp. 582, 583; 1891b, pp. 66, 360; 1891c) correlated the Paradoxides beds at Manuels with those of New Brunswick, Massachusetts, and Great Britain. In 1890 Dr. Matthew (1890, p. 187) correlated the upper Paradoxides beds at Manuels with the Menevian of Wales, “Etage 1d” of Norway, and the “Upper Paradoxides Beds” of Sweden; and the two lower “zones” of the Para- doxides beds at Manuels with the ‘Lower Paradoxides Beds” of Sweden, a part of the “Upper Sparagmite-Htage 1c” of Norway, and a part of the Solva of Great Britain. In the following year he correlated the Newfoundland faunas with those in other parts of the world, as shown in table I] (page 15 of the present paper). In 1896 he divided them into the following subzones: 99 PARADOXIDES SECTION 99 3. “Subzone of P. davidis.” 2. “Subzone of P. abenacus” (doubtfully identified in Newfoundland). . “Subzone of P. eteminicus.”’ and eS these subzones with what he considered to be their equivalents in Massachusetts, New Brunswick, France, Spain, Bohemia, Sweden, Norway, and Wales (Mat- thew, 1896, pp. 193, 194). In 1897 Frech correlated the “Zone mit P. Davidis und Ptych. Linnarsoni Brogeg.” at Manuels with the ‘Middle Menevian mit Par. davidis” of Wales, the zone of “Par. Davidis (Microdiscus u. Par. Tessini)” of Sweden, and the “Schiefer m. Par. rugulosus, Con. Sulzeri, Agnostus nudus”’ of Norway; and correlated the “Schichten mit Paradoxides Hicksi” at Manuels with the “Lower Menevian mit Par. Hicksi und aurora” of Wales, the zone of ‘Par. Tessini (u. Par. Hicksi, Ellips. granulatus, Conoc. exsulans, Ptych. Lin- narssoni, Agnostus rex)” of Sweden, the “Schiefer mit Par. Tessini’” of Norway, and the “Sandstein mit Par. Tessini” of Sandomir, Poland (Frech, 1897, pp. 34-38). In 1899 Dr. Matthew (1899c, p. 67) subdivided the Paradoxides beds of Newfoundland and correlated them with beds in New Brunswick, as shown below (the notes in parenthesis refer to New Brunswick) : Horizon of Paradoxides davidis (unknown in New Brunswick). Horizon of Paradoxides tessini (—subzone of Paradoxides abenacus). Horizon of Paradoxides spinosus (—subzone of Paradoxides eteminicus). Horizon of the Conocoryphine (=—subzone of Paradoxides eteminicus). In 1914 Professor vanIngen correlated the Manuels series with the “Acadian” of “N. H. America” and the Menevian and Solva of “Western Europe,” as mentioned above (vanIngen, 1914). A more detailed correlation table, drawn up by the writer, is shown in table V . 100 BULLETIN 43 _ 1Ge As far as can be told at the present time, the Para- doxides faunas of southeastern Newfoundland are more nearly related to those of New Brunswick, Massachusetts, Great Britain and Scandinavia, than to those of France, Bohemia, Poland and southern Europe. The fossils of the Paradoxides beds of Vermont will probably also prove to be closely related to those of Newfoundland, but they are at present too little known to make a satisfactory comparison possible (see pagero6). The succession of early Paleozoic faunas in southeast- ern Newfoundland, from the Lower Cambrian to the Lower Ordovician, appears to have been very much the same as it was in Massachusetts, New Brunswick, Great Britain, and Scandinavia, as is shown in table VII (p.1ro1). Recent dis- — coveries in Shropshire and Warwickshire, England, also indicate the presence there of “Callavia’” and “Protolenus” faunas very similar to those of southeastern Newfoundland (Cobbold, 1910; 1919; 1921; Illing, 1913; 1914a, 1916). One of the most interesting things about the Paradoxi- des fossils at Manuels is their very close resemblance to the contemporary fossils of Europe, in general, and of Great Britain, in particular. The number of identical or nearly identical species found at Manuels and in Europe, and the similarity of their ranges on the two sides of the Atlantic, are illustrated in table VI and on pages 117-132. What we know of the species occurring above and below the Para- doxides beds at Manuels, including those found in the beds of Kelly’s, Little Bell, and Great Bell Islands, in Conception Bay, indicates that they, too, are closely related to forms of the same age in Europe. Altogether, they and the fos- sils of the Paradoxides beds exhibit one of the most interest- ing examples known of a succession of very similar faunas on opposite sides of the Atlantic Ocean. The Paradoxides faunas of Scandinavia seem also to be very similar to those of Manuels, but they do not seem to show quite so close a resemblance as do those of Great Brit- ain. They are compared on pages 125-132. Interesting ° ee vy Tvs . 4 a . . =) a) . =] a . + Moe “J » a . . 7] ae ae soles eee qe se co se Eire .cwlev + , ‘ Se ae fe Gaetan BD ee q * mm weoiveiarter ai ete we dae . Sato cay, s —$—$— $$ $< $ $$ } BAT BRITAIN : SCANDINAVIA aunes, with Didy- :"Arenig” feunes, with Didy- 8,Tetragraptus,etc. ;mograntus,Tetregraptus,etc. —_- : ond Upper Lingula :"Sremadoc” faunas, with ; as, with Niobe, sHaratopyge, Shumardia,Parabolina, »Bryograptue,Parab- :and "Dictyograptus” @ Orusia lenticular-:{Dictyonema). — ee yacterized by Ag= sFaunsa characterised by Agnos= siformis obesus situs pisifomis obesus and es of Olenus. :specias oF Olervse be Jacking in the : in some districts : Agnostus pisifcrms. ) Britain, but part 3; llariy some cf the g ldes forohharmeri : iy prove tc ba s_Agnostus leevigstus. ‘lin other districts. °; sParadexides forchhammert fauna. | Gis net I eee \des davicig fauna. :; Paradoxides davidis fama. B des hichsi feune. s Paradoxides hicksi fenna. . — ee re ere \des groomi fauna. \loxides Glandievs fauna. Erte ER bbs] op ee 2&8 of ne) te} | eee = } to be present. Pos= resented in the mastone” of Comley jot known to be present. D me os Cf GO am ae | :Mot known to bo present 2Feuna or faunas characterized :by Holmia and Mesonacis. re net See ower Ordovician faunas and Scandinevia. { ; 4 \ } t wi cea SOUTHEASTERN NEWFOUNDLAND MASSACHUSETTS : NEW BRUNSWICK _ 3 GREAT BRITAIN : SCANDINAVIA mei = =SO*S*~—*—*~—<“ O”*:i OO = : 2Fauna or faunas characterized: Arenig f : mien : ; : nig faunes, with Didy- :"Arenig" feunas, with Did "grenig" faunas, ; zby Didymograptus and Tetra- : mograptus,Tetregraptus,etc. :m tus, p 4 Didymograptus » Sues : sgraptus (C 3 d). 3 eee” seer wor cera Fie ee rcs ee eee ee a = == See : > : ; : » games, with Niobe, 2 + :Famsas characterized by :Premadoc and Upper Lingula :"Sremadvc" faunes, with ST eGeaCl a Bry : a sCtenopyge, Perabolina,Bryo- :Flag faunas, with Niobe, of eratopyge , Shumardia Parabolina, parabo and Orusie ienticuleris. ; @ seraptue, and Dictyonem :Shumardia, Rryograj sand "Dictyograptus" 5 : ' graptus, s a 3(C Zaq-c}. colina, am Orusia La>-:{Dietyonema), 3 é 8 sis. 3 a Be ae ———---___ 7 7 2 a :{Probably represented in sFaun® characterized by Ag- : no ehee g racterized by Agnostua y repre racterized by Ag sFaunsa characterised b nO0g= 1 ae obeave and specics : 2 jpart, at least, by the snostus pisiformis obesns stus pistfoxmis aNsauevam Rta = 2558 re i/favuna or famas chearacter- sand species o? Olenus. s6pecies oF Olenvsa. of Olenus. z 8 2)ized by Acentholenus 2(Known to be jacking in the : zz = siformis. $ a 77 Qnd varieties of Agnostus s\sections in some districts +: Agnostus pisifcrma. Agnostus pi 2 apisiformis (C2). sJin Great Britain, but part : 2 e 6 3 Se # (particularly some cf the g a apes ae : = x $ Be radeeides fo rehhamnmeri g leevi atins ? 8 ¢ No nown to be present. sf Tauna) may prove to be Q Agnostus Lleevige’ : =: ee = : *\preaent in other districts. : forchhammeri fauns? : s sNot certainly known to be 8 sParadoxides Paradoxides = spresent. Possibly represented: B ; sin Cape oe —_—_——— Paradexides dayicuis fauna. Peradoxides hichsi feune. Paradoxides crooml fauma, : Paratoxides hicksi fauna. joxidee Olandicus fauna. sNot known to be presenti. Pog=;Not known to be present. :Sibly ren ented in the :"bilack limestone” of Comley :Quarry, Suropshire. sProtoicnus faunas :Callevia fama. Paredoxides davidis fauna. é a tNot certainly known to be 8 + 4 spresent. 3 — — lS eae 42 : g Paradoxides bioksi fauna, A S :Paradoxides abenecus fauna B 3 3{Cla). 8 So reser = ee eee £ Paradoxides bennetti taune. ¢ Paredoxides harlani fama- :Peradcxides regina f: 5 8 sP. lamellatus end P. 8 = 3 oteminicus {Gle1.2). : Catadoxides fauna, 8 3Not know to be pragent. ° 3 ° x 2 3 i do Z 8 & 0 3 a oo 3 ° ee Protolenus fauna. ; Se sProtolenvs fauna. Se a en = sae — rece Callavia fama. :Callavis feuns. :Cailavia fauna. 2 ere SE Se 3 ° ° 3 e G + known to bo present 2Fauna or faunas characterized :by Holmia and Mesonacis. fable VI]. Table to show the similarity in the succession of Cambrian, Ozarkicn, snd Lower Ordovician faunas in southeastern Hewfoundland, Massachusetts, New Brunswick Great Britein, and Scandinavia. a ies vee te os § a s jue nap Dm p Ci 418 OF 3 yt i — : i tuft, 38 MHown * t4 ben v rent le igh om ited STEVEN Bob : a Agdcadh ee ae jie rad yadoxR S ansst eivalrepcsiiavis fawne., ete ee Bie ee % Se ‘te ot asad ate = tereptus a) he ~~ bh ne ae % “2, , ox eee eae aim ” er Portal a ee — mea fo ee ee gt «9 A bbe oe Perera: res Cape BRS RO Q——earaei ear eS poe ena = ; pre gent Che anemia rh analete = ar ed agate “; : + Sy: : = ‘ibia) redo? ra * at oS 1 qa os wee ~- vaoeacktais apres ee - oe vs rae isl Bhi ps aah have) ee be fe sist sebh i Prot olents Pens. C ~~ Bet | cep te pcm aan a . a en i Sel IOI PARADOXIDES SECTION IOI correlations might also be made between the Manuels faunas and those of Bohemia, Poland, France, and the Iberian Peninsula; but, as they can be made with much greater accuracy after the Paradoxides faunas of all of southeastern Newfoundland have been more thoroughly studied, they are not attempted in this paper. The great similarity between the Paradoxides faunas of North America and Europe naturally raises the question of migration between the two regions. Did any or all of the faunas originate in one of these regions and migrate to the other; or did they come in from some outside source? If they came from outside, did they occupy one of the two regions for any appreciable period before advancing to the other; or did they invade both regions quite, or nearly quite, simultaneously? These are difficult questions, most of which probably cannot be satisfactorily answered with the infor- mation now at hand. In the regions where the Paradoxides faunas and those which preceded them have been most care- fully and thoroughly studied, none has been found which can be said to have been the immediate ancestor of the earliest known Paradoxides assemblage. This indicates either that the first known Paradoxides faunas entered those regions from outside, or that, if they evolved in those regions, conditions at the time of their evolution and since have destroyed all records of their earlier stages. There are unconformities between the Paradoxides beds and the beds underlying them in some regions, but there is certainly no angular unconformity at the bottom of the Paradoxides beds at Manuels, and, as has been stated before, it is doubtful whether there is any well marked disconformity present. The first Paradoxides faunas seem to have come in with an advancing sea in most of the regions in which they have been found, and appear to have been followed by successive invasions of later and slightly different forms; for while the changes from one fauna to the next succeeding one are usually gradual, in the sense that old species slowly drop out as new ones come in and that some stems gradually 102 BULLETIN 43 ; = ez change in character, nevertheless, many entirely new types enter which could have had no antecedents in this region; so that there must certainly have been open an outside source of new life during most of known Paradoxidian time. In almost every section where any two of the three sets of faunas, the “Olenellus,”’ ‘Paradoxides,” and “Olenus,’’ have been reported in the regions where Paradoxides is known to occur (Europe and northeastern North America) there are present either the “Olenellus” and ‘‘Paradoxides,”’ “Olenellus,” ‘““Paradoxides” and “Olenus,” or ““Paradoxides” and “Olenus” faunas—very seldom the “Olenellus’ and “Olenus” faunas together without the “‘Paradoxides.” Probably the following tenative deductions are the ones which can most logically be drawn from the facts now at hand: 1. No fauna immediately ancestral to the earliest known Paradoxides faunas has been found. The nearest ancestral faunas now known are probably those characterized by Catadoxides in south- eastern Newfoundland and by Metadoxides in Sardinia. 2. The earliest known Paradoxides faunas probably invaded the regions where their remains are now found from some outside area, and the new types which distinguish the succeeding faunas also came in from the same or some other outside place or places, those after the earliest one being composed of a combination of these new types and sur- vivals and derivatives of the fauna or faunas which preceded them. Some or all of the ances- tors of the Paradoxides faunas may have devel- oped in the regions where Paradoxides is now found, but, if they did, no record of their presence there has been discovered. 3. The migrations of the Paradoxides faunas over the European and northeastern North American 103 PARALOXIDES SECTION 103 regions appear to have been mainly in the nature of direct invasions from some outside region. 4. There is a great faunal break between the latest Paradoxides fauna and the next succeeding well- developed one at every known place where they occur. The similarities and differences betwen the Para- doxides faunas found at Manuels and other regions are roughly indicated in tables IV and V, and are briefly described under separate headings in the remaining part of this chapter. COMPARISON OF THE PARADOXIDES FAUNAS OF MANUELS WITH THOSE OF BRAINTREE, MASS. The Braintree fauna and the fauna of the lower 21 feet and 7 inches (beds 1-13) of the bennetti zone at Manuels, so far as they are known at the present time, are essentially alike. The two faunas are listed below, species which are known to be identical or very similar being placed opposite each other. The Braintree list is taken partly from Wal- cott (1884) and Grabau (1900). The Manuels list, like all the other lists of fossils from Manuels in this chapter, includes only species which the author has found at that locality. The numbers in parenthesis following the names of the species in the Manuels list indicate the particular beds of the lower 21 feet and 7 inches of the bennetti zone in which each species has been found. BULLETIN 43 104 104 ‘98V SUVS OY} Ap}OVXO 10 SOW)? Jo 9q A[UTY}10D JSNU OMY 9Sey} SB oes OY} A[IBEU OS SVUNVJ Jey} Woos p[NOM 4] (I) ‘epun ‘ds pue ‘ues osuodg neqeiy sisuepreMmAvy (2) soyypoAH qoojeM Ltopeys seyzoAP (OT) gf Peq UL punoy culo dy} Wort YUoTOYIp “opun ‘oeds pue ‘ues pryy[oAH (

of Chester County. 2nd Geol. Survey of Pennsylvania. Repor* C4, the geology of Chester County, edited by J. P. Lesley. pp. 215-345, map. Harrisburg, 1883. Frech, F. 1897. Lethza geognostica, theil 1, band 2, lieferung 1. Stuttgart, 1897. Grabau, A. W. 1900. Paleontology of the Cambrian Terfanes of the Boston Basin. Occasional Papers of the Boston Soc. of Nat. Hist., IV, pp. 601-694, 1900. Groénwall, K. A. 1902. Bornholms Paradoxideslag og deres fauna. - Geol. Undersogelse, II Rekke, No. 13, Copenhagen, Hawle, I., and Corda, A. J. C. 1847. Prodrom einer Monographie der boéhmischen Trilobiten. Prague. 1847. 136 BULLETIN 43 136 Hayes, Albert O. 1914. Geology of the Wabana Iron Ore of New- foundlend. Bull. Geol. Soc. America, vol. 25, 1914, p. 74. Hayes, A. O. 1915. Wabana Iron Ore of Newfoundland. Geol. Surv. Canada, Memoir 78, No. 66, Geological Series; 1915. Hicks, H. 1872. On some undescribed fossils from the Menevian group. Quart, Jour. Geol. Soc. London, vol. 28, 1872, pp. 173-185. Howell, B. F. 1920a. Correlation of the Middle Cambrian of New- foundland and Great Britain. Bull. Geol. Soc. Amer., vol. 31, No. 1, 1920, p. 214. Howell, B. F. 1920b. The Middle Cambrian beds at Manuels, New- foundland, and their relations. Science, New Series, vol. 51, No. 1330, 1920, p. 644. Howley, J.P. 1889. The Taconic of Eastern Newfoundland. Amer- ican Geologist. vol. 4, 1889, pp. 121-125. Howley, J. P. 1907. Geological map of Newfoundland compiled from the most recent and authentic sources, the coast line cor- rected to date from the admiralty surveys, the interior chiefly from the work of the geological survey, by James P. Howley, F. G. S. Director. 1907. Illing, V. C. 19138. Recent discoveries in the Stockingford Shales near Nuneaton. Geol. Mag., dec. 5, vol. 10, 1913, pp. 452, 453. Illing, V. C. 1914a. Recent discoveries in the Stockingford Shales near Nuneaton. Report of the British Ass’n for the Ady. of Science, 19138 (Birmingham) meeting, 1914, pp. 498, 499. Illing, V. C. 1914b. The Par-doxidian fauna of a part of the Stockingford shales. Geol. Mag., dec. 6, vol. 1, no. 8, 1914, pp. 379. 380. Illing, V. C. 1916. The Paradoxidian fauna of a part of the Stock- ingford shales. Quart. Jour. Geol. Soc. London, vol. 71 (for 1915). 1916. pp. 386-450. Jackson, C. T. 1859a. [On a Paradoxides from St. Mary’s Bay, Newfoundland.] Proc. Boston Soc. Nat. Hist., vol. 7, 1859. p. 54. Jackson, C. T. 1859b. [Comparison of specimens of Paradoxides from Braintree, Massachusetts. and St. Mary’s Bay, Newfound- land.] Proc. Boston Soc. Nat. Hist.. vol. 7, 1859, p. 75. Jekel, O. 1909. Ueber die Agnostiden. Zeitschr. der. Deutsch. geol. Gesell.. vol. 61. 1909. pp. 380-401. Jukes, J. B. 1839. Report on the geology of Newfoundland. St. John’s. Newfoundland, 1839. Jukes, J. B. 18402. Report on the geology of Newfoundland. Edin- burg Phil. Jon~., New Series. vol. 29. 1840, pp. 103-111. Jukes. J. B. 1849%b. Renort of the progress of the geological sur- vey during 1840. St. John’s Newfoundland. 1840. Jukes. J. B. 1842. Exeursions in and about Newfoundland during the years 1829 and 1840. Two volumes. London. 1842. Jukes. J..B. 1848. General report of the geological survey of New- foundland, executed under the direction of the government and legislature of the colony during the years 1839 and 1840. Lon- don. 1843. Keith, A. 1923. Cambrian succession of northwestern Vermont. American Journal of Science. 5th series. vol. 5. 1928. pp. 97-139. Lake. Philip. 1906. A monograph of the British Cambrian Trilo- bites. part I: 1906. 137 PARADOXIDES SECTION 127 Lapworth, C. 1891. On Olenellus callavei and its geological rela- tionships. Geol. Mag., dec. 3, vol. 7, No. 12, 1891, pp. 529-536. Linnarsson, J. G. O. 1869. Om Vestergothlands Cambriska och Siluriska aflagringar. Kongl. Svenska Vetenskapsakademiens handlingar. Band. 8, No. 2, 1869. Logan, W. E. 1865. Geological Survey of Canada. Report of prog- ress from its commencement to 1863, atlas, 1865. Marcou, J. 1889. The Taconic of Eastern Newfoundland. Ameri- can Geologist, vol. 4, 1889, pp. 121-125. Marcou, J. 1890a. The Lower and Middle Taconic of Europe and North America, I. American Geologist, vol. 5, 1890, pp. 356-375. Marcou, J. 1890b. The Lower and Middle Taconic of Europe and North America, II. American Geologist, vol. 6, 1890, pp. 78-102, 221-233. Matthew, G. F. 1865. On the Azoic and Paleozoic Rocks of South- ern New Brunswick. Quart. Jour. Geol. Soc. London, vol. 21, 1865, pp. 422-434. Matthew, G. F. 1885. Illustrations of the fauna of the St. John group continued: on the Conocoryphea, with further remarks on Paradoxides. Proc. and Trans. Roy. Soc. Canada for 1884, vol. 2, sec. 4, 1885, pp. 99-124. Matthew, G. F. 1886a. Illustrations of the fauna of the St. John group continued, No. 3—Descriptions of new genera and species (including a description of a new species of Solenopleura by J. F. Whiteaves). Proc. and Trans. Roy. Soc. Canada for 1885, 1st ser., vol. 3, sec. 4, 1886, pp. 29-84. Matthew, G. F. 1886b. Abstract of a paper on the Cambrian faunas of Cape Breton and Newfoundland. Canadian Record of Science, vol. 2, 1886, no. 4, pp. 255-258. Matthew, G. F. 1887. On the Cambrian faunas of Cape Breton and Newfoundland. Proc. and Trans. of the Roy. Soc. of Canada for 1886, vol. 4, sec. 4, 1887, pp. 147-157. Matthew, G. F. 1888a. On a basal series of Cambrian rocks in Acadia. Canadian Record of Science, vol. 3, 1888, pp. 21-29. Matthew, G. F. 1888b. On Psammichnites and the early trilobites of the Cambrian rocks in eastern Canada. American Geologist, vol. 2, 1888, pp. 1-9. Matthew, G. F. 1888c. Illustrations of the fauna of the St. John Group. No. 4—Part 1, Description of a New Species of Para- doxides (Paradoxides regina). Part 2. The smaller trilobites with eyes (Ptychoparide and Ellipsocephalide). Proc. and Trans. 28 Soe. Canada for 1887, vol. 5, sec. 4, 1888, pp. 115-166, pls. Matthew, G. F. 1888d. On the classification of the Cambrian rocks in Acadia. Canadian Record of Science, vol. 3, 1888, pp. 71-81. Matthew, G. F. 1890. On Cambrian Organisms in Acadia. Proc. and Trans. Roy. Soc. Can. for 1889, vol. 7, sec. 4, 1890, pp. 135-162. Matthew, G. F. 1891. On some causes which may have influenced the spread of the Cambrian faunas. Can. Ree. Sci., vol. 4, 1891, pp. 255-269. Matthew, G. F. 1896. Faunas of the Paradoxides beds in eastern North America, No. 1. Trans. N. Y. Acad. Sci., vol. 15, 1896, pp. 192-247. 138 BULLETIN 43 138 Matthew, G. F. 1899a. A Paleozoic terrane beneath the Cambrian. Annals N. Y. Acad. Sci., vol. 12, No. 2, 1899, pp. 41-56. Matthew, G. F. 1899b. A new Cambrian trilobite. Bull. Nat. Hist. Soc. New Brunswick, No. 17, 1899, pp. 136-142. Matthew, G. F. 1899c. Studies on Cambrian faunas, No. 4—Frag- ments of the Cambrian faunas of Newfoundland. Proc. and Trans. Roy. Soc. Canada for 1899, 2nd ser., vol. 5, sec. 4, no. 3, 1899, pp. 67-95. Murchison, R. I. 1872. Siluria, 5th edition, London, 1872. Murray, Alexander, and Howley, James P. 1881a. Geological Sur- vey of Newfoundland. London, 1881. [A re-publication of the Reports of Progress of the Geological Survey of Newfoundland for the years 1864 to 1879.] Murray, Alexander, and Howley, James P. 1881b. Map of the Peninsula of Avalon, showing distribution of formations. Scale, 4 miles to 1 inch. 1881. Nicholas, T. C. 1914. The trilobite fauna of the Middle Cambrian of the St. Tudwal’s Peninsula (Carnarvonshire). Geol. Mag., dec. 6, vol. 1, no. 8, 1914, p. 380. Nicholas, T. C. 1915. The geology of the St. Tudwal’s Peninsula (Carnarvonshire). Quart. Jour. Geol. Soc. London, vol. 71, 1915, pp. 83-1438. Nicholas, T.C. 1916. Notes on the Trilobite fauna of the Middle Cambrian of the St. Tudwal’s Peninsula (Carnarvonshire). Quart. Jour. Geol. Soc. London, vol. 71, 1916, pp. 451-472. Peach, B. N. 1913. The relation between the Cambrian faunas of ’ Seotland and North America. Report of the eighty-second meet- ing of the British Association for the Advancement of Science, Dundee, 1912, pp. 448-459; 1913. Perkins, G. H. 1908. Preliminary Report on the Geology of Frank- lin County. Report of the State Geologist on the Mineral Indus- tries and Geology of Certain Areas of Vermont, 1907-1908; pp. 189-209; 1908. Raymond, P. E. 1913. Some changes in the names of genera of trilo- bites. Reprint from the Ottawa Naturalist, vol. 26, 1913, pp. 137-142. Raymond, P. E. 1914. Notes on the ontogeny of Paradoxides, with the description of a new species from Braintree. Mass. Bull. Mus Comp. Zool., Harvard, vol. 58, No. 4, 1914. pp. 225-244, 1 pl. Raymond, P. E. 1917. Beecher’s classification of trilobites after twenty years. Amer. Jour. Sci., 4th ser., vol. 438, 1917, pp. 196-210. Raymond, P. E. 1920. The pygidium of the trilobite. Geol. Mag., vol. 57, 1920, pp. 22-25. Rogers, W. B. 1859. [Comparison of Paradoxides bennetti with Paradoxides harlani.] Proc. Boston Soc. Nat. Hist., vol. 7, 1859, p. 86. Salter, J. W. 1859. On the fossils of the Lingula-flags or “Zone Primordiale.” Paradoxides and Conocephalus from North Amer- ica. Quart. Jour. Geol. Soc. London, vol. 15, 1859, pp. 551-555. Schuchert, C. 1910. Paleogeography of North America. Bulletin of the Geological Society of America, vol. 20, 1910, pp. 427-606. pls. 46-101. Tullberg, S. A. 1880. Om Agnostus-arterna de kambriska aflagrin- 139 PARADOXIDES SECTION 139 garne vid Andrarum. Sveriges geol. undersdkning, Ser. C, No. 42, 1880. Ulrich, E. O. 1911. Revision of the Paleozoic Systems. Bull. Geol. Soc. Amer., vol. 22, 1911, pp. 281-680. vanHise, C. R., and Leith, C. K. 1909. Pre-Cambrian geology of North America. U. S. Geol. Surv., Bull. No. 360, 1909. vaniIngen, Gilbert. 1914a. Cambrian and Ordovician faunas of southeastern Newfoundland. Bull. Geol. Soc. Amer., vol. 25, 1914, p. 138. vanIngen, Gilbert. 1914b. Table of the Geological Formations of the Cambrian and Ordovician Systems about Conception and Trinity Bays, Newfoundland, and their Northeastern-American and Western-European Equivalents, based upon the 1912-1913 field work. Princeton, 1914. yonToll, E. 1899. Beitrage zur Kenntniss des Sibirischen Cambrium. I. Memoires de l’Academie Imperiale des Sciences de St. Peters- burg, 8th Series, vol. 8, No. 10 et dernier, 1899, pp. I-IV, 1-57. vonZittel, K. A., and Eastman, C. R. 1913. Text-book of paleon- tology, vol. 1, 1913. Walcott, C. D. 1884. On the Cambrian faunas of North America, preliminary studies. U.S. Geol. Surv., Bull. No. 10, 1884. Walcott, C. D. 1887. Second contribution to the studies of the Cambrian faunas of North America. Bull. U. S. Geological Sur- vey, No. 30, 1887. (Dated 1886, but said by Dr. Walcott, in Amer. Jour. of Science, 3rd ser., vol. 37, 1889, p. 377, footnote, not to have been issued until 1887). Walcott, C. D. 1888a. The Taconic system of Emmons and the use of the name Taconic in geologic nomenclature. Amer. Jour. of Sci., 3rd ser., vol. 35, 1888, pp. 229-242, 307-327, 394-401. Walcott, C. D. 1888b. The Stratigraphical Succession of the Cam- brian Faunas in North America. Nature, vol. 38, 1888, p. 551. Walcott, C. D. 1889a. Stratigraphic position of the Olenellus fauna in North America and Eprope. Amer. Jour. Sci., 3rd ser., vol. 37, 1889, pp. 374-392; vol. 38, 1889, pp. 29-42. Walcott, C. D. 1889b. Description of new genera and species of fossils from the Middle Cambrian. Proc. U. S. Nat. Mus., vol. 11, 1889, pp. 441-446. Walcott, C. D. 1891a. The fauna of the Lower Cambrian or Olenel- lus zone. Tenth Annual Report of the U.'S. Geol. Surv., part 1, pp. 509-774; 1891. (Report dated 1890, but not issued until 1891.) Waleott, C. D. 1891b. Correlation Papers—Cambrian. U. S. Geol. Surv., Bull. No. 81, 1891. Walcott, C. D. 1891c. [The stratigraphic succession of the Cam- brian faunas in North America.] Congres geologique interna- tional, compte rendu de la 4me session, Londres, 1888, pp. 223-225; London, 1891. Walcott, C. D. 1891d. The North American continent during Cam- brian time. Twelfth Annual Report of the Director of the U. S. Geol. Surv., Part I—Geology, pp. 523-568; 1891. Walcott, C.D. 1898. Cambrian Brachiopoda; Obolus and Lingulella, with description of new species. Proc. U. S. Nat. Mus., vol. 21, 1898, pp. 385-420. Walcott, C. D. 1899. Pre-Cambrian fossiliferous formations. Bull. 140 BULLETIN 43 140 Geol. Soc. Amer., vol. 10, 1899, pp. 199-244. Walcott, C. D. 1900a. Lower Cambrian terrane in the Atlantic Province. Proc. Wash. Acad. Sci., vol. 1, 1900, pp. 301-339. Walcott, C. D. 1900b. Random, a Pre-Cambrian Upper Algonkian Terrane. Bull. Geol. Soc. America, vol. 11, 1900, pp. 3-5. Walcott, C. D. 1902. Cambrian Brachiopoda: Acrotreta; Linnars- sonella; Obolus; with descriptions of new species. Proc. U. S. Nat. Mus., vol. 25, 1902, pp. 577-612. Walcott, C. D. 1905. Cambrian Brachiopoda, with descriptions of new genera and species. Proc. U. S. Nat. Mus., vol. 28, 1905, pp. 227-337. Walcott, C. D. 1910. Cambrian Geology and Paleontology, No. 6, Olenellus and other genera of the Mesonacide. Smithson. Mise. Coll., vol. 53, No. 6, 1910. Walcott, C. D. 1912. Cambrian Brachiopoda. U. S. Geol. Surv., Monogr. 51; 1912. Walcott, C. D. 1916a. Cambrian Geology and Paleontology, II, No. 3, Cambrian Trilobites. Smithson. Misc. Coll., vol. 64, No. 3; 1916. Walcott, C. D. 1916b. Cambrian Geology and Paleontology, III, No. 5, Cambrian Trilobites. Smithson. Misc. Coll., vol. 64, No. 5; 196: Walcott, C. D.. 1920. Cambrian Geology and Paleontology IV, No. 6—Middle Cambrian Spongize. Smithson. Misc. Colls., vol. 67, No. 6, pp. 261-338; 1920. Weston, T. C. 1896. Notes on the geology of Newfoundland. Pro- ceedings and Transactions of the Nova Scotia Institute of Science, vol. 9, 1896, pp. 150-157. Whiteaves, J. F. 1878. On some Primordial Fossils from South- eastern Newfoundland. Amer. Jour. Sci., 3rd ser., vol. 16, 1878, pp. 224-226. Willis, B. 1912. Index to the Stratigraphy of North America. U.S. Geol. Surv., Professional Paper 71; 1912. FINIS BULLETINS OF AMERICAN PALEONTOLOGY Vol. 11 No. 44 JURASSIC CEPHALOPODA FROM MADAGASCAR By L. F. SPATH, D. Sc., (Lond.), F.@.S. . December 21, 1925 Harris Co, Cornell University, Ithaca, N. Y. U.S: A. ye a AN Piel = ae oy CONTENTS I. INTRODUCTION Il. SPECIFIC DESCRIPTIONS A. ORDER AMMONOIDEA Family Phylloceratidzx, Zittel Genus Phylloceras, Suess Family Hecticoceratidz, Spath Genus Lunuloceras, Bonarelli Genus Hecticoceras, Bonarelli Family Bonarellidx, Spath Genus Chanasia, Rollier Genus Sindeites, gen. nov. Family Haploceratidx, Zittel, emend. Spath Genus Haploceras, Zittel Family Macrocephalitidx, S. S. Buckman Genus Macrocephalites (Sutner MS.), Zittel Genus Pleurocephalites, S. S. Buckman Genus Catacephalites, S. S. Buckman Genus Kheraiceras, Spath Family Proplanulitide, S. S. Buckman, emend. Spath Genus Grossouvria, Siemiradzki Family Ataxioceratidx, S. S. Buckman, emend. Spath Genus Torquatisphinctes, Spath Family Peltoceratidx, Spath Genus Peltoceras, Waagen Family Simoceratidz, Spath Genus Hemisimoceras, gen. nov. B. ORDER NAUTILOIDEA Genus Nautilus, Breyn C. ORDER BELEMNOIDEA Genus Belemnopsis, Bayle Ill. CONCLUSIONS INTRODUCTION The collection of Madagascan cephalopods described in the following pages consists of some forty ammonites, two nautili, and two indeterminable belemnites, and the writer is indebted to Prof. J. Stansfield of the University of Il- linois,: for entrusting him with the description of the ma- terial he collected.2. The fossils evidently come from two sets of Upper Jurassic beds, and include a Callovian assem- blage, preserved partly in a yellowish-brown limestone, partly as limonitic casts; further a Kimmeridgian series, the matrix of which is a glauconitic, sandy limestone of a greenish-grey colour. Six localities are represented and the specimens collected at each are listed and discussed in the concluding chapter of the present paper. Some of the ammonites have already been referred to in connection with descriptions, by the writer, of Jurassic faunas of Kachh, India, and of Somaliland, and while it will not be necessary again to go into a detailed comparison of the Jurassic ammonites of Madagascar with those of the Af- rican continent and of India, a discussion of the Madagas- can forms described by Lemoine * and Newton * may prove of interest. 1 Now of Montana State School of Mines, Butte, Montana. 2See Amer. Jour. Sci., Fifth Ser., vol. X, 1925, p. 1. 3“Ammonites du Jurassique Supérieur du Cercle d’Analalava (Madagascar).” Paléontologie de Madagascar, VIII. Annales de Pal., vol. V, 1910, pp. 1-32, pls. I-V; vol. VI, 1911, pp. 33 52, pls. VI-VIII. 4“Notes on Fossils from Madagascar, &c.” Appendix to Baron: “Notes on the Geology of Madagascar.” Quart. Journ. Geol. Soc., vol. XLV, 1889, p. 334. Also: “On a Collection of Fossils from Mad- agascar, &c.” Ibid., vol. LI, 1895, p. 78. 6 BULLETIN 44 : 146 II SPECIFIC DESCRIPTIONS A. ORDER AMMONOIDEA Fami y PHYLLOCERATIDA, Zittel Genus PHYLLOCERAS, Suess Phylleceras aff. disputabile, Zittel 1852. Ammonites tatricus, Kudernatsch (non Pusch): “Die Am- moniten von Swinitza.” Abhand. K. K. Geol. Reichsanst., vol. I, part 2, pl. I, figs. 1—4. 1869. Phylloceras disputabile, Zittel: ‘“‘Bemerkungen tiber Phylloce- ras tatricum Pusch sp. und einige andere Phylloceras-Arten.” Jahrb. K. K. Geol. Reichsanstalt, vol. XIX, pt. 1, p. 68. This species is represented by an example (No. 7) of only about 40 mm. in diameter, somewhat corroded, but showing the suture-line, striation, and characteristic con- strictions. In whorl-shape the example agrees more with the Tanganyika specimen figured by Dacque ' than with the Kenya form, recorded by the writer as Phylloceras cf. dis- putabile.2, Of Waagen’s? Kachh examples, in the synon- ymy of which erroneous reference is made to Kuder- natsch’s figs. 4-6 of pl. I (= Phylloceras kudernatschi, Zittel, pars.), probably only one specimen, namely the orig- inal of pl. VI, fig. 3, can be attached to the present species. Through the kindness of Dr. E. H. Pascoe, Director of the Geological Survey of India, I have been able to study not only Waagen’s types but some well preserved additional Kachh Phylloceras of the group to which Ph. disputabile belongs. Pending detailed description of these it may be briefly mentioned that the species represented by Waagen’s figs. 1-2 differs from the typical Ph. disputabile in showing greater compression and more numerous constrictions which in the not very successfully drawn original of fig. 1 are wider than in the example of fig. 2. The constrictions are also more conspicuous across the venter in this flat- tened form than in the original of Waagen’s fig. 3, and 1“Dogger und Malm aus Ostafrika.” Betr. z. Pal. & Geol. Osterr.- Ung., vol. XXIII, 1910, p. 34; pl. V, figs. 3a-—e. 2“On Jurassic Ammonites from East Africa, &c.” Geol. Mag., vol. LVII, 1920, p. 318, pl. V, figs. 4a—d. 3 “Jurassic Fauna of Kutch.” I Cephalopoda. Mem. Geol. Survey India, Pal. Indica, Ser. IX, No. 2, 1875, p. 31, pl. VI, figs. 1-3. 147 MADAGASCAR CEPHALOPODA 7 the striation is finer. There are transitional forms, how- ever, including the two Phylloceras disputabile recorded by myself ' from bed No. 4 (of the Lower Chari Group) at Khera, Kachh, and it seems inadvisable at the present stage, to separate the two forms. A second specimen (No. 1) in the present collection, rather poorly preserved, probably belonging to the flat- tened and more finely costate type above referred to, may be identical with the Madagascan form described by Le- moine’? as Phylloceras lodaense, Waagen. In the Indian types of the latter species, the course of the radial line is far more sigmoidal. Localities and Horizon. Ankidabé (localities I and II), Callovian, macrocephalus zone. Phylloceras, sp. ind. 1910. Phylloceras sp. du groupe de Ph. mediterraneum, Neumayr; Lemoine: “Ammonites Jurassiques d’Analalava.” Loc. cit., p. 4, pl. I, fig. 4. a A small fragment of a cast (No. 28) without striation but a typical, deep rursiradiate constriction, probably be- longs to the “Upper Oxfordian” form described by Lemoine as being extremely close to Ph. mediterraneum of the Callo- vian. Waagen’s Indian examples* which are before me, together with other Kachh specimens, are probably not identical with the later species here discussed, as stated al- ready by Lemoine. On previous occasions‘ I expressed 1“On the Blake Collection of Ammonites from Kachh, India.” Mem. Geol. Survey India, Pal. Indica, New Series, vol. IX, No. 1, 2924, p. 22. 2“Ammonites du Jurassique Supér. d’Analalava.’’ Pal. de Mad- agascar, VIII. Annales de Pal., vol. V, fase. 4, 1910, p. 5, pl. I, fig. 2. 8 Loe. cit. (1875), p. 34, pl. V, figs. la, b; pl. VII, figs. 3a—c. 4“Jurassic Ammonites from Jebel Zaghuan (Tunisia).” Quart. Journ. Geol. Soc., vol. LXIX, 1913, p. 561. Also: Geol. Mag. (loc. eit. 1920), p. 320. 8 BULLETIN 44 148 doubt whether Ph. mediterraneum really persisted from the macrocephalus zone to the uppermost Jurassic, as is gen- erally assumed. The present material, however, is insuffi- cient for comparison with the Argovian forms recorded by de Riaz! and myself and the Kimmeridgian fragment de- scribed and figured by Canavari.? Locality and Horizon. Antsalova, Kimmeridgian. Family HECTICOCERATIDZ, Spath? This family is taken to include the genera Hecticocerus, Bonarelli (genotype: Ammonites hecticus, Reinecke, 1818, Mar. Prot. Naut. et Argon., &c., pl. IV, fig. 37) and Lunulo- ceras, Bonarelli (genotype: Ammonites lunula, Zieten, 1830, Verst. Wiirt., pl. X, fig. 11). The latter is earlier than Hec- ticoceras and cannot therefore be considered to be a sub- genus of Hecticoceras. The genus Brightia, Rollier (geno- type: B. nodosa, Quenstedt sp. = Ammonites hecticus no- dosus Quenstedt: Amm. d. Schwab. Jura, 1887, pl. LX XXII, fig. 10) aiso belongs to this family, further Hecticocera- toides, Spath (genotype: H. suborientalis, Spath = Oppelia orientalis, Waagen, non d’Orbigny sp.; Pal. Indica, 1875, pl. XI, fig. 5). This was described as probably a develop- ment of the “subpunctata * group of Hecticoceras’”’ and since the latter genus must now be restricted to the group of evo- lute forms with typical hecticus ornamentation, it is neces- sary to separate with a new name: KHERAITES gen. nov.,’ 1“Description des Ammonites des Couches a Peltoceras transver- sarium de Trept (Isére).” 1898, p. 40, pl. XVI, figs. 9, 10. 2“Rauna degli Strati con Aspidoceras acanthicum del Mte. Serra.” Pal. Ital., vol. II, 1896, p. 38. 3“The Ammonites.” Part VII of J. W. Gregory’s: “Somaliland.” Monographs of the Hunterian Museum, Glasgow, 1924, p. 114. 4 This name was used in error without being marked as new and explained. 5 Genotype: Harpoceras crassefalcatum, Waagen, loc. cit. 1875, pl. XII, figs. 6, 6a. 149 MADAGASCAR CEPHALOPODA 9 5 the forms grouping themselves round “Harpoceras’’ crasse- falcatum, Waagen, and K. SUBPUNCTATUS, sp. nov. (= “Harpoceras” punctatum, Waagen non Stahl sp., loc. cit. pl. XII, fig. 9). These latter forms, however, never lose the keel entirely, whereas Hecticoceratoides develops an almost subconeave periphery, with costation continuous across it, in the form of forwardly directed chevrons. The genus Putealiceras 8. 8. Buckman (genotype: Ammonites putealis, Leckenby, 1859, Quart. Journ. Geol. Soc., vol. XV, pl. Il, fig. 3; S. S. Buckman: Type Ammonites, vol. IV, 1922, pl. CCXCVII) which includes such rectiradiate Kachh species as ‘“Harpoceras’” trilineatum Waagen,' Genus LUNULOCERAS Bonarelli Lunuloceras cf. lunuloides (Kilian) 1887. Ammonites hecticus compressus, Quenstedt: “Ammon. d. Schwab. Jura,” pl. LXXXII, figs. 31-32. 1911. Hecticoceras lunuloides (Kilian) Tsytovitch: “Hecticoceras du Callovien de Chézery.” Mém. Soc. Pal. Suisse, vol. Reece Vl p. 70; pl. VIII, fig. 7? Four immature examples (Nos. 19, 21-23) with smooth inner areas and fine and close crescents on the outer half of the sides may be tentatively referred to this species though it is very doubtful whether they are identical. The largest of the specimens has a diameter of only 16 mm. Localities and Horizon. Ankidabé (localities III and IV), anceps zone ? Genus HECTICOCERAS, Bonarelli Hecticoceras, sp. juv. ind. A small ammonite (No. 17) and the fragment of an- other (No. 17a) represent the inner whorls of forms like Hecticoceras svevum Bonarelli, as figured by Mlle. Tsyto- vitch.2, The ventral area is still rounded and the keel is ‘ Loe. cit. (1875), p. 71, pl. XIII, figs. 2a, b. also belongs to this family Hecticoceratide. emoc eit. (1911); p. 51, pl. V, fig. 7. IO BULLETIN 44 150 hardly visible at the diameter of 14 mm., but the umbilicus is perhaps less open in the Madagascan examples than in the Wiirttemberg types figured by Quenstedt.! Locality and Horizon. Ankirihitra (locality V), Cal- lovian, anceps zone ? Family BONARELLIDA, Spath ? This family includes the genera Bonarellia, Cossmann » as Gea (= Distichoceras, Munier-Chalmas; genotype: B. bico- — stata Stahl. sp. Corresp. Blatt. Wiirtt. Landw. Ver., vol. VI, 1824, p. 49, fig. 9), Horioceras, Munier-Chalmas (geno- type: Amm. baugieri, d’Orbigny, 1846, Pal. Franc., Terr. Jurass., p. 445, pl. CLVIII, figs. 5-7) and Chanasia, Rollier (genotype: Hecticoceras chanasiense, Parona and Bona- relli, “Faune du Callov. Inférieur de Savoie,’ Mém. Acad. Sci. Savoie (4), vol. VI, 1897, p. 134, pl. IV, fis. 2) the new genus SINDEITES, gen. nov., proposed for a stock that has affinities with Chanasia as well as with Hectico- ceratoides. Its resemblance to the latter is indicated by the inclusion, by Waagen, in his Oppelia orientalis (non Sowerby) of a doubtful, small, Kachh example (Waagen’s fig. 6 of pl. XI, and figs. 8, 8a, b of pl. XII) which is dis- tinguished from a new species of Sindeites before me, from Jessulmir, Sinde, India (B. M. No. C. 23545), chiefly by its rounded whorl-shape and retention of a keel. From Bonarellia and Chanasia the new genus differs in having more recticostate ornamentation and in not showing a sub- division of the lateral area into distinct inner and outer halves. The costation of the new form (and genotype) of Sindeites described below as S. madagascariensis, nov., almost resembles that of the outer whorl of Peltoceratoides semirugosum (Waagen).? In Hecticoceratoides also the 1 Loe. cit. (1887), pl. LXXXII, figs. 3-5. 2Loc. cit. (Somaliland, 1924), p. 114. If it is considered that there was no need to change Distichoceras into Bonarellia (on ac- count of supposed preoccupation by Distichocera, Kirby) the names Distichoceratidz and Distichoceras will have to be substituted for Bonarellide and Bonarellia respectively. 3Loce. cit... (1875), p. oo, pl. XhVE fies I5I MADAGASCAR CEPHALOPODA 11 ribs are not rectiradiate and on the outer whorl at least they are not flattened as in the genera of the family Bonarellide. Rollier ' separated Bonarellia (including Horioceras as its supposed male), as an Oppelid stock, from Hect'coceras and its allies (comprising Chanasia) which were consid- ered to be offshoots of a “Ludwigian” group. It is true that at least some of the Kachh forms comprised in Waagen’s “Oppelia subcostaria”’ ? (which I included in the genus Alcidia), are very close to Chanasia and thence to Bonarellia, but it seems more probable that Bonarellide are ofishoots of Hecticoceratide (e.g. Lunuloceras). It also appears advisab!e to restrict the family to those genera in which the costze show that peculiar flattening which is found again in the Lower Albian genus Leymeriella and to exclude del Campana’s genus Taramelliceras (compris- ing the typically Callovian “flexuosi’’), which with Phlycti- ceras, Haug, may be grouped in a family PHLYCTICERATIDA, nov. Genus CHANASIA, Rollier Chanasia, sp. juv. ind. A small example (No. 20), fragmentary but well pre- served and resembling similar limonitic casts from Euro- pean localities, e. g., the “Ornatenthon” of Gammelshausen, Wiirttemberg, can be attached to this genus. As its diam- eter is only 18 mm., specific determination is difficult, but Ch. chanasiensis, Parona and Bonarelli, above referred to, from the Callovian of Chanaz, Savoy, France, appears to be very close, though it is considerably less compressed. Locality and Horizon. Ankidabé (locality IV), Callo- vian, anceps zone ? 1“Phylogénie des Ammonites.” Ecloge Geol. Helvet., vol. XVII, 1922, pl. XXII (table). 2 Loc. cit. (1875), p. 48, pl. X, fig. 2, non fig. 1. U2)) os, BULLETIN 44 152 Genus SINDEITES, gen. nov. Sindeites madagascariensis, sp. nov. This form is based on a fragment (No. 24) showing only a portion of a septate whorl of 8.5 mm. height and 7 mm. thickness (fig. 6c, pl. I, magnified 2) and less well preserved remains of the almost smooth next inner whorl. The sides are compressed, flattened; the whorl- section is subhexagonal, the venter subtabulate, with the median line elevated but not actually keeled. The ribs — are almost rectiradiate, with distinct inner and outer tu- bercles and a median bulla, as indicated in fig. 6a, (pl. I). They are alternately long and short, the latter ending at the median tubercle. The outer portions of the ribs up to the peripheral clavus are flattened (see figs. 6a and 6d, pl. I). The suture-line is closely comparable to that of typical Bonarellia and of a new form of Sindeites (to be figured in the forthcoming Revision of the Jurassic Am- monites of Kachh) in which there is no trace of a keel, but costation across the ventral area as in Kosmoceras (pl. I, fig. 6b). There is some resemblance to the form figured by Quenstedt! as Ammonites cf. bipartitus, Zieten, but its keel is distinct and the ribs have no median tubercle. On the other hand the vigorously ornamented Hecticoceras fortocostatum, Tsytovitch? from the Middle Callovian of the Hautes in the Jura, with a wide ventral area bearing three keels, appears to be a true Hecticoceras. The ferm referred by Petitclerc? as a variety boginense to Hecii- coceras hecticum (Reinecke) also is not closely comparable to the form here described and, in any case, has a high keel. Locality and Horizon. Ankidabé (locality IV), Callo- vian, anceps zone ? 1 Loe. cit. (1887), pl. LXXXII, fig. 14. 2 Loe. cit. (1911), p. 38, pl. Il, figs. 10, 11. 3“Haune du Callovien du Department des Deux Sévres.” Contrib. a l’Etude des Terr. Jurass. dans. l’W. de la France, Vesoul, 1915, p. 26, pl. I, fig. 4, pl. XIII, fig. (4) 38. 153 MADAGASCAR CEPHAIOPODA ue Family HAPLOCERATIDA Zittel emend. Spath Genus HAPLOCERAS Zittel Haploceras elimatum (Oppel) Plate I, Figs. la—c 1868. Ammonites elimatus, Oppel; Zittel: ‘“Cephalopoden der Stramberger Schichten,” Pal. Mitteil. Mus. K. Bayer Staates, vol. II, pt. 1, p. 79, pl. XIII, figs. 1-7. 1924. Haploceras deplanatum (Lemoine, non Waagen) Spath: loc. cit., Pal. Indica, p. 6. 1924. Haploceras elimatum (Oppel) Spath: loc. cit. Monogr. Hunter- ian Museum, p. 160. This form is represented by eight examples (Nos. 35— 42) of which three are here figured. It seems probable that Lemoine’s “Lissoceras’’ deplanatum' which was _ be- fore him in a large number of specimens, is not identical with Waagen’s Indian type. The latter is even more com- pressed than the figure,? whereas Lemoine considered the section of his Madagascan form to be very close to that of Uhlig’s Haploceras indicum.* In the examples here de- scribed the umbilical border most decidedly marks the region of greatest whorl]-thickness as in Zittel’s Stramberg and Koniakau examples. 2 Locality and Horizon. Antsalova, Kimmeridgian. Family MACROCEPHALITIDA, S. 8. Buckman Genus MACROCEPHALITES (Sutner MS.) Zittel Macrocephalites aff. madagascariensis, Lemoine. Plate I, Fig. 7 1911. Macrocephalites macrocephalus (Rein.) race noetlingi, Le- moine, Ann. Pal. loc. cit., p. 31, pl. III, fig. 3. 1911. Macrocephalites madagascariense, Lemoine, ibid., p. 51. 1924. Macrocephalites madagascariensis (Lemoine) Spath: Pal. Indica, loc. cit., p. 7. The example (No. 2) of which the outline section is Bioe, cit. (1911), p. 13, fig. 8 on p. 14. * Waagen, loc. cit. (1875), pl. XI, fig. 9. (The inner whorl in fig. 9b is wrongly restored, i.e., too inflated.) _* “Fauna of the Spiti Shales.” Mem. Geol. Surv. India, Pal. In- pa XV, Himalayan Fossils, vol. IV, fase. 1 (1903), p. 21, pl. » NE. 2. 14 BULLETIN 44 . 154 here figured (pl. I, fig. 7) is completely septate and it is believed that the more rounded periphery shown in Le- moine’s figure, as in all the Indian forms of the formosus group, characterises only the larger and outer whorls. In its whorl-section the present example closely resembles typical European examples of M. macrocephalus (Schlot- heim), e.g. a Chanaz specimen in the British Museum (No. C. 10564). Locality and Horizon. Ankidabé (locality I), Lower Callovian, macrocephalus zone. Genus PLEHUROCEPHALITES, S. S. Buckman Pleurocephalites, sp. ind. Five badly preserved examples (Nos. 3, 8, 12-14) seem to be referable to this genus, but exact identification is impossible. Pleurocephalites folliformis, S. S. Buckman * is more inflated. The largest of the Madagascan speci- mens has some resemblance to the Indian forms of the group of Pl. ? grantanus (Oppel) Waagen, and Pl. ? chrysoolithicus (Waagen),? but owing to its being crushed obliquely, and to its otherwise defective preservation, its identity with the other four examples must remain doubtful. Localities and Horizon. Ankidabé (localities I [No. 3] and II [No. 8]) and Ankirihitra (Nos. 12-14). Lower Callovian, macrocephalus zone. Genus CATACEPHALITES, S. 8, Buckman Catacephalites, sp. ind. Plate I, Figs. 3, 4 An example (No. 4) of which the whorl section is here figured (pl. I, fig. 4), unfortunately not well preserved, has costation resembling that of Macrocephalites colcanapti Lemoine? or of Catacephalites durus, 8S. S. Buckman.* 1 Type Ammonites, vol. IV, 1922, pl. CCCXLVIII. 2 Loe. cit. (1875), pls. XXXVI, fig. 6, and pl. XXX, fig. 1. 3 Loe. ‘cit. (1911), p. 33, pl. II, figs. 1, 2. 4 Type Ammonites, vol. IV, 1922, pl. CCLXXXIII. 155 MADAGASCAR CEPHALOPODA 15 The suture-line represented in fig. 3, pl. I, was taken from a small example (No. 9) of only 13 mm. diameter, which has a similar whorl-section, but may not belong to the same species, and a third specimen (No. 11) is only a doubtful fragment. Localities and Horizon. Ankidabé (localities I [No. 4] and II [No. 9]) and Ankirihitra (No. 11). Lower Callo- vian, macrocephalus zone. Genus KHERAICERAS, Spath Kheraiceras ? stansfieldi, sp. nov. Plate I, Figs. 2a, b This species is based on a completely septate cast (No. 5) which may belong to a new genus, but is here attached to Kheraiceras, created! for the scaphitoid Macrocepha- l tes, i.e. the so-called bw'lati of the Callovian. The type of Kheraiceras is K. cosmopolita, Parona and Bonarelli ° (= Stephanoceras bullatum, Waagen non d’Orbigny sp.) and as the original example is now before me, it may be mentioned that Waagen’s Indian form is even more de- pressed that Quenstedt’s figures 21-238 of pl. LXXVIII (“Ammonites platystoma,”’ pars, non Reinecke) included in K. cosmopolita by Parona and Bonarelli, but that it has similar rectiradiate costation. In the Madagascan form, on the other hand, the ribs are very strongly prorsiradiate, the umbilical border is merely rounded, and the suture- line is extremely complex, finely divided, and interlocking, so that probably several distinct stocks within the Macro- cephalitide produced these scaphitoid endforms that may eventually have to be separated. If the present form is now left in Kheraiceras, it is partly on account of insuffi- ciency of comparable European material and partly be- cause forms like Kheraiceras ? platystomum (Reinecke) * and Kh. ? globulatum (Quenstedt) ‘ bridge the gap between the type of Kheraiceras and the present form. 1Jn Pal. Indica, loc. cit. (1924), p. 7. 2 Loe. cit. (1897), p. 146. 3In Quenstedt, loc. cit. (1887), pl. LX XVIII, fig. 25. *Thid., fig. 2. 16 BULLETIN 44 : 156 On the side of the Madagascan form not figured the remains of the umbilical suture of the (missing) outer whorl are visible, as indicated in fig. 2a, pl. I, by the white line. This is comparable to that shown in Quenstedt’s fig. 1, pl. LX XVIII, but is even more eccentric. The suture- line is considerably more complex than that of Kh. ? platy- stomum (Reinecke) Quenstedt (pl. LX XVIII, fig. 25) and the saddles are deeply divided, whilst the second lateral lobe is less wide and the whole of the suture-line closely interlocking with the preceding and succeeding lines. Kh. ? QUENSTEDTI, sp. nov. (= Ammonites bullatus Quenstedt non d’Orbigny ') also has a wider second lateral lobe and less individualized auxiliary elements of the suture-line, and a less strongly prorsiradiate costation, but comes close to the present species in whorl-shape. Kh. ? globulatum (Quenstedt) is much more inflated and differs in ribbing. Excentrumbilicate shells somewhat similar to Kherai- ceras were successively produced by Spheroceratidxe and Morphoceratidx in the Bajocian, and by Tulitide in the Bathonian (including the true Ammonites builatus d’Or- bigny) and correct identification of incomplete specimens is very difficult. The apertures of Kheraiceras cosmopolita and Kh. ? quenstedt have neither collar nor lip. I have much pleasure in dedicating this species to Prof. J. Stansfield, its discoverer. Locality and Horizon. Ankidabé (locality I), Lower Callovian, macrocephalus zone. Family PROPLANULITIDA, 8.8. Buckman, emend. Spath — Genus GROSSOUVRIA, Siemiradzki Grossouvria, sp. ind. cf. anomala (Loczy) 21875. Perisphinctes curvicosta (Oppel) Waagen: Pal. Indica, loc. cit., pl. XX XIX, fig. 5 (non 4-6). 1915. Perisphinctes anomalus, Loczy: “A Villanyi Callovien-Am- monitesek Monografiaja.” Geol. Hungar., vol. I, p. 347, pl. - VIII, figs. 8-11, pl. XIV, fig. 5. A fragment (No. 10) is doubtfully referred to this 1 Quenstedt, loc. cit. (1887), pl. LX XVII, figs. 7 and 8, as repre- sented by a Neuffen specimen in the British Museum (No. 22366). 157 MADAGASCAR CEPHALOPODA 17 Lower Callovian form, but its mode of preservation is un- satisfactory. The rursiradiate character of the secondary costz is pronounced but there are more single coste than in Waagen’s original of his fig. 5. It may be added that Loczy was misled by the incorrect drawings of Waagen’s types. The three Indian forms are all different and only fig. 5 shows a peripheral aspect comparable to that of Loczy’s fig. 10 (pl. VIII). The other two forms also dif- fer considerably in whorl-section from the compressed form figured by Loczy (text—fig. 92, p. 346) as Perisphinc- tes curvicosta. Locality and Horizon. Ankidabé (locality II), Callo- vian, macrocephalus zone? Grossouvria ? cf. waageni (Teisseyre) Loczy sp. 271915. Perisphinctes waageni, Teisseyre; Loczy: Geol. Hungar., loc. cit., p. 356, pl. XIII, fig. 4. A fragmentary example (No. 15), the whorl-section of which is figured (pl. I, fig. 10), may be tentatively at- tached to Loczy’s P. waageni (non Teisseyre ?). Its um- bilicus, however, may have been smaller since the outer whorl appears to have increased in thickness comparatively more rapidly. Perisphinctes recuperoi (Gemmellaro) Waagen' has finer secondary costation, also P. waageni, Teisseyre in Petitclerc.2 The latter author’s P. cardoti,’ however, in coarseness of costation, resembles the form here discussed, but is too evolute and far too rounded- whorled. Locality and Horizon. Ankirihitra. Callovian, macro- cephalus zone ? Grossouvria, sp. juv. ind. 1887. Ammonites convolutus dilatatus, Quenstedt: “Amm. des Schwab. Jura,” pl. LXXXI, figs. 1-9. Three immature specimens (Nos. 18, 26, 27) probably memencit. (1875), p. 172, pl. XLITI, fig. 1. * Callovien des Deux Sévres. II, 1915, pl. IX, fig. 1. ona. p. 64, pl. V; fic. 2. pl. XIII, fig. 2 (62). 18 BULLETIN 44 ‘ 158 belonged to forms of Grossouvria comparable to Quen- stedt’s species, but specific identification is impossible. The largest example has a diameter of only 16 mm. and is slightly more compressed and more closely costate than the smaller examples. Localities and Horizon. Ankidabé (localities III and IV) Callovian. Family ATAXIOCERATIDA, 8.8. Buckman emend. Spath Genus TORQUATISPHINCTES, Spath Torquatisphinctes ? cf. bangei (Burckhardt) Plate I, Fig. 5 1921. Perisphinctes (Aulacosphinctes) bangei, Burckhardt: Faunas Juras. de Symon, &c. Bol. Inst. Geol. Mexico, No. 33, vol. IL (Atlas), pl. IX, figs. 5-9. 1924. Torquatisphinctes bangei (Burckhardt) Spath: Pal. Indica, loc. cit., p. 15. Four fragments of Perisphinctoid ammonites (Nos. 29-— 32), the whorl-section of one of which is here figured (pl. I, fig. 5), show great resemblance to the Mexican form from the Middle Kimmeridgian. Their reference to Tor- quatisphinctes, however, must remain uncertain. Some fragmentary Lithacoceras, another Middle Kimmeridgian genus, probably belonging to L. ? andranosamontx, Le- moine, in the British Museum and included (with other forms) in Newton’s Perisphinctes polygyratus (Reinecke),' are distinguished from the examples here described by their finer and tri- or multiplicate ribbing. Lemoine’s “Perisphinctes” colcanapi,2 with numerous constrictions, differs from the typical Torquatisphinctes and should per- haps be referred to Subplanites or to Lithacoceras, but is represented only by worn fragments (B. M. C. 3587a, C. 3580f). Locality and Horizon. Antsalova, Kimmeridgian. 1Loe. cit. (Q. J. G. S., 1889), p. 334. 2 Loe. cit., (1911), p. 48, pl. VIII, figs. 2a, b. 159 MADAGASCAR CEPHALOPODA 19 Family PELTOCERATIDA, Spath Genus PELTOCERAS, Waagen Peltoceras, sp. juv. ind. A small fragment of a limonitic cast (No. 25) probably belongs to a form of Peltoceras but cannot be attached to any described species. The inner whorls of Peltocera- toides, at the same diameter are very finely costate. P. madagascariense Lemoine? does not belong to this genus, but is apparently an immature Subdichotomoceras. The fine and close costation of other small examples mentioned by the same writer and compared to Amm. arduennensis, d’Orbigny, and A. eugeniti, Raspail, points rather to species of Peltoceratoides. On the other hand the same author’s Peltoceras cf. syriacum, Noetling ? may be close to the pres- ent form, though not Noetling’s Argovian type? which like G. Boehm’s P. tjapalului+ is probably a Peltoceratoides. In the absence of outer whorls Peltoceratids are often dif- ficult to distinguish. Locality and Horizon. Ankidabé (locality IV), Callo- vian? Family SIMOCERATIDA, Spath Genus HEMISIMOCERAS, gen. nov. This genus is established for the two Madagascan forms described below, H. semistriatum, nov. to be genotype. On a previous occasion ® I referred to these forms as “pe- culiar (undescribed) Aspidocerates,” but their resemblance to Aspidoceratidx is probably quite superficial and based on the resemblances of the cross-section to that of typical Pihoc. cit. (1911), p. 14, pl. VII, fig. 2 (non 1!). Eeeoid.. p. 15, pl. V, figs. 3a, b. 3“Der Jura am Hermon.” 1875, p. 31, pl. V, fig. 3 only. The ex- ternal saddle of this form is extremely wide and the suture-line differs also in the umbilical elements. 4“Beitr. z. Geol. ye Niederland.-Indien.” Pal. Suppl. (1904-7), pl. XXIX, figs. 2a-b, 5 Loe. cit. (Pal. Indica, 1924), p. 6. 20 BULLETIN 44 | : 160 forms of Aspidoceras, although there is also similarity in suture-line. Sutneria, another somewhat similar genus, though having a reduced suture-line, shows a change in ornamentation comparable to that noticed in the new forms here described, namely from merely costate or striate to tuberculate. Comparison with this genus is difficult, be- cause both the Madagascan species are septate casts and there is no indication of a modified body-chamber. It seems possible, of course, that like Sutneria the new genus is an abnormal offshoot of some regular, perisphinctoid, stock, though here the resemblance ends. Both the new forms show peculiar, deep, prorsiradiate constrictions, es- pecially where the striate ornamentation definitely changes to a tuberculate one. At the same time the whorl-thickness increases abruptly (after the deep constrictions). This points to affinity with Simoceratide, especially those early forms like Pseudosimoceras, Spath, which were at one time included in Holcostephanus. In a specimen of Simoceras volanense (Oppel) from Catria, Apennines (B. M. No. C. 8365) agreeing with the inner whorls of Zittel’s! type- figure but in a better state of preservation, there can be seen the perisphinctoid inner whorls, resembling in orna- mentation and whorl shape such a Pseudosimoceras ? as e.g. “Perisphinctes acer’ (Neumayr). Peripheral tuber- cles, however, soon appear on the ribs and at a diameter of 15 mm. already, these become more blunt and distant, until finally (at 35 mm.) only very indistinct ribs but strong outer tubercles remain. The inner tubercles appear only at still larger diameters. The suture-line of Simo- ceras volanense does not show a dependent umbilical lobe, but other Simoceratids and especially Pseudosimoceras have suture-lines closely comparable to those of the forms here described. Since the present species, however, are quite unlike any forms previously described and so far are known 1“Fauna d. Alt. Cephalopoden-fiihrenden Tithonbildungen.” Pale- ontgr. Suppl. 1870, pl. XXXII, fig. 7. 16. MADAGASCAR CEPHALOPODA 21 in only two entirely septate specimens, the reference of the new genus to the family Simoceratide must remain tentative. For the same reason a Middle Kimmeridgian age is assigned to these forms merely on the strength of the associated ammonites. Hemisimoceras semistriatum, sp. nov. Plate I, Figs. 8, b, text—Fig. 1 The holotype has dimensions 42—.30—.36-.45.' The inner whorls are depressed, subcoronate, with subtubercu- late primary ribs and very faint and fine secondary ribbing across the broad and almost flat ventral area. Later the whorl-section becomes less depressed, with wider sides and longer primaries bearing a faint tubercle at the point of involution. Then the whorls become rounder, though re- maining wider than high, the primary costze almost dis- appear and only uniform single strize remain, somewhat lytoceratid, with only at intervals a single tubercle near the ventral part of the lateral area (only developed on the side not figured). These tubercles remain and become increasingly conspicuous, whilst all striation disappears. The primary ribs, leading from the tubercles towards the umbilical suture, tend to reappear on the outer whorl. The suture-line is characterized by its trifid lateral lobe and dependent umbilical lobe. Its saddles are less slender than those of the comparable suture-line of “‘Perisphinctes”’ acer Neumayy.? 1 Diameter in mm.; whorl-height, thickness and umbilicus in per- centages of the diameter. 2 As figured in Canavari’s: “Fauna d. str. con Aspidoceras acan- thicum di Mte. Serra.” Pal. Ital., vol. III (1897), p. 209, text— figs. 17, 18. 22 BULLETIN 44 162 =f (<4) Fic. 1. Hemisimoceras semistriatum, gen. et sp. nov. Middle Kim- meridgian, Antsalova. Suture-line (x 4) near end of outer whorl. E = external lobe; L = lateral lobe; U = umbilical su- ture. (Compare fig. 8a, pl. I.) Locality and Horizon. Antsalova, Kimmeridgian. Hemisimoceras nodulosum sp. nov. Plate I, Figs. 9a, b This species is closely allied to that last described but cannot be included with it. While whorl-shape and suture- line are very similar, the present species differs from H. semistriatum in showing a perisphinctoid type of ribbing to a diameter of 30 mm. where there is a deep constriction. The primary ribs then become distant and nodate at the outer ends, with four or five very faint secondaries to each, crossing the evenly rounded venter. The siphonal line is smooth on the last half-whorl. The costation of the earlier whorls resembles that of the forms of the acanthicus beds, figured by Canavari.' Sutneria evoluta (Quenstedt) ? had a flattened venter and if it is as close to S. reineckeiana as Quenstedt thought, its suture-line would be quite different from that of the present form. Costate forms of Sutneria like S. galar 1 Loe. cit. (1897), pls. VII, VIII. 2 Loe. cit. (1888), pl. CXII, fig. 19. 163 MADAGASCAR CEPHALOPODA 22 (Oppel), S. cyclodorsata (Moesch) de Loriol,' and S. nu- splingensis Fischer ? suggest that the stock here described is derived from quite a different branch of perisphinctoids. Locality and Horizon. Antsalova, Kimmeridgian. B. ORDER NAUTILOIDEA Genus NAUTILUS, Breyn Nautilus cf. calloviensis, Oppel. 1875. Nautilus calloviensis Oppel. Waagen: Pal. Indica, loc. cit., Pets. Ll fie. 2: A poorly preserved specimen (No. 6) seems to agree with this species which was described as the ‘‘most com- mon of all the Kachh nautili.”” Waagen’s type is now be- fore me and shows a similar whorl-shape. Locality and Horizon. Ankidabé (locality I), Callo- vian. Nautilus cf. kumagunensis Waagen. 1875. Nautilus Kumagunensis, Waagen, Pal. Indica, loc. cit., p. 19, pl. III, fig. 1. A small and fragmentary specimen (No. 16) has a sub- sulcate periphery and the high umbilical wall of this spe- cies. Nautilus hexagonus, Sowerby, with a similar whorl- shape, has a wider umbilicus. Locality and Horizon. Ankirihitra, Callovian. C. ORDER BELEMNOIDEA Genus BELEMNOPSIS Bayle Belemnopsis, sp. ind. 1895. Belemnites hastatus, Blainville. R. B. Newton: “Fossils from Madagascar.” Quart. Journ. Geol. Soc., vol. LI, p. 78 (pars.). 1921 ?. Belemnites tanganensis (non Futterer?) Morand: Bull. Soc. Géol. France (4), vol. XX (1920), p. 158. 1“Monogr. Paléont. des Couches de la Zone a A. tenwilobatus de Baden.” Mém. Soc. Pal. Suisse, vol. V, 1878, pp. 90-93, pl. XV, figs. 3-5. *“Ueber einige neue oder in Schwaben bisher unbekannte Ver- steinerungen d. Weiss. & Braun. Jura.” Jahresb. Ver. Vat. Vaturk. Wirttemb., vol. LXIX (1913), p. 54, pl. V, fig. 23. 24 BULLETIN 44 164 Two fragmentary examples (Nos. 43, 44) are com- parable to the majority of the specimens recorded by New- ton (e.g. B. M. No. C 4925-6) but specific determination is probably impossible. Mlle. Morand recorded Belemnop- sis ef. semisulcatus (Mstr.), Bel tanganensis; Futterer. and Hastites claviger (Waagen), from blue shales at An- dranosomonta, and mentioned that B. tanganensis occurred already at the horizon of the sandy limestones associated with lamellibranchs of the Inferior Oolite. The identifica- tions undoubtedly require revision, and there may be no need to invoke the aid of faults to explain apparently anomalous superposition. Locality and Horizon. Antsalova, Kimmeridgian. Ill CONCLUSIONS The cephalopods described in the foregoing pages may be arranged in the following assemblages: I. From “Hill approaching hollow west of and before reaching Ankidabé”’: Phylloceras aft. disputabile, Zittel. Macrocephalites aft. madagascariensis, Lemoine. Pleurocephalites sp. ind. Catacephalites sp. ind. Kheraiceras ? stansfieldi sp. nov. Nautilus cf. calloviensis, Oppel. This is clearly a Lower Callovian fauna, from the mac- rocephalus beds (sensu lato). Il. From: “Coming down the Hill from Ankidabé, going towards Maevatanana”’: Phylloceras aff. disputabile, Zittel. Pleurocephalites, sp. ind. 165 MADAGASCAR CEPHALOPODA 25 Catacephalites sp. juv. ind. Grossouvria cf. anomala (Loczy). This assemblage is apparently of the same age as that er Locality I. - Ill. From 1 Km. East of Ankidabé: Grossouvria sp. juv. ind. Lunuloceras ef. lunuloides (Kilian). A middle Callovian age may be suggested for these forms but they are immature and no definite conclusions could be based on them. IV. From “Approaching Ankidabé, on trail to Maeva- tanana”’: Chanasia sp. juv. ind. Lunuloceras cf. lunuloides (Kilian). Sindeites madagascariensis sp. nov. ' Peltoceras sp. juv. ind. Grossouvria sp. juv. ind. This assemblage also is probably of Middle Callovian age (anceps zone in the wider sense) but again consists merely of small and indefinite examples. Ve from “1 hr. Hast of Ankirihitra” : Pleurocephalites sp. ind. Catacephalites ? sp. ind. Grossouvria ? cf. waageni (Teisseyre) Loczy sp. Nautilus cf. kumagunensis, Waagen. Hecticoceras sp. juv. ind. The first four forms are probably from the “‘macro- cephalus beds,” like those of localities I and II; the last, resembling the small limonitic casts from III and IV, may tentatively be grouped with the Middle Callovian (anceps zone). VI. Near Antsalova: Phylloceras sp. ind. (mediterraneum group). Haploceras elimatum (Oppel). Torquatisphinctes ? ef. bangei (Burckhardt). Hemisimoceras semistriatum gen. et sp. nov. 26 BULLETIN 44 : 166 Hemisimoceras nodulosum sp. nov. Belemnopsis sp. ind. This assemblage is probably of the Middle Kimmerid- gian age, and would have been included in the ‘‘zone of Amm. acanthicus” or “Lower Tithonian” of the older authors. There is no evidence for definite reference of this fauna to one of the three corresponding zones pre- viously listed for the Middle Kimmeridgian.' It is prob- able that it differs from Middle Kimmeridzian faunas of Kachh and Somaliland because the exact subzone or horizon to which the Madagascan forms belong is not represented in India or on the African continent; but Toucas’s “Diph- yakalk”’ of the South of France ? which includes both Haplo- ceras elimatum (the commonest ammonite in the present collection) and similar perisphinctoids may possibly com- prise a corresponding and as yet undefined horizon. No zonal collecting, unfortunately, has been done in Mediter- ranean countries, and the more detailed subdivision of the geological time-scale given in previous papers* must re- main somewhat speculative. Considering, however, for the present only the larger zones, it is clear that the Jurassic sequence of Madagascar must be very incomplete. Of Lemoine’s list of Madagascan species, forms not already referred to above include not- ably Reineckeia anceps, Obtusicostites, and Kinkeliniceras, which were recorded by Waagen and myself from the Middle Callovian (anceps zone) of Kachh, India; further Hildoglochiceras kobellt (Oppel) which is an important Middle Kimmeridgian form, occurring in the Spiti shales as well as Kachh and Tanganyika. But Lemoine also re- 1 Spath, loc. cit. (Somaliland, 1924), p. 158. 2See Spath: “Ammonites from New Zealand,” Quart. Journ. Geol. aap vol. PEXXIX (1923), p. 305, and loc. cit. (Somaliland, yee) . 160. a cit. (New Zealand, 1923), p. 304 and Somaliland, 1924, p. 158. 4 See also Spath, loc. cit. (Somaliland, 1924), p. 160. 0 Pad 167 MADAGASCAR CEPHALOPODA 627 cords a number of more doubtful forms, like “Oppelia,”’ Lytoceras rex, and some Phyilloceras, which may indicate the presence of Oxfordian beds, and above all some Mayaites (“Macrocephalites” of the group of “M.”’ maya and “M.” transiens, Waagen), and Dhosaites (including his “Macrocephalites” elephantinus Waagen). These indi- cate beds of Argovian age. Some comparable examples in the British Museum, recorded in two papers by Mr. R. B. Newton, have already been referred to on a previous oc- cas.on.! The example figured in Newton’s pl. XIV, figs. 1-2 (loc. cit. 1889) collected by the Rev. J. Richardson, is not in the Baron Collection in the British Museum. The Stephanoceras macrocephalus, St. herveyt, and St. callo- viense, are poorly preserved, in a peculiar reddish, soapy matrix, and are probably Argovian Mayaites and Dho- saites, and not Callovian Macrocephalitids. Another speci- men, from North of Andranosamonta (B. M. No. C. 3586) preserved in a sandy limestone of a light colour, I have pre- viously described as possibly a new globose form of Dho- saites, with the blunt ribs that characterise the (equally Argovian) genus Tornquistes. ; An example of Haploceras elimatum (B. M. No. C. 3585b) and an indeterminable fragment of a Phylloceras (B. M. No. C. 3585a). were not referred to by Mr. Newton. Among his twelve fragments of “Perisphinctes polygyratus (Reinecke) those that can be determined are referable to Lithacoceras ? andranosamontx (Lemoine) (B. M. Nos. C. 3588a and c), L.? cf. colcanapi (Lemoine) (B. M. No. C. 3587a, 3588f) and doubtful Subdichotomoceras (C. 3587b, 3588e). The “Perisphinctes sp. (probably allied to P. polygyratus, Rein.)”’ recorded by Mr. Newton in 1895? I have previously * described as a Lithacoceras of the group of L. eystettense—fruticans (Schneid). 1 $path, loc. cit. (Pal. Indica, 1924), p. 10. Soe eit. (QO. J. G. S., 1895), p. 78. 8 Loe. cit. (Somaliland, 1924), p. 160. 28 BULLETIN 44 168 There is, then, nothing yet recorded that would defi- nitely indicate the presence, in Madagascar, of the Upper Callovian, the Divisian, a good deal of the Argovian, the. Lower and Upper Kimmeridgian, and, of course, the up- permost Jurassic (Portlandian and Tithonian). Whilst it is improbable that the Jurassic succession of Madagascar is more complete than that of other areas and some gaps are expected, it might yet be claimed that the absence of e.g. Callovian Cosmoceratids or Argovian Cardioceratids is due to differences of “province” and not to the non- representation of strata of those ages. The discovery re- cently of Virgatitids in Somaliland may justify the expecta- tion that e. g. Cosmoceratids also may yet be found in more southern deposits if there are anywhere beds that contain the true Peltoceras athleta and its zonal associates. Perisphinctoids, however, are still the commonest and often only representatives of Oxfordian faunas and offer great difficulty in specific identification. Reference also has already been made to some doubtful Oppelids, recorded by Lemoine, so that the probable gaps in the Madagascan sequence, in the present state of our knowledge, cannot be definitely indicated. In conclusion, it may be repeated that when only strictly contemporaneous formations are compared, the question as to whether there is affinity with forms of the Indian or Mediterranean Provinces ceases to have importance. Haug! pointed to the presence of the peculiar Sequenzi- ceratid genus Bouleiceras in Madagascar as possibly indi- cating a separate zoological province. The discovery, by the writer,? of a specimen of Bouleiceras nitescens, Theve- 1 Traité de Géologie, vol. II, fase. 2 (1907), p. 995. 2“Cretaceous Cephalopoda from Zululand.” Ann. South Afr. Mu- seum, vol. XII, part VII, No. 16 (1921), p. 272. 169 MADAGASCAR CEPHALOPODA 29 nin, in a Domerian-Toarcian collection from the Valley of Kelat, Baluchistan, dealt the death-blow to this speculative “ province,” as the record of Virgatitids, from Somaliland throws doubt on the value of the ‘Ethiopian Province” of Neumayr, Uhlig, Dacqué, and Krenkel. Upon the analogy between the Jurassic faunas of India and Madagascar. insisted on by Boule and Lemoine, more will be said in connection with the forthcoming revision of the Ammonite Fauna of Kachh. 30 FIG. 2. BULLETIN 44 EXPLANATION OF PLATE 170 PAGE Haploceras elimatum (Oppel). Middle Kimmeridgian, Ant- salova. a. Side-view of a small example; b. septate frag- ment, showing suture-lines; c. peripheral view of another completely septate specimen £0 chee onc ee Kheraiceras ? stansfieldi, sp. nov. Lower Callovian, (mac- rocephalus beds) Ankidabé. a. side-view of holotype, with umbilical suture of (missing) outer whorl drawn in from traces on side not shown; 6b. sectional outline of outer whorl. ghia: tar Jeg Bee Sa ein ee Catacephalites sp. ind. External suture-line (x 5) of a small specimen from near Ankidabé (macrocephalus beds). E = external lobe; L=lateral lobe; U = umbilical su- ture. The dotted lines represent the rib-curve. ; Catacephalites sp. ind. Sectional outline of another example from the same bed. «ao Say SS Torquatisphinctes ? cf. bangei (Burckhardt). Middle Kim- meridgian, Antsalova. Sectional outline of a fragmentary example. 6a, c,d. Sindeites madagascariensis gen. et sp. nov. Callovian, 6b. 10. Ankidabé. a. external suture-line (xX 6); c. sectional out- line (X 2); d. side view, diagrammatic (x 1%). Sindeites sp. nov. Callovian, Jessulmir, Sinde, India. (British Museum, No. C. 23545.) External suture-line (x 5) for comparison with fig. 6a. PS Macrocephalites madagascariensis, Lemoine. Lower Callo- vian, nr. Ankidabé. Sectional outline of a completely sep- tate specimen. oa) Sr Hemisimoceras semistriatum gen. et sp. nov. Middle Kim- meridgian, Antsalova. a. side-view; 6. peripheral view of holotype (genotype). so wl}. Ete Hemisimoceras nodulosum sp. nov. Same locality and bed. a. side-view; 6b. peripheral view of holotype. Paes Grossouvria ? ef. waageni (Teisseyre) Loezy sp. Callovian, Ankirihitra. Sectional outline of a fragmentary example. 15. 14 14° 18 10 10 ig 21 22 17 Des DicTutheite ONT. “INOWIVd “MHWY “T1Ing LL 10, ‘bd nd SW Sac seston: [ZTNOSIVY GSW W “Tad Tee Fay a ET Id G 3 . > eve . jee Wee ty ‘ ; 4 * ® \ Ms Me. . hae + en” ’ a A me VOL. IL III. Ty. VI. Vil. VIII, “ mollusea, Ep 38 eee ene roils NG: 4 Claiborne” fossils, 1 phy 52. pp. 5 “Tertiary mollusea, 5. pl.,. 30 p.; No. ‘Miss. faunas, Ky. and Ind., 110: pp.; No, 13, Ca ‘valley, 4 pl. : 26 pp.; No. 14, ‘Crown Point Sect -Hocene mollusea, Southern States, 5 pl, 26 p “12. pp.;, No. 26; Buda. Echinoidea, 9) pk, _~ Miocene mollusea,’ 3 pl., 82 pp.;_ No, TOs pes Ne. 33, New Tert. moll., 5 ae 32. PB. : recent-Pliocene mollusca, ae pl., 116 pp.;. -to. No. a 7140- Pp-}. No. 39, ets -mollusea,. ae pla 260 Bee oe eet eer Tex., 4 pl., 82 pp.; No. 4, Midway fossils, 1 No. 5: ‘Say’s: paleont. sepplthes te ply 104. No. e Ithaca fossils, 2 pl., 56 pp.; 3 No. 7, 110 ; No. 8, New Eoc. species, 0 pl., 26 pp.; Nc biva feu 14 pl., 102 pp.; No. 10, Atlantic slope’ for: 63 pp. Pea aiestet cae aigia mraeeseemiat ea aN No. 11, Lignitic gastropoda, EZ: pl: 128 p No. 15, Oligocene of West Europe. an 118 pp. ee es No. 16, Eocene of Georgia, 8 PP-; No. 17, and Newport, 17 -pp.; No, 18, “Mohawk. Calei r 4>pl., 24 pp.; No, 19, Helderberg invasion of the 9 pl., 27 pp.; No. 20, Jefferson. limestone, 4. 3e New Oligocene shells ‘of Florida, ts pl, 40 p 14 pp.; No: 24, Neocene, J vonian fossils, 2 pl, No. ‘25, Cranium | of. Bat esp plain, 5° pl..34 ‘pps No. 29 and '30, ie Donne 296. pp. No. 31, Mid-Eocene Poleeypoda, 5 59. ae ‘i Fa Guppy triary fossils, 10° pl., 198 pp.; aoe oe Recent and No.<87, New Eocene ‘Shells, 8 pl. 32 ay ica 208 48) ¢ a aey es cha RC Radi Bc Rota Costa: Riga curators 2 fs 18. pp: No. and Camb. fossils, 9 pl., 132 PP; No, 42, T BULLETINS OF AMERICAN PALEONTOLOGY Vol. x No. 45 VENEZUELAN AND CARIBBEAN TURRITELLAS Wirn a List OF VENEZUELAN TYPE STRATIGRAPHIC LOCALITIES ( Presented to the Graduate School of Cornell University in partial fulfillment of the requirements for the degree of Doctor of Philosophy) ‘ By FLOYD HODSON, A. M. Sune 17, 1926 Harris Co. Cornell University, Ithaca, N. Y. Wes. Ag INTRODUCTION This article is based on more than two years of field work in Venezuela unraveling the stratigraphy and paleontology of certain areas for one of the American companies operat- ing in that country. We propose here to describe some of the new material collected during this investigation with general reference to type localities and general age de- termination. At some future time when the interests of the company for whom the work was done permit, we hope to publish more definite information as to the exact locatity of each collection, the type sections, and the stratigraphic range and age determination of the species described. STRATIGRAPHIC FORMATIONS AND THEIR TYPE LOCALITIES The formations which we have found it convenient to use are recorded in alphabetical order and the type locality for each is indicated: ; 1. Agua Clara series: Chiefly black clay-shale. Type locality—around Agua Clara, District of Democracia, State of Falcon, Venezuela. 2. Cerro Pelado series: Alternating conglomeratic sand- stones, black shales, and coals. Type locality—Cerro Pe- lado range, Districts of Democracia and Miranda, State of Falcon. . 3. Codera formation: Variegated clays and sands. Type locality—just south of Codera de Dentro, District of Demo- cracia, State of Falcon. 4. Damsite series: Chiefly limestones, clays and soft sandstones. Type locality—-Damsite at Caujarao, south of Coro, on Rio Coro, District of Miranda, State of Falcon. 5. El Paraiso shales: Black shales with interbedded coals and quartzitic sandstone layers. Type locality—E] Paraiso, in Quebrada El] Paraiso, District of Bolivar, State APR 97 1939 4 BULLETIN 45 ee of Falcon. 6. Guarabal conglomerates: Conglomeratic sandstones, black shales and siliceous limestones. Type locality—just southwest of Guarabal, south of Coro, on the Coro-San Luis trail, District of Miranda, State of Falcon. 7. La Puerta series: Variegated, gypsiferous clays and sands. Type locality—-La Puerta, southeast of Dabajuro, District of Buchivacoa, State of Falcon. 8. La Vela series: Variegated, gypsiferous clays and sandstones; the latter are sometimes conglomeratic. Type locality—In Rio or Quebrada La Vela (also called Mata- ruca) about one mile northeast of La Vela de Coro, District of Colina, State of Falcon. 9. Misoa Trujillo series: Massive sandstones, frequently conglomeratic, interbedded with black shales. Type local- ity—Misoa Trujillo ranges, in eastern part of State of Zulia. 10. Querales shales: Chiefly black clay-shales with some sandstone members. Type locality—Querales, (southwest of Sabaneta), District of Miranda, State of Falcon. 11. Patiecitos series: Conglomeratic sandstones, black shales and siliceous limestones. Type locality—Quebrada de Los Patiecitos, near Las Alambiques, on the Coro-Cabure trail, Districts of Petit and Miranda, State of Falcon. 12. Pauji shales: Chiefly black shale with interbedded dense sandstone layers. Type locality—Rio Pauji, eastern part of State of Zulia. (Pauji shales of authors in part). 13. San Luis series: Foraminiferal limestones, inter- bedded with shales and sandstones; limestones and sand- stones often conglomeratic. Type locality—San Luis range at point where the Coro-Cabure trail crosses it, District of Petit, State of Falcon. 14. Socorro series: Gypsiferous sandstones and varie- gated gypsiferous clay-shales. Type locality—Socorro, south of Urumaco, District of Democracia, State of Falcon. 175 VENEZUELAN TURRITELLAS eS ACKNOWLEDGMENTS Mr. S. H. Williston, of the Venezuelan Sun Company, brought his collection of Falcén fossils to Cornell Univer- sity and described them here in the summer of 1923. His manuscript was not published, but where feasible, we have redescribed his species using his name and have given him credit for the species. Where possible, we have quoted his exact descriptions, but in most cases, due to our much larger collections, we have been able to supplement his work. In the cases where we have figured his specimens, acknowl- edgment is given in the explanation of plates. es Among the many people to whom acknowledgments are due for help and suggestions in this work, the following deserve special mention: Professor G. D. Harris, for help in the field and in his splendid laboratory at Cornell University where there are extensive collections and library facilities for the study of the Tertiary formations of the Western Hemisphere. Among the geologists to whom we are grateful for col- lections and help are Messrs. B. Hubbard, R. A. Liddle, C. R. Nichols, A. Andreas, Jr., W. A. Argabrite, L. G. Weeks, A. L. Owens and P. O. Olcott. My assistant, during the latter half of the field work, Mr. G. A. Weaver, was invaluable in the instrument and map work. Above all, I am indebted to my wife for invaluable aid in the field and in the laboratory, as well as in checking my work and manuscript. rel rcl Neel deek ve BULLETIN 45 aig TURRITELLA HOLOTYPES The following holotypes are figured in this Bulletin: PAGE wuliana’’(n.. sp.) Pl. lj tig. Alice eeeccacd cee 8 zuliana palmeri (n. subsp.) Pl. 2, figs. 1, 4.........2...... i) larensis (n. Sp%)sPIi, 3), files, Dec..eeie- ee eeatecnec eee 10 larensis santiagana (n. subsp.) PI. 4, fig. 3.............. ‘12 larensis carrizalensis (n. subsp.) PI. 6, fig. 3.............. 12 larensis guaratarensis (n. subsp.) PI. 10, fig. 3.......... 12 robusta Grzybowski fredeai (n. subsp.) PI. 7, fig. 7.... 18 hubbardi (nm. sp.) Plo 8) fig. Siccc.cciceccsce cece eee 14 hubbardi weeksi (n. subsp.) Pl. 10; fie, 4.2 eee 15 cauredalitoensis (Williston MS.) Pl. 11, fig. 3............ is cauredalitoensis liddlei (n. subsp.) Pl. 11, fig. 6........ 16 cauredalitoensis dabajuroensis (n. subsp.) Pl. 11, fer Al Bedi cnadesseweSeveadianlonete mean seaise Ack eee 17 cauredalitoensis filensis (n. subsp.) Pl. 11, fig. 2...... 1 gilbertharrisi: (m. sp.) Pl 13) fies 3h. ee 1% gilbertharrisi staufferi (n. subsp.) Pl. 15, fig. 6........ 19 gilbertharrisi aguavivensis (n. subsp.) Pl. 15, fig. 1 19 gilbertharrisi falconensis (Williston MS.) Pi. 14, PG. :saccnntoewseiiadiansaapadeeont Aan seeeeeee en eee 19 montanhitensis (n. sp.) Pl. 16, fig. 10; Pl. 17, fig. 8.... 20 montanitensis olcotti (n. subsp.) Pl. 20, fig. 9............ PA! montahitensis saladilloensis (n. subsp.) Pl. 16, fig. 2 22 curamichatensis (n. sp.) Pll 17, fig. 2i..c.21-c-.- esse 22 gatunensis Conrad lavelana (n. subsp.) Pl. 19, fig. 7 23 gatunensis Conrad willistoni (n. subsp.) Pl. 18, fig. 2 25 gatunensis Conrad taratarana (n. subsp.) Pl. 18, Pi IO soe ccc ad boas od Secasakensseeeoeeee eee eetnaee ee 25 berjadinensis (mn. sp:) PI. 19, ties 4...-2...---..ee 26 berjadinensis colinensis (n. subsp.) Pl. 19, fig. 8...... 27 berjadinensis warfieldi (n. subsp.) Pl. 21, fig. 6......... 27 77 VENEZUELAN TURRITELLAS 7 altilira Conrad urumacoensis (n. subsp.) Pl. 26, fig. 7 44 altilira Conradmirandana (n. subsp.) Pl. 28, fig. 10.. 45 PAGE T. berjadinensis socorroensis (Williston MS.) Pl. 21, Pate, BR Oe ee eS 28 T. berjadinensis cocoditana (n. subsp.) Pl. 19, fig. 5...... 24) Semeinyee (m. Sp.) Pl 28) He. Vee ccsctscseccsconsesarenes 2 T. variegata Linné es (iasSubSsps), elaecl PURSUE eRe Ceo as TSA ct deca euay ohceta ee aaeeacvine uae saseereont 31 T. plebeia Say A-L-Owensi (n. subsp.) Pl. 28, fig. 2...... 1 Pemenarucana Cn. sp:) Plo, fie. Oinc.....c..sccctedessscoonseseees Bie Pemenentelrina (n.vsp.) Pl. 22) fio! 6....:...2c0.ctsieccnccneses (oese ae T. venezuelana quirosana (n. subsp.) Pl. 22, fig. 9........ 34 T. venezuelana watkinsi (n. subsp.) Pl. 22, fig. &.......... 34 Beeerenvednvert (a. Sp.) Pl. 22) fi 13..c.sccscseecenccesenemeeeess 34 meenrnellana (nm. sp.) Pil. 24, fig. 14... ccccpececcceeeee cooenees 3) T. cornellana bolivarensis (n. subsp.) Pl. 24, fig. 13...... 335 BIRTHS I) bE let 24 TIP, Oxacces cast chose Sockawsedte Soceeeseaoanent 335) meagoareasi (Williston MS.). Pl. 24, fig. 8.............0.-...00 on Melacarinenensis (nm. sp.) Pl. 25, fig. G............c..ceceeeene ) Beene hivacoana (n. Sp.) Pl. 25, f1@. 9.2..2....ccccccaseecesoeeesees 3) T. buchivacoana cafionensis (n. subsp.) Pl. 25, fig. 1.... 40 Beeunencnsis (n. sp.). Pl. 25, fig. L1...2...... cess. eceeeceeeseeeeee Al Beeienoyiecossmann, Pl. 27, fio. Lic iscci..c..cccceecceccoseaseneweass 42 T. guppyi Cossmann morantensis (n. subsp.) Pl. 26, SIE. 1S) qe g coh CSCI EEN cA EE Ro ep on ee £4 PeecemarremCme Spa) Plo27, fig. 1k acscccc.cccesssdssisenssescesssss 43 aT. iT. MS TER Te GTI 9) Py AB TID. Oooo. cece a noc st edendeseedavestocehosnace ce 46 T. vistana nicholsi (Williston MS.) PI. 29, fig. 5........... 48 T. bifastigata Nelson maracaibensis (n. subsp.) Pl. 30, UML PSE SPIN ns eg hS Sonos cc oSaad sindaaneds as whahuarmoteartic 48 T. bifastigata Nelson democraciana (n. subsp.) PI. 30, CLL. 3 2S) MASSA CA ORNs Aan Dini oe ee SS Lame 50 LECTOTYPE The following lectotype is figured: murritella bifastigata Nelson, PI. 30, fig. 1.....1.............00 48 8 BULLETIN 45 178 DESCRIPTIONS OF SPECIES MOLLUSCA CLASS GASTROPODA Order Ctenobranchiata Schweigger Suborder Platypoda Lamarck Superfamily Tzenioglossa Bouvier Family Turritellidz Gray Genus Turritella Lamarck Turritella zuliana, n. sp. Pl. 1, figs. 1-6; Pl. 2, figs. 2) 3;)5aNGn Shell of medium size, with large apical angle; whorls with scalloped carinz, very concave in adult stages. Protoconch consists of about 114 smooth round whorls. First two nepionic whorls are sharply medially carinate, angularly convex and almost smooth; the next three nepionic wnorls become less convex, of which the first 114 whorls carry two strong spiral threads about equally spaced behind the keel; the next 114 whorls show sub-microscopic spiral threads in addition to four primary spirals; the primaries include,— the two spirals posterior to the keel, the thread which forms the keel in earlier whorls, and a thread which is developed on the anterior of the whorl just behind the suture. The sides of the first neanic whorls are almost flat but soon be- come concave; the basal or first anterior primary cord rap- idly increases in size and begins to form the carina which is so conspicuous in the larger whorls; the concave neanic whorls carry four, about equally spaced, strongly beaded, large spiral cords with weaker threads of varying strength appearing in the concave portion between the posterior and anterior spirals as well as behind the posterior cord. 179 VENEZUELAN TURRITELLAS 9 The adult whorls are decidedly concave with a produced scalloped flange or carina decorated with a reticulate, super- ficial ornamentation which likewise covers the remaining surface of the whorls; the spiral ribbing in the concave por- tion is less pronounced and more subdivided; the sigmoidal, close-set, prominent growth lines are strongly retractive on the upper half of the whorl, swing forward to the edge of the keel and are again strongly retractive on the base of the whorls. The sutures are compressed and rather incon- spicuous, except in the gerontic stages when a wide V- shaped depression may lead down to the suture. The keel or carina in the adult whorls usually projects out straight from the base, but frequently, even in the larger whorls of the same specimen, the keel may turn down anteriorly or up posteriorly or may be moved posteriorly from the base almost a third of the whorl length. In the latter case, the base of the keel is slightly concave and ornamented with numerous spiral threads. This is one of the most ornate of our Venezuelan species. It dces not bear a very close resemblance to any. other form. The large, scalloped, knobby keel is very characteristic. This species is easily distinguished from the more slender T. montanitensis (n. sp.) by the much longer whor! lengths of the latter, as well as by an entirely different character of spiral ribbing in all stages. Age: Oligocene-Miocene. Locality: District of Miranda, State of Zulia. Locality Numbers: 6, 1140. Turritella zuliana palmeri, n. subsp. Pl. 2, figs. 1, 4. This subspecies is characterized by the more numerous (6-7) rather prominent, heavy, spiral cords instead of 4, the number of primary spirals usually found on the closely related species. This increase in the number of spiral cords is probably due to the fact that the second anterior cord of T. zuliana IO BULLETIN 45 180 (n. sp.) became divided into two primaries and to the fact that some of the secondaries have increased in strength. Named in honor of Dr. Katherine Van Winkle Palmer. Age: Oligocene-Miocene. Locality: States of Zulia and Falcon. Locality Num- bers: 6, 1140, 2027. Turritella larensis, n. sp. Pl. 3, figs. 1-5; Pl. 4, figs. 1, 2, 4, 5; Pl 5. fig. 4. This Turritella is very abundant, very variable, and sel- dom well preserved. In examining several hundred speci- mens, we have found great variations in all its features; to try to establish close varieties or subspecies from the many variations would be hopeless. A few subspecies have been made to mark the limit in variation of some particular feature. Intermediate forms are considered as intergrada- tions between species and subspecies. Shell is rather large, heavy, and keeled in adult stages. Protoconch and first nepionic whorls were not found; later nepionic whorls are almost flat or slightly convex, with one strong, sub-medial, spiral cord, which in some specimens is double due to the appearance of a cord of variable strength just posterior to it. Immediately behind the su- ture, there is a faint anterior spiral which rapidly increases in strength in the succeeding whorls; there is a strong upper spiral on the posterior fifth of the whorl; on each side of the medial spiral rib there is a concave interspace in which weaker spiral lines appear; the medial or posterior inter- costal area rapidly increases in size and carries 2 or 3 smaller, beaded, secondary spirals and some tertiary threads. In the adult whorls, the weak, anterior, post-sutural spiral of the younger whorls becomes a wide flange or keel of varying width, ornamented with fine revolving threads. The adult whorls are more or less concave and the sides are ornamented with 3—5 (usually 4) stronger, beaded, spiral ribs and weaker intervening spirals; the strong sig- 181 VENEZUELAN TURRITELLAS Il moidal growth lines are retractive from the suture to the medial concavity, then curving forward, are protractive to the keel or carina and retractive on the base of the whorl; the superimposition of the growth lines over the spiral ribs forms beads or nodes on all the spiral cords; the finer super- ficial ornamentation has usually disappeared in slightly weathered specimens; the suture is overhung by the keel which generally projects more or less straight from the base of the whorl, but may turn up or down; if the keel turns up posteriorly, it is rounded on the base; if it droops anteriorly, it is concave on the base and tends to conceal the suture; although the keel may be very pronounced, it is regular in width and does not become scalloped as in T. zuliana (n. sp.); well preserved specimens show fine, superimposed, spiral threads on the keel. It slightly resembles the pagoda-like T. cawredalitoensis filensis (n. subsp.) but is so different that a comparison is not worth while. It is much closer to T. systoliata Dall } from the Tampa silex beds and probably is derived from the same stock. The flange or keel on our specimens is more pronounced, the shape of the whorl and sculpture is somewhat different, but the ensemble shows a striking sim- ilarity. It must be remembered that there are all degrees of varia- tion between the species and its subspecies as well as between the subspecies themselves. The subspecies are established only to mark the limit of variation in certain directions. Age: Oligocene-Miocene. Locality: States of Lara, Falcén and Zulia. Locality Numbers: 10 (variation), 797, 811 (slender variation), 814 (with variations), 815, 822 (with variation), 1627 (and subspecies), 1760, 1761, 1940, 1955, 1958, 2018, 2019, 2019A, 2021 (and variations), 2022 (and variations), 2023 (and variations), 2024 (with variations), 2027, 2048 1W. H. Dall, Bull. 90, U. S. Nat. Mus., p. 99, Pl. 9, fig. 6, 1915. I2 : BULLETIN 45 182 (variation). Turritella larensis santiagana, n. subsp. Pl. 4; figs. Sher 5, fis This subspecies differs in that in the later ephebic and gerontic stages it tends to lose its spiral ribbing and to be- come more nearly smooth. The whorls are concave and show more or less faint spiral striation even on weathered specimens. We are figuring a few intermediate forms to show its relation to the species, otherwise, it might be hard to believe that they belong to the same stock. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcén. Lo- cality Numbers: 1760, 1761, 2019, 2023 (a variation), 2027, 2048. Turritella larensis carrizalensis, n. subsp. Pl. 65 figss Ieeoeeas This subspecies is characterized by having three very prominent primary spirals on the adult whorls; it is the antithesis of the new subspecies guaratarensis in respect to the number of strong primary cords; it is a very large heavy form, however, and the anterior primary spiral tends to develop into a strong keel. It is the limit in variation of the species toward a few, very strong, pronounced spiral cords on the sides of the whorl; this subspecies is generally very similar to the species, but tends to be slightly larger and to have larger spiral cords. Age: Oligocene-Miocene. Locality: States of Zulia and Falcon. Locality Num- bers: ‘10, 1290, 1898 (2), 2022, 2023) 2027@eees 2049 (cir Turritella larensis guaratarensis, n. subsp. Pl. 10, figs: due This subspecies is coarser and heavier in size and sculp- ture. The whorls are more nearly flat-sided behind the basal flange. The adult whorls usually carry 5 or more strong, primary, spiral ribs and sevéral weaker spirals of varying strength. 183 VENEZUELAN TURRITELLAS 13 This subspecies marks the limit in variation toward havy- ing a large number of heavy spiral cords and developing a heavy shell. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcon. Lo- cality Numbers: 1988, 2021 (a variation), 2023, 2027. Turritella robusta Grzybowski fredeai, n. su sp. Ik, By tess ily Bs 1 Gs mess, Ay He IR 5 wes. IL Bp 8 1k O. ines, Well Aki, ies, (6. The characteristics of T. robusta Grzybowski from Peru have been well reviewed by Spieker.' Our form is very closely related to this species, but is easily distin- guished by the flatter, straighter-sided whorl behind the keel and by the much larger size of the shell. Depend- ing on the size of the whorl, there are 6—9 spiral cords of varying strength and spacing behind the sharp keel; behind the anterior suture, there are 3 pre-carinal cords in the younger whorls; but in the large adult whorls there are 3 or 4 strong spirals with several weaker interven- ing ones in well preserved specimens. Behind the keel the growth lines are nearly straight, running diagonally from the posterior suture almost to the keel where they curve and take a vertical direction to the anterior suture; on the base of the whorl they are slightly protracted and cross about a dozen spiral threads of varying strength, giving a faint reticulate ornamentation. The pre-carinal spiral cord ad- jacent to the suture is rather faint in the younger whorls, but in the larger adult whorls it becomes very strong and produces a prominent shoulder which projects over the suture. Named in honor of Mr. M. F. Fredea, of the Standard Oil Company of Venezuela, who collected some of our larg- est specimens. 1K. M. Spieker: The Paleontology of the Zorritos Formation of the North Peruvian Oil Fields, pp. 84-85, Pl. IV, fig. 5. The Johns Hopkins Press, 1922. 14 BULLETIN 45 184 This subspecies somewhat resembles T. larensis (n. sp.) and its subspecies, but is easily distinguished by the flat straighter sides of the whorl behind the keel and by the straight growth lines. It attains a much greater size than any of the other Venezuelan species. Age: Miocene. Locality: Common in the State of Falcén. Locality Numbers: 70A, 70C, 72, 72C, 81, 83, 84, 86, 87, 90, 100, 149A, 149B, 169A, 210, 215, 307, 1238, 1398, 1600, 1856, 1875, 1892, 1928. Turritella hubbardi, n. sp. Pl. 7, figs. 2-5; Pl. 8, figs. 1-6; Pl 9) fies Sees Shell rather small, turreted, with two, strong, equal or subequal, basal, carinating, spiral cords, and a weaker pri- mary spiral near the top of the whorl in the neanic and later stages. Between the upper spiral and the posterior basal cord, except in the youngest whorls, there is a pro- nounced concavity in which lie two secondary and one or more tertiary spiral threads. No protoconch was found; the nepionic whorls carry a prominent medial rib which develops into the posterior basal carina; also, a weak rib is present or suggested adjacent to the anterior suture and another one on the upper fourth of the whorls. In well preserved specimens, the whole surface of the shell is” highly ornamented by numerous, strong, sub-microscopic, spiral lines, and crossed by prominent growth lines which cause transverse nodes or prominences on all the spiral cords; in the larger whorls, the strength of the nodes in- creases and the stronger carine have a decidedly beaded appearance; the strong growth lines have only a moderate curve on the posterior portion of the whorls. The nepionic portion of the shell is smoother and less ornate. The suture is not conspicuous and is slightly overhung by the anterior basal carina; the carina is ornamented with conspicuous, spiral, sub-microscopic lines (which are less strong near 185 VENEZUELAN TURRITELLAS 15 the suture) and crossed by forward swinging growth lines. The ornamentation of this Turritella is somewhat similar to that of T. zuliana (n. sp.) and T. gilbertharrisi (n. sp.). This species is very easily distinguished from T. zuliana which has a wide, scalloped, basal keel, and from T. gilbert- harrisi which has less concave whorls and a more aciculate spire. The two, strong, equal or subequal, basal cords easily distinguish this species from other Venezuelan Turritellas. The holotype is a fragment of about 514 whorls which: measures about 3 mm. at the upper suture of the uppermost whorl and 3.7 mm. through the middle of it; maximum diameter of the largest whorl is about 8.8 mm.; the great- est length of the fragment is 22 mm. Named in honor of Dr. B. Hubbard, Chief Geologist of the Standard Oil Company of Venezuela. Age:. Oligocene-Miocene. Locality: Rather common in District of Buchivacoa. State of Falcén. Locality Numbers: 15, 16, 25,. 1216, 1629, 1762 (cf.), 1943 (cf.), 2018, 2019, 2019A, 2040. Turritella hubbardi weeksi, n. subsp. PING) nes bls LOmheS. 4.n0: This subspecies differs in having a tendency to double the uppermost, primary, spiral cord; also, the posterior of the two basal carinze becomes weaker in the adult whorls. Named in honor of Mr. L. G. Weeks, Geologist for the Standard Oil Company of Venezuela, who helped collect the material. Age: Oligocene-Miocene. Locality: States of Zulia and Falcon. Locality Num- bers. 2018, 3250. Turritella cauredalitoensis (emended from Williston’s MS.) PL (5 kes (HG IL sree 2, Zls 124k, Al() pes PAS IEMA wal, stresy al, By G. Shell of medium size and taper; whorl lengths compar- atively short; whorls concave in adult stages. Later ne- 16 BULLETIN 45 186 pionic whorls are ornamented with three, strong, regularly spaced, subequal spirals; the anterior one is adjacent to the suture; the middle one is slightly stronger than the others, with a concave interspace on each side of it; the spiral ribs and interspaces even in these young whorls are ornamented with many, fine, microscopic, spiral lines and crossed by sub-microscopic, curved, growth lines which are retractive to the posterior interspace and then protractive to the suture; in all except the youngest whorls, one or two strong secondary spirals appear in the posterior concavity and several sub-microscopic spiral lines in the anterior con- cavity ; the whorls rapidly become concave and the anterior ~ rib forms a small keel which projects over the compressed and inconspicuous suture; at the points where the growth lines cross the spiral cords, beads or nodes are formed, but unless the specimens are well preserved, this superficial ornamentation of the spiral cords does not show; in the adult specimens, the posterior concavity of the nepionic stage has become medial and carries two or three secondary spirals; behind the posterior spiral rib on the adult whorls, there sometimes appear one or two weaker spirals; the keel is usually distinctly ornamented with fine, superficial, spiral threads and crossed transversely by elevated growth lines. This species somewhat resembles T. zuliana (n. sp.) but does not have the scalloped, knobby keel of the latter. The widely concave whorls, the rather small size of this Turri- tella, the comparatively short whorl lengths easily distin- guish it from T. larensis (n. sp.) and other Venezuelan forms. Age: Oligocene. Locality: District of Buchivacoa, State of Falcon. Lo- cality Numbers: 1084, 1139, 1159, 1159A, 1159B, 1159C, 1159H, 1217, 1953 (one specimen). Turritella cauredalitoensis liddlei, n. subsp. PI. 9, fig. 3; Pl. 11, fig. 6. This subspecies differs in the adult whorls in that the 187 VENEZUELAN TURRITELLAS 17 base of the keel or flange is rounded and the edge of the keel turns up posteriorly, causing the whor! to be more con- eave. The entire shell carries many revolving spirals of varying strength. Named in honor of Mr. R. A. Liddle, now Divisional Geologist with the Pure Oil Company. Age: Oligocene. Locality: District of Buchivacoa, State of Falcén. Lo- cality Numbers: 1159K, 1987. Turritella cauredalitoensis dabajuroensis, n. subsp. TAL, abil, anes Ae This subspecies differs in having much larger and coarser primary spiral cords. It approaches some of the varieties of T. larensis (n. sp.) in the character of ribbing in the neanic stages, but the younger and older whorls are not confusable. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcén. Lo- cality Numbers: 1627 (?), 2027. Turritella cauredalitoensis filensis, n. subsp i fer 2 124k alle, artes, ln By Ge This subspecies differs in favine pagoda-like keels or ex- tremely wide carinating flanges. The growth lines are prominent; the entire shell is ornamented with fine spiral threads. The concave whorls with extremely wide, pagoda- like, sharp keels make this a unique subspecies. Age: Oligocene. Locality: District of Buchivacoa, State of Faleén. Lo- cality Numbers: 1217, 1979. Turritella gilbertharrisi, n. sp. IPAL, IVA. reds, Pa, ZL (Be TEAL ALS fies. l—Geble 14 fess ae ye Shell slender, of moderate size; spire very much attenu- ated. Protoconch consists of 134, convex smooth whorls, which continue into the nepionic stage; the first 4 nepionic whorls carry a single, sharp keel near the anterior third of the whorl; the total length of the 4 whorls is about 1.7 mm. Each of the succeeding 7 whorls has a strong spiral on the 18 BULLETIN 45 ; 188 upper quarter, with a posterior slope or bevel between it and the suture; in this stage, the keel of the earlier whorls becomes more nearly sub-medial with a concave interspace on each side of it; a slightly weaker spiral appears just above the suture at the base of the visible portion of these whorls; the posterior concave interspace begins to show one or more secondary spirals; the spirals even in this young stage have a tendency to be beaded at the intersection of the growth lines; the length of the 7 later nepionic whorls is slightly less than 8.5 mm., with a maximum diameter of about 2.8 mm. The neanic whorls continue with approxi- mately the same taper; the posterior concave interspace of the nepionic whorls becomes the slightly concave medial por- tion of the later whorls, and usually carries two, beaded, subequal, secondary spirals, and finer, intervening, tertiary threads. In the later adult whorls, the sides are slightly concave between the two posterior cords and carry 2—4 sec- ondary, and numerous tertiary, spiral threads; just behind the posterior cord, there sometimes appear one or two, fainter, spiral cords; the basal cord varies tremendously in strength in different specimens; usually it is low and weaker than the other two posterior, larger, primary spi- rals; all the larger cords and interspaces are ornamented with several, smaller, superimposed, finer spiral threads. The whole surface of the Turritella, in all stages except the very youngest, is covered with irregular, elevated, growth lines, which have a tendency to slightly offset the numerous, fine, microscopic or sub-microscopic, costal and intercostal threads and grooves which cover the whole surface of the whorl, including the spiral cords. The suture is compressed and inconspicuous; the close-set growth lines are strongly retractive from the posterior suture to the concave portion of the whorl, in which they curve forward and are pro- tractive to the anterior spiral cord, which they cross at al- most right angles. This rather long, slender Turritella shows a wonderful 189 VENEZUELAN TURRITELLAS 19 microscopic ornamentation. The species is not likely to be confused with any of the others. It has much less prominent spirals than T. hub- bardi (n. sp.) and is, also, much more aciculate or attenuate. Named in honor of Professor G. D. Harris of Cornell University. Age: Oligocene-Miocene. Locality: Rather common in the States of Falcén and Zulia. Locality Numbers: 6, 741, 777 (cf.), 919, 1140, 1414, 1415, 1419 (cf.), 1901 (one specimen only), 1944A, 2019, 2023 (7). Turritella gilbertharrisi staufferi, n. subsp. 124K ali), figs. 6, 9. This subspecies differs from the species only in that the middle rib, or second anterior spiral, is doubled and is formed by two, complementary, united, spiral cords. This is scarcely more than a variation but represents a geo- graphical difference. Named in honor of Dr. Stauffer of the Caribbean Petro- leum Company. Age: Oligocene-Miocene. Locality: District of Miranda, State of Zulia. Locality Number: 6. Turritella gilbertharrisi aguavivensis, n. subsp. Pl. 15, figs. 1, 8. This subspecies differs in having an intervening riblet between the two, anterior, primary, spiral cords. This in- tervening rib sometimes attains enough strength to give the appearance of three instead of two anterior primary spirals. Age: Oligocene-Miocene. Locality: States of Falcon and Zulia. Locality Num- bers: 6, 2019. Turritella gilbertharrisi falconensis (Williston MS.) 12 , uigs. 2,6, o; Pl. 1b, figs. 8, 4, 6, 7; Pl 28, fir. 5: This subspecies is much larger and coarser in appearance 20 BULLETIN 45 ~ 190° than, the species and has larger spiral ribs. There are four, large, spiral cords instead of three in all except the younger stages; the various weaker spirals found between the two, posterior, primary cords in T. gilbertharrisi (n. sp.) have been replaced in this subspecies by a fairly strong spirai, which gives this Turritella the appearance of having four main spiral ribs; there is sometimes the suggestion of a fainter secondary spiral near the middle of the whorl. In the adult whorls, between the posterior cord and the suture, there are 1-3 weaker secondary threads (usually 2 in num- ber). Well preserved fragments show the same kind of superficial ornamentation observed in T. gilbertharrisi (n. sp.), the beading being more accentuated in the larger speci- mens of the subspecies. The anterior spiral frequently pro- jects over the suture as a weak keel. Age: Miocene. Locality: Common in the State of Falcén. Locality Num- bers: 100, 101A, 104, 273, 339, 341, 342, 798 (ef-); 10Ges 1067, 1127, 1265, 1398, 1436, 1704, 1810 (7), 18025 shea 1820 (poorly preserved), 1825, 1827, 1900, 1901 (also con- tains one specimen of T. gilbertharrisi, sensu stricto), 1903, UGPAS, IU BINE, USIMGS We WS sb Turritella montanitensis, n. sp. Je WW sates Bio IRI, ilsy, ines, is IP, it. figs. 4, 5, 6, 8, 10; Pl. 17, figs. 3, 7, 8, 10, 11. Shell turreted, slender, with a knobby basal carina on larger whorls. Protoconch is missing; first nepionic whorls are convex with a sub-medial, V-shaped carina; they are smooth except for faint, microscopic, posterior and anterior, spiral cords, which can be seen very early in these whorls; the posterior spiral cord is located approximately on the upper fifth of the whorl; the anterior spiral cord lies im- mediately behind the suture; the cords are beaded by the intersecting, curved, growth lines. In the subsequent whorls, the anterior spiral forms a strong cord, and in the adult whorls becomes a prominent, basal, knobby carina; finer, intervening, spiral threads are found on all except the IgI VENEZUELAN TURRITELLAS 21 youngest whorls; between the posterior primary and second anterior primary cord, there is a wide concave area with ‘minor spiral lines which are usually beaded; the primary and secondary cords are beaded by the intersecting growth lines; the suture is moderately well exposed below the over- jutting knebby keel; the strong primary spirals, and es- pecially the first anterior of these which forms the keel or carina, are ornamented with many, fine, revolving spirals crossed by prominent growth lines; on the sides of the whorl, the growth lines recede sharply to the middle of the concave area, then swing forward to the keel, and then retract on the base of the whorl, giving the incremental lines a sigmoid shape. The strength of the primary spirals is rather variable as well as the spacing between them. The keel is usually basal and sometimes droops anteriorly, more or less covering the suture. The scallops or knobs on the. keel are more numerous but smaller than on T. montaniten- sis olcotti (n. subsp.) ; the knobs are more conspicuous on the base of the keel than on the posterior side of it, because they are prolonged as prominences, taking the direction of the growth lines to the suture, and giving the base a fluted appearance. The number of knobs on the keel varies, but as an average, there are about 16 knobs on a whorl 15 mm. in diameter. This species has longer whorl lengths than any of our other Venezuelan forms. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcén. Lo- cality Numbers: 25, 52, 56, 811 (cf.), 1114 (cf.), 1939, 1957, 1996, 2015A, 2031, 2050, 2052. Turritella montanitensis olcotti, n. subsp. IPAL, Ia, anes, A123 Pl. 20, fig. 9. This subspecies is less slender than 7. montanitensis, and is particularly characterized by having fewer but larger knobs on the keel. The number of knobs on the keel of a whorl varies somewhat in different specimens, but as a gen- eral average, there are about 10 knobs on a whorl 20 mm. 22 BULLETIN 45 192 in diameter. Named in honor of Mr. Perry Olcott of the Standard Oil Company of Venezuela. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcén. Lo- eality Numbers: 1639, 2050, 2051, 2052. Turritella montanitensis saladilloensis, n. subsp. Pi. 116; fetes This subspecies differs in having more concave whorls, in having the posterior, primary, spiral weaker, and in hay- ing the second, anterior, primary, spiral cord stronger, than in T. montanitensis, sensu stricto; also, the two an- terior primaries tend to be closer together. The carina overhangs and tends to conceal the suture. Age: Oligocene-Miocene. _ Locality: District of Buchivacoa, State of Falcon. Lo- cality Number: 1957. Turritella curamichatensis, n. sp. Pl. 16; figs: 1,35) 72)93ee Pl. 17, figs. 1, 3, 4=65588 Shell medium sized with strong spirals and a deeply ex- cavated suture. Early neanic whorls carry two, strong, subequal, sub-medial, spiral cords, the lower of which grad- ually becomes more prominent and forms the second an- terior spiral of the adult whorls, while the upper one be- comes gradually weaker, changing from a primary cord in the neanic and early ephebic stages to a strong secondary one in the lower part of the medial concavity of the later ephebic and gerontic stages; the posterior part of each whorl is beveled and slopes into the excavated suture; the anterior margin of the beveled area is limited by a strong spiral cord which is found in all stages except possibly the nepionic which is not represented in our collection; the bev- eled area usually carries one or two, beaded, spiral threads; the anterior spiral, just behind the suture, becomes stronger in the adult whorls and forms a basal keel which projects over the excavated suture; the later nepionic and earlier 193 VENEZUELAN TURRITELLAS 23 neanic whorls are convex with concave interspaces between the spirals; in the adult whorls there is a wide, more or less concave, medial depression between the second anterior and the posterior primary spirals; this concavity is present in most specimens and carries two or more secondary and some tertiary threads; the whole surface of the shell is crossed by prominent growth lines which cause knobs or beads at their intersection with the spirals; the two, an- terior, primary spirals are frequently covered with numer- ous, sub-microscopic, superimposed, fine threads with nar- row intervening grooves; the crossing of these threads and growth lines gives a reticulate structure especially notice- able on the base of the first primary cord or basal flange; the growth lines are sigmoid in shape, being retractive on the upper part, protractive on the lower half of the visible portion of the whorls, and retractive on the base of the whorls; the knobby ornamentation of the ribs is emphasized by the protuberance of the growth lines over the primary spiral cords, particularly on the basal flange. This species, when well preserved, shows somewhat the fancy, superficial ornamentation found on T. hubbardi, T. gilbertharrisi and other of our Venezuelan Turritellas; it is easily distinguished from these and other similar species by the deeply excavated suture. It differs from T. larensis, with somewhat similar ribbing, by being smaller in size, and having differently shaped whorls: in T. larensis the adult whorls are concave, do not have the excavated suture, and have a much wider keel. . The medial concavity is variable in this species and in a few cases, the sides of the whorls have become almost flat. Age: Oligocene-Miocene. Locality: Common in the State of Falcon. Locality Numbers: 15, 440 (cf., poorly preserved), 760 (cf.), 775 mae (10 (cf.), 919, 1070, 1143, 1408, 1414 (cf.), 1436, 1437, 1754, 1761 (one specimen), 1944A. Turritella gatunensis Conrad lavelana, n. subsp. JPA alts, dikes (Oe Wb aR) onary, Ye 24 BULLETIN 45 194 This subspecies is distinguished by slight but constant variations from JT. gatunensis Conrad and T. gatunensis caronensis Mansfield. In all except the earliest whorls, this subspecies is distinguished from gatunensis, sensu stricto, by a stronger medial carina; the strong posterior spiral just behind it is closer and usually without the intervening spiral threads found in the species; in a few cases, we have found a very weak intervening thread; the proximity of the medial carina and the strong spiral just posterior to it gives a strong medial prominence, which causes the lower part of the whorl to be more nearly perpendicular to the base; behind the medial prominence there is a sharp uniform con- striction to the suture; in front of the prominence there is a pronounced, depressed, spiral cingulation, limited anteriorly by a primary spiral which forms a shoulder overjutting the suture; this gives the Turritella a bicarinate appearance; the cingulation is ornamented with more numerous, smaller, subequal spirals than in the species. The visible portion of the base of the whorl usually carries one or two strong spirals with wide interspaces which sometimes carry a faint thread; the corresponding area in most of our adult specimens of T. gatunensis Conrad from Gatun shows three or more, more closely spaced spirals between the anterior suture and the anterior primary spiral; also, the upper fourth of the whorl tends to be slightly more concave than in our subspecies. This subspecies is distinguished from T. gatunensis caronensis Mansfield * by the weak medial carina and the more uniformly convex whorls of the latter. Age: Miocene. Locality: Common in the State of Falcén. Locality Numbers: 67, 70A, 100 (?), 150A, 150B, 184, 185, 273, 298, 314 (7), 1033, 1067, 1078, 1265, 1901. 1W. C. Mansfield, Proc. U. S. Nat. Mus., Vol. 66, Art. 22, pp. 51-52, Plate 8, figs. 12-14, 1925. 195 VENEZUELAN TURRITELLAS to on Turritella gatunensis Conrad willistoni, n. subsp. Pl. 18, figs. 2—4, 8. We have not yet found the nepionic stages of this form, but it seems to be a subspecies of 7. gatunensis Conrad. The constriction at the top of the neanic whorls is narrower and consequently seems deeper; our subspecies is a smaller form and has fewer secondary and tertiary spiral threads; in some specimens the whorls become very roundly convex in the ephebic and gerontic stages with a rounded area be- tween the sides and base of the whorl; in this case they approach T. gatunensis caronensis Mansfield (l. c.) in the general shape of the whorl! but not in the character of the ribbing. Our subspecies has three or four strong primaries with fewer secondary and tertiary spirals than occur in either the species or any of the other subspecies. The pri- mary spirals are stronger and the intervening threads weaker than in similar forms. Named in honor of Mr. S. H. Williston of the Venezuelan Sun Company. Age: Miocene. Locality: Common in the State of Falcén. Locality Numbers: 80 (cf.), 93, 94, 97, 149A (?), 178, 184, 225A, 291, 303, 317 (cf.), 325, 1031, 1033, 1255, 1265, 1856, 1858 (cf.), 1866, 1900, 2054. Turritella gatunensis Conrad taratarana, n. subsp. Pl. 18, figs. 5, 7. This subspecies is a further development of the new sub- species, lavelana. It is characterized by gentle scallops on the primary spirals which gives a knobby or beaded appear- ance. The scallops or beads are elongated transversely, 7. e., in the direction of the spiral ribbing. All of the strong spirals may show the scalloped character, but it is more pronounced and sometimes found only on the first strong anterior spiral. The number of scallops on a spiral varies, but usually there are 16-18 on a whorl 7 mm. in diameter. The knobby character begins to appear early in the neanic whorls. Age: Miocene. 26 BULLETIN 45 196 Locality: Districts of Colina, Miranda, and Democracia, State of Falcén. Locality Numbers: 80, 93, 184, 185, 193, 201083, 123255. LOM Turritella berjadinensis, n. sp. Pl. 18, fics 15" Piggy figs. 1, 3, 4, 6. Shell is of medium size; sides of the whorl] are almost flat except for two spiral prominences or keels, the posterior of which is a strong, medial or sub-medial, spiral rib with wide, flat or very shallow, concave areas on each side of it; the other spiral prominence forms a basal swelling on the anterior portion of the whorl, just behind and overhanging the suture; this basal prominence is decorated with several spiral ribs of varying strength; the flat or concave areas behind each of the keels, also, are ornamented with spiral ribs and on the larger whorls have weaker spiral threads alternating with the stronger; the total number of spirals on the visible portion of the whorls varies in different indi- viduals and in different stages of the same individual; for example, the smallest whorl of the type specimen has 18 visible spirals and the largest has about 30. In some speci- mens there is a fairly strong spiral developed in the anterior concave area just behind the basal prominence. The base of the adult whorls is ornamented with spiral ribbing sim- ilar to that on the sides of the whorls, but usually shows a greater number of weaker intervening spirals between the stronger. The growth lines are not very conspicuous; but, on weathered specimens, they are straight and retractive on the upper portion of the whorl; they swing to an axial direction on the lower part of the whorl and cross the base at approximate right angles to the spiral sculpture. The base of the whorls is flat or slightly convex. The inside of the shell has several strong lirze which leave spiral grooves on the casts. The upper part of the whorl] is sometimes more or less constricted just in front of the posterior suture. This is an extremely variable species which grades into its subspecies, which in turn merge into each other, and 197 VENEZUELAN TURRITELLAS BF sometimes approach 7. mimetes Brown and Pilsbry ' and similar forms which have been previously described from the Caribbean region. T. berjadinensis, sensu stricto, most closely resembles T. boweni (n. sp.) but the latter is distinguished by the fewer, larger, more widely spaced spirals, which are more uniform in size on the upper half of the whorl. The whorls are shorter and flatter than in T. mimetes Brown and Pilsbry. Age: Miocene. Locality: Common in the State of Falcén. Locality Numbers: 24, 74A, 82, 90, 93, 150B, 270, 317, 830, 1000, 1078, 1111, 1115, 1231, 1232, 1447, 1630, 1757, 1855, 1856, 1858, 1861, 1892, 1911, 2036. Turritella berjadinensis colinensis, n. subsp. Pls howe Plo iiss This subspecies is characterized by the lack of carinating spiral threads. The whorls are almost flat and have the same spiral ribbing as the species but less accentuated. This subspecies bears a close resemblance to T. mimetes Brown and Pilsbry (loc. cit.) in all stages, but the young whorls show it to be more closely related to T. berjadinen- sis. It is distinguished from 7. mimetes by its shorter flat- ter whorls and finer spiral ribbing. Age: Miocene. Locality: Common in the State of Falcon. Locality Numbers: 71, 74 (variation), 79, 80, 82, 83A, 84, 85, 90, 93, 97, 100, 103, 118, 149A, 150A, 150B, 180, 184, 185, 187, 189, 193, 204, 206, 215, 229, 298, 299, 300, 307 (variation), 1007 (variation), 1033, 1043, 1064, 1255, 1335, 1507, 1552, 1901, 1928. Turritella berjadinensis warfieldi, n. subsp. 12415 PAD aikes (E TAL, Pals sakes, (33: This subspecies differs from the species, sensu stricto, in that the medial portion of the whorl] carries two, instead of 1Proc. Acad. Nat. Sciences of Philadelphia, 1911, p. 357, EL XXVIII, fig. 1. 28 BULLETIN 45 : 198 one, prominent spirals; also, the development of the sec- ond strong spiral on the posterior third or fourth of the whorl tends to make the upper part of the whorl more con- stricted than in the species. The development of a strong secondary spiral just behind the basal prominence is more common than in T. berjadinensis, sensu stricto. The adult whorls approach T. mimetes Brown and Pilsbry (loc. cit.) in general appearance; the ribbing on the lower half of the whorl is, indeed, very similar, but on the upper part of the whorl, T. mimetes has three or more strong spirals, instead of two. T. mimetes has longer adult whorls and the younger whorls are much more convex than in T. berjadinensis war fieldi. Named in honor of Mr. Wm. Warfield, formerly with the Standard Oil Company of New Jersey. Age: Miocene. Locality: Common in the State of Falcén. Locality Numbers: 83B, 84, 88, 98, 96, 98, 118, 187, 150B, 152, 155, 169A (cf.), 178, 184, 185, 187, 193 (ef.), 206, 215, 225; 270} 307, 830, 1007, 1033, 1043, 1066, 1111 (cf.), 1232, 1335, 1447, 1449, 1450, 1509, 1552, 1757, 1870 (cf.), 1892, 1905. Turritella berjadinensis socorroensis (Williston MS.) Pl. 21, fig: 3; Pl. 23, fic. 83 Ela This subspecies is a further development in the direction taken by the new subspecies, 7. berjadinensis warfieldi. It tends to be a larger, heavier variety, having three, strong, spiral cords and a fourth weaker spiral between the two anterior primaries. There is a pronounced constriction around the upper fourth of the whorl in front of the pos- terior suture. The spiral ornamentation of the base is poorly preserved. This subspecies is retained because it marks the limit in the variation of this stock of Turritella toward a large, heavy, strongly ribbed shell. Age: Miocene. Locality: Principally in the Districts of Democracia and 199 VENEZUELAN TURRITELLAS 2y Miranda, State of Falcén. Locality Numbers: 100, 103 (cf.), 104 (variation), 122B (cf.), 149A, 149B, 169A, 171, meeei'74, 270 (cf.), 301, 314, 325, 1017, 1127, 1931, 1932, 1934 (cf.). Turritella berjadinensis cocoditana, n. subsp. TEL, ae), sale, He IL, Bo). AR, Bi Ty UE This subspecies is distinguished by having the upper half of the whorl much more constricted and by having a wider basal flange just behind the suture. The whole surface of the shell, when well preserved, bears more numerous and finer intervening spiral threads. The young stages show nodes on the upper half of the whorls elongated in the di- rection of the growth lines. The base bears about eight equally spaced spiral ‘‘welts’’ with three or more smaller intervening threads. Age: Miocene. Locality: District of Falcon, State of Falcon. Locality Number: 2207. Turritella planigyrata Guppy Pl. 19, figs. 2, 9. Turritella planigyrata Guppy, Sci. Assoc. Trinidad Proc., Vol. 1, Pt. 3, pp. 169-170, 1867. Turritella planigyrata Guppy, Geol. Mag., London, Vol. 1, N. S., p. 408, pl. 18, fig. 5, 1874. Turritella planigyrata Guppy, Quart. Journ. Geol. Scc., London, Vol. 32; p. 519, 1876. Turritella planigyrata Guppy, Agr. Soc. Trinidad and Tobago (So- ciety Paper No. 440) p. 11, 1910. Turrritella plangyrata Guppy, Agr. ‘soc. Trinidad and Tobago, Society Paper No. 444), Vol. 10, p. 451, 1910. (fide Mansfield). Not Turritella planigyrata Guppy, Maury, Bull. Amer. Pal., Vol. 5, pp. 293-294, pl. 48, fig. 14, 1917. Turritella planigyrata Guppy, Maury, Bull. Amer. Pal., Vol. 10, eco, pl. 42. fies. 6, 7) 8, 1925. Turritella planigyrata Guppy, Mansfield, Proc. U: S. Nat. Mus., 1925. No. 2559, Viol. 66) Art, 22, pp. 55-57, pl. 9; figs: 1, 9: We have a single, small, young specimen of this species. It shows a slight variation in the early neanic whorls in developing slightly stronger spirals than are found in the corresponding whorls of our specimens from Trinidad. It is the only Venezuelan specimen which we can refer un- questionably to T. planigyrata Guppy. 30 BULLETIN 45 - | 2a Age: Miocene. Locality: State of Delta Amacuro, Eastern Venezuela. Locality Number: 1131. Turritella maurye, n. sp. JI Pasi, ines, lil Turritella planigyrata Guppy, Maury, Bull. Amer. Pal., Vol. 5, 1917, pp. 293-294, Pl. 48, fig. 14. Shell rather small, turreted; spirals scalloped or beaded. Protoconch consists of about 134, smooth convex whorls; the first 3 nepionic whorls are convex with a medial carinating spiral; in front of the carina, there are about three equally spaced primary spirals with intervening spiral threads; be- tween the medial carina and the posterior suture there are about three primaries, the middle one of which is stronger than any of the other spirals on the whorl] except the medial carina; between the primaries, there are intervening, weaker, spiral threads. The number of spirals increases on the succeeding whorls, and all of them, especially the pri- mary ones, become scalloped or knobby; the number of scal- lops varies, but on the last whorl (8.5 mm. in diameter) there are about 17 or 18 small scallops. On the larger whorls, there are three about equally spaced spirals which are stronger than the others; between these, there are in- tervening, secondary spirals, with still weaker tertiary threads; on the posterior slope, there are three or four less prominent spirals. The sides of the larger whorls are mod- erately convex; the base is almost flat and is spirally striate with 2—5 weaker spirals between the stronger ones. The growth lines are not conspicuous. The inside of the shell is reinforced by about eight spiral lire of varying strength. Attention was drawn to this species in the synonymy given by Mansfield! for T. planigyrata Guppy. The finely scalloped spirals easily distinguish this species from T. planigyrata Guppy from Trinidad. Named in honor of Dr. C. J. Maury. 1W. C. Mansfield: Proc. U. S. Nat. Mus., Vol. 66, 1925, Article 22, p. 256; 201 VENEZUELAN TURRITELLAS 31 Age: Miocene. ‘ Locality: Bluff 2, Cercado de Mao, Santo Domingo. Col- lected by Dr. Maury, expedition of 1916. Turritella variegata Linné paraguanensis, n. subsp. PI. 21, figs. 2, 7. This subspecies is very close to T. variegata Linné which is so common in the recent faunas along the north coast of Venezuela. It differs in having stronger ribbing, especially on the base of the whorls, which are corrugated with about six very prominent ridges; the corrugations and interspaces are overridden with numerous, finer, spiral threads. This subspecies increases in size more rapidly than the recent species shown in PI. 22, fig. 7. Age: Quaternary. Locality: District of Falcon, State of Falcon. Locality Number: 1504. Turritella plebeia Say A-L-Owensi, n. subsp. PI. 20, figs. 1, 2, 5, 6; Pl. 23, fig. 2; Pl. 28, fig. 1. This subspecies differs principally in attaining a much greater size; the larger whorls are less convex; the posterior slope is much longer and less inclined. The spirals become more nearly equal in size, the channeled interspaces being as wide or wider than the spirals. The base of the whorls is somewhat flattened and ornamented with many fine spirals. T. plebeia Say from Jones Wharf, Maryland, is figured for comparison, PI. 21, fig. 5. Named in honor of Mr. A. L. Owens, Geologist for the Standard Oil Company of Venezuela. Age: Miocene. Locality: Districts of Colina and Democracia, State of Falcon. Locality Numbers: 71, 84, 90, 96, 150A-B, 150B, mit (?), 193 (and variations), 225A, 298. Turritella matarucana, n. sp. VEAL PAU), ante Abe eal, ils ineasy. iL. (8): Shell turreted, with long flattened whorls, ornamented with many spirals. Protoconch and nepionic whorls are missing. Neanic whorls are roundly convex; the posterior BULLETIN 45 202 iS) N slope is more gentle than the anterior; whorls are orna- mented with about twelve flattened spiral ribs of varying strength, wider than the interspaces. The succeeding whorls become less convex and almost flat-sided except for the rounded posterior and anterior slopes. In the large adult whorls the number of spirals increases due to the interpolation of intervening spirals, or the width of the spirals increases, or both may occur on the same whorl; that is, some of the flat spirals become uniformly wider on the succeeding whorls, or narrower, intervening, secondary ~ spirals may appear, or the ribs may become wider with only a few intervening weaker spirals; the proportion of the wide ribs to the narrow is not at all constant, but in well preserved specimens all of the ribs are flat, and they are wider than the groove-like interspaces; in the adult whorls the number of ribs varies from 15—22 depending on their width. The growth lines are decidedly curved on the visible portion of the whorls, but we have not seen the sculpture on the base. The suture is excavated. The wide spirals with narrow interspaces on almost flat- sided whorls make this a distinctive species. Age: Miocene. Locality: Districts of Colina and Democracia, State of Faleén. Locality Numbers: 71, 194A, 197, 199, 1027, 1250. Turritella venezuelana, n. sp. Pl. 21, figs. 4, 8; Pl. 22; figs) ieee Shell is small, turreted, and rather slender. Protoconch consists of 134, small round whorls. Nuclear whorls are continuous with the nepionic whorls which are sub-medially bicarinate, with a slightly convex, long, posterior slope; an- terior slope, short, steep, and slightly concave; the second nepionic whorl shows a trace of a microscopic spiral on the anterior third of the posterior slope; the later nepionic whorls develop another spiral thread on the upper third of the posterior slope; the spiral ribs are sharply elevated; the 203 VENEZUELAN TURRITELLAS 33 interspaces are wide and concave. The neanic and suc- ceeding whorls of this abundant species begin to show many variations in the shape of the whorl, and in the character and spacing of the spirals. The commonest form has a rather flat tapering whorl with an overjutting beveled base sloping down sharply to the suture; the second anterior primary spiral forms the apex of the angle between the long, tapering, posterior part of the whorl! and the narrow, steep, anterior slope leading down to the suture; the long pos- terior part of the whorl carries four elevated spirals with wide concave interspaces; the interspace between the 2 large posterior primaries is usually wider than the others; just in front of the suture on the posterior slope, there are one or two secondary spiral threads; about the middle of the steep anterior slope there is a fairly strong spiral. The surface of the shell is usually covered with a shiny enamel. The growth lines are very inconspicuous, but are slightly retractive on the sides and base of the whorl. The base is flattened and ornamented with numerous fine spirals. There are innumerable variations in this species. The shape of the adult whorls varies tremendously, even in the same individual. The larger whorls frequently betome roundly convex instead of being angulated behind the an- terior slope. In some specimens, secondary threads appear in some of the wide interspaces, and other threads near the sutures. This small form somewhat resembles the much larger species, 7. subgrundifera Dall1 from the Chipola beds, which has its nuclear whorls oblique to the axis of the shell. The inside of the shell gives very faint traces of internal liree. Age: Oligocene-Miocene. Local’ty: District of Miranda, State of Zulia. Locality Numbers: 6, 810 (?), 811, 814, 815, 822, 1628, 1754 (cf.), 1Wm. H. Dall: Trans. Wagner Free Inst. of Science, Vol. III, Part 2, 1892, pp. 313-14, Pl. 22, fig. 23. 34 BULLETIN 45 204 1761, 1942, 1955, 2010, 2019, 2019A, 2021, 2022, 2023, 2027. Turritella venezuelana quirosana, n. subsp. Pl 225) fisss Ose in Pl 24 ticks This subspecies is characterized by the tendency to inter- polate an intervening secondary spiral thread between the primaries. The whorls vary tremendously in convexity; the moderately flattened, convex, tapering, adult whorls, are more common than the very roundly convex ones; the secondary threads may appear between most or all of the primary spirals. Fig. 1 on Pl. 24 shows a freak in which in- tervening secondary spirals suddenly appear after an injury to the shell. Age: Oligocene-Miocene. Locality: States of Falcon, Lara, and Zulia. Locality Numbers: 6, 814, 815, 822, 2019, 2022, 2027. Turritella venezuelana watkinsi, n. subsp. Pl 22) ficiaes This subspecies represents a common variation toward roundly convex whorls without intervening secondary threads between the primaries. Adult whorls become less convex than the younger but do not show intervening spirals. Named in honor of Mr. W. A. Watkins, formerly of Cor- nell University, who helped collect some of this material. Age: Oligocene-Miocene. Locality: District of Miranda, State of Zulia. Locality Number: 6. Turritella G-A-Weaveri, n. sp. Pl. 22; figs. 3) 5s sele2er fig. 9; Pl. 24, fig. 4. Shell rather small, slender, with convex whorls in young stages, and somewhat flattened whorls in adult stages. Protoconch and first nepionic whorls are missing ; succeed- ing nepionic whorls have a medial carina with the anterior and posterior slopes about equal in length and steepness; posterior slope carries 2 (and later 3) about equally spaced spiral threads; anterior slope carries 3 (and later 4) spirals 205 VENEZUELAN TURRITELLAS 35 of which the middle one soon becomes stronger than the others and approaches the strength of the medial carina. The convex neanic stages are bicarinate, one keel being medial and the other weaker one half way between it ana the anterior suture; secondary spirals appear between the primaries; typically, the intervening thread just behind the medial keel is very faint or lacking. The spiral sculpture of the neanic whorls resembles that mentioned in the ne- pionic stage with additional threads of varying strength appearing on the upper half of the whorl. The succeeding whorls become less convex with flattened sides, and bear 7 or 8 subequally spaced strong spirals, some of which are larger than others. The base of the whorl, ornamented with several revolving lines, is rather flat and projects over the succeeding whorl. Growth lines are inconspicuous. Largest whorl found measures 10 mm. in diameter. This Turritella most closely resembles some of the flat- sided adult whorls of T. venezuelana (n. sp.). It is easily distinguished by the single medial carina of the nepionic whorls in place of the double of the latter species. Named in honor of Mr. G. A. Weaver who was an invalu- able collaborator in the field work during 1924—5. Age: Oligocene-Miocene. Locality: District of Buchivacoa, State of Falcén. Lo- cality Numbers: 47, 51, 52, 2010. Turritella cornellana, n. sp. Pl. 22, figs. 2, 4; Pl. 24, figs. 11, 14. Shell is very slender, ornamented with many spirals of varying strength; shape of whorl and prominence of rib- birg are very variable. Protoconch and earliest whorls are m'ssing. Later nepionic whorls are convex and medially carinate; posterior slope is flat or convex and ornamented w:th several spiral striz of varying strength; anterior slope, flat or concave with spiral ornamentation similar to that on the posterior slope. On the neanic and succeeding whorls the sides become almost flat and carry four about equally spaced primaries with two or more intervening secondary 36 BULLETIN 45 206 threads; anterior slope is shorter and steeper than the pos- terior; both slopes carry a spiral ornamentation similar to that on the sides; the suture is excavated; in the adult whorls, the overjutting base eliminates an anterior slope; the base of the whorl is flattened and seems to have a spiral sculpture similar to that on the sides of the whor!. The growth lines are retractive on the upper half of the whorl and take an axial direction over the lower half; prominent beads or nodes are formed on the upper half of the whorl at the intersection of the spirals and the growth lines; de- pressions in the spirals and interspaces are sometimes noted on either side of the elevated growth lines which cause the nodes. : Age: Oligocene-Miocene. Locality: District of Miranda, State of Zulia. Locality Number: 3222. Turritella cornellana bolivarensis, n. subsp. Pl. 24, fies! 25.3. oles This subspecies is less slender, more strongly ribbed, and has a more pronounced overjutting shoulder at the base of the whorl. The posterior slope is slightly concave in the adult whorls; all the spirals may be strongly beaded in this subspecies. Age: Oligocene-Miocene. Locality: District of Miranda, State of Zulia. Locality Number: 3222. Turritella boweni, n. sp. Pl. 24, figs: 5; 6, 10= P25 ieee Shell small, bicarinate except in the younger stages; adult whorls almost flat with about 5 equally spaced pri- maries on the upper half of the whorl. Protoconch and early nepionic stages are missing. Later nepionic whorls are sub- medially carinate; the posterior slope is longer and more gentle than the anterior; both are ornamented with a few spirals. Neanic whorls are convex and bicarinate; the an- terior carina is less prominent and located just behind the anterior suture; there is a wide concave interspace between the 2 carinee, which usually shows a secondary intervening 207 VENEZUELAN TURRITELLAS 37 thread; posterior slope is almost flat and carries 4 equally spaced, elevated, primary spirals behind the medial carina. Succeeding whorls have about the same sculpture as the neanic, but are less convex and become nearly flat-sided ; the bicarinating spirals of the earlier whorls are not so prominent in the adult stages, and form the 2 anterior primary spirals with a wide flat interspace between them; this interspace usually carries one intervening secondary spiral; between the anterior primary and a secondary thread that lies near the anterior suture, there is a narrow. beveled or slightly concave, steep, anterior slope with a faint spiral near its middle. The growth lines are seldom seen, but are slightly retractive on the visible portion of the whorl. The scuipture on the base of the whorl has not been seen. This species somewhat resembles T. venezuelana (n. sp.) and T. berjad‘nensis (n. sp.); it is distinguished from the former by the wide flat interspace between the 2 anterior primaries of the adult whorls; from the latter, by the ab- sence of the numerous, intervening, secondary threads and by the fewer spirals in the adult. Named in honor of Mr. C. F. Bowen, Chief Geologist of the Standard Oil Company of New Jersey. Age: Oligocene-Miocene. Locality: Districts of Buchivacoa and Acosta, State of Falcon. Locality Numbers: 725, 1217. Turritella andreasi (Williston MS.) Pl. 24, figs. 7-9, 12; Pl. 25, fig. 2. Shell is of moderate size with convex whorls. Protoconch and nepionic whorls were not found; earliest whorls pre- served have a diameter of about 2.3 mm.; they are convex, ornamented with about seven primary and secondary spirals with some weaker microscopic intervening threads; the an- terior slope is slightly steeper than the posterior and car- ries several fine spiral threads. The adult whorls are almost uniformly moderately convex, with 9-11, sub- equally spaced, primary spiral cords of varying strength; there are usually one or two intervening threads between 38 BULLETIN 45 2a the spirals on the lower half of the whorl; in some speci- mens an intervening thread is found between some of the spirals on the upper part of the whorl; the primaries tend to be somewhat more closely spaced on the posterior than on the anterior slope; the anterior slope always carries at least one primary spiral about midway between the anterior suture and the first anterior primary on the sides of the whorl; the wider interspaces frequently carry two or more intervening, sub-microscopic, spiral threads. The growth lines are s.ckle-shaped, being strongly curved on the upper half of the whorl and crossing the lower half and base of the whorl more or less at right angles to the spiral ribbing. Sutural line is rather inconspicuous. In the intermediate stages it is sometimes difficult to tell this species from T. filacarmenensis (n. sp.) but the young and adult stages are easily distinguished; in the young stages, T. filacarmenensis is much more convex and has stronger, fewer, primary spirals; in the adult whorls, T. filacarmenensis has a pro- nounced excavated anterior slope without prominent spiral ribbing. Named in honor of Mr. A. Andreas, Jr., Geologist for the Standard Oil Company of Venezuela. Age: Oligocene. Locality: Common in the State of Falcon. Locality Numbers: 31, 32, 40, 66, 508 (and variation), 802, 1048, 1058, 1061, 1084, 1130 (?), 1135 (variation), 1139, 1159, 1159B, 1217, 1638 (variation), 1664 (and variation), 1664B (variation), 1951 (variation), 19538, 1963, 1964, 1987. Turritella filacarmenensis, n. sp. Pl. 23, figs. 1,°53 3B) figs. 4, 6, 10, 12. This Turritella closely resembles T. andreasi (Williston MS.) in the intermediate whorls, but is very different, es- pecially in the younger stages. Protoconch is missing: nepionic whorls are sub-medially bicarinate, with a long gentle posterior slope carrying a faint spiral, which later forms a third primary. spiral on the upper part of the 209 VENEZUELAN TURRITELLAS 39 neanic whorls; in the later neanic whorls there are four, about equally spaced, strong, primary spirals with an inter- vening sub-microscopic spiral thread in the interspaces; some weak secondary spirals appear on the posterior slope and the deeply concave anterior area carries very faint spiral threads. The number of primaries increases in the adult stages and the whorls become more gently convex. In the larger whorls there are about 9 primaries of vary- ing strength and many weaker intervening threads; the visible portion of the base of the whorl forms a concave, narrow, anterior slope which does not carry any prominent spiral ribbing and is limited anteriorly by a faint spiral just behind the inconspicuous suture; this spiral forms a sharp shoulder between the flat concealed base of the whorl and the sides. The sculpture on the base of the whorls is not well preserved but has spiral ribbing more or less like that on the sides of the whorl. The growth lines are not conspicuous, but on our weathered fragments they seem to resemble the incremental lines of T. andreasi (Williston MS.). It is distinguished from T. andreasi (Williston MS.) which it most closely resembles, by the absence of any strong spiral ribbing in the anterior concave area in the adult stages and by the fewer and stronger spirals in the younger stages. Age: Oligocene. Locality: District of Buchivacoa, State of Falcon. Lo- cality Numbers: 19538, 1964, 1987, 2000. Turritella buchivacoana, n. sp. 12, sip nes, WS IL Bs, aires, 7/8). Shell is small; whorls are ornamented with crowded minute spirals and 3 strong primary spirals,— the first of which forms the basal carina, the other two are on the convex portion of the whorl. Protoconch consists of about 2 smooth round whorls, continuous with the succeeding stages. Nepionic whorls are medially carinate, with a pos- 40 BULLETIN 45 210 terior spiral appearing on the midgle of the posverior slope and an anterior spiral at the base of the visible portion of the whorl just behind the suture; the whole surface of these and succeeding whorls is covered with minute or micro- scopic spirals of varying strength. In weathered speci- mens a deep groove frequently is seen at the posterior base of the anterior spiral. The neanic and succeeding adult whorls change in shape due to the development of a sharp basal carina which juts over the sutural line at the base of the whorl; the sides of the whorls are slightly convex with two primary sp.rals occurring near the medial line; the in- terspaces between the spirals are concave; the area between the posterior primary spiral and the posterior suture is somewhat concave and about equal in width to the anterior concave interspace between the middle and basal primary spirals; the medial concavity is slightly narrower than the others; a strong secondary spiral is sometimes developed in the anterior concavity. Weathered specimens often seem to have four primaries, instead of three (with a strong secondary in the anterior interspace); the sutures fre- quently appear deeply excavated in weathered specimens. This species does not closely resemble any of the others. It bears a superficial resemblance to T. gatunensis Conrad, but the numerous fine spirals easily distinguish it, as does the sharper basal flange. Age: Oligocene-Miocene. Locality: Common in the State of Falcon. Locality Numbers: 257, 400, 1042,.1070, 1282, 1408, 1627, 1628, 1629, 1722, 1761, 1935 (cf.), 1936, 1942, 1955, 2013.20 2019A, 2022, 2023, 2027, 2031. Turritella buchivacoana canonensis, n. subsp. Pl. 25, figssoemes This subspecies differs in having less prominent micro- scopic spiral striation, in having a less pronounced basal keel and in having a less deeply excavated suture. The diminutive basal keel is the most characteristic and dis- t_nctive feature. The sides are somewhat more convex 211 VENEZUELAN TURRITELLAS 4I than in T. buchivacoana, sensu stricto. Age: Miocene. Locality: District of Miranda, State of Falcon. Local- ity Numbers: 341, 1630, 1792. Turritella elmenensis, n. sp. VAL, PAS) wake, able Whorls finely striate; the strong spiral on the anterior third of the whorls gently scalloped; apical angle large; proteconch and earliest whorls missirg. Smallest whorl on the holotype is 2 mm. in diameter; it carries a strong, sub-medial, carinating spiral, with a concave interspace on each side; the posterior of these interspaces is wider than the anterior and is limited by a spiral on the posterior fourth of the whorl; the anterior concave interspace is limited by a spiral which is adjacent to the sutural line. The posterior slope is steeper in the young stages. In the succeeding whorls the medial carina increases in relative strength; it moves forward to the anterior fourth of the larger whorls where it becomes gently scalloped; the pos- terior primary remains on the posterior fourth of the whorls and the anterior one remains just behind the suture. The whole surface of the Turritelia, including the primaries, is covered with minute spirals of varying strength; there _are about 50 of these minute spirals on a whorl 9 mm. in diameter. In the largest whorls, some of the minute spirals have become relatively stronger than the others, and might be termed secondary spirals. The base of the whorl is slightly concave and ornamented with numerous, fine, revolving lines. The abundant, fine, spiral striation and the gently scal- loped sub-medial keel make this a unique species. Collected by the Miranda Exploration Company. Only one specimen has been found. Age: Oligocene-Miocene. Locality: Near Quir6s, District of Miranda, State of 42 BULLETIN 45 ; 212 Zulia. Turritella guppyi Cossmann morantensis, n. subsp. Pl. 26, figs. 3, 5, 6, 83 Plo 28) homes Protoconch consists of 114 smooth convex whorls, fol- lowed by two mono-carinate nepionic whorls in which the carina is located on the anterior third of the whorl; the posterior slope is longer and more gentle than the anterior ; length of the first eleven whorls, 7.7 mm. ; all the succeeding whorls are bicarinate with single, beaded ribs; the upper half of the wide medial concavity carries three or four . spirals of varying strength; the lower half of the concavity seldom shows any prominent spiral ornamentation. The posterior slope carries a secondary spiral thread about mid- way between the upper cord and the posterior suture. The shell is thin, and rather attenuate; the growth lines are not very prominent, but are retractive to the middle of the whorl and protractive on the lower half; the beads on the spiral ribs and threads are at the points where the growth lines cross them and are elongated in the direction of the growth lines. Well preserved specimens show minute spirals over the whole surface of the shell including “he tops of the spiral carine. This subspecies seems to differ very little from the cast and photograph (Pl. 27, fig. 1) which Dr. R. 8S. Bassler of . the U. S. National Museum very kindly made for us of the type of T. guppyi Cossmann! (equals T. tornata Guppy) from Cumana, Venezuela. It may be that they are identical but we believe it advisable to consider this Bowden form as a subspecies until better material is~ collected from Cumana. The distinctive characters are very slight, but the pos- terior slope in T. guppyi is more gentle and is not as con- cave as in the subspecies morantensis. The shell of the subspecies seems to be more delicate and attenuate than suggested by the cast of the type of T. guppyt. 1M. Cossmann: Revue Critique de Paléozoologie, 1909, p. 225. Br3 VENEZUELAN TURRITELLAS 43 Age: Miocene. Locality: Bowden, near Bowden Wharf, Morant Bay, Jamaica. Locality Number: 1109. Turritella carlottz, n. sp. IRL PAG), (ne, AAS 12 Pirig aie, all Turritella tornata Guppy, Maury, Bull. Amer. Pal., 1917, Vol. 5, p. 294, Pl. 48, fig. 15. Shell is thin, turreted, rather small, with two spiral car- inating prominences. Protoconch is missing; first two nepionic whorls, smooth, very convex due to a strong angu- lated carina on the basal third of the whorls; on each side of the carinating ridge, there is a slope to the sutures; the posterior slope is much more gentle than the anterior; the two succeeding nepionic whorls are not angulated but are roundly convex with a strong spiral on the basal third, and with two or three fainter intervening ones distributed be- tween it and the posterior suture; all the succeeding whorls carry a medial, concave, spiral depression with a promin- ence on each side of it; the anterior prominence is a single, spiral, carinating rib near the anterior third of the whorls; it is higher and sharper than the posterior -prominence which is a wide, rounded ridge occupying most of the upper half of the whorl and surmounted by two separate beaded spirals of about equal strength; near the middle of the medial concave area, theré are two or three secondary spir- als; between the lower cord and the anterior suture there is a concave area. The whole shell carries microscopic spiral ribbing, which is especially noticeable in the anterior concave area and on the posterior slope. In the larger whorls, there is a spiral formed at the base of the anterior concave area, just behind the anterior suture. The growth ‘lines are inconspicuous. ° This Turritella is distinguished from T. altilira Conrad by having a greater apical angle and by having a lower, wider, rounded, posterior, carinating prominence sur- mounted by two weak spirals. It is distinguished from T. guppyi Cossmann and its new subspecies morantensis by A4 BULLETIN 45 214 carrying two equal spirals on the rounded upper promin- ence, instead of one strong spiral with a weaker one on the posterior slope. It further differs from T. guppy? Coss- mann morantensis (n. subsp.) in having a greater apical angle. Named in honor of Dr. C. J. Maury. Age: Miocene. Locality: Rio Gurabo at Los Quemados, Santo Domingo. Specimens collected by Dr. Maury, expedition of 1916. Turritella altilira Conrad urumacoensis, n. subsp. Pl. 26, figs. 4, 7; Pl. 27,, figs: 3siipae This abundant Twrritella differs so much from our adult specimens of T. altilira Conrad collected from Gaiun, C. Z., that we were inclined at first to make it a new species. Unfortunately, these long slender forms are seldom pre- served except as fragments. We did not find the proto- conch or the earlier nepionic whorls of the Venezueian form, but the later nepionic whorls agree so closeiy w.th some of the Gatun specimens that we are considering this as a subspecies of T. altilira Conrad until the young stages are found to establish its relationship with certainty. The adult whorls show a closer similarity to T. guppyi Coss- mann or its new subspecies morantensis than to T. altilira Conrad, but it is distinct from each of them. The greatest diameter of the smallest well preserved whorl in our collection is about 3.7 mm.; the posterior spiral rib has an auxiliary spiral about the middle of its lower side and this tends to make it somewhat larger than the anterior keel in the young stages; there is a secondary spiral in the center of the concave area in the middle of the whorl; near the posterior kase of the anterior spiral there is another secondary spiral; in addition to the two primary carinating ribs, there are three secondary spirals with fainter intervening threads in the medial concave area; sub- microscopic spirals are to be found over the entire surface of the whorl in well preserved specimens; the later nepionic 215 VENEZUELAN TURRITELLAS 45 whorls bear a close resemblance to the corresponding stages of T. altilira Conrad, sensu stricto. For comparison we are figuring specimens of T. altilira from Gatun, C. Z. (Loc. No. me) in Pl. 26, fig. 1; Pl. 28, fig. 3; Pl..29, fig. 1. The neanic and ephebic stages of the subspecies do not show a very close similarity to the species because in the former the two carinating spirals are always single; the posterior cord does not show any tendency to become double or to be strengthened by a strong auxiliary cord on its lower side; the two large spiral keels are about equal in size, but the anterior one tends to be slightly smaller. The medial concavity is slightly wider because the upper rib is narrower than in T. altilira Conrad; three or four weaker secondary spirals are found in the middle of this concavity and on the lower part of the posterior keel; the upper side of the anterior keel does not seem to bear any prominent spiral — ornamentation. It differs from T. guppyi Cossmann in having a larger and heavier shell, in being more strongly beaded, and in lacking the spiral thread on the posterior slope, It is Lke T. guppy? in having two single keels. Age: Miocene. Locality: Common in the State of Falcon. Locality Numbers: 78, 79, 83, 84, 90, 93, 96, 118, 120A, 120B, 121, Peat 2Ab, 123, 127, 129,180, 131, 188, 150, 150A, 154, 159(?), 162, 163, 176, 178, 180, 189, 204, 205A, 206, 206A, me. 220; 222, 261, 290, 295, 297, 325, 354, 719 (7), 1004, mone 2), 1012, 1034, 1067, 1233, 1322, 1447, 1450, 1852, 1856, 1858, 1863 (cf.), 1870, 1892, 1908. Turritella altilira Conrad mirandana, n. subsp. IPL, AS, aie, Oe We4l, Pts}, Tne, 2. This subspecies stands very close to T. a/tilira Conrad, sensu stricto, and to its variety chiriquiensis Olsson,' but there is a slight difference which is constant in the Ven- ezuelan forms, viz., the components of the double upper 5 A. Olsson, Bull. Amer. Pal., Vol. 9, pp. 322-323, Pl. 17, fig. 4, 46 BULLETIN 45 ~ 2G spiral rib are almost equal, or the weaker supplementary cord is posterior to or above the larger anterior one. In all of our specimens from Gatun, C. Z., the upper rib, if double, has the weaker cord below, or just anterior to the larger posterior cord. The two main cords forming the strong posterior rib vary in strength; frequently, the two cords attain an almost equal size but in no case does the posterior component become stronger than the lower com- ponent, as is the general case in typical T. altilira Conrad. This subspecies differs from variety chiriquiensis Olsson in that the upper carinating rib is double; in variety chiri- quiensis, it is single; in the latter, there occurs on the pos- terior slope a weaker spiral thread which is not a supple- mentary cord of the upper keel. The posterior slope is much steeper than in variety chiriquiensis due to the greater width of the upper double keel. Age: Miocene. Locality: Common in the States of Falc6n and Zulia, Venezuela, and at Cartagena, Columbia. Locality Num- bers: 1038 (?), 122B, 149, 154, 171 or 172 (variation), 178 (variation), 179, 277, 304, 1004, 1031 (?), 1100, 1799 (cf.), 1800, 1802, 1805, 1859 (cf.), 1860 (2), 1901, 1905, IGBHUR, WGBIG, Ge, B22. Turritella vistana, n. sp. Pl. 23, figs. 3, 4; Pl. 27, ficss08, oem Pl. 28, fig. 4. Shell is turreted, bicarinate, attenuate. Protoconch and early nepionic whorls are missing. Later nepionic whorls are bicarinate; upper spiral rib is weaker than the lower; the concave interspace between the keels is ornamented with about three secondary spirals; all the spirals are beaded in well preserved specimens. In the neanic stages, the two single keels are almost equal in size, and later, the posterior finally becomes the larger of the two. In the succeeding whorls, the keels are always single, very pronounced in transverse height, rather narrow in axial width, with deep, narrow, concave interspaces; when the keels are much ex- 217 VENEZUELAN TURRITELLAS 47 tended, they are frequently concave on either side, but more so on the sides facing the sutures; the deeply excavated area between the keels usually is ornamented with about three beaded secondary spirals and occasionally finer spiral threads; one of the spirals usually lies on the lower side of the upper rib; the sigmoid growth lines cause heavy beads or oblique nodes elongated in the direction of the lines of growth over the keels. In well preserved specimens, the whole surface of the shell shows minute spiral ornamentation; in the adult whorls below the concavity under the anterior keel, there is a strong prominence or shoulder, ornamented with sub- microscopic threads, which juts over the excavated suture. This species closely resembles 7. altilira Conrad but dif- fers in having the upper keel weaker than the lower in the young stages; in having single carinating ribs, the pos- terior of which becomes stronger in adult stages; and in having deeper, narrower interspaces between the carinat- ing spirals. It is not confusable with T. perattenuata Heil- prin or T. altilira Conrad mirandana (n. subsp.), both of which have a double upper keel. It differs from T. guppyi Cossmann morantensis (n. subsp.), and from T. altilira Conrad urumacoensis (n. subsp.) in having the weaker posterior keel of the young stages become the stronger in the adult whorls, and in having more pronounced keels with a deeper, narrower, medial concavity between them. It further differs from T. guppyi Cossmann and its new sub- species morantensis in lacking the spiral thread on the pos- terior slope of the latter species and subspecies. Age: Miocene. Locality: Common in the State of Falcon. Locality mummpers: Sab, 93 (cf.), 97, 121 (cf.), 122B, 146 (or mar), 149, 149B, 154, 155, 163, 169A, 171, 174, 180, 182, 185, 187, 193, 194A (cf.), 204, 216, 229, 270, 277 (or subsp. meavoist), 317, 1007 (cf.), 1017, 1031, 1033, 1043, 1111, 1255, 1335, 1552, 1850, 1855, 1856, 1866, 1867, 1873, 1911, 1928. 48 BULLETIN 45 218 Turritella vistana nicholsi (Williston MS.) Pl. 23, fiss6; PI 29) ficsse2eeoe This subspecies is distinguished by the pronounced size of the extended keels which cause them to become unwieldy, irregular in spacing, and slovenly in appearance. The keels may droop down anteriorly or turn up posteriorly on even a single volution. Age: Miocene. Locality: Districts of Colina and Miranda, State of Falcén. Locality Numbers: 140, 148, 148BB, 149A, 149B, PO Ailes, BX0IL, 2A, IOS (Ci), IIIS USS (Gis) = Turritella bifastigata Nelson maracaibensis, n. subsp. Pl. 30, figs. 2, 4, 6. This subspecies is very close to T. bifastigata Nelson and we would refer it to this species, sensu stricto, except for slight but constant differences which make it advisable to separate it as a distinct subspecies. Dr. Carl O. Dunbar of the Peabody Museum at Yale University very obligingly sent us Nelson’s specimens from which T. bifastigata was originally described. Two of these Nelson ! mentioned especially, giving the dimensions. The first one mentioned (length 61 mm., greatest diameter 19.1 mm., diameter of smallest whorl 7 mm.) we have se- lected for the lectotype of the species; Spieker ? reviewed these species quite thoroughly and figured one of the speci- mens, but did not state whether it was Nelson’s specimen or not. Upon examining the lectotype, we find it to be the one which Spieker figured, but for clearness we are re- figuring it, Pl. 30, fig. 1. There is only one specimen in Nelson’s collection which corresponds to each he described and measured. As Spieker states (loc. cit.), Grzybowski described and figured this Turritella as a new species which 1 Trans. Conn. Acad., Vol. 2, p. 189, 1870. 2H. M. Spieker. The Paleontology of the Zorritos Formation of the North Peruvian Oil Fields, pp. 63-65, Pl. 3, fig. 1, The Johns Hopkins Press, 1922. 219 VENEZUELAN TURRITELLAS 49 he called T. gothica.* In 1912, Pilsbry and Brown ” described as a new species three whorls of a Turritella from Cartagena, Columbia, which is evidently of the T. bifastigata type, under the name of 7. cartagenensis; as described, it is somewhat larger than most of our specimens and in the description of the base they say: “The base is somewhat convex, and shows four very low, wide spiral welts, with the same finer spirals as the upper surface.” * This would seem to indicate that it is inter- mediate between the Peruvian and Venezuelan forms, as the typical bifastigata shows about six sharp spirals or “welts” on the base, while our form is almost smooth and orna- mented by numerous finer spirals of varying strength. In 1922, Olsson * described as a new species, a form oc- curring in Costa Rica which is very similar to T. bifastigata Nelson, but until more specimens are collected to show its genetic relationship, it is hard to say whether this is of specific or subspecific rank. It differs chiefly in having a very prominent posterior swelling which overlaps the an- terior suture of the preceding whorl. ‘ The distinguishing feature in the new subspecies mara- caibensis is the ornamentation of the base. The typical Peruvian form has five or six very prominent folds on the base of the adult whorls which in this subspecies have been changed to many small spiral threads of varying strength. On the part of the base covered by the succeeding whorls there are six or seven widely spaced spirals, somewhat stronger than the weaker intervening spiral threads. This change is probably due to a locality difference. It is easy to separate the Venezuelan forms of the bifastigata stock by this difference alone, because T. bifastigata 1 Neues Jahrbuch fiir Min. Geol. etc., Beil. Bd. 12, p. 645, Pl. 20, fig. 10, 1899. 3 2 Proc, Acad. Nat. Sciences of Philadelphia, 1917, pp. 34-85, Pl. 5, = Be Soc. cit., 35. Bios A. picker: Bull. Amer. Pal., Vol. 9, pp. 324-325, Pl. 17, fig. 1, 50 BULLETIN 45 220 Nelson, sensu stricto, has the base of the whorls corrugated by very strong spiral ribs; this ornamentation makes the Peruvian form quite distinct from ours which has a’ smoother base with more numerous weaker spirals. The base of the whorl is somewhat convex with a rounded, carinating, anterior prominence just behind the anterior suture. The posterior shoulder tends to be prominent and to overlap the suture to a certain extent, but less than is com- monly found in the Peruvian form. In other words, the suture in this subspecies tends to be slightly more open or gaping than in bifastigata, sensu stricto, and less gaping than in the new subspecies democraciana. Age: Miocene. Local ty: Common in the northern part of the State of Falcén. Locality Numbers: 70A, 70B, 71, 72, 81, 83, 86, 87, 90, 1233, 1757, 1856, 1858 (cf.), 1872, 1874, 1875, 1883, 1884, 1892, 2036. Turritella bifastigata Nelson democraciana, n. subsp. PI. 28, fig. 3; Pl. 30, figs. 3, 5. This subspecies differs from the other subspecies, mara- caibensis, in having a nearly flat base, slightly concave in some specimens, slightly convex in others, which projects a little beyond the sides of the whorls as a basal keel; the base is almost at right angles to the sides of the whorls; the rounded area between the base and sides of the whorls in maracaibens?s is lacking in this subspecies and replaced by the angulated anterior keel. The base of the whorl is ornamented with twenty or more small spirals of varying strength; usually there is one or more weaker, intervening, spiral threads between the stronger. The sutures are more open and gaping than in any of the similar forms. Age: Miocene. Locality: Common in the northern part of the State of Falcon. Locality Numbers: 70A, 70B, 70C, 80, 81, 82, 83, 83B, 84, 85, 86, 87, 90, 94, 96, 97, 108, 149B, 150B, 163, 184, 219, 1117, 1231, 1233, 13885, 1757, 1850; 18525 1858, 1859, 1860, 1861, 1862, 1869 (?), 1872, 1883, 1908, SHG F Ia. PLATE 1. PAGE Turritella zuliana, n. sp. Paratype from Loc. No. 6, col- lected by Mr. W. A. Watkins, length 20 mm., greatest diam- Sten CIB SMI iswscncecs Siete cwaestancwoencecsveuncseetroecoecercace eee Mee eee cere eaees Turritella zuliana, n. sp. Paratype from Loc. No. 6, proto- conch, length 3.7 mm., greatest diameter of next to last whorl, 1.38) Mats csesads ieccesedesaceseseade teemetah eects ceceeseceeee eee eee eReeEEee Turritella zuliana, n. sp. Paratype from Loc. No. 6, young whorls, length 3 mm., greatest diameter 1.9 mm.........:........... Turritella zuliana, n. sp. Holotype, from Loc. No. 6, length 60 mm., greatest diameter 22.5 MM. ..............cceccecceecceeceeeeeeees Turritella zuliana, n. sp. Paratype from Loe. No. 6, length 388 mm., greatest diameter 22 MM................cscseccscescnsceccsscnceees Turritella zuliana, n. sp. Paratype from Loc. No. 6, col- lected by Mr. W. A. Watkins, length 33 mm., greatest diam- eter 2A WMisc. o.cocshassccacevtes cn sees eu ceneenoas cere estan cesar eee ee eee Le i PL. 5, Vou. 11 BULL. AMER. PAL. No. 45, PL. F Ia. ile ILI). PAGE T. zuliana palmeri, n. subsp. Holotype from Loc. No. 6, length 21 mm., greatest diameter 10.7 mm.................2.-0.-00e00 T. zuliana, n. sp. Paratype from Loc. No. 6, length 15 mm., ereatest diameter 12) minie.d:ccccscosccscosccecseesneeneee eco eee eee eee T. zuliana, n. sp. Paratype from Loc. No. 6, enlarged to show young stages, length 13.2 mm., greatest diameter 7 mm. T. zuliana palmeri, n. subsp. Same specimen as figure 1, On ar Sede) | 2s ha des sacasnshasoadens soosee SeSeaseouss Sas ee Cece see eR CRO EEEE Ree eee eee T. zuliana, n. sp. Paratype from Loc. No. 6, length 7 mm., ereatest dlameter 3.3) MIM. ..<.cass.cnsscacsncecce cease nese see eeereeeteeee eee T. zuliana, n. sp. Paratype from Loc. No. 6, enlarged to show young stages, length 9.4 mm., greatest diameter Alii, IYDTINE. -sasseaint aden dcwels te smenclec enetsnepeeeadeee tossed elses See ee ER ORE: EERE EERE EERE i) 8 PL. 6, Vou. 11 BuLi. AMER. PAL. No. 45, PL. 2 en Hy Ls Fic. 5. JPanel) SS, PAGE Turritella larensis, n. sp. Paratype from Loc. No. 815, length 35 mm., greatest diameter 20 mm...............c..:.cc0e0ee000 10 Turritella larensis, n. sp. Paratype from Loc. No. 815, length 30 mm., greatest diameter 21 mm.........................00000 10 Turritella larensis, n. sp. Paratype from Loc. No. 2019A, length 12.5 mm., greatest diameter 4.6 mm....................0060 10 Turritella larensis, n. sp. Paratype from Loc. No. 2019, length 39 mm., greatest diameter 15.2 mm........................055 10 Turritella larensis, n. sp. Holotype from Loe. No. 815, length 46 mm., greatest diameter 18.1 mm.......................0065 10 Bee 7, VOL. 11 BULL. AMER. PAL. No. +5, PL. 3 Fic. OU PLATE 4. PAGE T. larensis, n. sp. Paratype from Loc. No. 2019A, young whorls enlarged, length 15 mm., greatest diameter 6.5 mm. 10 T. larensis, n. sp. Paratype from Loc. No. 1761, length Gi sania, PAK ASIG GlieMMN@ter ZAG IMM, SooeconscosbonccocHsneoscb 8s 22605005250: 10 T. larensis santiagana, n. subsp. Holotype from Loc. No. 2048, length 45 mm., greatest diameter 26.5 mm..................... Lz T. larensis, n. sp. Paratype from Loc. No. 2019, length AS Tain, SSeS Chiammewere Goll WAN, coos .snosaese0cn.0sdnEHAenSnHsONIO500200 10 T. larensis, n. sp. Paratype from Loc. No. 815, young whorls much enlarged, length 11.8 mm., greatest diameter On mex tO) Wastin wihiorl (Gein eeaeseeeeseeseceeeene eee eee eer ee ee eee eee res 10 Form intermediate between T. larensis, n. sp., and T. larensis santiagana, n. subsp. Length 48 mm., greatest diameter 27.3 mm. From Loc. No. 2019 PL. 8, VoL. 11 Buti. AMER. PAL. No. +5, PL. + FIG. il. PLATE 5. PAGE T. robusta Grzybowski fredeai, n. subsp. Paratype from Loc. No. 72, young whorls enlarged, length 12.6 mm., great- est diameter 8:6. mimiiiciie.acsssesaecs oes see obese ses essseeeeedse ee eee ee ee eee Form intermediate between T. larensis, n. sp. and T. larensis santiagana, n. subsp. Length 26 mm. From Loc. No. 2027. T. robusta Grzybowski fredeai, n. subsp. Paratype from Loc. No. 70A, enlarged, length 41 mm.................6....ecceseeeeeees T. larensis, n. sp. Paratype from Loc. No. 2027, length 34 mm., greatest diameter 20.5 mm. For comparison with Jel, 4k, ines, By unl ies, 2 Oi Wawls JOVEMc cocococnocossoscHooscosesesoocoscconce: 13 Wz Pre 9, VOL. ti BULL. AMER. PAL. No. 45, PL. 5 FIG. (J) JeGVMINS; 6, PAGE T. larensis carrizalensis, n. subsp. Paratype from Loc. No. 2022, length 36.5 mm., greatest diameter 20.5 mm................. 12 T. robusta Grzybowski fredeai, n. subsp. Paratype from ‘Loe. No. 72, -youngest whorls found, enlarged; length 21.5 mm., greatest diameter of second largest wnaorl Berea 000 00 eee nee ee Lam ree nee er Meee ent ea AlN Anat sci nqaboomocoocooccce 13) T. larensis carrizalensis, n. subsp. Holotype from Loe. No. 2027, length 41 mm., greatest diameter 21 mm....................... 12 T. larensis carrizalensis, n. subsp. Paratype from Loe. No. 2023, length 64 mm., greatest diameter 23 mm....................5 We T. robusta Grzybdowski fredeai, n. subsp. Paratype from Loc. No. 70A, very large whorl, greatest diameter 503 mm..... 12 T. cauredalitoensis (Williston MS.). Paratype from Loe. Pi. 10, Vou. 11 BULL. AMer. PAL. No. 45, PL. 6 Fig. “I GAINS, 7, PAGE T. robusta Grzybowski fredeai, n. subsp. Paratype from Loe. No. 72C, showing sculpture on base of whorl, greatest diameter 28 mm T. hubbardi, n. sp. Paratype from Loc. No. 2040, length 22 mm., greatest diameter 10 mm. ................ RPE ee onboblesacasc T. hubbardi, n. sp. Apex of holotype from Loc. No. 2040, enlarged to show the spiral ribs, length 8 mm., greatest diameter 2.9 mm er wee we eww e tee eee eee e ee eee esas eae rete nneeeeseseeeaaeneasesecnsess T. hubbardi, n. sp. Paratype from Loc. No. 2040, length 19 mm,, greatest diameter 7.5 mm T. hubbardi, n. sp. Paratype from Loc. No. 2040, enlarged, showing earlier whorls with strong medial rib, length 12 mm., greatest diameter 4.9 mm Cem e ccc eee e cena eee e arias earn c es enenntnne T. robusta Grzybowski fredeai, n. subsp. Paratype from Loc. No. 70A, length 89 mm., greatest diameter 30 mm T. robusta Grzybowski fredeai, n. subsp. Holotype from Loc. No. 70A, length 105 mm., greatest diameter 39 mm 14 14 PL. 11, VoL. 11 BuLL. AMER. PAL. No. 45, PL. 7 FIG. 1. PLATE 8. PAGE T. hubbardi, n. sp. Paratype from Loc. No. 2040, greatest diameter 13 mm. Shows ornamentation on lower part of basal carina and slightly knobby character of the lower cord. 14 T. hubbardi, n. sp. Paratype from Loc. No. 2019A, young whorls enlarged, length 8.8 mm., greatest diameter 4 mm..... 14 T. hubbardi, n. sp. Holotype from Loc. No. 2040, length 30 mm., greatest diameter 10 MIM. .................cccecceccnesceeeeeeseeeeees 14 T. hubbardi, n. sp. Same specimen as fig. 6, enlarged......... 14 T. hubbardi, n. sp. Paratype from Loc. No. 2040, length 18.7 mm., greatest diameter 6.5 MMM...............:cccccceeeeeceseeees 14 T. hubbardi, n. sp. Paratype from Loc. No. 2019A, length 20 mm., greatest diameter 8 MIM................ccccccecceeceees eee ese ness 14 T. hubbardi weeksi, n. subsp. Paratype from Loe. No. 3250, collected by Mr. L. G. Weeks. A slender variation, length 26 mm., greatest diameter 9.4 MM ................cc0.ccceeceeeceeeeeee eee 15 PL. 12, VoL. 11 Buti. AMER. PAL. No. 45, PL. 8 FIG. il, 6. eIbaitia; 2). PAGE T. hubbardi, n. sp. Paratype from Loc. No. 2040, length 20 mm.. ereatest diameter 8.3) mintte-s-..c:cces-ceee sees eee T. cauredalitoensis (Williston MS.). Paratype from Loc. No. 1217, length 12 mm., greatest diameter 6.8 mm............. T. cauredalitoensis liddlei, n. subsp. Paratype from Loc. No. 1987, length 19 mm., greatest diameter 16.6 mm. ........ T. cauredalitoensis (Williston MS.). Paratype from Loc. No. 1217, length 15 mm., greatest diameter 8.5 mm............. T. hubbardi, n. sp. Paratype from Loc. No. 2018, length 19.7 mm., greatest diameter 6.5 MMM.................:2eeceeeeneeeeeeee ees T. hubbardi, n. sp. Holotype. Same specimen as PI. 8, fig. 8. Length 30 mm., greatest diameter 10 mm............... T. robusta Grzybowski fredeai, n. subsp. Paratype from Loc. No. 72, enlarged to show basal sculpture, greatest diam- Sheree Al hy Anas ee ect enna teeta eee she Saincid ois SAG SSE SC Se Re See eee 14 15 16 PL. 13, VoL. 11 BULL. AMER. PAL. No. 45, PL. 9 FIG. ie PLATE 10. PAGE T. laremsis guaratarensis, n. Subsp. Paratype from Loe. No. 19388, length 17.5 mm., greatest diameter 13.1 mm......... 12 T. cauredalitoensis (Williston MS.). Paratype collected by Mr. S. H. Williston, length 21 mm., greatest diameter 9 mm. 15 T. larensis guaratarensis, n. subsp. Holotype from Loc. No. 1938, length 40 mm., greatest diameter 19 mm....................... 12 T. hubbardi weeksi, n. subsp. Holotype from Loc. No. 325@, collected by Mr. L. G. Weeks, length 37.3 mm., greatest Giameter °925> Wes .ds.cccsFsas eee uae deencesee soensmosenlaseesee ee eee eee eRe REE 15 T. hubbardi weeksi, n. subsp. Paratype from Loc. No. 3250, collected by Mr. L. G. Weeks, length 04 mm., greatest diam- C=) = os BS 000 0 eR eee Rael pra Or eptor sa seRcemacucdnar dodenacddcocccocatascqcace- 15 PL. 14, VoL. 11 BULL. AMER. PAL. No. 45, PL. 10 Sains ne avon y peer » . Pe ed WN REAR >: errs erry aggre FIG. 1. pee PAGE T. cauredalitoensis (Williston MS.). Paratype from Loe. No. 1217, base of keel enlarged to show sculpture. Width Of k@el! 225. WM... secdisctnecenedoncsaseice cgusuioeueeseoe nace tee wane cee ae See eee eee eeeeeee 15 T. cauredalitoensis filensis, n. subsp. Holotype from Loc. No. 1979, length 33 mm., greatest diameter 20 mm............. Wy T. cauredalitoensis (Williston MS.). Holotype collected by Mr. S. H. Williston. Length 21 mm., greatest diameter LOM, ooh hecnedeue asics cuteseecteleseveasedeameoesinccnes gece ss tes oe eee eee eee Cee R eee eee eaee 15 T. cauredalitoensis dabajuroensis, n. subsp. Holotype from Loc. No. 2027, length 28 mm., greatest diameter 14.4 mm. 17 T. cauredalitoensis (Williston MS.). Same specimen as fig. 1, length of fragment 13 MM..................cc.cceeceecceeceeseeeee sees 15 T. cauredalitoensis liddlei, n. subsp. Holotype from Loc. No. 1159K, length 28 mm., greatest diameter 21 mm........... 16 PL. 15, Vou. 11 Buti. AMER. PAL. No. 45, Pu. 11 Fic. 1. iw) PACE S22 PAGE T. cauredalitoensis filensis, n. subsp. Paratype from Loc. No. 1979, length 37.5 mm., greatest diameter 15 mm........... 1L'7/ T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, length 37.mm., greatest diameter 13.5 MIM. ................0cccecceeececeenccesenes 17 T. cauredalitoensis filensis, n. subsp. Paratype from Loc. Nos 1979; Teneth 3825. min iiece-csccsccseccsnsecoose see seen eee eee ie T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, proto- conch, length 4.51 mm., greatest diameter 1.27 mm............. 17 T. cauredalitoensis filensis, n. subsp. Paratype from Loc. No. 1217, enlarged to show sculpture, greatest diameter a GS is 60000 Reese ene ere eee eer sor racn cceanretinecebosencdsocactacoccosc0000c 17 T. gilbertharrisi, n. sp. Paratype from Loc. No. 6. young whorls enlarged, length 11 mm., greatest diameter 3.1 mm. 17 PL. 16, VoL. 11 BULL. AMER. PAL. No. 45, PL. 12 PLATE 13. FIG. PAGE 1. T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, length IS mimes ereatest diameter 9 mimes css eceresee ses teeeeee eee eeeeee eee 17 . T. gilbertharrisi, n. sp. Same specimen as fig. 1, enlarged 17 3. TT. gilbertharrisi, n. sp. Holotype from Loc. No. 6, length 60 mm., greatest diameter 15 mm. The longest single specimen found. (The part that had been hopelessly crushed is replaced with modeling wax).............c.scccecsseeeseeeees 17 4. T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, length Z20;mms:, oreatest dlameterai) minhes-eesessseeeee see ere tee teee eee eee ee eee 17 5. T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, proto- conch, length 4.13 mm., greatest diameter 1.22 mm........... Ly 6. T. gilbertharrisi, n. sp. Paratype from Loc. No. 6, length 32 mm., greatest diameter 9 MIM..................cccccceseeeeeeceeeseeseeees WT PL. 17, Vor. 11 BULL. AMER. PAL. No. 45, PL. 13 FIG. il, (Je) PLATE 14. PAGE T. gilbertharrisi, n. sp. Paratype from Loc. No. 1415, young whorls enlarged, length 14.5 mm., greatest diameter 6 mm. T. gilbertharrisi falconensis (Williston MS.). Paratype from Loe. No. 1928, showing common appearance, length 52 mm., greatest diameter 23.2 mm T. montanitensis, n. sp. Paratype from Loc. No. 19389, young whorls enlarged, length 13.5 mm., greatest diameter PED: AMIN. -oecsacatetesdccscudedectlbag wor ae sec tic oot seat ee EEE Oe T. gilbertharrisi, n. sp. Same specimen as Plate 13, fig. 6. Length 32 mm T. hubbardi, n. sp. Paratype from Loc. No. 2040, young whorls enlarged, length 9.5 mm., greatest diameter of smali- est whorl 1.4 mm., greatest diameter of largest whorl 3.9 mm. T. gilbertharrisi falconensis (Williston MS.). Holotype from Loc. No. 1928, length 59 mm., greatest diameter PAU e110 0 eae een mee el SE Ree RRP GE ade eden AGbedobdns500c0000000C T. gilbertharrisi, n. sp. Paratype from Loc. No. 1415, en- larged to show sculpture, length 21 mm., greatest diam- eter 12 mm T. gilbertharrisi falconensis (Williston MS.). Paratype from Loc. No. 1127, length 38.5 mm., greatest diameter 18.7 mm. iL) 20 14 19 PL. 18, Vot. 11 Buti. AMER. PAL. No. 45, PL. 14 acer il. (J) er) = ¥), PLATE) is: PAGE T. gilbertharrisi aguavivensis, n. subsp. Holotype from Loc. No. 2019, length 22 mm., greatest diameter 11 mm............. 19 T. montanitensis, n. sp. Paratype from Loc. No. 1939, length 20.5 mm., greatest diameter 9.5 mm................cc00ecc00e 20 T. gilbertharrisi falconensis (Williston MS.). Paratype from Loe. No. 341, length 29.5 mm., greatest diameter WQS SMAI...” sained Meaaisswonsquecedismoesonaacnead Gee nates coteancee Selec shee Reeeeee eeeeeaeee 19 T. gilbertharrisi falconensis (Williston MS.). Paratype from Loe. No. 1067, length 41.1 mm., greatest diameter DD AMM! Sede adsheacuesedosanetesencoeee a ecaae use eee keen coe een ee See ene eee il) T. gilbertharrisi falconensis (Williston MS.) Paratype from Loc. No. 1265, length 17 mm., greatest diameter ;o ey 0.0 00 Ee Ree Rer ere nner eee eer orion ten cericcrcoakrnmrsticardsecaccagascacacccocq0asdo00c 19 T. gilbertharrisi staufferi, n. subsp. Holotype, length 17 mm., greatest diameter 7.5 MM. ...........ccccecceceececceseneeeceeen 19 T. gilbertharrisi falconensis (Williston MS.). Paratype, from Loc. No. 1436. Enlarged to show sculpture, length 31.5 mm., greatest diameter 16 MIM................cccseccceceeseneeneeees 19 T. gilbertharrisi aguavivensis, n. subsp. Paratype from Loc. No. 6; intervening riblet is weaker than in holotype, length 21 mm., greatest diameter 14 mm.......................0000000s 19 T. gilbertharrisi staufferi, n. subsp. Same specimen as TU Sen Oa eee Rie ei oe er es oor em oer ana gp ngn ciseSeGadaooooccocconecec 19 Pu. 19, VoL. 11 BuLL. AMER. PAL. No. +5, PL. 15 wet Poutee : ‘Ge PLATE 16. PAGE T. curamichatensis, n. sp. Paratype from Loc. No. 1436, length 30 mm., greatest diameter 13.5 mm..................c0cc.000 T. montanitensis saladilloensis, n. subsp. Holotype from Loc. No. 1957, length 16 mm., greatest diameter 8.2 mm..... T. curamichatensis, n. sp. Paratype from Loc. No. 1436, length 25 mm., greatest diameter of smallest whorl 8.2 mm. T. montanitensis, n. sp. Paratype, a slender variation from Loc. No. 52, length 80 mm., greatest diameter 16 mm......... T. montanitensis, n. sp. Paratype from Loc. No. 19389, length 17.8 mm., greatest diameter 11 mm.......................00c00 T. montanitensis, n. sp. Paratype from Loc. No. 1939, length 23.5 mm., greatest diameter 9 mmM...................0.s0.00e000e T. curamichatensis, n. sp. Paratype from Loc. No. 1436, length 15.1 mm., greatest diameter 6.8 mmM.....................00008 T. montanitensis, n. sp. Paratype from Loc. No. 2031, com- mon appearance when weathered, length 18 mm., greatest diameter 628) MiIMies...3.cfscc)el ib ac es da sswe ed sae e cae se eon no ee T. curamichatensis, n. sp. Paratype from Loc. No. 1070, length 18 mm., greatest diameter 10.7 mm..................0..0000200 T. montanitensis, n. sp. Holotype from Loc. No. 1989, length 25 mm., greatest diameter 10 mm.......................00.0000 T. curamichaiensis, n. sp. Paratype from Loc. No. 1408, length 17 mm., greatest diameter 7.3 mM..,.............scs0cce-0ecees T. montanitensis olcotti, n. subsp. Paratype from Loc. No. No. 2052, length 24 mm., greatest diameter 18.6 mm........... 22 20 20 20 22 20 PL. 20, VoL. 11 BULL. AMER. PAL. No. 45, PL. 16 10. Wil. lelGyevild} A. PAGE T. curamichatensis, n. sp. Paratype from Loc. No. 1436, showing a variation with flatter whorls. Length 16.5 mm. 22 T. curamichatensis, n. sp. Holotype from Loc. No. 1070, length 27 mm., greatest diameter 138 mm......................00c0ee0ee 22 T. montanitensis, n. sp. Paratype from Loc. No. 1939, length 27 mm., greatest diameter 11 mm......................c00c000e 20 T. curamichatensis, n. sp. Paratype from Loc. No. 1408, length 9 mm., greatest diameter 5 MM...............cc0ecceceeeeesneees 22 T. curamichatensis, n. sp. Paratype from Loc. No. 1754, length 10 mm., greatest diameter 5.5 mm. Young whorls SNMAP GSM.) eieed as aeeactie salnsen oemnde ws emea sec eemeaeeee sce aa: eecar ee eee ae eae eee 22 T. curamichatensis, n. sp. Paratype from Loc. No. 1436, length 34 mm., greatest diameter of smallest whorl 6 mm. 22 T. montanitensis, n. sp. Same specimen as Plate 16, fig. 6. Length), 2325) smmicicscscuiaes coscbcshasroen se tnseeceaesenctecena seen geeeee ee eee 20 T. montanitensis, n. sp. Holotype from Loc. No. 1989, length 25 mm., greatest diameter 10 mm. Same specimen as Pl. VG; fe WO ihe wccsabecdeaasusnnceccc tee gecseoeee seat cee sne cee eeee eee eee ee eee 20 T. curamichatensis, n. sp. Paratype from Loc. No. 1436, length 22 mm., greatest diameter 12.5 mm.................cceccseeeees 22 T. montanitensis, n. sp. Same specimen as fig. 3............0.c006 20 T. montanitensis, n. sp. Paratype from Loc. No. 1939, length 13.5 mm., greatest diameter 7.2 mm. Same speci- mencas PL. 14. filo. Silessee! iscae ack caksieceds dove eee os eae Gas ae eee eee eee 20 PL. 21, Vor. 11 BuLi. AMER. PAL. No. 45, PL. 17 FIG. iL. PLATE 18. PAGE T. berjadinensis, n. sp. Paratype from Loc. No. 67, length 15 mm., greatest diameter 8 MIM..................cs0c0scesesecsenecceseesess 26 T. gatunensis willistoni, n. subsp. Holotype from Loc. No. 1033, length 28.8 mm., greatest diameter 7.6 mm................. 25 T. gatunensis willistoni, n. subsp. Paratype from Loc. No. 1033, length 20.5 mm., greatest diameter 8 mm..................... 25 T. gatunensis willistoni, n. subsp. Paratype from Loc. No. 1856,-deneth, V8.7. mlm... ia) i. ccecc eccste sae noesoseo 2s sees eee eee eon ee cee 25 T. gatunensis taratarana, n. subsp. Paratype from Loc. No. 1038, length 30 mm., greatest diameter 9.1 mm............... 25 T. gatunensis lavelana, n. subsp. Paratype from Loc. Ne. 183, length 17.5 mm., greatest diameter 10 mmi..................... 23 T. gatunensis taratarana, n. subsp. Holotype from Loc. No. 185, length 27.3 mm., greatest diameter 10 mm........... 25 T. gatunensis willistoni, n. subsp. Variation with shallow- er constriction at top of whorl, from Loc. No. 938, length 19 mm., greatest diameter 7 MM. ............ceccecccseeceeeseecenscneeens 25 T. berjadinensis colinensis, n. subsp. Paratype from Loc. No. 1033, length 43 mm., greatest diameter 15 mm............. PAT PL. 22, VoL. 11 BuLL. AMER. PAL. No. 45, PL. 18 FIG. il. (Se) ] §). Jeibyaille, 18). PAGE T. berjadinensis, n. sp. Paratype from Loc. No. 1078, length 17 mm., greatest diameter 8 mM.................2-02:e0eeeeees 26 T. planigyrata Guppy, from Loc. No. 1131, length 15.5 mm., greatest diameter 6) MUM: ...00.).cr..csaceseeen= se eeneue- te doecbemeoaseeeee meee 29 T. berjadinensis, n. sp. Paratype from Loc. No. 1078, length 14.5 mm., greatest diameter 9.8 mm..................2-...00000 26 T. berjadinensis, n. sp. Holotype from Loc. No. 1757, length 27 mm., greatest diameter 13 mm................. "daatabpeaeeeeee 26 T. berjadinensis cocoditana, n. subsp. Holotype from Loe. No. 2207, length 35 mm., greatest diameter 9.5 mm............. 29 T. berjadinensis, n. sp. Paratype from Loc. No. 1757, length 19 mm., greatest diameter 9.6 mm....................s2::0000 26 T. gatunensis lavelana, n. subsp. Holotype from Loc. No. 298, length 21 mm., greatest diameter 8 mm.......................+. 23 T. berjadinensis colinensis, n. subsp. Holotype from Loe. No. 1033, length 37.2 mm., greatest diameter 8 mm............. PATI T. planigyrata Guppy, collected by Professor G. D. Harris from Manzanilla, Trinidad. Figured for comparison with fig. 2, length 17 mm., greatest diameter 6.8 mm..................- 29 PL. 23, Vou. 11 Buti. AMER. PAL. No. 45, PL. 19 FIG. il, 10. 11. 2 PLATE 20. PAGE T. plebeia Say A-L-Owensi, n. subsp. Paratype from Loc. No. 90, length 19 mm., greatest diameter 10.5 mm............... 31 T. plebeia Say A-L-Owensi, n. subsp. Paratype from Loc. No. 298, length 14 mm., greatest diameter 10.5 mm............. 31 T. berjadinensis cocoditana, n. subsp. Base enlarged show- ing sculpture, greatest diameter 12.1 mm. From. Loc. ON fc eit 0 1 eae ri Ue Oe avn Berean He aacneca dodiaoodaodadabesococ 29 T. matarucana, n. sp. Paratype from Loc. No. 197, great- est diameter of middle whorl 16 mm.....................cccceceeeeeceeeees 31 T. plebeia Say A-L-Owensi, n. subsp. Paratype from Loc. No. 298, length 15 mm., greatest diameter 6 mm................. 31 T. plebeia Say A-L-Owensi, n. subsp. Paratype from Loc. No. 90, length 19 mm., greatest diameter 9 mm................... 31 T. berjadinensis cocoditana, n. subsp. Paratype from Loe. No. 2207. Young whorls enlarged. Greatest diameter F2}) OX 0) D743 17) 001 00 GRRE sain era eoe I RgdeE ne ABoocdcSaner concuanaScadsHouodadousoooccone 29 T. berjadinensis warfieldi, n. subsp. Paratype from Loe. No. 93, length 26 mm., greatest diameter 10.5 mm............... Pall T. montanitensis olcotti, n. subsp. Holotype from Loc. No. 2050, length 41 mm., greatest diameter of next to last WOK] MG: MIM. e coe ceiscesle see soc one Gesageneeee dene eee el aoe eee 21 T. berjadinensis cocoditana, n. subsp. Paratype from Loe. No. 2207, length 32 mm., greatest diameter 9.3 mm............. 29 T. berjadinensis, n. sp. Paratype from Loc. No. 1856, length 39.3 mm., greatest diameter of next to last whorl MEA IMM.” levawecadstestiiacee ste taseeerscsesceaact Seton e esses eee eee ee 26 T. berjadinensis warfieldi, n. subsp. Paratype from Loc. No. 1048, length 25, greatest diameter 14.4 mm...................-- Pat No. 45, PL. 20 Buty. AMER. PAL. PL. 24, VoL. 11 FIG. 1. 9. PGA 28 PAGE T. matarucana, n. sp. Paratype from Loc. No. 197, great- est diameter D4 mimeicecsscccccaishcensensaeesssheae doses secs sce ee pee eee eee eeaeeeeeee T. variegata Linné paraguanensis, n. subsp. Paratype from Loe. No. 1504, base enlarged to show sculpture, greatest diameter “OVS mami ses sc sede cbcc aseowe sc becuse cawaaueeu cued e rosette ceeae ene eee REEee T. berjadinensis socorroensis (Williston MS.). Holotype, collected by Mr. S. H. Willisten. Length 31 mm., greatest diameter U2) Minss.c53.e.ccssesece setene astionanneccteesecpenoemccee see eee eRe eRe EEE Eee T. venezuelana, n. sp. Paratype from Loc. No. 6, length 18 mm., greatest diameter 7 MM.................ceeeeeee eee eereeee eee ecene T. plebeia Say from Jones Wharf, Md. Figured for com- parison. Length 27.3 mm., greatest diameter 9 mm............. T. berjadinensis warfieldi, n. subsp. Holotype from Loc. No. 118, length 26 mm., greatest diameter 10 mm................. T. variegata Linné paraguanensis, n. subsp. Holotype from Loc. No. 1504, length 78.5 mm., greatest diameter 21 mm..... T. venezuelana, n. sp. Paratype from Loc. No. 6, showing protoconch. Length 2.07 mm., greatest diameter 1 mm....... T. matarucana, n. sp. Holotype from Loc. No. 197, length 63.5 mm., greatest diameter 15 mm....................022006- ixeueeeteenee 31 21 No. 45, PL. MER. PAL. BuLL. A PEATE, 22: No. 6, length 14.6 mm., greatest diamete . 34 Fig. : 1. T. venezuelana, n. sp. Paratype from Loc. 16.5 mm., greatest diameter 7 mm.................. 2. T. cornellana, n. sp. Paratype from Loc. No 16 mm., greatest diameter 4.5 mm.................. 3. T. G-A-Weaveri, n. sp. Holotype from Loc. i 27 mm., greatest diameter 7.9 mm.................. ! so a! ¥ 4. T. cornellana, n. sp. Paratype from Loc. No he Ve f 20.5 mm., greatest diameter 6.3 mm............... Sew a 30) i 5. T. G-A-Weaveri, n. sp. Paratype from L : length 19.5 mm., greatest diameter of next - HOE TIVINYS We cenasna tech Sesea anecsen nena eee aae eect ene eae . — .. 34m 6. T. venezuelana, n. sp. Holotype from Loe. : 7 19 mm., greatest diameter 7 mm . 32 7. T. variegata Linné. Recent from north coas ¥ Figured for comparison. Length 75 mm., gr geno) 0a Orlane trae Mie esa Memrr snes NO em BN Se eco bs foo. Gl : 8. T. venezuelana watkinsi, n. subsp. Ho : ; No. 6, length 29 mm., greatest diameter . ; j 9. T. venezuelana quirosana, n. subsp. Holot; : Fi No. 6, length 20 mm., greatest diameter 11 1 . 3 10. T. venezuelana quirosana, n. subsp. Parat} - Fe i PL. 26, VoL. 11 Buti. AMER. PAL. No. 45, PL. 22 Fic. il. 10. iil, IVAN, 233, PAGE T. filacarmenensis, n. sp. Paratype from Loc. No. 1964, length 13 mms ereatest, diameter Sy misses esse eee 38 T. plebeia Say A-L-Owensi, n. subsp. Holotype from Loc. No. 90, length 27 mm., greatest diameter 16 mm................... 31 T. vistana, n. sp. Holotype from Loc. No. 185, length 25 MM Sreavest, diameter dale Oy miniesseessssee eee ss sees ee ee eee eee ee eee A6 T. vistana, n. sp. Paratype from Loc. No. 1552, length 22.2 mm., greatest diameter 19.5 mm.................cccecceceeseeeeeees 46 T. filacarmenensis, n. sp. Paratype from Loc. No. 1987, length 22, greatest diameter 11.8 mm....................cccc0cceeeeeeeees 38 T. vistana nicholsi (Williston MS.). Composite of 2 speci- mens collected by Mr. S. H. Williston. Upper specimen: length 44 mm., greatest diameter 15.5 mm. Lower speci- men: length 50 mm., greatest diameter 21 mm................. 48 T. buchivacoana, n. sp. Paratype from Loc. No. 1761, young whorls enlarged, length 7.7 mm., greatest diameter 7a a 000 9 ORR ERE EHO p oan erie paeeeberrncr re rcr eaccebnccrcnanec dr biceonécdotooSococcosoncuco: 39 T. berjadinensis socorroensis (Williston MS.). Paratype from Loc No. 100, length 32 mm., greatest diameter gS BE yo 0 080 3 eRe eee aac encroaceeaaacanee soaco nc os oeeanecscecrrbucosoddecoscosocosan00 28 T. G-A-Weaveri, n. sp. Paratype from Loc. No. 52, length 11.6 mm., greatest diameter 5.4 MM................0.ceccseceeceecceesees 34 T. altilira mirandana, n. subsp. Holotype from Loc. No. 154, length 22.5 mm., greatest diameter 13 mm..................... 45 T. mauryez, n. sp. Holotype, collected by Dr. Maury from Cereado de Mao, Santo Domingo. (Same specimen as in Bull. Amer. Pal., Vol. 5, Pl. 48, fig. 14.). Length 22.5 mm., greatest diameter 8. MMs “.....csccctscsetececscaeds cdeco sac -sc eeeneeee eee 30 PL. 27, VoL. 11 Buti. AMER. PAL. No. +5, PL. 23 FIG. PLATE 24. PAGE T. venezuelana quirosana, n. subsp. Injured specimen from Loc. No. 6, length 13.2 mm., greatest diameter 8 mm........... 34 T. cornellana bolivarensis, n. subsp. Paratype from Loc. No. 3222, length 17 mm., greatest diameter 6 mm................. 36 T. cornellana bolivarensis, n. subsp. Paratype from Loe. No. 3222, length 11 mm., greatest diameter 5 mm............... 36 T. G-A-Weaveri, n. sp. Paratype from Loc. No. 52, length Al inn, CAPONE ChAMMMASIGere BK IMT ocncsssoasnsoconsossenososcounmtese: 34 T. boweni, n. sp. Paratype from Loc. No. 1217, length 11 mm., greatest diameter 4.5 mM................cc0.cscccecceseecceeereeers 36 T. boweni, n. sp. Holotype from Loc. No. 1217, length i smmes, onreatest, diameter Sy antisera tees stesee ence eee eee 36 T. andreasi (Williston MS.). Paratype from Loc. No. 1217, length 12 mm., greatest diameter 5 mM................0..s0cceeeeee eee Bie T. andreasi (Williston MS.). Holotype, collected by Mr. S. H. Williston, length 22.8 mm., greatest diameter 12.1 mm. 37 T. andreasi (Williston MS.). Paratype from Loe. No. 1217, length 10 mm-., greatest diameter. 4 mim... 37 T. boweni, n. sp. Paratype from Loc. No. 1217, length 3 mm, greatest diameter S immense ee 36 T. cornellana, n. sp. Paratype from Loc. No. 3222, length 16 mm., greatest diameter 4 mM..............c0..cseccceeccceeceseecescceass 35 T. andreasi (Williston MS.). Paratype from Loe. No. 1217, length 18) mm, greatest diameter 7 mim-.....-..2...-.-..- eee 37 T. cornellana bolivarensis, n. subsp. Holotype from Loc. No. 3222, length 21 mm., greatest diameter 8 mm................. 356 T. cornellana, n. sp. Holotype from Loc. No. 3222, length 15.5 mm-, greatest diameter 3:5) mim.......4.:..-.-- eee 35 PL. 28, VoL. 11 Buti. AMER. PAL. No. 45, PL. 24 Fig. 1. ey) Ve) 10. iil. 12. PLATE 25. PAGE T. buchivacoana canonensis, n. sp. Holotype from Loe. No. 341, length 13 mm., greatest diameter 11.6 mm......... 40 T. andreasi (Williston MS.). Paratype from Loc. No. 1139, length 15.4 mm., greatest diameter 8.3 mm..................2..000008 aT T. boweni, n. sp. Paratype from Loc. No. 1217, length 10 mm., greatest diameter 6.1 MM.................cccccceeceeceeeceeeeeeee 36 T. filacarmenensis, n. sp. Paratype from Loc. No. 1979, greatest diameter of upper whorl 6.5 mm..................2.cssec00eee 38 T. buchivacoana canonensis, n. subsp. Paratype from Loc. No. 341, length 13.5 mm., greatest diameter 7 mm............. 40 T. filacarmenensis, n. sp. Holotype from Loc. No. 1987, length 16 mm., greatest diameter 12 mM................s.cseseeeeeeeeeee 38 T. buchivacoana, n. sp. Paratype from Loc. No. 2028, length 15 mm., greatest diameter 9 mM..................0.c0cceseeeeee 39 T. buchivacoana, n. sp. Paratype from Loc. No. 1942, length 14 mm., greatest diameter 9.8 MM.................c0cecseeeeuees 39 T. buchivacoana, n. sp. Holotype from Loc. No. 2019, length 29 mm., greatest diameter of next to last whorl a 000 0 ee Ren eran ar aera irae eerie are ore once ron cbeceondocudcccoocoodaeancce 39 T. filacarmenensis, n. sp. Paratype from Loc. No. 1964, length 16.1 mm., greatest diameter 7.4 mm..................::0:000+ 38 T. elmenensis, n. sp. Holotype collected by the Miranda Exploration Company, length 23 mm., greatest diameter Co abs alge ene ae Mere Rp ane renee ner ne Re rer ee suremusucudosconHDosco0ncn072 Al T. filacarmenensis, n. sp. Paratype from Loc. No. 1979, length 4.6 mm. Young whorls enlarged..................cceccceceoees 38 PL. 29, VoL. 11 Buti. AMER. PAL. No. 45, PL. 25 FIG. il. 6. PEATE 26: ? T. altilira Conrad, from Gatun, C. Z., Loc. No. 1101, fig- ured for comparison. Length 25 mm., greatest diameter 5.1 mm. T. carlottz, n. sp. Paratype collected by Dr. Maury from Rio Gurabo at Los Quemados, Santo Domingo. Length 21 mm., greatest diameter 7 mm T. guppyi Cossmann morantensis, n. subsp. Paratype with protoconch from Bowden, Morant Bay, Jamaica, Loc. No. 1109. Length 7.85 mm., greatest diameter 1.7 mm............. T. altilira Conrad urumacoensis, n. subsp. Paratype from Loc. No. 206A, length 14.8 mm., greatest diameter 8 mm. T. guppyi Cossmann morantensis, n. subsp. Paratype from Bowden, Morant Bay, Jamaica, Loc. No. 1109. Length 22.2 mm., greatest diameter 4.9 mm T. guppyi Cossmann morantensis, n. subsp. Paratype from Bowden, Morant Bay, Jamaica, Loc. No. 1109. Length 27 mm., greatest diameter 12.3 mm T. altilira Conrad urumaccensis, n. subsp. Holotype, col- lected by Mr. S. H. Williston, length 72 mm., greatest diam- eter 18.8 mm T. guppyi Cossmann morantensis, n. subsp. Holotype from Bowden, Morant Bay, Jamaica, Loc. No. 1109. length 59 mm., greatest diameter 12.6 mm PAGE 45 43 42 44 42 42 44 PL. 30, VoL. 11 BuLL. AMER. PAL. No. 45, PL. 26 FIG. il, (SS) =i d Ws rae ie PAGE T. guppyi Cossmann. Holotype, photograph by Dr. R. S. Eassler of the U. S. National Museum. Length of cast 17.5 mm., greatest diameter 8 Wim............0.-ce-----2+-ceeceseceeeeeene 42 T. berjadinensis socorroensis (Williston MS.). Paratype, collected by Mr. S. H. Williston, length 23 mm., greatest diameters: 18). samimstaceet ies cireiSeesas. seiwestacenaies seats eee 28 T. altilira Conrad HAASE SS. n. subsp. Paratype from Loc. No. 118, length 19 mm., greatest diameter 13.5 mm. 44 T. altilira Conrad urumacoensis, n. subsp. Paratype from Loc. No. 118, length 31 mm., greatest diameter 16 mm..... 44 T. altilira Conrad urumacoensis, n. subsp. Paratype from Loc. No. 118, length 38.5 mm., greatest diameter 8 mm..... 44 T. altilira Conrad urumacoensis, n. subsp. Paratype from Loc. No. 121, length 28 mm., greatest diameter 9 mm......... 44 T. altilira Conrad urumacoensis, n. subsp. Paratype from Loc. No. 121, length 14 mm., greatest diameter 5 mm......... 44 T. vistana, n. sp. Paratype from Loc. No. 1033, length 30 mm., greatest diameter 8 mMM.................cccececeeeeceeneceeecneeeees 46 _T. vistana, n. sp. Paratype from Loc. No. 1033, length 22 mime Seacrest diameters) Miimeececsssesssesce ces eceeee estes nee eee 46 T. altilira Conrad urumacoensis, n. subsp. Paratype from Loe. No. 121, length 22 mm., greatest diameter 6.2 mm......... 44 T. carlottz, n. sp. Holotype, collected by Dr. Maury from Rio Gurabo at Los Qucmados, .Santo Domingo. (Same specimen as figured in Bull. Amer. Pal., Vol. V, Pl. 48, fig. 15.) Length 20.5 mm., greatest diameter 5.5 mm....... 43 T. vistana, n. sp. Paratype from Loc. No. 122B, length 34 mms, Sreatest diameter 7-5) Wiessssse-sseesseste see eee ee eee eeeeeee AG PL. 31, Vout. 11 BuLi. AMER. PAL. No. 45, PL. 27 FIG. il PLATE 28. PAGE T. plebeia Say A-L-Owensi, n. subsp. Paratype from Loe. No. 298, length 12 mm., greatest diameter 7.3 mm............... Bil T. altilira Conrad mirandana, n. subsp. Paratype from Loc. No. 154, length 18 mm., greatest diameter 9.1 mm....... 45 T. altilira Conrad. Collected from Gatun, C. Z., Loe. No. 1101. Figured for comparison. Length 47 mm., greatest diameter U0 sammie: 2525. o eecec sec cates nace sesuucecctes eee ee 45 T. guppyi Cossmann morantensis, n. subsp. Paratype from Loe. No. 1109, Bowden, Morant Bay, Jamaica. Length 60 mm., greatest diameter 14.6 mim...................00.c0ecceeceeececeeees 42 T. gilbertharrisi falconensis (Williston MS.). Holotype. Same specimen as Pl. 14, fig. 6. Length 59 mm.................--.- 9) PL. 32, VoL. 11 Buti. AMER. PAL. No. 45, PL. 28 Fig. PLATE 29. PAGE T. altilira Conrad, from Gatun, C. Z., Loc. No. 1101. Fig- ured for comparison. Length 23 mm., greatest diameter aa Bd iD Os eee eter ein fee ine ner enn Pa EEE PRB ETA a ncpoaoooonene= 45 T. vistana nichelsi (Williston MS.). Paratype from Loe. No. 1552, length 27.5 mm., greatest diameter 17 mm........... 48 T. bifastigata Nelson democraciana, n. subsp. Paratype from Loc. No. 86. Base enlarged showing scultpure; ereatest. diameter, 1629) aMmimnh.ncsesecstsaccecsosees eee econo ee eee DO T. vistana, n. sp. Paratype from Loc. No. 1033, length 10 mm., greatest diameter 3.8 MM................ccccccecceeseececseeceeeee 46 T. vistana nicholsi (Williston MS.). Holotype from Loc. No. 140, length 33 mm., greatest diameter 21.5 mm............... 48 T. robusta Grzybowski fredeai, n. subsp. Paratype from Loc. No. 1600, length 157 mm., greatest diameter 51 mm..... 13) PL. 33, VoL. 11 BULL. AMER. PAL. No . 45, PL. 29 FIG. il, to 6. PLATE 30. PAGE T. bifastigata Nelson. Lectotype kindly lent by Dr. Carl O. Dunbar of the Peabody Museum at Yale University. Length 61 mm., greatest diameter 19.1 MM............. ccc cece cee eeeeeeceeeeees T. bifastigata Nelson maracaibensis, n. subsp. Paratype from Loc. No. 70A, length 62 mm., greatest diameter 18 mm. T. bifastigata Nelson democraciana, n. subsp. Holotype from Loe. No. 1233, length 46 mm., greatest diameter 1K 001) 06 Baca neem enter Rn mn nce nan Cr rnc acaesne rican Gsdsodene paacanancossosaadesooce T. bifastigata Nelson maracaibensis, n. subsp. Paratype from Loc. No. 86, base enlarged, greatest diameter 21 mm. T. bifastigata Nelson democraciana, n. subsp. Paratype from Loe. No. 12838. Base enlarged to show sculpture; BROMGIS GChieyoNewere W4LG TWAT os .occcocpcac0b conc oobHanooDOHSAONDADEDDODDESGHOOHEOD T. bifastigata Nelson maracaibensis, n. subsp. Holotype from Loc. No. 72, length 58 mm., greatest diameter 20 mm. 48 48 50 48 50 48 PL. 34, Vou. 11 Buti. AMER. PAL. No. 45, PL. 30 satiate: ee 2.00 BULLETINS OF AMERICAN PALEONTOLOGY_ 2NA\ Vol. 11 No. 46 VENEZUELAN DEVONIAN FOSSILS ( Presented to the Graduate School of Cornell University in partial fulfillment of the requirements for the degree of Master of Arts) By NORMAN E. WEISBORD December 9, 1926 Harris Co. Cornell University, Ithaca, N. Y. Ge SA INTRODUCTION AND ACKNOWLEDGMENTS Although Mesozoic and Cenozoic deposits are well repre- sented:in northern Venezuela, knowledge of Paleozoic his- tory has hitherto been quite obscure. Rather recently, however, and mainly through the efforts of Mr. Charles W. Yeakel, a collection of Devonian fossils was secured near the headwaters of the Cachiri River in the northwestern part of Venezuela near the Colombian border. Inasmuch as the literature is nearly barren of information concerning these older deposits, this provides a comparatively new field of investigation in Venezuela for Paleozoic sedimentation. That the Devonian series were found in place I am as- sured by various investigators, though I am informed in a communication from Mr. Liddle, who accompanied Mr. Yeakel, that the majority of our fossils were collected from fioat, and that many of the specimens are from the shale phases in the series. A number of the specimens are in a fairly good state of preservation, but often the inaccessibil- ity of the cardinal areas and other diagnostic parts, whose characters define criteria for identification, prevents pre- cise determination. There is some variation in the rock matrix containing the fossils and in the absence of any notes relative to the stratigraphic succession of the beds, I cannot be assured of the homogeneity of the entire fauna, though as a general statement it seems reasonable to assert that the fossils show closest relationships to upper lower and lower middle Devonian species from the type localities of eastern United States. The types of rock represented are: 4 BULLETIN 46 224 1. A gray, fine-grained sandstone, containing fine flakes of mica and occasional patches of bright brown iron stains. 2. A dirty gray, rather compact shale, weathered to a dull brown color with numerous Bryozoa. 3. A gray-black, semi-crystalline limestone. These various rocks are all fossiliferous, and the fact that the lithological characters of some tend to grade into that of the others, together with the occurrence of some of the same species in the different matrices, warrants the temporary assumption that the beds may be considered as a stratigraphic unit. In addition to the above named rocks, a few fragments of a hard, rather coarse-grained, metamor- phosed sandstone are included in the collection, but it con- tains very few specifically identifiable forms. In the pale- ontological discussion, the type of rock in which each species occurs will be noted. To Prof. Harris of Cornell University, for whose interest and co-operation I have always been indebted, I tender my sincerest gratitude. I am also under obilgations to Mr. J. E. Brantly and Mr. C. R. Rider of the Venezuelan At- lantic Refining Company for their many kindnesses in facilitating the shipment of the fossils to me and for per- mission to publish this article. And finally to Mr. Yeakel, whose loss is keenly felt by all who knew him, we offer our belated appreciation. DESCRIPTION OF SPECIES COELENTERATA ANTHOZOA Cyathophyllum venezuelense, n. sp. Pl. 1, Figs. 1—5 Corallum simple, varying somewhat in outline. Some specimens are sub-cylindric, curved, elongated, while others are broadly conical in shape. The more elongate speci- mens bear slight circular irregularities or swellings at dis- 225 DEVONIAN OF VENEZUELA on tant intervals; epitheca thin; calice ovate, circular, or el- liptical, probably with a moderately deep cavity, the depth of which cannot be ascertained on our specimens because of adhering material; septa rather numerous, varying with the size of the calix, almost equal, straight or slightly bent, extending nearly to the center where they twist together to produce the appearance of a pseudo-columella. A longi- tudinal section shows that the tabule are quite prominent, their proximity to each other varying in different speci- mens. The distance between septa in vertical section av- erages slightly less than a millimeter. Dimensions: Alt. 105 mm., diam. of calice 50 mm. Alt. 55 mm., diam. of calice 46 mm. Alt. 45 mm., diam. of calice 32 mm.; 25 mm. In many respects this species is rather closely akin to C. pocillum Davis from the middle Devonian of Kentucky. One specimen suggests a relationship to the C. zenkeri Billings from the Onondaga formation of New York, while C. robustum and C. galerum Hall from the Hamilton forma- tion of western New York also exhibit similarities. Des- pite this seeming relationship to middle Devonian forms, there are also a number of lower Devonian corals, especially from Kentucky, that are not unlike this in their characters. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Diphyphyllum vermetum, n. sp. Tab diy LE @y WS I Bi ne, a This species virtually comprises the limestone in which it is found imbedded. Coralla occurring in clusters, but as far as observed they are simple and not branching, sub- vermiculate in appearance, with numerous growth lines and irregular constrictions and swellings of low relief; longi- tudinal sculpture consisting of rather numerous septal fur- rows; longitudinal sections show the visceral chambers 6 BULLETIN 46 > 26 crossed by subequally spaced tabule; calyx probably shallow. sah Dimensions. Diameter of average corallum 4—6 mm. There seems to be no alliance between this species and any hitherto described from South America, though in North America the form D. verneuilanum from the middle Devonian has a suggestive relationship. From the latter, the present species is distinguished in the smaller size of the coralla. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Pleurodictyum venezuelense, n. sp. JG i, lies. Bh, 2 Corallum massive, rather small, subhemispherical, the upper surface somewhat unevenly convex, the under side slightly concave; in the superficial characters of the ex- terior it resembles Favosites forbesi of England, but dif- fers in its internal composition. In polished section, the center of the base presents an irregular, undefined mass, recrystallized by secondary calcite which seems to indicate that the corallum may have been attached to some foreign body during life. Befoie polishing the base it was seen to have a thin epitheca, but so weathered as to obscure its structure; corallites polygonal, in close contact throughout, not surrounded by thickened margins; septa fairly numer- cus, about 35 on our specimen, which in longitudinal sec- tion diverge somewhat and rather frequently bifurcate; tabule faintly displayed, extending across the visceral © chambers, sometimes slightly flexuous, at other times straight, but not vesicular in character; mural pores small, arranged in a single irregular row upon each of the pris- matic faces of the corallites. D'mensions. Long diameter of corallum, 26 mm.; short diameter, 16 mm. 227 DEVONIAN OF VENEZUELA 7 I am not entirely satisfied with the generic determination of this form, though .ts characteristics indicate a Favositic relationship. Specifically, this tabulate coral seems not unlike the Ar- gentinian Plewrodictyum sp. Thomas (Zeit. Deut. Geol, Gesell, Vol. 57, p. 267, pl. 12, figs. 21, a, 1905) from the beds at Cerro del Fuerte, but neither the description nor figure of his poorly preserved form warrant accurate com- parison. The Brazilian P. amazonicum Katzer (Geol. Unt. Amazon., p. 192, pl. 2, figs. I-Id, 1903) has larger Michelinia-like calices. The shape of the corallum of Favosites argentina Thomas (op. cit., p. 268, pl. 12, figs. 20, a, b, and plate accompanying p. 268, 1905) is quite un- like this species, though F. forbesi, with which Thomas compared his form, is much more alike, the present species differing in its septal arrangement. ,Michelinia transitoria Knod (N. Jahr. f. Min., Geol. u. Pal., Beil. Bd. 25, p. 561, pl. 30, figs. 1-4, 1908) from the Icla beds of Bolivia is another South American species, but here again the Vene- zuelan form ‘s at variance in the chaarcters of the septa and in its smaller corallites. The small size of the calices distinguish this species from such Devonian forms as the Hamiltonian P. stylophorum Eaton, and P. problematicum Goldfuss from the Coblent- zian of the Rhine region and Onondaga of Illinois (Meek and Worther). Occurrence. In the limestone series. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. MOLLUSCOIDEA BRYOZOA Fenestella venezuelensis Nn. Sp. Plie2) esse 2eeo Zoarium a somewhat undulating flabellate expansion; o) BULLETIN 46 "aa fronds large, compact; branches very slender, ridge-shaped, gently curved, bifurcating at distant intervals; dissepi- ments short, horizontal or arched, from half to less than half the width of the branches. Zoecia in two ranges. Fenestrules tabulate, rectangular, about 20 in one centi- meter, the measurements taken in the direction of the longer axes; cell apertures,°as indicated by the filling of the cells, somewhat irregular in shape, about three to each fenestrule; the non-celluliferous faces of the fronds have a fibrous structure composing the more solid portions of the branches, and these branches are occasionally nodose, and carinated. , Dimensions. A fairly complete frond measures 50 mm. in altitude and 30 mm. in width. This species closely resembles the Upper Heldebergian (Onondaga) F’. parallela Hall (Geol. Sur. N. Y., Vol. 6, p. 107, pl. 44, figs. 8-18, 1887) from the vicinity of Buffalo, N. Y., and may subsequently prove to be identical, but in the absence of branches showing the celluliferous faces in- tact, and in the occasional nodosity of the non-celluliferous faces (which feature is not mentioned or figured by Hall) it seems advisable to consider this as new. In his list of fossils from the Devonian of Curua, Brazil, Katzer men- tions the presence of F. parallela but does not figure it, though it is probably a very similar species to the present one described. Occurrence. In the gray shale. Occasional impressions of a similar or identical form is also found in the dark- gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Polypora cachirita n. sp. Pl. 2, Figs. 4-6 Occurs with the preceding and differentiated from it by its broader branches and with a gerater number of ranges of cell apertures; zoarium an undulating expansion, in- fundibuliform, fronds large; branches strong, angular and 229 DEVONIAN OF VENEZUELA 9 generally carinate, with an average width of .6 of a milli- meter; bifurcations distant. Interstices slightly wider than the branches. Dissepiments somewhat narrower than the branches. Fenestrules subquadrangular to subovate; cell apertures usually in three to five ranges as indicated by the fillings impressed on occasional specimens. There are about nine fenestrules in the length of one centimeter. Dimensions. An incomplete portion of a frond measures 40 mm. in altitude and about 20 mm. in width. This species suggests a relationship to a number of Onon- daga Polyporid forms in western New York, of which we may cite P. robusta Hall as a typical representative (Geol. SrnmiNe Ye. VOl 6, p. 156, pl. 34, figs. 4—7, 1887). Not-un- like this in all probability is a specifically unidentified Fenestella recorded by Katzer (Bol. Mus. Paraense, Vol. 2, p. 210, 1897-98) as an impression in the Devonian rocks of the Maecurt River of Brazil, which he says is somewhat of the same type as P. cultella Hall. This is of the same group as robusta previously mentioned above. Occurrence. In the gray shale. The ceil impressions have afforded easy access of weathering agents and hence many of the specimens are largely stained with iron. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. BRACHIOPODA Dalmanella (?) venezuelensis n. sp. PAL BR he (7) Shell quite small, suborbicular. Cardinal angles rounded. Greatest width of shell somewhat greater than length of the hinge-line. Beak small, projecting slightly above the cardinal area. Dorsal valve from which the specimen is described is depressed-convex, with a shallow mesial de- pression extending the length of the shell, rapidly widening anteriorly. Surface covered by 24 strong longitudinal riblets. 10 BULLETIN 46 230 Dimensions. Alt. 5 mm., long. 6 mm. The generic position of this species is doubtful. The form is characterized by its broadly circular outline and unusually strong ribbing. It is somewhat the same type of shall as Orthis ? sp. Reed (Ann. 8S. Afr. Mus., Vol. 4, p. 175, pl. 21, fig. 6, 1908) from the Bokkeveld beds of Africa, but differs in its broader medial sinus and short, straighter hinge-line, though from Reed’s figure it would appear as if the Bokkeveld shell were weathered posteriorly, hence ac- counting for the apparently sloping hinge-line. Occurrence. In the gray, micaceous, fine-grained sand- stone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Stropheodonta (Leptostrophia) caribbeana Nn. sp. Pl. 3, Fig. 1 Shell semi-elliptical; pedicle valve slightly inflated, the maximum inflation coming at the umbos; length somewhat less than the width; hinge-line equal to the width of the shell below; anterior margin well rounded, the lateral mar- gins nearly straight; cardinal angles sub-acute; beak ap- pressed, scarcely rising above the hinge-line. The surface is covered with a series of fine, unequal, radiating threads, gently undulating, increasing by bifurca- tion, and crossed by fine, even, concentric striz; in addi- tion there are about 16 subregular, concentric undulations which follow the outline of the valve; these are weaker at the apex but become more prominent anteriorly. Of the fine, radial riblets there are about two hundred in number. Dorsal valve partially and obscurely represented. Dimensions. Alt. 24 mm., long. 31 mm. To South American species, this is closest perhaps to S. argentinus Thomas (Zeit. Deut. Geol. Gesell., Vol. 57, p. 261, pl. 13, figs. 27, 28, a, 1905) from Cerro del Fuerte; the latter differs in its weaker concentric undulations, and 231 DEVONIAN OF VENEZUELA II in its rather pronounced anterior genuflection, though this may have been caused by distortion. This is also like the form Knod referred to as S. perplanas (probably not of Conrad) from the Icla and Conularia beds of Bolivia (N. Jahr. f. Min., Geol. u. Pal., Beil. Bd. 25, p. 540, pl. 27, fig. 4, 1908) but here again the strong concentric undulations are wanting; the Bolivian form is also more finely ribbed. Nevertheless the relationship is very close, as is that of Strophomena sp. Ulrich also from Bolivia, which Knod considers syonymous with his S. perplana. In North America this species stands intermediate between the Lower Heldebergian S. becki Hall and the typical Hamil- tonian S. perplana (Conrad). The concentric undulations are very similar to those on beck?, but the latter has coarser radial striz than has the Venezuelan species. To perplana this exhibits a striking resemblance, but the North Ameri- can form is generally less strongly undulate and with some- — what finer radial striz. Still, the Venezuelan species may be particularly well developed in its sculptural details and a more complete suite of specimens might prove it to be identical with forms of perplana. In South Africa, S. cf. concinna (Morris and Sharp) reported by Reed, is a simi- lar species but lacks the undulations of the surface. Occurrence. In the gray weathered shale. The dark gray limestone also has a form that may be the brachial valve of. this species. Locality. Upper course of the Cachwi River. Type. Cornell University Paleontological Laboratory. Stropheodonta zuliana n. sp. Pl 2 hioss Species represented by several external molds of the brachial valve; shell rather small, nearly flat, hemispheral, with a well rounded anterior margin and straight hinge- line having the full width of the shell; cardinal area very narrow; sculpture consisting of a series of distant, sub- regular, rather fine, elevated filiform striz, between which are still finer threads with an occasional coarser one. The 12 BULLETIN 46 major radii are more pronounced in the center of the valve than laterally, especially near the hinge where they become obsolete. Dimensions. Alt. 6 mm., long. 9 mm. This very pretty species at once recalls the Oriskanian Brachyprion majus Clarke from Becraft Mountain, N. Y. (See Mem. N. Y. State Mus., No. 3, Vol. 3, p. 56, pl. 8, figs. 1-7, 1900). Comparison with the brachial valve, how- ever, recalls that the Oriskanian species is more elongate than the Venezuelan shell, and considerably larger. Oc- casional forms of Leptostrophia interstrialis (Vanuxem) of the Portage (Upper Devonian) are not unlike this spe- cies as illustrated in the Maryland Geological Survey Re- port, p. 554, pl. 48, fig. 7, 1913. Occurrence. In the gray, fine-grained, micaceous, sand- _ stone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Schuchertella (?) sp. aff. sullivani (Morris and Sharpe) Pl.3, Fig. 2 Orthis Sullivani Morris and Sharpe. Quart. Journ. Geol. Soc. Lon- don, Vol. 2, p. 275, pl. 10, fiig. 1, 1846. Strophomena sullivani Sharpe. Trans. Geol. Soc. London, Vol. 7, p. 209, pl. 26, figs. 18, 19, 1856. Strophomena sullivani Reed. Ann. S. Afr. Mus., Vol. 4, p. 170, pl. 20, fig. 8, 1903. Schuchertella sullivani Clarke. Foss. Dev. Parana, p. 279, pl. 23, figs. 16-23, 1913." A fragment of the basal part of a large subovate form closely resembles the above named species in the character of its ribbing. S. sullivani has been reported from the Falkland Islands, the Bokkeveld beds of South Africa and from Parana, Brazil. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Specimen. Cornell University Paleontological Labora- 233 DEVONIAN OF VENEZUELA 13 tory. Chonetes (?) zuliensis n. sp. IEG By ges Shell small, subelliptical, wider than high; hinge-line straight, equal to or greater than the width of the shell be- low; cardinal angles subacute, anterior margin well rounded; pedicle valve very convex umbonally, flattening somewhat toward the front; umbos broad and full, beaks projecting slightly over the hinge line, the spines of which are obscured (if any). Surface ornamented with about twenty-five strong, subequal, radiating ribs which seem to increase by bifurcation. The ribs are high, subangular, separated by well channeeld interspaces. Careful inspec- tion reveals in addition a very fine series of concentric growth striz. Dorsal valve missing. Dimensions. Alt. 7.5 mm., long. 10 mm. This rather scantily ribbed form very strongly suggests a more than casual relationship to Chonetes stubeli Ulrich 4 from the sandstone of the Rio Sicasica between Oruro and La Paz, Bolivia. Both Ulrich’s description and figure of the species, however, show a medial sinus on the pedicle valve, which is lacking on the present species. The con- vexity of the Venezuelan form is also greater’ than that of the Bolivian but in other respects the two are nearly iden- tical. From the Bokkeveld beds of Africa, Reed’s C. aff. settiger ? Hall, is also allied to zuliensis, but here again the presence of a median depression and less gibbosity of the pedicle valve serve to temporarily differentiate the two. The South African form also has a few more ribs than our shell. With the exception of its somewhat greater number of ribs, Reed’s species compares very favorably with Ulrich’s. In Brazil, the Erere sandstone produces a small Chonetes, named herbert-smithi* which is not unlike zuliensis, but is 1N. Jahr. f. Min., Geol. u. Pal., Beil. Bd. 8, p. 80, pl. 5, figs. 3, 4, 1902. 2 Ann. S. Afr. Mus., Vol. 4, p. 174, pl. 21, figs. 4, 5, 1908. 3 Bull. Buff. Soc. Nat. Hist., Vol. 1, p. 251, pl. 10, figs. 39-42; 44-47, 1874. I4 BULLETIN 46 Cael ‘more finely ribbed and less robust. Clarke’s figure 4, pl. 24, of Chonetes falklandicus (Dev. Foss. Parana, 1913) is very much like the present shell except for its more numerous ribs, but I am not prepared to call the Venezuelan species falklandicus, for even though it is much like one of the varietal phases of what Clarke interpreted as falklandicus, it is clearly different than the original portrayal of that species by Morris and Shape. True, there may be the gra- dations from finely to coarsely ribbed forms of falklandicus, but in the absence of intergrading shells of the Venezuelan species I proposed for ours the new name zuliensis. Haug — (Doe. Scientifiques Saharienne, pl. 14, figs. 4, 5, partim, 1905) has figured an unnamed form on a slab of rock also containing Stropheodonta (Leptostrophia) oriskania Clarke from the lower ? Devonian sandstone of Tassili which superficially compares favorably with this species. Occurrence. In the gray to black limestone and in the gray, weathered shale. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Chonetes venezuelensis Nn. sp. BINS, Higss 4 5os Onn (cme Shell’ of medium size, hemispherical in outline, with evenly rounded margins and straight hinge line equal to the width of the shell below or slightly shorter; cardinal ex- tremities obtusely angular; pedicle valve strongly convex, with or without a slight medial depression; the beak rises barely above the hinge-line, is small and appressed; hinge area very narrow. The sculpture consists of about eighty radial riblets, subequal to equal in size, separated by inter- Spaces equally as wide. Where a portion of the original shell has been retained at the sides, the intercostal spaces are observed to be punctate, while the ribs, which reflect the sculpture of the interior of the valve, are with small spicules. The punctations and spinosities are better de- veloped laterally than centrally; cardinal spines of hinge- 235 , DEVONIAN OF VENEZUELA 15 line obscured, if any are present. Brachial valve nearly flat with a slight central inflation which is barely sulcate. An impression of the internal mold shows the ribs to bear small spines as on the opposite valve. Dimensions. Alt. 14 mm., diam. 17.5 mm. It seems probable that the specimens figured as C. cf. coronatus Conrad by Reed (Ann. 8. Afr. Mus., Vol. 4, p. 172, pl. 20; figs. 11, 12; pl. 21, 1908) is somewhat similar to our species. The Venezuelan form, however, is much more nearly allied to the species Reed has doubtfully re- ferred to as coronatus than to typical forms of coronatus from New York which are accessible to me for comparison. Fig. 8, pl. 3 shows a slightly distorted mold with a medial incision suggesting the characters of Hodevonaria. I am not sure, however, that the specimen represents this species. Occurrence. In the gray, fine-grained, micaceous sand- stone. A similar form also occurs both in the limestone | and shale but their identification is uncertain. Locality. Upper course of the Cachiri River, State of Zulia. : Type. Cornell University Paleontological Laboratory. Chonetes (Eodevonaria) subhemispherica n. sp. _ PI. 3, Figs. 9; 10 ? Shell semi-elliptical to semi-circular. Ventral valve ex- tremely gibbous, especially below the beak; anterior mar- gin well rounded; hinge-line straight, apparently denticu- late on the external border; umbo more or less gibbous, beak appressed and incurved. Dorsal (?) valve (Fig. __) nearly flat, with a medial elevated rib extending from the beak down the center of the valve. External border of hinge clearly crenulate. Surface marked by about fifty low, equal, evenly rounded riblets, crossed perhaps by concentric striations which are not clearly visible on the specimen. There seems to be a tendency for the extremities of the hinge-line to become ex- tended or auriculate. 16 BULLETIN 46 - 236 Dimensions. Ventral valve, alt. 14 mm., long. 16 mm. This species shows unmistakable affinities to the Oris- kanian and Onondagan C. hemispherica and C. arcuata Hall, though it is somewhat smaller than mature specimens of these two forms. The presence of the genus Hodevo- naria is rather good evidence of the upper lower or lower middle position of the Venezuelan Devonian. Occurrence. In the dark gray to black limestone. The dorsal valve of what seems to be this species is found in the gray shale. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Conchidium (?) sp. indet. Ply 3 higaae General characteristics as figured. The valve is strongly convex, and ornamented with a series of strong, elevated plications which become nearly obsolete laterally. Below the umbo, some concentric wrinkles suggest the assumption that the radial plications are crossed by rather strong con- centric striations. A rather wide sulcus appears umbonally but the appearance may be deceptive due to crushing of the valve. Cardinal area partially weathered and obscured on one side of the beak and broken on the other making even the generic identification doubtful. Dimensions. Alt. 29. mm., long. 34 mm. Occurrence. In the dark gray limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Amphigenia (?) sp. indet. Pl. 4, Fig. 1 The umbonal portion of a large species is very doubt- fully referred to this genus. The partial contour of the shell indicates that it is probably triangularly elliptical in outline. Beak low, appressed; cardinal margins sloping rather acutely from the beak and making an angle of about 120 degrees with each other. The ventral (?) valve is strongly convex, especially umbonally where it is subangu- 237 DEVONIAN OF VENEZUELA 17 larly elevated into a rather high, rounded prominence. Al- though this eminence continues with undiminished strength to the anterior portion of our specimen, it probably becomes depressed anteriorly on a complete shell. Cardinal area obscured by adhering material, but rather broad for the genus. The surface is marked by rather distant concentric weak lamellz which are broadly V-shaped, and a series of radiating riblets which are just barely discernible on our specimen. Dimensions. Alt. ?, long. 50 mm. This very interesting shell very unfortunately has the cardinal area partially broken and obscured, hence making the generic determination extremely doubtful. From the general contour of the cardinal area, it would seem entirely too broad for the genus, but the characteristic angulation of the valve, and the radial ribbing (unfortunately nearly all weathered) suggests a Rensselarid relationship of the form. We await better material for more accurate identi- fication. Atrypa cf. reticularis (Linne) AR oie 2. Be * Anomia reticularis Linne. Syst. Nat., ed. 12, p. 1132, 1767. Terebratula reticularis De Verneuil. Geol. of Russia and Ural Mts., Vol. 2, pl. 10, fig. 12, 1845. Atrypa reticularis Morris. Cat. British Fossils, p. 132, 1854. Atrypa reticularis Hall. Nat. Hist. N. Y., Vol. 4, pt. 1, p. 316, pls. 51-53a, 1867. Atrypa reticularis Schuchert. Bull. 87, U. S. G. S., p. 154, 1897. See also for complete bibliography up to 1897. Atrypa reticularis Weller. Geol. Surv. N. J., Vol. 3, p. 236, pl. 21, figs. 35-37, 1903. Atrypa reticularis Schuchert and Maynard. Md. Geol. Surv., p. 392, pl. 67, figs. 26-28, 1913. Venezuelan shells rather large, subcircular; pedicle valve moderately gibbous about the umbo, depressed anteriorly ; cardinal angles well rounded; beak fairly prominent, rising somewhat above the hinge; Brachial (?) valve similar to 18 ‘BULLETIN 46 238 the preceding but somewhat less inflated. Surface of valves marked by coarse, subequal, radiating riblets, of which there are about five to five millimeters at the base of the shell. These are somewhat rounded and separated by interspaces of nearly equal width. Though not dis- tinctly shown, the valves are crossed by concentric, lamel- lose extensions of the shell which have been partially oblit- erated by weathering. Interior of valves inacessible. Dimensions. Alt. 32 mm., long. 35 mm. Alt. 36 mm., long. 39 mm. Curiously enough, this is the first time this widespread and cosmopolitan species has been recorded from South America, and is as yet unknown in the Bokkeveld beds of South Africa as far as lam aware. Our specimens, though somewhat the worse for wear, appear to be the normal type of reticularis, but in the absence of stronger sculptural details and with the interior inaccessible, identity with the above is provisional. Geographically and stratigraphic- ally, this form is known for its efflorescence. Ranging from England, the Continent, North and now South Amer- ica, with a vertical span from Silurian to uppermost De- vonian (Chemung), this species is of little value in discrim- - inating horizons. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Specimen. Cornell University Paleontological Labora- tory. Spirifer meridioamericana N. sp. PI. 4, Figs. 4, 5 This species is represented by a poorly preserved ventral valve and a fragment of another specimen which shows to better advantage the sculpture of the form. Shell of me- dium size, transversely subelliptical, rather compressed ; beak fairly high, prominent, but small and slightly in- curved; anterior margin of the valve broadly rounded; car- dinal margins concavely sloping from the beak; cardinal e 239 DEVONIAN OF VENEZUELA 19 angles rounded; sinus shallow but well defined, broadening anteriorly. The major radial plications are relatively low, rounded, broad, about five or six in number, decreasing in prominence from the sinus to the periphery. On the more complete specimen which seems to be fairly mature, a se-. ries of fine, but rather strong striz longitudinally traverse the valve and are interrupted occasionally by small nodular growths. In addition to these secondary threads, numer- ous, fine, concentric growth lines cross the valve. The sculpture on the fragment is as follows: (1) a number of strong, radial undulations like those described above; (2) these major plications are crossed by widely spaced, overlapping, concentric lamelle, and (3) a finer series of concentric growth strize; (4) on the lamellae are a number of elongated pustules or spiny processes. Al- though these latter radial lineations are disconnected from those on the lamella above they are in alignment, and it seems probable that it is the growth of these that form the secondary radial lines observed on the larger specimen. Dimensions. Alt. 15 mm., long. 23 mm. This is a very interesting species, and I am by no means satisfied that my diagnosis is entirely correct, having to reconstruct the whole from unsatisfactory fragments. It is apparently of much the same type as S. kayserianus Clarke (Foss. Dev. Parana, p. 252, pl. 19, figs. 1-15, 1918) from Brazil, a species which presents numerous different aspects, depending upon its stage of growth. Occurrence. In the gray weathered shale. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Museum. Spirifer venezuelensis n. sp. : Pl. 4, Fig. 6 Shell transversely subelliptical, of medium size, moder- ately gibbous, with a straight hinge-line which represents the greatest width of the shell. Ventral valve (from which this species is described) inflated centrally, regularly curv- 20 BULLETIN 46 240 ing to the front and sides; sinus well defined, rather deep and rounded or subangular; beak small but sharp, rather high, somewhat incurved over the narrow cardinal’ area. Surface marked by five strong plications on either side of the medial sinus; the plications are high, rounded, sepa- rated by narrow interspaces, and become weaker laterally ; these are crossed by imbricating lamellose strie rather closely spaced and studded with very fine, sometimes slightly vermicular lines, arranged in parallel bands ac- cording to the distance of the concentric lamelle. The sculpture is very similar to that of Reticularia fimbriata — (Conrad) from the Hamilton formation of North America, but the shape of the shell and strong longitudinal folds serve to differentiate the South American species. Dimensions. Alt 16 mm., long. 19 mm. This beautifully sculptured species may be identical with the previous form, but it seems advisable to consider them distinct for the present. Strikingly similar in sculptural details is the Eifelian S. aculeatus of Europe. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Spirifer audaculus zulianus n. subspecies Pl. 4, Figs. 7, 8; Pl. 5, Figs. 1-8 The most characteristic form of the Venezuelan Devonian in our collection is a species very closely allied to the S. au- daculus,macronata type of shell from the middle Devonian deposits of eastern United States. Ventral valve larger and deeper than the dorsal, with a high, prominent, slightly incurved beak. Cardinal area high, triangular, slightly concave, divided in the middle by a deltoid fissure which is twice as high as wide, and reaches the apex of the area. Cardinal line straight; mesial sinus rather wide, of moderate depth, widening fairly rapidly 241 DEVONIAN OF VENEZUELA 21 from the apex to the anterior margin, and generally rounded. Cardinal margins obtusely sloping from the beak. Dorsal valve less convex than the pedicle valve and not as high; beak small, appressed, and slightly incurved; area obscured but probably linear. Mesial fold well defined, convex, rounded on top (sometimes angular), rapidly ex- panding from the apex. Sculpture of both valves consisting of an average of about twenty to thirty equal plications on each half of the valve, rounded, separated by narrow interspaces, and prob- ably crossed by numerous growth strize which are not vis- ible on the specimen. Both valves are occasionally inter- rupted by heavier concentric growth bands. , Dimensions. Ventral valve: alt. 31 mm., long. 46 mm; alt of cardinal area 15 mm. Dorsal valve of another speci- men: alt. 30 mm., long. 58 mm. This species closely resembles the Marcellus to Hamil- tonian S. audaculus Conrad, and indeed:some of the dorsal valves are practically identical. The ventral valves of the Venezuelan specimens, however, have a somewhat higher cardinal area and slightly less incurved beak than the typi- cal audaculus. Nevertheless there are varietal forms of the latter with high areas that are closely allied to the present species such as the form known as the variety eatont. In the height of its cardinal area our form recalls such species as S. manni and S. granulosus but the fewer ribs of the latter two forms, and the bipartite mesial fold of granulosus serve to differentiate them. Figure 7, Plate 5, shows a form which apparently grades into this species, but also resembles the Brazilian S. pedro- anus Hartt and Rathbun from the Erere Devonian of Bra- zil. Of pedroanus, Rathbun (Bull. Buff. Soc. Nat. Hist., Vol. 1, p. 237, pl. 8, figs. 1-9, 18, 14, 16-20, 1974) writes this interesting conclusion: “Prof. Hall, who examined a 22 BULLETIN 46 Beis small number of specimens of this species of Spirifer after the above description was written, thinks that in its differ- ent varieties it is very closely related to several American Devonian Spirifere; S. varicosus, Corniferous limestone; S. medialis (audaculus) Hamilton group, which varies much in form; S. angusta, Hamilton group, perhaps only a variety or young form of S. medialis; and S. macrus of the Corniferous limestone, which last species, however, has gen- erally a narrower and more curved hinge area. S. pedro- anus, therefore, appears almost like a connecting link, unit- ing the above named species in a single series.” Occurrence. In the dark gray to black limestone; in the gray micaceous sandstone; and possibly in the weathered shale. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Spirifer sp. cinder 12, Bi, Iie, 12 General features as feuved: The obliteration of the sculptural details does not allow specific characterization. Dimensions. Alt. 14 mm., long. 22 mm. (approxima- tion). Unlike the majority of the other forms, this shell occurs in the metamorphosed sandstone. Locality. Upper course of the Cachiri River, State of Zulia. Specimen. Cornell University Paleontological Labora- tory. Vitulina (?) venezuelensis n. sp. | Pl. 5, Figs. 9-13 This very interesting little species does not adequately fit into any well known genus of Brachiopod, and may possibly be a new genus, in which case I propose the new name of Venezuelia. In its general aspects it appears to belong to the Coelospiridz# showing affinities to such genera as Lep- tocoelia and Anoplotheca, though its straight hinge-line to on 2 ar DEVONIAN OF VENEZUELA more strongly suggests Vitulina. Here again, however, the character of the ribbing is somewhat diser epant and makes reference to that genus temporary. The form is relatively abundant but is represented in the collection only by pedicle valves. These vary in gibbosity, some being very ventricose, others less so. The shells are generally hemispherical in outline with well rounded an- terior margins, and relatively long, straight hinge-lines; cardinal margins depressed; umbos generally full, beaks small. Surface marked with eight to eleven strong, rather distant, angular plications. ‘There is generally one medial plication, flanked by others of equal size, which play out rather sharply toward the lateral margins; concentric lamelle and a few unusually strong concentric riblets tra- verse the radial ribs. The pustular markings character- istic of the genus may be present but have not been observed. Dimensions. Alt. 6.5 mm., long. 7.5 mm. This species is quite different from the well known North and South American V. pustulosa or Leptocoelia flabellites. It is readily recognized by its straight hinge, heavy radial and centrally located ribs, and strong concentric sculpture. Occurrence. In the gray, micaceous, fine-grained sand- ‘ stone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological.-Laboratory. Athyris aff. spiriferoides (Haton) JPA ( Teigee, AL _ Terebratula spiriferoides Eaton. Amer. Journ. Sci., Vol. 21, p. 137, 1831. Spirifera spiriferoides Hall. Tenth Rep. N. Y. State Cab. Nat. Hist., Planes. t, 2) 187. Athyris spiriferoides Hall. Pal. N. Y., Vol. 4, pt. 1, p. 285, pl. 46, figs. 5-31, 1867. Athyris spiriferoides Schuchert. Bull. 87, U. S. G. S., p. 149, 1897. Athyris spiriferoides Schuchert and Maynard. Md. Geol. Surv., p. 24 BULLETIN 46 2478 Pall joe, Blu, ess, it, @, ess Two shells in our collection may be compared with the common spiriferoides of the Onondaga and Hamilton groups of North America. One of our specimens is an ex- tremely broad and somewhat distorted individual but is probably the same species as portrayed in Fig. 1, which is a normal shell. The valve is transversely ovate, well in- flated, with a short hinge line and rounded cardinal ex- tremities. Maximum convexity umbonally and medially, the valve exhibiting a tendency to become widely sulcate toward the front. Beak small, somewhat incurved. Sur- face marked by concentric lamelle. Dimensions. Alt. 17 mm., long. 22 mm. Alt. 19 mm., leng. 32 mm. This genus has hitherto been unrecorded from South American Devonian deposits. The species is not unlike S. fultonensis (Swallow) which has the same stratigraphic range as spiriferoides. The relationships of this Venezue- lan shell again-show the probable lower middle Devonic affinities of the fauna. Occurrence. In the gray shale. Locality. Upper course of the Cachiri River, State of Zulia. Specimens. Cornell University Paleontological Labora- tory. MOLLUSCA PELECY PODA Aviculopecten yeakeli n. sp. Pl. 6; figs Shell rather large, sub-ovate, slightly oblique, length and height nearly equal; anterior and ventral margins regularly rounded; posterior margin of the disk proper forms a nearly even, oblique line to the beak. Left valve slightly inflated in the umbonal region, flatter ventrally. Hinge-line straight, approximately two-thirds the length of the shell. Umbo slightly elevated, sloping rather abruptly to the wings, the umbonial line forming an angle of somewhat 245 DEVONIAN OF VENEZUELA N un more than ninety degrees; ears unequal, the anterior smaller, the posterior alate and triangular; beak located anterior to the middle of the cardinal line; terminal mar- gins of the ears missing but probably somewhat concave. Sculpture consisting of about thirty-eight strong, ele- vated, subtriangular, radiating ribs, subequal in size with perhaps an occasional weaker riblet, separated by equally wide interspaces; on the posterior area of the valve there are about twelve fine, radiating riblets, more closely spaced and less pronounced than those on the disk proper. In ad- dition the valve is crossed by numerous concentric growth striz, but these are only faintly suggested on the specimen. Dimensions. Alt. 44 mm., long. 46 mm. This species is very closely allied to Katzer’s A. coel- hoanus from the “Spirifer sandstone” of the Rio Maecuru, Brazil, the latter differentiating primarily in having alter- nating stronger and finer ribs, and wide, flat interspaces. To A. pecteniformis and A. princeps (Conrad) the present species shows similar characteristics, but its broadly ovate outline and stronger ribs differentiate it from the afore- mentioned middle Devonian species of North America. Named in honor of Mr. C. Wesley Yeakel. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Aviculopecten sp. indet. TPL Sy, Leis A partial specimen doubtfully referred to Aviculopecten, shows surface markings of numerous, subequal, radiating riblets crossed by a series of fine, regular, closely spaced concentric striz. Characters of the upper portion of the valve not observable. The specimen probably represents an immature shell. Dimensions. Alt. 9 mm., long. 9 mm. (approximately). The form appears to be or the same general character as A. coelhoanus Katzer (right valve) from the Brazilian De- 26 BULLETIN 46 +246 vonian of the Rio Maecurit, but is very much smaller. Occurrence. In the gray weathered shale. Locality. Upper course of the Cachiri River. Specimen. Cornell University Paleontological Labora- tory. Conocardium sp. PI. 6, Figs. 4, 5 Two partial specimens, indicating a shell of generous pro- portions; is probably subovate-trigonal in outline when complete. Length greater than the height; beak very high, rather narrow, situated on the posterior two-fifths of the valve, and is strongly recurved as well as sharply keeled medially. Posterior area of the shell rather acutely trun- cated; umbonal slope continuing to the ventral margin; an- terior dorsal margin acutely sloping to form the cuneate anterior. Markings consist of unequal to subequal radiat- ing ribs, not clearly defined on our specimens, and numer- ous, fine, concentric growth striz. Probable dimensions. Alt. 50 mm., long. 60 mm. : This genus has hitherto been unrecorded from other De- vonian deposits of South America. The obscurity of cer- tain features and distortion of the shell, however, render the determination doubtful. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Specimen. Cornell University Paleontological Labora- tory. Cypricardinia subindenta Nn. sp. PEG Hick Cypricardinia indenta (Conrad) Hall. Geol. Surv. N. Y., Vol. 5, Tle Ie jo) 4S, joll, "WG, imss, GG, Ieee. Specimen small, subrhomboidal in outline, anterior mar- gin sharply rounded, ventral border nearly straight, with a slight sinuosity; posterior extremity acutely curved below, obliquely truncate above. A well defined but rounded um- bonal ridge extends from behind the beak to the posterior margin, marking off the post-cardinal depression ; dorsal 247 DEVONIAN OF VENEZUELA 27 margin straight. Left valve slightly inflated umbonally, flattening out below. Beaks anterior, small, appressed, probably rising slightly above the hinge-line. Sculpture consisting of about ten or twelve rather distant, concentric undulations, which, anterior to the middle of the valve, curve slightly toward the umbo, and which follow the con- tour of the valve’s outline on the depressed area behind; though not distinctly visible on the specimen, a series of finer concentric striz occur between the major undulations; some oblique radial markings are also barely discernible. Dimensions. Alt. 6.5 mm., long. 9.5 mm. In many particulars this shell closely resembles C. in- denta Conrad which ranges from the Oriskanian to Hamil- tonian horizons in North America. Characteristic of the latter species, however, is its chevron-like sculpture visible under conditions of suitable preservation; although the South American form shows some indistinct radial mark- ings, which may suggest the ornamentation of indenta, they are not clear enough to warrant positive comparison with the North American species, and hence I suggest the name subindenta until a later collection affords better material for comparison. Though Clarke’s Macrodon ? sp. from Parana (Foss. Dev. Parana, p. 184, pl. 16, fig. 8, 1913) has fewer major undulations, is more visibly depressed cen- trally and has a less oblique posterior margin, the natural variability of such a form as this, may indicate a closer re- lationship to our form than comparison of only one shell indicates. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. GASTEROPODA Diaphorostoma neveritanum N. Sp. Pl 6, bie ela, bigssel—s Shell subovate, globose, rather large, naticoid in outline; 28 BULLETIN 46 248 spire slightly elevated, obtuse; sutures linear, weakly in- cised; whorls about four, convex, increasing rapidly in size, the apical volution seemingly papilliform; body whorl glo- bose, convex laterally but with an apparent slight flatten- ing below the suture on what would ordinarily correspond tc the shoulder of the whorl; aperture nearly ovate, large; columellar lip thickened and folded, and seemingly reflexed over the umbilicus which is closed to view. Surface marked by a series of fine, subequal, flexuous, longutudinal, thread-like strize, much like those often occur- ring on the genus Pleurotomaria, and in addition crossed by equally fine, but perhaps slightly wider spaced revolving lines giving a beautiful reticulate appearance to the whorls. This latticed sculpture is observable on the less weathered portions of the shell, the longitudinal markings being otherwise the more prominent; as the longitudinal strize leave the suture they are curved convexly toward the aper- ture, but upon approaching the madial portion of the body whorl they are flexed rather sharply inward, again sug- gesting a relationship to the scu!pture of Pleuwrotomaria. The species, then, is at once distinguished by its Neverita- like shape, obtuse spire, g!obose whorls, reticulate sculpture and ovate aperture. Dimensions. Alt. 32 mm., diam. 30 mm. In affinities, the species is most nearly related to the Oriskanian D. desmatum Clarke of eastern United States, the latter differing, however, in its shorter and wider body whorl. The sculpture of the two forms, on the other hand, is very similar. In shape, the present species would ap- pear somewhat allied to Holopea furmaniana Hartt and Rathbun from the Ereré Devonian of Para, Brazil, but the diagnosis “surface as determined by external molds, smooth,” of that species does not harmonize with the lat- ticed ornamentation of the Venezuelan species. HExamina- tion of the type specimens of furmaniana show also that the shape of the shell is discrepant from that of our species. Littorina ? (Holopea) bainii Sharpe from the Devonian of 249 DEVONIAN OF VENEZUELA 2y South Africa is higher and more prominently spired. In Argentina, a weathered Naticopsis ? sp. Kayser, is a form of this type of shell. From Pebble Island, West Falkland, Clarke’s D. allardycei, though of the same order, is a less compact and less prominently reticulate species. In many characteristics this form is also like Plewroto- maria keyseri Ulrich from the Conularia beds of Bolivia, but the latter has a rather angulate body whorl with a nar- row circumventing band. Finally, there is a suggestion of similarity to Platyostoma lineatum (Conrad) which ranges from the Onondaga formation to the. Chemung in New York, but lineata differs in being more finely sculptured. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Type. Cornell University Paleontological Laboratory. Pleurotomaria venezuelensis n. Sp. IPA i, Innes, Al 1 Shell poorly preserved and somewhat _ distorted, depressed-trochiform; spire probably moderately elevated, but subdued in this particular specimen; apex small. Vo- lutions about five, depressed-convex on the upper side; last whorl ventricose and angulate; sutures incised, but prob- ably not as much as the weathered specimen indicates; sur- face ornamentation very obscure, but there seems to be a narrow peipheral sulcus on the spire whorls at the sutures; the penultimate whorl shows a slight revolving sulcus me- dially, but this, however, is obsolete on the other whorls and may simply represent an irregularity in the shell. There are evidences, nevertheless, of regular, rather strong radiating striz visible near the sutures, which bend back at the angulation of the whorl. Body volution with a rather large, subcircular and eee umbilicus underneath; aperture subquadrate. Dimensions. Alt. 138 mm., diam. 22 mm. There is much to be desired in the description of the spe- cies, but it seems advisable to include any generically rec- 30 BULLETIN 46 250 ognizable forms in a treatise dealing with a new area. Occurrence. In the gray, fine-grained, micaceous sand- stone. Locality. Upper course of the Cachiri River, State of Zulia. . Type. Cornell University Paleontological Laboratory. ARTHROPODA Phacops argentinus ? Thomas. Plo 7, Bisss7Gs a Phacops ef. rana Kayser. Zeit. Deut. Geol. Gesell., Vol. 49, p. 284, pl. 11, figs. 8-10, 1897. Phacops argentinus Thomas. Ibid, Vol. 57, p. 246, pl. 11, figs. 8—9b, 1905. A coiled form showing some of the thoracic segments and the pygidium appears to be closely allied to argentinus, though the imperfect preservation and absence of parts of both this and the Argentine specimens do not assure un- doubted identity. Discussing the characteristics of his species, Thomas mentions the pentahedral shape of the glabella, the well developed and projecting eyes, and the subtriangular shape of the pygidium with its six segments on the axis and on the sides. Unfortunately the cephalic portion of the Venezuelan specimen is missing; the thorax shows about seven seg- ments, which on the moderately broad, gently arched axis are rather strongly elevated, of moderate width, separated by well channeled furrows and firmly anchylosed with the pleura by articulations visible on the peeled carapace; dor- sal furrows not as pronounced as the distortion of the speci- men indicates; pleural areas somewhat wider than the axis; the maximum width of the thorax comes at the fourth or fifth segment from the pygidium. Pygidium subtriangular, rather short and of moderate breadth, with a characteristic narrow, depressed-convex, rather rapidly tapering axis, sculptured with about six or seven segments, of which the anterior four are the more 2) in DEVONIAN OF VENEZUELA ai prominent, the posterior members being indistinctly de- veloped; pleura broad, convex, carrying about seven seg- ments, the upper four well defined, flattened, rather broad, and separated by narrow sulci, the remaining smaller, nar- rower and indistinctly divided; border narrow and flat. Dimensions. Probable length of specimen 25 mm., diam. 18mm. Pygidium, length 7 mm., width 14 mm. The thorax and pygidium of this species is similar to that of a number of Phacopidx, and this taken in conjunction with the absence of the cephalon, makes differentiation diffi- cult. To P. rana Green (Onondaga and Hamilton of North America) this form shows a striking similarity and indeed it was originally referred to it by Kayser (op. cit.) ; Thomas, however, recognizes that argentinus has the frontal part of the glabella flatter, but I cannot attest to this difference because of the absence of the glabella on my single specimen. At present it seems best to separate argentinus from rana until more complete specimens are available, though it is interesting to note that Groth (Bull. Geol. Soc. France, (4), Vol. 12, p. 607, pl. 19, figs. 2, 2a, 1912) has referred to Phacops from the Icla beds of Bo- livia to rana. Other allied species are P. latifrons Bronn from the Eifelian of Europe. This form has also been re- ported by Salter from Oruro, Bolivia, but Thomas conjec- tures it may be his argentinus. P. africanus Salter (pars) as figured by Lake differs in its much broader pygidial axis, as does P. scirpeus Clarke from the Rio Macurt, Brazil. Again, as far as the parts are comparable, the Venezuelan species is similar to the Oriskanian P. logani of North America. In its affinities, then, it would seem to apper- tain to the upper lower or middle Devonian series of rocks. Occurrence. In the dark gray to black limestone. Locality. Upper course of the Cachiri River, State of Zulia. Specimen. Cornell University Paleontological Labora- tory. is>) No BULLETIN 46 252 CONCLUSIONS That the Venezuelan fossils have a Devonic expression seem unmistakable, though at present it is somewhat pre- . mature, with the limited number of forms we have studied, to state definitely with what known horizon they may be correlatable. Considering the fossils as a whole, however, and assuming the homogeneity of the fauna despits its oc- currence in somewhat varying rock matrices, it would seem as if the stratigraphic position of the beds ought to come between the Oriskanian and Onondagan formations as known from the type section in New York. Of the ascer- tained relationships with described species whose strati- graphic range is known, our fauna shows that the greater number of species are related to Onondaga forms, a some- what smaller have both Oriskanian and Onondagan affini- ties, while a few are akin to Hamiltonian species. The Corals, Cyathophyllum venezuelense and Diphyphyl- lum vermetum, exhibit resemblance to known forms in the middle Devonian beds of eastern United States. The tabu- late Pleurodictyum venezuelense contributes nothing defi- nite toward ascertaining the age of our fauna, but the Bryozoa Fenestella venezuelensis and Polypora cachirita which occur abundantly in the shale phase of the series, seem to be strongly suggestive of the Onondaga formation, the former being nearly identical with F. parallela Hall, the latter representing the P. robusta type. In the Brachio- pods, Leptostrophia caribbeana seems to lies between the Heldebergian becki and the Hamiltonian perplana, while Stropheodonta zulianum seems to be akin to Brachyprion majus from the Oriskany of Becraft Mountain, New York. The Hodevonaria division of the Chonetes group strongly reminds one of such characteristic Oriskany and Onondaga forms as C. hemispherica and C. arcuata Hall. Chonetes zuliensis suggests more than a casual relationship to the Bolivian C. stubeli described to be lower-middle Devonian inage. Atrypa reticularis is of no diagnostic value because B53 DEVONIAN OF VENEZUELA ae of its great stratigraphic range, but it is of interest in be- _ ing recorded for the first time in South America, giving it a further known geographical distribution. Spirifer me- vidioamericana and S. venezuelensis are fimbriate forms much like S. kayserianus Clarke from the Devonian of Parana, Brazil, and like the Hifelian S. alculeatus of Europe. It is in this genus that we get the closest affinity to Hamiltonian species, for S. audaculus zulianus is nearly identical with S. awdaculus, the well known middle Devonian form of eastern United States, Vitulina (?) venezuelensis is a very distinct form whose relationships have not been ascertained. Athyris aff. spiriferoides ranges from Onondaga to Hamilton in North America. The representatives of the Pelecypods are four in num- ber—A viculopecten yeakeli is of the same order as A. coel-- hanus Katzer from the Maecurt’ beds of Brazil... If the generic determination of Conocardium is correct, this will be the first notification of this genus from South America, though the distortion and obscurity of the specimen makes its identification uncertain. The most characteristic of the lamellibranchs, however, is the small Cypricardinia subin- denta which may subsequently prove to be identical with C. indenta (Conrad), a shell which ranges from Oriskanian to Hamiltonian in the United States. Because of the weathering of the sculptural details which prevents close identification, I have deemed it advisable to consider the Venezuelan species temporarily distinct. The Gasteropod Diaphorostoma neveritanum seems to -come between the Oriskanian D. desmatum Clarke and ihe middle Devonian Platyostoma lineatum (Conrad). With its regularly globose body whorl and beautiful sculpture it makes a most outstanding form. Pleurotomaria venezue- lensis is too poorly preserved for satisfactory comparison. The single imperfect Trilobite, Phacops argentinus ? Thomas has been found in the middle Devonian of Argen- tina, but the Venezuelan shell awaits better material for 34 BULLETIN 46 254 more accurate identification. . The facies of the Venezuelan fauna indicates that the deposits are of a marine, littoral type, laid down in warm waters of rather shallow depth. The abundance of Corals and Bryozoa together with the occurrence of certain types of Brachiopods lend themselves to this supposition. Such forms as Vitulina pustulosa, Tropidoleptus carinatus and Leptocoelia flabellites which are so common in the other Devonian deposits of South America, have not as yet been found, nor have other rather characteristic forms made their appearance, but it is to be hoped that further explora- tion and collection will afford sufficient material to aid us in correlating the Venezuelan deposits not only with those in North America but with the scattered deposits in South America, and with the Bokkeveld beds of South Africa. Dis BLATES 36 BULLETIN 46 256 PLATE 1 FIGURE PAGE 1. Cyathophyllum venezuelense n. sp. External view of coral- Time Slieshibliy; i enallain, Cea e ai toe teases eects san nas ca aee ee cere eee 4 2. Longitudinal section of same species, showing tabule and septa. 3. Transverse section. Note how the septa twist together at the center to form a pseudo-columella. 4, External view of a broad, conical specimen. About natural size. 5. Adult specimen showing variation in shape this species may assume. 6. Diphyphyllum vermetum n. sp. Single corallum showing septal markings) anid) |COMSTRICEIONS iscsi s-see cesses ee eee 5 7. Another specimen isolated from the cluster. , 8. Pleurodictyum (?) venezueclense n. sp. x2. Corallites. are best idisplayed! at, the sides ieee ee ea eee 6 9. Enlarged longitudinal section showing the disposition of the disseptiments, septa, and mural pores. The crystalline mass in the center may represent the former location of attachment. BuLL. AMER. oe Ne es ee 38 BULLETIN 46 258 PLATE 2 FIGURE PAGE | 1. Diphyphyllum vermetum n. sp. Rock specimen showing the typical occurrence of this species in ClUSteYS.................:cces01e 5 2. Fenestella venezuelensis n. sp. Natural size of zoarium. The specimen is|\an (internal /mold.c nei er ec sesakeessee eee 7 3. Enlarged view of same showing character of fenestrules and double row of cell fillings. 4, Polypora cachirita n. sp. Internal mold of portion of zoarium. 8 5. Enlarged view of same showing the several ranges of cell fillings between the fenestrules. 6. Fibrous, non-celluliferous face of frond of Polypora cachi- TEED Ts SD a2. ee ee UD TLE g oe 8 7. Dalmanella (?) venezuelensis n. sp. Dorsal (?) valve. x 2.5. 9 8. Stropheodonta zuliana n. sp. The rather distant filiform radiations have finer striz between them..................::00.-cc0sss-eeee00 4 ~ ne ee a pes ERATE) ccs 40 BULLETIN 46 . 260 PLATE 3 FIGURE PAGE 1. Leptostrophia caribbeana n. sp. Specimen showing the con- centric) undulations) ec lessee eae cane pe eRe aa 10 2. Schuchertella sp. aff. sullivani (Morris and Sharpe)................. 12 3. Chonetes zuliensis n. sp. x1.8. Ventral valve.............:cccse--es 13 4. Chonetes venezuelensis n. sp. Somewhat enlarged view of in- ternal mold showing portions of shell on the sides with punc- tate markings (ce RO 6) ee 14 5. Internal mold of a somewhat broader ventral valve than the preceding. 6. Internal mold of the ventral valve of a form very doubtfully identified as Chonetes venezuelensis. 7. Dorsal valve of Chonetes venezuelensis ? 8. Ventral valve, slightly distorted, of Chonetes venezuelensis ? 9. Chonetes subhemispherica n. sp. Highly convex ventral Valve.) (ox DBs ca ee a Te) 15 10. Brachial valve of Chonetes subhemispherica ? x1.8. Lis sConehidiunm) (2) spy ind ete) exc een eaten nan ne eee co pea teens 16 12.) Sparcfer Sp.) TNGEty ose hac sdsesces cnc deva cose ee eee eee Cee whee oie Shel PLATE 4 42 BULLETIN 46 268 PLATE 4 FIGURE PAGE LV Amphigenta (7) (Spe) ING eb ae ae Sane a eee ee 16 2. Atrypa cf. reticulams (Manne). x 1A eee 1 38. Ventral valve (?) of same species. About natural size. 4. Spirifer meridioamericana n. sp. Ventral valve about nat- UTaliverzenie yen coe em eon PEEP UC MMA RNER ELL yf) A 2) 18 5. Gutta percha cast of external surface. 6. Spirifer venezuelensis n. sp. Ventral valve. x1.8. Note the fine hair-like radial striz on the plications............cccccceesee 19 7. Spirifer audaculus zulianus n. var. Dorsal valve......cc.ccceeeeeee 20 8. Ventral valve of same species. x 1.4. BuLL. AMER. PAL. PRAT ES 44 BULLETIN 46 264 PLATE 5 FIGURE PAGE 1. Spirifer audaculus zulianus n. var. Ventral valve................... 20 2. High cardinal area of same species. 3. Ventral valve of the same species occurring in the micaceous sandstone. 4. Gutta percha cast of Spirifer audaculus zulianus with fewer ribs than the typical form. This also occurs in the sandstone phase. 5. Internal mold of ventral valve. 6. Cardinal area of a somewhat less elevated ventral valve. 7. Dorsal valve of probably the same species but similar to Spirifer pedroanus Hartt and Rathbun from Brazil................. Zi 8. Internal mold of dorsal valve of a scantily ribbed form of this species. 9. Vitulina (?) venezuelensis n. sp. Rather compressed ventral valve, with ribs not well defined because of shell covering. RAZiD a ews alcenacsventoatweceddryntsncc cause ivadavad tate eevencese ee eee ap tine wales 22 10. Ventricose ventral valve of same species showing the heavy concentric riblets. x 2.2. . 11. Less gibbous ventral valve. x 2.2. 12. Same species. x 2.2. 13. ) x 1-G..8 ... ee eee 23 2. Aviculopecten sp. indet. Right valve (7). x 1.3... 25 3. Aviculopecten yeakeli n. sp.. Left valve.................:0:cccccceseeeereees 24 4. Conocardium (7?) sp. indet. View showing beak and keel....... 26 5. Conocardium (?) sp. indet. Lateral view. x 1.3. 6. Cypricardinia subindenta n. sp. Left valve. x 2.2................. 26 7. Diaphorostoma neveritanum n. sp. Enlarged view of a dis- tOrtbed: ‘SPITe.) shocss.csessss bets Sas Ses ee 27 BULL. “AnmeR. PAT: Se/, Fea WE a 48 BULLETIN 46 268 PLATE 7 FIGURE : PAGE 1. Diaphorostoma neventanum n. sp. Front view showing fine reticulate. sculptures. x VeBe eee es eee ee aa Parl 2. Rear view of same specimen. Front view of a somewhat distorted specimen. 4. Pleurotomaria venezuelensis n. sp. View of spire whorls. Be De iossMeulencs vcasie wes venley canes Seals Gin WENU Deu aais acs | ULDAMAN UE UU Sata: eae 29 5. View of same species showing umbilicus. x 1.8. 6. Phacops argentinus ? Thomas. View of thoracic segments. Be Bi ae eee UU OAT LO OG CE 30 7. Pygidial view of same species. x 1.8. eg JILL. AMER. INDEX TO VOLUME XI Figures refer to consecutive plate and page numbering of the volume, not of the separate bulletins. Acrothele gamagei, 104 Acrothele matthewi, 44, 50, 51, 52, Sii5 56), Acrotreta gemmula, 52, 53 Acrotreta misera, 38, 40, 42, 48 Agnostus acadicus declivis, 37, 38, Bikes 44645, 146, 47, 48, 49; 50, 52 Agnostus barlowi definitus, 52, 53 Agnostus barrandi, 43, 44, 45, 46, 47, 48, 49 Agnostus bibullatus, 38, 39 Agnostus clare, n. sp., 3, 74 peters exaratus tenuis, 38, 40, Agnostus fallax, 36, 37, 39, 40, 41, AP V4. 46, 48.49 Agnostus fallax trilobatus, 51, 52, 55 Agnostus fissus, 44, 45, 46, 47, 48 Agnostus fissus perrugatus, 40, 42, 43 Agnostus gibbus, 42 Agnostus gracilis, 38 Agnostus gibbus hybrida, 48 Agnostus granulatus, 39, 40, 41, 42, 44, 48, 49, 52, 53 Agnostus incertus, 36 Agnostus kjerulfi, 38, 39, 40 Agnostus levigatus, 36 Agnostus levigatus ciceroides, 36, Agnostus levigatus mamilla, 37 Agnostus levigatus terranovica, 36, 37 Agnostus levigatus terranovicus, 38, 39, 40, 41 Agnostus longifrons parvulus, 38 Agnostus longifrons parvulus, n. var., 3, 83 Agnostus parvifrons mammillatus, 38, 39, 40 Agnostus, parvifrons punctifer, n. Bold face figures refer to plates. Var i35 15 Agnostus parvifrons tesselata, 48, 53 Agmnostus nudus, 37, 40, 41, 45, 48, 49 Agnostus punctuosus, 36 Agnostus punctuosus, 38, 39, 40, 41 Agnostus pusillus, 38 Agnostus rectangularis, n. sp., 3, 85 Agnostus rex, 37, 38, 42, 46, 47, 48, AG Ole O24) Obs) OasnOD Agnotus sulcatus, 37, 38, 39, 40. 41 Agnostus umbo, 41, 42 Agnostus vaningeni, 39, 40 Agnostus vaningeni, n. sp., 3, 80 Agyraulus affinis, 55 Agyraulus quadrangularis, 104 Agraulus socialis, 42, 43, 44, 45, 46, 47, 48, 49, 50 Agua Clara series, 173 Athyris aff. spiriferoides. 40, 243 Avalon Peninsula, classification of rocks, 12 Atrypa reticularis, 38, 237 Aviculopecten sp., 40, 245 Aviculopecten yeakeli, 40, 244 Amphigenia, 38, 236 Bailiella baileyi, 50, 51, 52, 538 Bailiella venulosa, 37, 38, 42 Belemnopsis, 163 Bonarellidz, 150 Brown, Jukes, work of, 11 Buddington’s work, 23 | Catacephalites, 4, 154 Centropleura sp., 36, 37 Centropleura henrici, 36 Centropleura pugnax, 40 Centropleura venusta, 43 Cerro Pelado series, 173 Chamberlin’s brook form., 60 Chanasia, 151 On ° Chonetes subhemispherica, 37, 235 Chonetes venezuelensis, 37, 234 Chonetes zuliensis, 37, 233 Codera formation, 173 Conechidium, 37, 236 Conocardium, 40, 246 Conocoryphe equalis, 42, 43, 44, 45, 49 Conocoryphe bullata, n. sp., 3, 54, 87 Conocoryphe elegans, 50, 51, 52, 53 Corynexochus minor, 36, 38 Cryptozoon shale, 26 Cyathophyllum venezuelensis, 224 Cypricardinia subindenta, 40, 246 Dalmanella venezuelensis, 36, 229 Damsite series, 173 Devonian of Venezuela, by Weis- bord, 221 Diaphorostoma . Al, 247 Diphyphyllum vermetum, 35, 36, 225 35, neveritanum, 40, Distribution of Paradoxides, Table 4, opp. p. 56 EK! Paraiso shales, 173 Hodiscus punctatus, 37 Hodiscus punctatus, 37, 38, 39, 40, 43, 44, 45, 47, 49, 50 EKoorthis sp., 58, 54 Fenestella venezuelensis, 36, 227 Frech quoted, 30 Goniodiseus dawsoni, 54 Grossouvria ef. anomala, 156 Grossouvria cf. waageni, 157 Guarabal conglomerate, 174 Haploceras elimatum, 4, 153 Hartshillia inflata, 37 Harttella matthewi, 50, 51, 52, 53 Hectioceras, 149 Hemisimoceras, n. gen., 159 Hemisimoceras nodulosum, 4, 162 Hemisimoceras semistriatum, n. sp., 4, 161 Hodson, F., on Turritellas, 171 Holocephalina primordialis, 37 Howell, B. F., on Cambrian of New Foundland, 1 Howell’s work, 19 Hyolithes haywardensis, 104 Hyolithes shaleri, 104 BULLETIN 46 270 Hyolithes tenuistriatus, 35, 37 Tphidella ornatella, 54 Kelligrew brook form., 59 Kheraiceras stansfieldi, 4, 155 La Puerta series, 174 La Vela series, 174 Lingulella ferruginea, 40, 49, 50, Sil, 54, Sei, S44, 55 Liostracus globiceps jaculator, n. Wait, 6 42, 413), Bil Liostracus ouangondianus, 51, 52, BB, G4! Liostracus tener, 49, 51, 52 Long Pond form., 59 Lunuloceras cf. lunuloides, 149 Macrocephalites rensis, 4, 153 Madagascar, Cephalopods of, 148 Manuels Brook, view of, plates opp. p. 32 Manuels, geology about, 19 Manuels, sections at, 16 Marcou, Jules, work of, 15 Matthew, G. F., work of, 138 Micromitra ornatella, 54 Misoa Trujillo series, 174 Murray, Alexander, work of, 11 aff. madagasca- Nautilus calloviensis, 163 Nautillus cf. kumagunensis, 163 Obolus fragilis, 37, 38 Outecrop of Cambrian at Manuels, map. p. 24 Paradoxides abenacus, 47 Paradoxides beds, detailed section, Paradoxides bennetti, 51, 52, 53, 54, 55, 56, 104 Paradoxides bennetti fauna, see Pl. 1 and 2, opp. p. 32 Paradoxides bennetti zone, 60 Paradoxides davidis, 35, 36, 39, 40, 41 Paradoxides davidis, see pls. opp. p. 32 Paradoxides davidis zone, 59 Paradoxides, distribution of, table opp. p. 56 | Paradoxides eteminicus, 48, 50, 51, | 52, 538, 54 | Paradoxides faunas by Howell, 1 | Paradoxides harlani fauna, 104 Pag fl Paradoxides haywardi, 104 Paradoxides hicksi, 42, 48, 44, 45, 46, 47, 48, 49 Paradoxides hicksi fauna, see pls. opp. p. 32 Paradoxides hicksi zone, 59 Paradoxides lamellatus, 53 ee provoculus, n. sp., 3, 8 ‘ Paradoxides parvoculus, 52, 53 Paradoxides rugulosus, 36 Patiecitos series, 174 Pauji shales, 174 Phacops argentinus, 41, 250 Phylloceras aff. disputabile, 146 Pleurocephalites, 154 Pleurodictyum venezuelense, 35, 226 Pleurotomaria venezuelensis, Polypora cachirita, 36, 228 Protospongia fenestrata, 36 Ptychoparia rogersi, 55, 104 Querales shales, 174 San Luis series, 174 Schuchert, quoted, 18 Schuchertella aff. sullivani, 37, 232 Sindeites madagascarensis, n. gen. | and sp., 152 Socorro series, 174 Solenopleura applanata, 37, 38, 39, AO eee aA: Ay, Solenopleura aff. applanatus, 43, 44, 46, 48 Solenopleura communis, 36 Spath, L. F., on Madagascar Cephalopods, 148 Spirifer audaculus zulianus, 38, 39, 240 Spirifer 238 42, meridioamericanus, 38, Spirifer venezuelensis, 38, 239 Stenotheca cornucopia, 37 Stropheodonta caribbeana, 37, 230 Stropheodonta zuliana, 36, 231 Torby slates, 20 Torquatisphinctes cf. bangei, 4, 158 Turritella altilira mirandana, 27, Be Pals) Turritella ajltilira urumacoensis, | SA paile Pale! Turritella andreasi, 28, 29, 207 Turritella berjadinensis, 22, 23, 196 41, | DEVONIAN OF VENEZUELA 5! | Turritella berjadinensis cocoditana, 23, 24, 199 Turritella berjadinensis colinensis, Por, Pease, ANSM | Turritella berjadinensis socorroen- sis, 25, 27, 198 Turritella berjadinensis 2A, 25, LOT Turritella bifastigata ciana, 32, 34, 220 | Turritella bifastigata maracaiben- sis, 34, 218 Turritella boweni, 28, 29, 206 wartieldi, democra- Turritella buchivacoana, 27, 29, 209 Turritella buchivacoana canonensis, 29, 210 Turritella carlotte, 30, 31, 213 Turritella cauredalitoensis, 10, 13, 14, 15, 185 | Turritella caredalitoensis daba- jourensis, 15, 187 : Turritella cauredalitoensis filensis, Zs, ais) Turritella cauredalitoensis liddlei, 13, 15, 186 Turritella cornellana, 26, 28, 205 Turritella cornellana bolivarensis. 28, 206 : Turritella curamichatensis, 20, 21, 192 Turritella elmenensis, 29, 211 2 Turritella filacarmenensis, 27, 29, | 208 | Turritella gatunensis lavelana, 22, ray, 8B} Turritella gatunensis tartarana, yy Saas} Turritella gatunensis willistoni, 22, 195 Turritella G-A-Weaveri, 26, 27, 28, | 204 | Turritella gilbertharrisi, 16, 17, 18. 187 Turritella gilbertharrisi aguaviven- sis, 19, 189 Turritella gilbertharrisi falconen- sis, 18) 19; 32, 139 Turritella gilbertharrisi 19, 189 _Turritella holotypes, 176 | Turritella hubbardi, 11, 12, 13, 184 Turritella hubbardi weeksi, 12, 14, staufferi, | 52 185 Turritella larensis, 7, 8 9, 180 Turritella larensis carrizalensis, 10, 182 Turritella jarensis guaratarensis, 14, 182 Turritella larensis santiagana, 8, 9, 182 Turritella matarucana, 24, 25, 201 Turritella mauryz, 27, 200 Turritella montanitensis, 18, 19, 20, 21, 190 Turritella montanitensis 20, 24, 191 Turritella montanitensis saladillo- ensis, 20, 192 Turritella planigyrata, 23, 199 Turritella plebeia A-L-Owensi, 24, 27, 32, 201 . Turritella robusta fredeai, 9, 10, 11, lid 82. LSS olcotti, HND BULLETIN 46 272 | Turritella variegata paraguanensis. 15, 201 Turritella venezuelana, 25, 26, 202 Turritella venezuelana quirosensis, 26, 28, 204 Turritella venezuelana watkinsi, 26, 204 Turritella vistana, 27, 31, 32, 216 Turritella vistana nicholsi, 27, 33. 218 ‘ Turritella zuliana, 5, 6, 178 Turritella zuliana palmeri, 6, 179 Van Ingen’s Conception Bay sec- tion, quoted, 20 Venezuelan and Caribbeo. tellas, 171 Vitulina venezuelensis, 39, 242 Walcott, C. D., work of, 12, 17, 18, 32 Weisbord on Venezuelan Devonian, 221 Williston, cited, 175 Turri- OF VOL XI e aa ADR RAA AAA a AAA RARa\aaaAamarry) vy usemmeeBiala|.!.) ’ YY Yum ARAAAARAIY 1) VNBABPPYY TTY YY YY) SAAR aalalaall \ By . a oe ‘aahaaa |_|: | | ol ala » | ae 2 la ‘ DAD AQ a! A AannAAn ae alalAAAAAAAARAS AA ARAAAAAAARAW A AAAAAAL ; y AMAAAAAMAMAWAAMAAM~A RASA PA\ AAAA AAAAAS BAARAAAANS R WANVANIA = | AA AaAM SAMAR AANA aA) \Valala alal pAAAAAA AAP RAAT aAARAAAAR NAAIA ARAAAAAAAAAAA Aan fF LAA AAA TY AAAAAARAAY YY AAA Ag a ~aARDAAa.. anarnnannAaRe oc'> al MANAANA Aa AAAAA AAA AAIAIAAIA AAAAE MAAAAAAARAAAAAAAAR DAAAAAAAAH A NANAAAIA AAA 3 ee AAG AWARAAAAAAY JAAR iA | \ AAAAAAAA VE ‘an 5g Val AAiar\ |AVEN Va ARAAAY o_o ARaA AAROn AARAAAAAAY A aia AW aaAAn iAARAAAIAAAAAAA NA AA ana WARAA A \AAA ARAAP Be AAAAAAAAA, NA = I ARAN AP yi = “NAAAAR A ! | PY y | WY | | ey AAR AA WAAR) AAAAA, AAA | AAA eaaaaaeaaeaaaal | mene wy AAA AAA Ar AAA EAA | RERREAGREAE BARRA ARS pa A aon pn »>D DD» > »») D> > D D>» se »»> ») > » > > D> >» ») > D >» > D»» > Dp» » » »»» p PD» DD» >» >» > D>» DD» D> >» >y» 3D yy \ > yy» Dy» by» ) YY» > »» y » 7) yp) ? 2) > J — > + »» y »» y : ¥ y » >) ) yD) ) > >> y 2» i > 5 » >, » > FJ J Dp) » BE > )) > 5 7 2. yy, Z ) >»). ». yy») PB p») DD) » 2 yy). yp J y) D2 ». 2 )) )) D ») » >>> SD» >> 2d D»Y > D> _2»Y >») > »») >») » > > DPP22 » > »» »2 >. »») 5) > D. >> > yy) ») > yp _» ») yy) 2 3233 ) >» » x 32 oy om > »») 33 » ») » » » > >> > P)) Dy yy » » ») 5) Dp» ? 23> D>» I» BP Dy I y». >») DPD DMY_S » D DIY)? »»)»> > Dy >>) DDD)? I> Wp 5 ). DI») dD FZ »»> Dy D> YY Dy > yD >»> WD» WD» »D D> Dy») 2» p> > DD >» >») YP» D> Dw = »_ »> DDD» > y» yy» ~ > DDD 2 BD Yd» YD DDI WP yy >» 2. »y DD») > YP yy») DDD» »D Dd») BD > >: > yy)?» >» PDWd)) D»»o> > PD>))) >>»). > >» Wd?d) >») > BD») yp) >) ED Ip)» » ) >» DD) > 2» ») > » ) DDD DPD Py»>y ») DDD D DY» ) ». ; >») » yy. > »») » D> Pp » D> » » » ») yy) ) 3 3 D> » ) »)) BP >> > > » J yp »») ~ Sy J 7 Kr) af \ ed Weel L Aj vy) DD >» D>) >>. >> >> > > >>> >> >>. >». DD. DP. WAS dd