battyyy (hha ad 1 ER tay oH { Pit He ets | f Lebh hae ‘yhntis t eer a a pins) aN qilinie bea mith well ih sly ite rah en Tht “et Hii 4) Wane Mp enyt } eH iid bi Wi y ra iia ih tinetie ‘ Ni stet a rt iAbey { alr it! il ate site fh ah oh te ps vial ee ag These Bits ee) ibe tata th nil ie caine felis HPL LUE Hea a AVL Se) , Uiiat au at Hie 8 aia ae ae ty Heal staat sebet| f att oe a ee 4 Shieh 4) ‘ BV hGY Pe 4) i His diy ui ae eg = AIHRGSHESERRE ESHER CS oe a ” a ia ite pie a sens 16) ait nt {hit ad ;! Nt yt a 7 anit ate ay i Alt } yi Hint it i Vota ae io Wi eli } ieee ee Se ee Ay thsaahe Pe ~~ int ODA AAve BL ae POL ER ae ari < iby i RNS An ih Vad ary eet wie ith \ ilar i ‘ slant i; 4 ‘ Aeitiint sty! hie \ ity i Gy SS Sd ats ea poole ret as SS ees Se Se es Tiara iat Es Bee ez = So = 3 Soe es Seana eS stetetstt= == =5> ae CURT LEAT Sahar fi mitt 48. 9a Ca ot bt chest ete oe \ Ti iat ithe ai guid seers i ieee PeSWUasy Se tt 1 : ui 8 it MH Asal Mei ae a Hl wut eta at Se aah / Wee ahauit SiGe lee wth i) bey aa eee Gav aay } i i Ws ute Pate wit } rl List i H 4 oa; sp isy i nf 1) + (ay vit UyiE} Aras HI ih, i Baal ity cay a sical AWh nt helt \e i} Wn uy My yheeh ee ‘ uh he Hie iy f Wu i) editid ta ae! el ie it i Wht sien f ie rn) Hii RRRRARA AA AR AA RAAAAAA RAAAGAAA AAANAAAR AAA NANA A ma p WA AA AYNAAA Ai mm PA AAA AAAIALY \ AAAA AR AA AAA ANNAN ARAAAN f— ~ \ | AAA AA a A { | AAA on\ AA AAA A | ox AAR AAA : AAAAAA, ) A AAAAR aaa e AY WF ahaa AAAA ASA AAAA | ~AAAAAI- lAIN A = WAAR AA AAAAAL WA PiNAAAW ANA ) \A WA GAA laliM@ialaliaia’ _) AA, AAAAAaa Anan AAR aA AAAAAAA \ ~ <£ ECC @_ | CES Ki LOL ECE (ac ai ¢ Cc qk C (ae @aC KC CECE C CF aca < KX (€@ OC OE COC CE C 4 CCC OO CCE ( co ergeaie saucent a : C C aX « CEM" EG Ca cae C é fT ame Ca @CCOCO € & SE EE Ee ( Gia a ‘a é i aCe ac’ CT SO OC ac (a & @ gq aC “am 46 Q@Gac (COE @ e€<« @| Cad C@ qe EF ¢ ee COE Ce aaa «Ef «@ OS Greer ace uee { ( & ( eke CE, .< < ecg @ CCC COCO ECC Car @G ST OCG CEOS «Ce COL a a (@ € Ca ; “ CG cs is ces eS K SNA | ! CCE CE © CE € Ste Sista cts a ccacte aac “CCC@q es (aa aaa ar 5 Oe SEKGELG CC are CC@€ CCl €€ COC i “COE COE CC ae OE eee mae x= ( “ a CC@ es CO Ce © CC « COL Cg ar CCG KE Cae Wve «€ (OE (COL KC @CCCe* OE be 2 pes ETINS C)} AMERICAN Poke ONTO LOG Y YO RN GAe INS OG aercsie A. SA i Ji %390 CONTENTS OF VOLUME XXIV Bulletin No. Plates Pages 80. Nomenclatorial notes on Eocene Mollusca By Katherine VanWinkle Palmer — —— 1-7 81. Devonian crinoids from the Mackenzie Rice Basin: N. W. T., Canada By Winifred Goldring — a 1-2 8-34 82, The correlation of certain Devonian fava of eastern and western Gaspé By E. M. Kindle _. eg eS ie ee Fes BEA! 35-86 82A. Devonian Bryozoa of Gaspé Biya Miaidieliteniese Ar. = Bure G7 ees es eee eee 5-6 87-100 83. A Devonian fauna from Colombia By Kenneth E. Caster; Ineluding Stratigraphic Notes By Axel A. Olsson —— _ 7-20 101-318 84, Notes on Cyprea heilprini Dall id (Cone ‘chilona Dall with new species from the Pliocene of Costa Rica By William Marcus Ingram —_ Z : = Al B19=326 85. New fossil Cypreide from the Miocene of fie Damn inican Republic and Panama, with a survey of the Miocene species of the Dominican Republic By William Marcus Ingram -_- : - 22 327-340 86. Reprint of Conrad’s Jackson Eocene ie. as aie scribed and illustrated in the Philadelphia Academy of Natural Sciences, Proceedings for 1855, pp. 257-63 and Wailes’ Report on the Agriculture and Geology of Mississippi, 1854, pls. XIV-XVII _. 23-26 341-359 87. A group of Pennsylvanian crinoids from the vicinity of Bartlesville, Oklahoma By Harrell L. Strimple en : : 27-29 359-386 INDEX sage eee fe aden Finn hs f aN EUNTAN Paces: ATHSOMAN tiis7, oo wi nN \ MUSE MUS oo ape ny ne pe Ithaca, New York, eS AG 4g ryt BULLETINS OF AMERICAN PALEONTOLOGY Vol. 24 No. 80 Nomenclatorial Notes On Eocene Mollusca By Katherine VanW. Palmer July 31, 1938 Ithaca, New York, Wh eS; Ae = 2 . \ . x NOMENCLATORIAL NOTES ON EOCENE MOLLUSCA By Katherine Van Winkle Palmer At the time of the publication of the monograph on the Clai- bornian fauna!’ the reference showing Ce@latura Conrad, 1865? was preoccupied, had not been found. Since that time the author has located the citation which Conrad probably had in mind when he changed the name of the genus to «Ict@onema.” The confusion in connection with these two names has been discussed in detail by the writer in the work on the fauna of the Claiborne and will not be repeated here. Conrad himself used Celatura previously in 1853* for a Naiad. Acteonema is there- fore a substitute name for Celatura Conrad, 1865 and the prob- lem of its genotype falls in that category. The one interpretation which was suggested on p. 156 of the Claiborne work may be eliminated. Aldrichia Palmer? is preoccupied by Aldrichia Coquillett® in insects and Vaughan‘ in corals. Aldrichia Palmer is herein re- named Timothia. ; Attention is called to the reference by J. W. Taylor* on the dates of the publication of the various parts of Moquin-Tandon’s “Histoire Naturelle des Mollusques Terrestres et Fluviatiles de 1Palmer, K. V. W., Bull. Amer. Pal., vol..7, No. 32, p. 1154, 1937. 2Conrad, T. A., Amer. Jour. Conch., vol. I, pp. 28, 35, 1865. 3Conrad, T. A., ibid, p. 147. 4Conrad, T. A., Acad. Nat. Sci., Phila., Proe. vol. 6, p. 268, 1853. 5Palmer, K. V. W., Bull., Amer. Pal. vol. 7, No. 32, p. 262, 1937. 6Coquillett, D. W., Trans. Amer. Entomol. Soc., vol. XXI, p. 98, 1894. TVaughan, T. W., U. 8. Geol. Sur., Mon. vol. XX XIX, p. 70, 1900; Proce. Biol. Soc. Wash., vol. XVI, p. 101, 1903. Aldrichia Vaughan renamed. 8Taylor, J. W., Proc. Mal. Soe. London, vol. 6, p. 186, 1904. 4 BULLETIN 80 4 France.” The six parts were issued April 12, May 4, August 1, September 10, 1855; January 2, April 9, 1856 respectively. These dates definitely establish the priority of Papillina Conrad, Janu- ary, 1855° over Papillina Moquin-Tandon, 1855. F. Stearns MacNeil, United States National Museum pointed out to the writer that Herrmannsen?® designated a type for Cla- vilithes Swainson, 1840 previous to other designations. - Herr- mannsen’s statement of type was overlooked in the discussion of of the Claibornian Clavilithes by the author’? and the work of Grabau2 was followed. The problem of Clavilithes Swainson. 18402 begins with Clavella Swainson, 1835‘ for which it was a substitute name. However, it does not seem that with the excep- tion of one genus (p. 7,“Trochilea, type Trochus pileus. Auct.”) Swainson’s “Elements of Modern Conchology’”’ can be used for generic names without difficulty. He did not mention any specific names or references (exception p. 14, Mitreola, reference given). The generic names are descriptions without species and would therefore require special ruling as under Int. Rules Zool. Nomenclature Opinion No. 46. In case of Clavella, the descrip- tion reads, p. 20, “The genus Fusus, again, has no plaits; it is so closely allied to the fossil genus Clavella (here now first defined),, that there can be no doubt of its entering within the limits of this group’; p. 21, “Clavella Sw. Fuciform [fusiform] ; channel long; no plaits, but the tip of the spire enlarged. Fossil only.” It appears that even though Clavilithes Swainson, 1840 was a substitute name for Clavella Swainson, 1835, since Clavella has no species, Clavilithes must take its type from the species mentioned under its own description and the type so established becomes the type of both genera (Int. Rules Zool. Nomen., Art. 30, f.). ®Conrad, T. A., Acad. Nat. Sei., Phila. Proce. vol. 7, p- 262, 1855; Palmer, K. V. W., Bull. Amer. Pal. vol. 7, No. 32, p. 363; 1937. ; ee ee A. N., Indicis Generum Malacozoorum, vol. I, p. 246, 46. 11Palmer, K. V. W., ibid, p. 356. 12Grabau, A. W., Smith. Mise. Coll., vol. XLIV, p. 104, 1904. 13Swainson, Wn., A Treatise on Malacology ete., p. 304, 1840. 1<8wainson, Wm., Elements of Modern Concholog’, pp. 20, 21, 1835. on 5 NoMENCLATORIAL Notes: K. V. W. PALMER Herrmannsen designated Fusus noe (Chem.). This is a valid designation as /. no@ was listed in Swainson’s original descrip- tion of Clavilithes, 1840. The characters of F. noe disagree with the original description of Clavella because F. noe (Chem.) Lam. has plications on the columella during part of its life history, such disappearing with maturity. Mr. A. Wrigley, England’? who was consulted in this matter has expressed the opinion that such a noncomformity eliminates the use of F. noe (Chem.) Lam. as genotype of Clavilithes. This is the reasoning which Grabau’® also used. ' The writer does not favor this interpretation because of the vagueness of the type of Clavella Swainson, 1835. Using F. noe Lam. as the genotype of Clavilithes Swainson causes a change in the current idea of the genus as well as a conflict with the subsequently named genus Rho palithes Grabau,™ type F. noe Lam, Rhopalithes Grabau becomes synonymous with Clavilithes Swainson, 1840 and the forms of Clavilithes typified by C. par- isiensis (Mayer-Eymar)=C,. longevus (Desh.) non Solander are without a generic or subgeneric name. It is in this last nonplicate group that the Claibornian species belong. However, there is some doubt that Clavilithes needs to be separated on the character of the columellar plications, particularly when those specimens which do have plications in the young stages of growth lose them in the adult. The presence or absence of columellar plications cannot always be taken as a factor of generic differentiation in the gastropods. A typical example of a genus including plicate and nonplicate shells is the Claibornisn Eocene genus Mazzalina Conrad, 1860 =(Bulbifusus Conrad, 1865).‘* Conrad made two genera on characters which are now known to be only specific and in some cases may not even be specific. 15Personal letter, May 7, 1938. 16Grabau, A. W., ibid, p. 104. i7Grabau, A. W., ibid, p. 135. 18Harris, G. D., Ark. Geol. Sur., Ann. Rept. State Geol., vol. II, p. 165, 1892; Palmer, K. V.-W., ibid, p. 349. 6 BULLETIN 80 6 The problem of Clavilithes is already too complicated to be burdened with more names without further complete and thorough work. Until then, the author prefers to. use Clavilithes Swainson, 1840, (=Rhopalithes Grabau, 1904) genotype by sub- sequent designation, Herrmannsen, 1846, Fusus no@ (Chem. ) Lam. to include the nonplicate forms of C. parisiensis (Mayer- Eymar). Wrigley’s?® criticism of Grabau’s study of the phylogeny of Clavilithes shows that additional investigation must be made on the group before a satisfactory conclusion is reached. Since needless time is wasted in searching for Eulimella Forbes, 1846, as given by authors” it seems worthwhile to reiterate Iie- dale’s? affirmation that no such reference exists. One will find that there is an article by Forbes in the Ann. Mag. Nat. Hist., vol. XIV, p. 412 but the genus Eulimella is no where mentioned in the article. The date of the reference is 1844 instead of 1846. Malaconchologists as H. and A. Adams, Fischer, Bucquoy, Daut- zenberg and Dollfuss, Tryon, Sacco, Cossmann and many other standard authors continued the error in their work. According to Iredale, Jeffreys’? was the first to mention Eulimella in liter- ature. Jeffreys was followed by Gray** in the same year. Gray gave the name and selected a type. Iredale suggested that the name Eulimella as ascribed to Forbes was a manuscript name. Such a supposition appears reasonable for particularly Jeffreys was assisting Forbes** in the work on the British Mollusca. Ire- dale prefers to give Gray credit for publishing the genus. Thiele?° in his Handbuch assigns the genus to Gray. However, it seems to the author that Jeffreys’ reference of Euwlimella is legitimate but with no designation of type. He states “Evulimella (Forbes) t9Wrigley, A. G., Proc. Mal. Soe. London, vol. XVII, pp. 222, 234-237, 1927. 20Dall, W. H. and Bartsch, P., Bull. U. S. Nat. Mus., 2 AG: 68) pp On ie 1909s see) iiredale, Mr The Nautilus, vol. XXIV, No. 5, p. 53; 1910 for previous references. 21Tredale, I., ibid, p. 53. 22 Jeffreys, J. G., Ann. Mag. Nat. Hist., vol. XIX, p. 311, May, 1847. ie J. H., Proe. Zool. Soc. London, pt. XV, p. 160, Nov. 9 (read), 847. 24Jeffreys, J. G., ibid, p. 309 NOMENCLATORIAL Notes: K. V. W. PALMER 7 crassula, Mal. and Conch. J. E. MacAndrei, Forbes in Ann. Nat. EMS viele iy. pitt?’ «2% Eulimella gracilis... . .” (Followed by a description of this last species. ) Authors*® believe that the /. crassula Jeffreys equals E. Mac- Andrei Forbes and is the same as E. scille (Scacchi). E. scille (Scacchi) is the species which Gray used as the type of his Eulimella and it is the species which is commonly used when the genus is assigned to Forbes. JT ortunately therefore the same species can be cesignated as the type of Eulimella (Forbes) Jeffreys and the established characters of the genus need not be disturbed. To bring this about, in case the authority of the genus is granted Jeffreys, the genotype of Eulimella (Forbes) Jeffreys, 1847 is herein designated as E. crassula (Jeffreys)=E. Mac- Andrei (Forbes)=E. scille (Scacchi). Recent. Great Britian and Scandinavia. Pliocene and Pleistocene of Italy and Sicily. (oma) 25Thiele, J.. Handbuch der Systematischen Weichtierkunde, pt. 1, p. 236, 1929. 26Morbes, H. and Hanley, S., Hist. British Moll., vol. III, p. 309. 1851; Dall) W. Hand Bartsch, BP.) U.S; Nat. Mus., Bull; Nos 68, p. 10) 1909. PALEONTOLOGICAL RESEARCH INSTITUTION May 25, 1938. my i : Al a a a My iy WA a ro i ee a ‘w nee 1 eye SL) tg i 2 a he pie, ; BE ee ies Labi vy 2 mY, a f LALLA Vie * * = %7 cote : | hy es i d rd My A hie. ieee in i, ie a A Voy Ca ee pity ve i ee + re AU a . ce WY . : 7 7" a chee th Z ' Te i : rd ft 1 ‘a i 1 1 ov ay i} i ee BULLETINS OF AMERICAN PALKFONTOLOGY VOD NOE, 81 Ithaca, New York, CT SE AS BULLETINS OF AMERICAN PALEONTOLOGY Vol. 24 No. 8c Devonian Crinoids from the Mackenzie River Basin, N. W. T., Canada By Winifred Goldring Ithaca, New York, We 8. As DEVONIAN CRINOIDS FROM THE MACKENZIE RIVER BASIN. We - CANADA By Winifred Goldring New York State Museum, Albany, N. Y. Recently a collection of crinoids from the Great Slave Lake region, Mackenzie River basin, Northwest Territories, Canada, was submitted to the writer for study by Doctor E. M. Kindle, Victoria Memorial Museum, Ottawa. Of this collection he writes, “The horizon is probably not far from that represented by erinoids described and figured by Springer in two of our reports”. In 1921 Springer described two new species Melocrinus borealis and MM. canadensis collected by E. J. Whittaker from the Hay River section, the former below the Alexandra falls and the lat- ter above the falls (ref. cit., p. 17). The Hay River section from which the crinoids were obtained was studied by Kindle who re- ferred (1919, p. 4) the beds to the Upper Devonian, having found a characteristic Portage fauna in the Simpson shale below the strata from which the crinoids came. Melocrinus borealis is represented in the present collection from locality 7005, bed h, Lady Evelyn Falls section of the Kakisa River; a few plates from locality 7300, the gorge section of the Redknife River are doubt- fully referred to M. canadensis. Springer relates M. borealis to M. tersus, a Missouri form described by Rowley (1893, p. 303; 1894, pp. 151, 153) from shales considered of Middle Devonian (Hamilton) age by early geologists and by later authorities of younger age (see Keyes 1894, p. 43; 1902, p. 271-273; Greger 1909, p. 374; Schuchert 1903a, p. 143, 1903b, p. 545; Weller, 1909, p. 264; Branson 1923, pp. 44-46). Springer concludes, “‘it is clear that the fossils of the Missouri and Mackenzie Devonian belong to the same palaeontological province, and are of approx- imately the same age” (ref. cit. p. 15). 4 BULLETIN 81 12 In his later paper (1926) Springer adds three new species of Melocrinus: M. kindlei and M. mackenzie from the coral reef in limestone above the horizon of the Simpson shale, Root River section; and M. whittakeri from the beds at least 300 feet above the Simpson shale in the Trout River section. No specimens of M. kindlei appear in the collection, and only a single. specimen that can be referred to 1/7. mackenzie was found and this in the crinoid bed at the upper falls of the Redknife River. At least three specimens from the Redknife River section, in the crinoid bed at the upper falls, have been referred to M. whittakeri. Three new species, MW. subtilistriatus, M. sulcosutwra and M. humet, are here added from the Redknife River crinoid bed. Springer also describes (p. 132) one species of Hexacrinus (H. humei) ; but Melocrinus is the only camerate genus represent- ed to any extent in the three collections and it is so far represented by eight species. Except for /. borealis which shows close rela- tionship to species of Iowa, Missouri and Wisconsin, as pointed out by Springer (1926, p. 127), the species of Melocrinus are “not. only thcroughly distinct from that, but also from each other. And the interesting thing about them from a geological point of view 1s that in the characters by which they differ so completely from all other known American species, the three .. species [M/. kindlet, M. mackenzie and M. whittakeri] exhibit a tendency to an asym- metrical construction of the calyx which is not observed among the abundant species of the Eifel limestone of the Middle Devon- ian, but which developed in certain species belonging to the Fras- nian (or lower) member of the Upper Devonian in Belgium [M. konincki, M. hieroglyphicus (non Goldf.) Fraipont=M. dewal- quei von Koenen, M. benedeni and M. mespiliformis|” (see also p. 129; Fraipont, 1883; Von Koenen, 1886). WM. borealis has a similar, though less marked tendency to asymmetry of the calvx, and in the three new species described here it as well-marked as in the others. The single specimen of Hexacrinus in the collec- tion studied by Springer (1926) represents “another very prev- alent Middle Devonian genus in the Eifel, but rare in America, or 13 MACKENZIE CRINOIDS: GOLDRING .. of a type completely different from that of the Eifel, but which is also represented in the Upper Devonian rocks of Belgium” (p. 127). In the collection under consideration inadunate species of crin- oids are represented from Lake Kakisa, one half mile back from the south shore, west end; the gorge section of the Bouvier River and eight miles above the mouth; above the upper falls of the Redknife River, in the gorge section and at the third chute. A single specimen representing the Flexibilia was collected from the crinoid bed near the base of the coral zone, Jean Marie River, and a starfish was found loose in the Trout River section, at the foot of the long heavy rapid, one half mile below the lower cas- cade. Only portions of the vertical side of two arms of the star- fish are preserved, and not in very good condition, so that this species has not been placed. The collections studied by Springer were made by E. J. Whit- taker in the Hay River and Trout River sections; by G. S. Hume for the Root River section. In his later paper (1926, p. 128) Springer has incorporated notes by Mr. Whittaker and Mr. Hume, relative to the stratigraphy of the crinoid-bearing beds. The present collection, in so far as labelled, was collected by E. J. Whittaker. The species of crinoids described in all collections so far submitted for study are CAMERATA Melocrinus borealis Springer M. canadensis Springer M. kindlei Springer M. mackenzie Springer M. whittakeri Springer M. sulcosutura Goldring M. subtilistriatus Goldring M. humei Goldring Hexacrinus humei Springer 6 BULLETIN 81 tf) 14 MLIa NBIC VAN Synaptocrinus (?) rotundatus Goldring INADUNATA Undetermined sp. Decadocrinus spinobrachiatus Goldring Prininocrinus robustus Goldring Linocrinus kindlei Goldring DESECRIMTIONTORYS bles CAMERATA Melocrinus canadensis Springer Melocrinus canadensis Springer, Geol. Surv., Canada, Bul. 33, p. 17, pl. ih, tite, By, ETE Melocrinus canadensis was based by Springer upon a single ‘specimen lacking basal plates and only partly free from the ma- trix. In this collection, from the Trout River section at the third falls (loc. 6978), are a few radial and interradial plates, partly separated, which might be referred to this species and then only with doubt. The only description given with the figure is that this species is of a larger and more robust type with very low plates, in the flat- ness of which “this form is comparable with one from the Hamil- ton of western New York figured by Hall, but never described, under the name M. brewradiatus’ (ref, cit., p. 17). Whe form figured by Hall in 1872 (pl. 1, figs. 18, 19) has since been de- scribed and refigured by the writer (1923, p. 127-130, pl. 13, figs. I 2). There is even more resemblance to Melocrinus clarkei (Hall Ms) Williams from Genesee and Portage beds of the Upper De- vonian of western New York (see ref. cit., p. 132-136, pl. 13, figs. 3-5; pl. 14). The figured specimen of 1/7. canadensis shows a short anal tube, not characteristic of the other species under discussion nor found in any of the specimens in this collection. —— ee eee 15 MACKENZIE CRINOIDS: GOLDRING Melocrinus whittakeri Springer Melocrinus whittakeri Springer, Geol. Surv., Canada, Bul. 42, pp. 131, 132, pl. 24, figs. 14-17, 1926. Melocrinus whittakeri was based by Springer upon three well- defined specimens in which the characters are thoroughly con- stant. The holotvpe came from the Trout River section, about 15 miles above its confluence with the Mackenzie, “from beds at least 300 feet above the Simpson shale and thought to be some- what higher than the M. borealis horizon of the Hay River, Upper Devonian” (ref. cit., p. 132). In discussing the relationship of this species with M7. kindlei and M. mackenzie from the Root River section Springer states that it is “readily distinguished from them and all others known by its marked ovoid contour, and ex- tremely small column facet, which indicates a considerably differ- ent type of column from that of the genus as generally found. The tendency is usually to a broad base. None of the Belgian species is at all similar to this except in asymmetry” (ref. cit.). In the collection submitted by Dr. Kindle are at least three specimens that might be referred to WM. whittakeri, but from the same crinoid bed were collected three new species all showing a small column facet and two of them with elongate ovoid calyx. These new species are, however, easily distinguished from M. whittakert. One of the specimens referred to this species is abnormally large, measuring 27 mm. to the arm bases with a broken basal cup. The second specimen has a height for the calyx of 28.2 mm. (24 mm. to arm bases); the third 25.5 mm. (22.2 mm. to arm bases). All the specimens, therefore, are larger than those de- scribed by Springer; and the writer believes that they are more mature forms, as indicated by the character of the plates. Second- ary thickening of crinoid plates develops in older forms sometimes with quite striking difference in plate characters. Springer de- scribes the plates of the calyx as “smooth or slightly rugose, flat, with a slight tendency to pitting at the angles, but without convex- 8 BULLETIN 81 16 ity or median elevation either in dorsal cup or tegmen” (p. 131). In these three specimens the pitting at the corners is well shown. Thickening of the plates is seen at the margins, and there is a central raised area or flattened tubercle surrounded by a slight depression due to the thickening at the margins. In the second largest specimen an occasional plate shows a more prominent central tubercle. Horizon and locality.—From the crinoid bed at the upper falls, Redknife River, locality 7288. Melocrinus subtilistriatus n. sp. Plate 1, figs. 1-5 In the collection submitted by Dr. Kindle are a fairly large num- ber of specimens which in the shape of the calyx and the small column facet bear a strong resemblance to M. whittakeri. This appears to be a smaller species. The specimens are of medium size, average calices measuring between 20.2 mm. and 22 mm. high with a width at the arm bases from 15.7 mm. to 18 mm. Two particularly large specimens have heights of 24.6 mm. and 25 mm. All the specimens have asymmetrical, elongate ovoid calices contracting more or less strongly below to a very narrow base, with the characteristic small column facet, and also between the rays into the very low tegmen with subcentral anal opening without a tube. Specimens in this collection, if seen alone, might give the im- pression of being varieties of the species, or even different species, because of the presence in some of faint stellate ornamentation, in others of raised ridges and more tumid plates. However, the writer has picked out a series showing the relation of one stage to the next. One specimen shows the “smooth or slightly rugose, flat plates”, referred to by Springer (1926, p. 131) in his descrip- tion of M. whittakeri, with pitting at the angles. Closer inspection shows remnants of delicate carinee crossing the suture lines, par- ticularly well shown on the radials, first primibrachs and primary interbrachials. A second specimen shows beautifully a delicate ornamentation of groups of two or three fine carine extending 17 MACKENZIE CRINOIDS: GOLDRING 9 from center to center of the basals, radials, first primibrachs and primary interbrachials. Usually the center ridge of each group is stronger. The higher plates of the radial and interradial series have strong ridges, usually only one running from center to center with an accompanying deeper pitting at the angles. As the indi- viduals grew older changes in the character of the ornamentation took place, well-shown in the selected series of specimens. The carine thicken into ridges with the development of a low ridge following each radial series. Sometimes there is de- veloped a central blunt tubercle, marking the junction of the carine at the center of the plates, particularly the primary inter- brachials, and with this a rugose character is given to the plates. As this thickening process continues the plates become quite tumid, with the presence of carinz indicated, if at all, only at the sutures and on the higher interbrachials, especially of the anal interradius. Thickening of the plates at the margin tends to de- velop depressed sutures. Horizon and locality.—The cotypes are from locality 7288, the crinoid bed at the upper falls, Redknife River. “There are other specimens from the same locality and from locality 7291, bed marked e (field No. 267), %4 mile below the upper falls, Redknife River ; probably also from locality 7005, bed h, Lady Evelyn Falls section, Kakisa River. Remarks.—The specific name is given because of the finely striated character of the ornamentation which distinguishes this species from both M. whittakeri and M. sulcosutura. Melocrinus sulcosutura n. sp. Plate 1, fig. 6 From the same crinoid bed in which M. subtilistriatus 1s so abundant was collected a single somewhat crushed specimen of the same type, but with enough differences to warrant placing it in a new species. Nearly all of the dorsal cup and part of the tegmen are preserved. A fragment from the gorge section at the lowest chute, Redknife River (loc. 7300) has also been referred te this species. 10 BULLETIN 81 18 M. sulcosutura has an ovoid calyx as seen in M. whittakeri and M. subtilistriatuws but is more rounded at the basals, which have a lower, broader saucer-shape, and likewise broader at the arms base. This is a more robust form, larger and heavier than the average of the specimens of M. subtilistriatus with a height to the arm bases of 24.5 mm.; height of calyx 28 mm. The asym- metry of this species is masked by a certain amount of crushing. The plates are somewhat elevated, flat or slightly rounded, beveled at their margins and with the suture lines giving the appearance of being widely grooved. In this respect the species differs from M. whittakeri and M. subtilistriatus and bears a resemblance to M. bainbridgensis described by Hall and Whitfield 1875 (see Goldring 1923, p. 130, pl. 12, figs. 5-9) from the Upper Devonian (Huron shales) of Ohio. The interradial plates above the pri- mary interbrachial are usually quite tumid, sometimes even point- ed and the same is probably true of the tegminal plates, judging from the little that is preserved. Nothing is known of the anal opening. The column facet is small as in M. whittakeri and M. subtilistriatus, Horizon and locality—The holotype is from the crinoid bed at the upper falls, Redknife River, locality 7288. Remarks.—The grooved sutures suggested the specific name. These and the smooth, elevated bevel-edged plates distinguish this species from VM. whittakeri and M. subtilistriatus. An abnormality in the right postero-lateral interradius should be noted. It is comparable to the abnormality found in the right antero-lateral radius in Melocrinus hwmei, involving the right postero-lateral and right anterior interradii. As in the case of that specimen the condition apparently is connected with the asymmetrical development and occurs on the convex side. In this specimen the abnormality takes the form of an extra plate in the position of an anal plate between the right posterior and right antero-lateral radials. The radials in these two rays are smaller than average ; the first primibrachs are slightly larger than 19 MACKENZIE CRINOIDS: GOLDRING 11 average as also is the primary interbrachial in the right postero- lateral interradius. Melocrinus humei n. sp. Plate 1, figs. 7, 8 Melocrinus humei comes from the same bed as M. subtilistria- tus and M. sulcosutura. The description and figures are based upon a single, well-preserved calyx; but two other specimens from the same bed have been referred to this species. M. humei is a robust species which has the asymmetrical calyx characteristic of the other species from this region. The dorsal cup has a height to the arm base of 25.5 mm. and a width between 23 mm. and 24 mm. (calyx 28.8 mm. high). As in M. whittakeri, M. subtilistriatus and M. sulcosutura the dorsal cup narrows rapidly below the radials to a small column facet, not however, as small as in these three species. The basals flare slightly out- ward to their junction with the first columnal. Above the basals the dorsal cup expands rapidly giving an inverted pyramidal shape to the calyx. The tegmen is low, flat near the central portion and with the ambulacral areas raised into low ridges which give a depressed effect to the interambulacral areas. The anus is subcentral and there is no anal tube. The plates of the dorsal cup are generally flat with a suggestion on the radial series of a low, broad, longitudinal ridge. A slight thickening of the plates gives a beveled appearance to the edge of the plates and an appearance of grooving to the sutures, though this latter character is not conspicuous as in M. sulcosutura. The character of the plates suggests two Upper Devonian forms: M. clarkei from western New York and M. bainbridgensis from Ohio. The interradial series has the succession I, 2, 3, 3, 4 or I, 2, 3, 4, 4. A curious abnormality occurs in the right antero- lateral radius apparently in connection with the asymmetrical development of the specimen as it occurs on the convex side. The right postero-lateral and right anterior interradii have each three plates in the third rank so large that this series of the two interradii meet above the first primibrach in the right antero- 12 BULLETIN 81 20 lateral radius, separating it from the primaxil above. This char- acter does not occur in the other specimens referred to this species. The plates of the dorsal cup are further characterized by small scattered tubercles more numerous on the basals, radials, first primibrachs, and primary interradials. The plates of the tegmen are small, usually flat and sometimes bearing a central tubercle particularly on the ambulacral series. One of the other specimens referred to this species shows a slightly tumid condition in some of the plates, with the develop- ment of low nodes on the first secundibrachs. The tubercles have also thickened and sometimes have coalesced. The third specimen shows a different condition which might develop with maturity, but has led the writer to refer this specimen to the species with a query. The slightly grooved appearance of the sutures is shown and the flat character of the plates with the low ridge on each radial series. Only here and there is there a dis- tinct tubercle seen, probably because the plates are so weathered and also because there has been a thickening of the plates and coalescence of tubercles. The plates are bordered by a narrow, flat thickened area suggestive of M. bainbridgensis but found also by the writer in specimens of Megistocrinus depressus as a char- acter developing with maturity (1923, p. 231, pl. 33). Horizon and locality—The holotype, and the other two speci- mens as well were collected from the crinoid bed at the upper falls of the Redknife River (locality 7288). Remarks.—Vhe specific name is given in honor of G. S. Hume, one of the collectors in this area. The shape of the calyx togeth- er with the small column facet, the ornamentation and character- istic tegmen distinguish this species from all the others. FLEXIBILIA Synaptocrinus (?) rotundatus n. sp. Plate 1, fig. 9 There is only a single specimen of this species which has been referred with a query to the genus Synaptocrinus. The infra- 21 MACKENZIE CRINOIDS: GOLDRING 13 basals are entirely within the ring of the basals. One of the basals, assumed to be the posterior, is elongate; but it is not strikingly larger than the others and is overlapped by the radials. There is no radianal in the position of the inferradial as in /cthyo- crinus which this calyx otherwise resembles closely (see Springer, 1920, p. 264). One interradius shows an interbrachial plate above the first secundibrachs, two interradii do not show any and two are so poorly preserved and fractured as to show nothing of value. Springer’s genus was based upon one species, Synapto- crinus nuntius. In discussing his genus Springer (ref. cit., p. 301) writes, “But if as in the case of Wachsmuthicrinus inter- brachials should be found in some specimens, the genus would stand upon its other characters.” The arms are dichotomous, joined or closely appressed. The species in question does not exactly fit into either the genus I/chthyocrinus or the genus Synaptocrinus; but it more nearly fulfills the requirements of the latter and the genus may prove to be variable in the matter of interbrachials and the size of the posterior basal. The single crown representing this species is poorly preserved, with the left posterior ray in the best condition. The calyx is small, rounded, expanding distally, with a width of 12 mm. at the primaxils above which it expands to its greatest width at the second bifurcation of 14.1 mm. The total height preserved is 15-2 mm., and the indications are that the arms were incurved at this height. The base outside the column is 3.4 mm. A raised ridge, almost a tubercle, marks the sutures between the primi- brachs of adjoining rays, prominent on the first primibrachs and flattening out at the top of the primaxil. A depression marks each interradius from the first primibrach to the secundaxil giving a raised appearance to the brachial series at this level. A similar depression separates the two half rays at the level of the secund- axils and above. The plates of the dorsal cup up to the top of the primaxils are fairly flat. Above this the plates of the brachial series are rounded. 14 BULLETIN $1 22 The basals are small but proportionally larger than in S. nun- tius. The posterior basal is noticeably larger than the other basals, but like them does not extend to the full height of the radials. It also terminates in an acute angle, leaving no surface for attach- ment of anal plates. The radials are considerably wider above than below, and the left posterior one is somewhat larger than the others. There are two primibrachs, much wider than high. All brachial plates are much wider than long. There are three sec- undibrachs, followed in one ray by at least ten tertibrachs without another division. The brachials seem to be joined at least as far as the fourth tertibrach. There is a rapid increase in width from the radials to the primibrachs which have the same width as the combined two secundibrachs immediately above. All the sec- undibrachs have about the same width and the arms narrow gradually above and have rounded backs. The interbrachial in the left postero-lateral interradius is long and narrow, resting upon the shoulders of the first secundibrachs and extending up to the second tertibrachs where the brachials of adjoining rays close above it with no indication of higher interbrachials. The plates of the dorsal cup apparently have been smoothed in cleaning for there is indication from patches here and there that the surface of the plates originally was rugose. The brachials definitely have a rugose surface, giving almost a pitted appearance in patches. Horizon and locality—Crinoid bed near base of coral zone (station 1297), Jean Marie River. Remarks.—TVhe species is named from its decidedly rounded oval shape, particularly in the distal part of the crown. This species is readily distinguished from S. nuntius, among other characters, by the presence in the latter of nodes on the radials, primibrachs and all axillaries beyond and strongly elevated rays, angular in the middle; and by its own proportionally smaller posterior basal and the presence of at least an occasional inter- brachial. 23 MACKENZIE CRINOIDS: GOLDRING ; 15 INADUNATA Decadocrinus spinobrachiatus n. sp. Plate 2, figs. 1, 2 Among the inadunate forms is one which bears a resemblance, particularly in the character of the arms, to Decadocrinus multi- nodosus var. serratobrachiatus Goldring from the Hamilton (Mos- cow shale) beds of western New York (1923, p. 431, pl. 56, fig. 1). This species is based upon a crown in a fair state of preser- vation. A second specimen partially and poorly preserved is tentatively referred to it. The crown is preserved to a height of 43.6 mm., but the distal portion of the arms is missing. The dorsal cup is low and flares rapidly. It has a height of 4.5 mm., a width at the top of the radials of 9.6 mm. and a width at the column facet of 2.2 mm. The infrabasals are very small, almost hidden by the basals which seem to be somewhat thickened in the lower portion. The basals are comparatively small, an average one having a height and width of 2.4mm. The radials occupy the larger portion of the dor- sal cup with a height of 3 mm., a greatest width at the shoulders in a typical plate (anterior) of 4.6 mm. and a width at the radial facet of 4 mm. The radianal is pentagonal, smaller than the basals and the anal x» which is of about the size of the basals and projects above the radials. The first tube plate is not completely preserved but appears to be as large as the anal x and borders upon the radial and first primibrach in the right posterior radius, as well as the radianal and anal. Nothing more of the anal tube is preserved. The dorsal cup is unornamented except by deep pitting at the corners of the plates which gives the effect on the basals of short ridges crossing the suture lines to the radials above and the infrabasals below. There are two primibrachs; the first quadrangular and about twice as wide as high, the primaxil pentagonal and slightly wider than high. With the exception of the one in the anterior ray, each first primibrach bears a spinose tubercle at the center of the 16 BULLETIN 81 24 upper margin. In addition there are more or less conspicuous, small spinose tubercles at each of the four corners. Each prim- axil bears a short, sharp spine just below the point of bifurcation and in addition may have the small spinose projections at the four corners as do the first primibrachs. The brachials are wedge- shaped, not conspicuously so, but enough to give a zig-zag effect to the arms. Each brachial is provided with a short, sharp spine, the spines pointing alternately to one side and then to the other, giving a very pronounced serrated or saw-toothed appearance to the arms. In addition there may be spinose projections at the corners. Each brachial bears a pinnule on the higher side be- neath the spinose projection, thus giving an alternate arrange- ment, as is the case with the spines. The pinnules are long and slender, composed of long ossicles ; and they appear to be rounded on the dorsal side. The column appears to have been subpenta- gonal. Horizon and locality—From the bed marked gq, at the third chute, Redknife River (locality 7208). Remarks.—This species derives its name from the spiny char- acter of the arms and may readily be distinguished from the Hamilton form to which it bears a resemblance. The spines ornamenting the brachials are short, sharp and rounded while in D. multinodus var. serratobrachiatus they are angular and tooth-like. The latter lacks the spinose projections at the angles of the brachials. D. spinobrachiatus has no nodes on the basals and apparently no surface ornamentation of the plates of the cups and arms. “The second specimen, tentatively referred to this species, was collected from the bed marked J, of the gorge section, Bouvier River. The specimen shows a crushed dorsal cup and portions of three arms, two preserved above the primaxils. The speci- men, so far as preserved, agrees with D. spinobrachiatus in all characters except the presence of three primibrachs in the two rays preserved. The writer carefully examined the type for 25 MACKENZIE CRINOIDS: GOLDRING 17 anchylosis of a possible second primibrach with the primaxil, but there was no indication of this though the anterior ray does show anchylosis of the first primibrach and the primaxil. The second specimen is a larger perhaps older specimen, so, since there is no knowledge of the character of the other rays or variability within the species, the writer feels this specimen should for the present be placed with D. spinobrachiatus. PRININOCRINUS n. gen. Prininocrinus is a dicyclic inadunate crinoid belonging to the subfamily Proteriocrinine of the family Proteriocrinide. The genotype is Prininocrinus robustus, new species. The dorsal cup of the only species is bowl-shaped. The radial facet occupies the full width of the radial; and the arms are unbranched above the first axillary, composed of quadrangular brachials bearing pinnules alternately on each side. The ventral sac is unknown and the column appears to have been round. The radianal is in line with the radials, the anal x in large part above the radials and resting upon the radianal and left posterior radial. 5 Bo ead. C0” ae Figure 1. Analysis of dorsal cup of Prininocrinus robustus, genotype. TB, infrabasal; B, basal; A, radial; RA, radianal; x, anal; rt, right tube plate. Of the other genera belonging to this family Prininocrinus re- sembles Decadocrinus W. and Sp. (Devonian to Carboniferous) in the character of the arms; but bears most resemblance in the character of the dorsal cup to two Carboniferous forms, Zeacrinus Hall and Cromyocrinus Trautsch (see Springer, 1913, pp. 223, 224; Bather, 1900, pp. 180, 181). Both of the last two genera 18 BULLETIN 81 26 have the wide radial facets and in both the anal 4 is high in the cup, extending above the radials. In Cromyocrinus the anal + rests upon a short face of the posterior basal as well as upon the radianal and left posterior radial. This genus has arms un- branched above the first auxillary. In Zeacrinus, with branching arms, the anal x rests only on the radianal and left posterior radial, but with a different arrangement for the plates. The name is derived from the Greek prininos sturdy; krinon, lily. Prininocrinus robustus n. sp. Plate 2, figs. 3, 4 Prininocrinus robustus is represented by five specimens from the Redknife River section. The figures and description are based largely upon one specimen, the holotype, because the others show no additional characters. Of the four other speci- mens one is a crushed dorsal cup, another a portion of an arm; a third shows the dorsal cup and portions of three arms; the fourth a portion of the dorsal cup and parts of the anterior, right antero- lateral and left antero-lateral rays. In the holotype 31.8 mm. of the crown are preserved and there is no indication that the arms are anywhere nearly complete. The dorsal cup is narrowly bowl-shaped, widening gradually from the base and definitely rounded at the radials. It has a height to the top of the radials of 5.2 mm. and a width at the column facet of 1.3 mm. The width at the top of the radials is between 6 mm. and 8 mm. As the cup is crushed no accurate measurements can be made. The plates are apparently heavy, smooth and flat. The infrabasals are five in number, small with a height of 1.1 mm. The basals are comparatively large, all pentagonal except the right postero-lateral which is the largest in the cup and hexa- gonal, since it abuts with a short upper face upon the radianal. An average plate (right anterior) has a height of 3.1 mm. and a greatest width at the shoulders of 3 mm. The radials are the largest plates in the cup, an average radial having a height of 3 mm. and greatest width at the shoulders of 4 mm. The anterior 27 MACKENZIE CRINOIDS: GOLDRING 19 radial is slightly larger, the right posterior radial distinctly smaller than the average. The radial facet occupies practically the entire upper face. Two primibrachs are present in the holo- type in all rays except the anterior where only the first primi- brach is preserved. In the specimen showing parts of three arms one ray shows three primibrachs, and this could very well be the anterior ray. Some of the primibrachs have a suggestion of a crenulate margin. In this specimen the right anterior ray indi- cates unequal bifurcation on the primaxil, but this may be due only to the position in which the right branch is preserved. The arms are rounded on the dorsal side, heavy, composed of quadrangular brachials, wider than high, with a tendency in some to wedge-shape ; and they bear pinnules alternately on each side. The pinnules are apparently long, stout, composed of quadrangu- lar ossicles, longer than wide. There is no trace of ornamenta- tion on the arms. The column appears to have been round. Horizon and locality.—The holotype is from the bed marked g, at the third chute, Redknife River, (locality 7298). The para type and other specimens are from the bed marked 7, of the gorge section, Redknife River (locality 7294). LINOCRINUS n. gen. The genus is represented by a single known species, the geno- Gs eG) nOC0O. Figure 2. Analysis of dorsal cup of Linocrinus kindlei, genotype. IB, infrabasal; B, basal; Rk, radial; RA, radianal; x, anal; rt, right tube plate. 20 BULLETIN 81 28 type, Linocrinus kindlei, new species, so the generic diagnosis of necessity is drawn from information furnished by this form. Linocrinus is a dicyclic inadunate crinoid the affinities of which are closest to those of the family Cromyocrinidee as defined by Bather (1900, p. 181) which includes genera placed by Springer (1913, p. 224) in the family Poteriocrinidee Roemer, as emended by Wachsmuth and Springer. The dorsal cup is elongate obconical. The five infrabasals are comparatively small and inconspicuous, the basals very large and the radials considerably smaller than the basals. The radial facet is very slightly curved and occupies the entire upper face of the radial. A radianal, anal and right tube plate lie within the cup and the anal plate is in line with the radial. The right antero-lateral radial, crowded by the right posterior radial, is con- siderably reduced in size and appears to bear no arm. ‘The radial facet is very slightly curved and occupies the full width of the upper face of the radial. The arms are unbranched through- out their length, four in number and bear two pinnules to each” brachial. The column is stout and round, Linocrinus shows striking similarity to Cradeocrinus Goldring (1923, p. 347) in the character of the dorsal cup. It is distin- guished by the four unbranched arms and the irregularity shown by the right posterior and right posterior-lateral radials. In the Permo-Carboniferous genus Tribrachiocrinus M’Coy irregularity is found similar to that shown in this genus with only three radials known definitely to bear arms, and those single (see Rather, 1900, Delo SpLingery Tone" p22): The name is derived from the Greek linon, strand (of a rope) and krinon, lily, because of the resemblance given by the long stout arms to an unraveled rope. 29 MACKENZIE CrINOIDS: GOLDRING 91 Linocrinus kindlei, n. sp. Plate 2, figs. 5, 6 ma (his species is represented by two adult specimens, two young specimens and a small portion of two arms. The description is based entirely upon one specimen, the holotype. The crown is long and slender, measuring in the holotype 90 mm. with the arms incompletely preserved and showing no tapering as far as present. The dorsal cup is slender, elongate obconical in shape. The crushed condition of the cup permits no measurement in width, but the height to the top of the radials is approximately 11.3 mm. There are five infrabasals, 2.3 mm. high, which give the appear- ance of a collar at the base of the cup, of practically the same width as the proximal columnal. The cup gradually widens above this. The basals are elongate and by far the largest plates in the cup. The least crushed plates are the posterior and right postero-lateral. The former has a height of 6.2 mm. and a width at the shoulders of 3.3 mm.; the latter a height of 6.3 mm. and a greater width of approximately 3.6 mm. The radials are of moderate size; the left posterior one has a greatest width of 4.7 mm, and a height of 3.3 mm.; the anterior a height of 3 mm, and a width of 4.6 mm. The left antero-lateral radial is missing but the shoulders of the basals indicate that this plate was present. The radial facets are very slightly curved. The right posterior and right antero-lateral radials are not typical. The former has a width of 4.6 mm. and a height at the posterior side of 3.5 mm. Fetween this radial and the first quadrangular brachial is a small wedge-shaped plate extending from the anterior side a little more than half way across the upper face of the radial. The right posterior radial encroaches upon the right antero-lateral radial which is considerably smaller than any of the others and almost in the position of an infer-radial. It has a very short upper face that apparently is not followed by any brachials. This condition cannot be regarded as an individual abnormality since it is also shown in one of the young specimens, the only other specimen 22 BcLueTIn 81 0) showing the posterior side. The anal plates have the arrange- ment characteristic of most of the genera in this family. The anal a4 is in line with the radials. The pentagonal radianal, smaller than the anal x, borders upon the posterior and right postero- lateral basals, the anal # and right posterior radial and bears the right tube plate. The left tube plate is not preserved. The arms are preserved to a length of 82 mm. and apparently are not nearly complete. They are unbranched throughout their length, four in number, and composed of quadrangular brachials of approximately equal length and width or longer than wide. An occasional one is wider than long. The noticeably long ones are formed by the anchylosis of two shorter brachials. In some the suture is faintly discernible, but their nature is also disclosed by the presence of two pinnules on each side. There is a pair of pinnules to each simple brachial, one on each side. They are short (4 mm.), stout in the proximal portion, tapering rapidly to a pointed tip, and fit into special niches or rounded sockets in the shoulders of the brachials. The column is round, of practically the same diameter as the infrabasal ring, and composed of alternating thin and thick columnals with crenulate margins. The species is named in honor of Dr, E. M. Kindle. Horizon and locality.—The holotype was collected from the bed marked /, gorge section of the Bouvier River (locality 7275) ; the paratype and the other young specimen from station 122, of the Bouvier River, 8 miles above the mouth (locality 7281). The other two specimens are from the same locality as the holotype and the bed marked g at the third chute of the Redknife River (locality 7298). BIBMIOGRAPREM Bather, F. A. 1900. Lankester’s Treatise on Zoology, Pt. 3, Echinoderma; The Crin- oidea, pp. 94-204. Branson, E. B. 1923. The Devonian of Missouri. Mo. Bureau of Geol and Mines. v. 17, sul MACKENZIE CRINOIDS: GOLDRING 23 gnd ser., 277 pp., 71 pls. Fraipont, J. 1883. Recherches sur les crinoides du Famennien (Devonien Superieur) de Belgique. Annales de la Soc. géol. de la Belgique, t. X, pp. 45-68; pls. 2, 3, 4. 5. Goldring, W. 1928. Devcnian Crinoids of New York. N. Y. State Mus. Mem. 16, 669 pp., 60 pls. Greger, D. K. 1909. The Devonian of Central Misscuri. Amer. Jour. Sci., ser. 4, v. 27, pp. 374-378. Hall, J. 1872. Descriptions of New Species of Crinoidea from the Carboniferous Rocks of the Mississippi Valley. (Includes plate of New York Devonian Forms). N. Y. State Mus. Bul. 1, pl. 1. (Photographic plates distributed privately). Keyes, C. R. 1894. Paleontology of Missouri, pt. 1, Mo. Geol. Surv. v. 4, 271 pp., pls. 12-32. 1902. Devonian Interval in Missouri. Geol. Soc. Amer. Bul. 138, pp. 271-2738. Kindle, E. M. 1919. The Discovery of a Portage Fauna in the Mackenzie River Valley. Geol. Surv. Can. Bul. 29, pp: 1-8, pls. 1, 2. Koenen, A. von 1886. Die crinoiden des norddeutschen Ober-Devons. Neues jahrb. f. Min. 1886, Bd. 1, pp. 101-116, pls. 1, 2. Rowley, R. R. 1893. Description of Some New Species of Crinoids, Blastoids and Brachiopods from the Devonian and Subcarboniferous Rocks of Missouri. Amer. Geol. v. 12, pp. 303-309, il. 1894. New Species of Crinoids and Brachiopods from the Missouri Ham- ilton. Amer. Geol. v. 18, pp. 151-154, figs. 1-10. Schuchert, C. 1903a. On the Faunal Provinces of the Middle Devonic of America and the Devonie Coral Subprovince of Russia, with Two Paleogeogra- phic Maps. Amer. Jour. Sci., v. 32, pp. 1387-162. 1903b. Paleogeography of North America. Geol. Soc. Amer. Bul. 20, pp. 427-605, 51 pls. Springer, F. 1913. Zittel-Eastman Text-book of Paleontology, 2nd ed., v. 1, The Crinoidea. pp. 173-243. 1920. The Crinoidea Flexibilia. Smith. Inst. Pub. 2501. 486 pp. with Atlas of A. B. C. and 76 plates. 1921 New Species of Devonian Crinoidea from Northern Canada. Geol. Surv. Gan. Bul. 33; pps 15-17, pl. 1. 1926. Devonian Crinoids from the Mackenzie River Valley. Geol. Surv. Can. Bul. 42, pp. 127-132, pl. 24. Weller, S. 1909. Correlation of the Middle and Upper Devonian and the Mississip- pian Faunas of North America. Jour. Geol. yv. 17, pp. 257-285. EXPLANATION OF PLATES EXPLANATION OF PLATE 1 All photographs were made by J. A. Glenn, Albany, N. Y. Figure OO cS Melocrinus subtilistriatus n. sp. _——-----------------o (All eotypes from locality 7288, the crinoid bed at the upper falls of the Redknife River). Specimen showing the delicate ornamentation of fine radiating caring; anterior radius up. View of right antero-lateral radius of a specimen so weathered that the fine carine have almost disappeared. Remnants are shown crossing the suture lines. Asymmetry shown. A third specimen crushed. The plates begin to show tumidity and the carine are strengthened, having more the appearance of faint ridges; left antero-lateral radius up. View of left antero-lateral radius of a specimen showing tumid plates and strong radiating ridges on all plates. Mature specimen (right posterior radius up) showing strong- ly tumid plates with radiating ridges only on the higher inter- radial plates. Melocrinus sulcosutura n. sp. _.. eel Raat Keel AA ENE SE ea aa ee (Locality 7288; crinoid bed at the upper falls of the Red- knife River). ' View of right antero-lateral radius of holotype showing some- what elevated flattened plates and deep grooves along the su- tures. Note extra plate at left in position of anal. Melocrinus: huimed 1) Sp ese rN I ea (Loeality 7288; crinoid bed at the upper falls of Redknife River). View of left postero-lateral interradius of holotype; asymmetry well shown. Right antero-lateral radius, showing interradial plates inter- ealated between the first and second primibrachs. Synaptcerimnus retundatusin: sp. eee (Locality 1297; crinoid bed near base of coral zone, Jean Marie kiver). Holotype (left posterior radius) showing small rounded calyx witli imcurved aims; elongated posterior basal and single inter- bracuial in left posterolateral interradius, x 114. 11 12 Buu. AMER. PALEONT. No. 81, Pu. 1 PLATE 1, VOL. 24 oan s ai: oe « Ase eae 8; EXPLANATION OF PLATE 2 Figure Decadocrinus (spinobrachiatus) n> spy EEE [From the bed marked q, at the third chute of the Red- knife River (locality 7298) ]. 1. Anterior view (anterior radial at left of the crown showing low, flaring dorsal cup and saw-toothed appearance of the arms due to the wedge-shaped spine-bearing brachials. 2. Dorsal cup, x 1144; anterior radial at left. The relatively large radials are well shown and the deep pitting at the corners of the plates. EAGAN OCEINUS SE OD US EUS i Tes os ec a From the bed marked q, at the third chute of the Red- knife River 3, 4. Anterior and posterior views of the holotype x 14%. In the anterior view the pinnules are shown on the arm at the right. Linoerinus: kind] ei ins (spi ee ee ee aes [ Holotype from bed marked 1, gorge section of the Bouvier River (locality 7275); young specimen from station 122 of the Bouvier River, 8 miles above the mouth (locality 7281) ]. Lateral view of holotype (posterior interradius at left) show- ing the long, rather heavy arms with stout rapidly tapering pin- nules and the atypical right posterior and right antero-lateral radials. 6. Young specimen of the species x 1%. O1 18 21 PLATE 2, VOL. 24 BULL. AMER. PALEONT. No. 81, Pu. 2 nom AN ts AMERICAN |" age ang VOR KRW NO. 82 ed % PALEONTOLOGICAL KSEARCH INSTITUTION ) Ithaca, New York id ; U.S. A. . 4 bei BULLETINS OF AMERICAN PALEONTOLOGY Volume 24 Number 82 The Correlation of Certain Devonian Faunas of Eastern and Western Gaspe By E. M. KINDLE With APPENDIX, 82 A, Devonian Bryozoa of Gaspé By Madeleine A, Fritz December 2, 1938 ITHACA, N. Y. Ues. A hed a ‘ae in We in, CONTENTS Tisai reCOYSPKOLENVON ie Bee STE arg ee, DAP rents Dd ae a eee eee Shere lwo tawlhays tally GAISt Oy pees eee eee eee tee eee areas ave tence dee TiO GENG ofc ga BEST ee epee pe te Mee ey EE Re ge ee eee TEES yaCeumiay (GENS OSE cee ey Se ae I eee See ee ee Biraytoqies AW eee: Sek ee Rt ed a ee ce ee inabler ot Honma tronssim eB astern: Gris es ess seers ems eee See eee Strano alae Seve ats 15 eset: Sac lense = eee ee ee eee he 2 eee fem Or clowaeiariy VOC Kg pe meceta tee een eae eyes SNE RE toe as 8 (Gas pe wibimntes Tomes mee eammnneree ee nee ak See eet oe re een eet eee e Contact of Grand Gréve Limestone and Gaspé Sandstone ___.. Interpretation of Devonian sedimentation. .....o.2 20.0242 HW Acles tots hem Gasper SAMUS tO. -.\ se.) ee 2: eee eke eee ee eee cen SETS|O Say SE MAONSERO IT Gc acT.DICS: 5 x Eee BALE Ce aa PRIN 6 sates Poe eel oe a ROME a a Oe an UCM ASO CLALONS ete am een tee el Pie ee, duel Pe FSS. CornelationtorsheuG as peatatian seeeees sa a ees cece eee ew ce eee Gas pen Sal SLOM CRS G ChOS ties te len tet ee eet et 3s. o eee 2 ee eae eran Hinihetpore tat OM ON CV ICN CO. 6 ac eee cee tes res ona ee nn dapicat sass bee nncadenenennnete AVES GEIL Gicls pC mame See Ree Ra Oe ee ee ee eden Eee ihe ee ihewh oie MalesiBnoo mse ctl iy esse eee ses ely eee “SURMISE so ad a ee pa ee a ee eee ee ROT OTCTIC CM mee eee tem eae tah Viled ota ewer ett yee, Urge ae Bee aie POE ot ecnacce eee TDIGHVEG) cast ee bee ae ER ha Ee Ne Pe oe th 5 aren a ee eee Ae ne ee rt eS to bo ~ [WW WWNNnNN DW WW LDH cen ow © © co » <> THE CORRELATION OF CERTAIN DEVONIAN FAUNAS OF EASTERN AND WESTERN GASPE ae By E. M. KiInpbLe Ottawa, Canapa* INTRODUCTION The faunas and formations discussed in this paper represent areas in the most easterly and the most westerly parts of the Gaspé Peninsula, Province of Quebec. The eastern sections lie in the region adjacent to Gaspé Bay, while the western area studied is located about 120 miles west of the eastern end of the Peninsula in the Metapedia River valley. Both lie within the region of Appalachian Mountain folding, the eastern area repre- senting the northeastern terminal region of these structures on the continent. The faunas chiefly considered include the earliest Devonian fauna, the Gaspé Sandstone fauna of eastern Gaspé and the youngest Devonian fauna of western Gaspé. Devonian formations are known to be present in the interior which is a timbered mountainous region, for the most part with- out roads. Here relative inaccessability has thus far prevented detailed investigation except in the upper part of the Cascapedia River valley and in the upper valleys of the Dartmouth, York and St. John’s Rivers. In the latter region mapping by Dr. I. W. Jones of the Quebec Department of Mines (14a) has been in prog- 1ess during recent years. The collections on which the present study is mainly based were made by the writer assisted by Dr. C. H. Kindle during the season of 1935 and by Dr. V. J. Okulitch in summer of 1936. Faunules representing the interior region col- lected by Dr. Jones and studied by the writer are recorded in Dr. Jones’ Annual Reports (14a and 14b). Dr. M. A. Fritz has kindly determined the bryozoa. *Published by Permission of Director, Mines & Geology Branch, Dept. of Mines and Resources. 6 BULLETIN 82 40 In the present inquiry emphasis will be placed on the physical environment of the faunas considered. “In order to understand the life history of the globe it is necessary to first know the physi- cal history” (36). This viewpoint is helpful in dealing with Paleozoic history and often neglected in the study of fossil faunas. In the present study both the physical and the faunal history will be utilized with a view to using the evidence of each to clari- fy the record of the other. The two kinds of evidence bear a supplementary relationship and the use of either alone fails to give ithe complete story which stratigraphic paleontology should supply. In any study of fos il faunas the environment of the life represented by them is of iindamental importance. . The con- trasts between faunas of successive geologic epochs produced by extinction, the emigration of species or the development of new ones are themselves largely the products of changing environ- ments. The physical envirrnment of aquatic life includes such potent factors as temperatt ce, salinity, depth and nature of sedi- ments. Study of the sediuents enclosing fossil faunas gives much information regarding i:ifluences which have moulded and select- ed the species prese’t. The maintenance of different tempera- tures in different parts of the same sea or ocean ‘through ma- rine currents of northern and southern origin has doubtless been of vital ixsportance in moulding the composition of some unlike contemporaneous Devonian faunas in eastern North America. Contrasted faunas resulting from great differences in the height of tides were probably developed in Devonian seas as often as they are in modern seas. The oyster fauna, less than 200 miles from the Gaspé coast, in Northumberland Strait bordered by the cold water of the Gulf of St. Lawrence is a significant modern ex- ample of such contrasts. It is some 50 miles north of the Bay of Fundy where, owing to the high tides and consequent low aver- age water temperature, the Acadian or Virginian fauna cannot live. The Northumberland Strait colony of the Acadian fauna affords a striking example of sharply contrasted adjacent living faunas separated only by temperature differences of the waters in which they live. Here shells like Venus mercenaria and the oyster which elsewhere are rarely found north of the southern 41 DEVONIAN OF GASPE: KINDLE a New England coast, flourish in habitats adjacent to areas in which the subarctic Saxicava rugosa is found, separated only by barriers of temperature and depth. In the Devonian of Gaspé we appear to have analogues of these modern examples. SKEAUGENOE APE SICALYyHISTORY In times preceding and during Devonian sedimentation a land mass composed of Precambrian crystalline rocks occupied a con- siderable area in the region of the present continental shelf and the coast line of the Atlantic States. This land has been called Appalachia (33) or the Province of the Devonian highland. It extended northeasterly including either a great continuous land mass embracing much of the Atlantic sea-board states and the southeastern two-thirds of Newfoundland or a series of large islands corresponding more or less closel. in position with the east coast “Paleozoic positive elements” of Prof. Schuchert (24) south of Ungava. Between Appalachia on the southeast and an- other old land mass lying northwest of ihe St. Lawrence and central New York stretched a seaway wuich at times could be more appropriately called the Appalachiar: Gulf (38). This sea- way received the erosion products of latids north, east and west of it, thus becoming the cradle of the A vpalachian Mountain sys- tem which near the close of the Devor.an cycle of sedimentation began to develop. The erosion of the ] evonian highlands and the accumulation of their muds and sand» in the Appalachian Strait was followed by their deformation and uplift and the foundering of Appalachia. The net result of thes» itwo orogenic agencies was the Appalachian Mountains which replaced the Appalachian Strait and the Continental shelf which supplanted the Devonian highlands. The Appalachian Movzintains now terminate at the eastern end of the Gaspé Peninsula in the north and disappear in the south under Cretaceous rocks in central Alabama less than 200 miles from the Gulf of Mexico. A Lower Devonian fauna re- ported in western Newfoundland by Schuchert and Dunbar (25) which resembles those in Gaspé an| at Dalhousie, N. B. indicates 8 BULLETIN 82 4° that the early Devonian seaway continued beyond eastern Gaspé across the Gulf of St. Lawrence into southwestern Newfound- land. The Gulf between the Gaspé coast and St. George Bay, Newfoundland now shows depths ranging from 300 feet or more near shore to 1800 near the middle of the Gulf. Block faulting of post-Pennsylvania age has probably been responsible for bringing the Alleghany structures, which presumably once reached across the Gulf of St. Lawrence to Newfoundland, some hundreds of feet below sea level. It is probable that an area of Devonian rocks greater than that of Gaspé and New Bruns- wick combined is concealed below the Gulf of St. Lawrence. The faunas and formations here considered lie at or near the northeasterly end of a belt of sediments which stretch more than 1500 miles in a southwesterly direction near the eastern border of the Continent from the Gulf of St. Lawrence nearly to the Gulf of Mexico. The seas which the Devonian formations and faunas show to have successively occupied this eastern continental belt differed greatly in extent and shape at different stages of De- vonian time. From the long wide strait of the Lower Devonian sea, coastal emergence developed embayments or gulfs such as J. M. Clarke’s “Appalachian Gulf” of early Portage time (66). Wes, (CrS IIa IBvACSIUN| EASTERN GASPE PREVIOUS WORK In the Gaspé Bay region folding and erosion have conspired to display to the best advantage the entire succession of Devonian formations. Orogenic forces have developed three wide north- west-southeast open folds. Erosion has etched away the eastern limb of the northeastern most of these structures on the north- east side of Gaspé Bay, leaving a zigzag scarp with sea cliffs facing the Gulf of St. Lawrence on the north side of the Forillon Peninsula and further north facing a narrow coastal plain shelf 43 DEVONIAN OF GASPE: KINDLE 9 cut in the black shales and other sediments of the older rocks. The several thousand feet of Devonian sediments exposed in the cliffs on the northeast side of the Forillon Peninsula and the dip slopes of the southwest side have given this finger like peninsula exceptional importance in the development of the geological in- vestigations made in eastern Gaspé. Sir Wm. Logan prepared the first description of the forma- tions of this area which was published in his 1844 report (17). In this he first described the Gaspé limestones and calcareous shales of the Cape Gaspé promontory. In the 1863 report (18) appeared his detailed section of the Devonian sediments. The importance of Logan’s work is well indicated in Dr. J. M. Clarke’s statement that “in all the years since elapsed from 1844 and 1845 little has been added to and naught subtracted from his achievements.” To the eastern Devonian faunas, first described by E. Billings (3), J. M. Clarke (7) added many new species and contributed a discussion of the geology and correlation of the faunas. A map compiled by Dr. F. J. Alcock was issued in 1931 by the Geological Survey of Canada (No. 259A) which assembled the available geological data for all of the Gaspé Peninsula on a map scale of 1 inch to 8 miles. In 1931 Prof. Parks (19) pub- lished a summary of the work done up to that date on the geol- ogy of the Gaspé Peninsula. A map (12 miles to one inch) which accompanied it presented a generalized conception of the geology of the entire peninsula without discriminating between the littke known or unknown and the intimately known areas. Reference to most of the work done up to the date of publicaltion appear in it but without bibliographic references. Dr. F. J. Alcock’s Memoir on the Geology of the Chaleur Bay region issued in 1936 includes a seven page bibliography embrac- ing nearly all of the literature touching the Gaspé Peninsula. The eastern Gaspé formations (see pl. I) which will be con- sidered in ‘this paper are indicated in the following table. Ad al io 2) b € BULLETID 10 = ae = Os Bs = WwW oF es Saal rs (as) Or S36 sayeys Jatsoyy ade UOULIOF S1IqG1OP]OT] TOC UISUT TOS NON a) ine saa ae eae ee Speq JdATY 9AOD UOFLIT) jOQI UPIS1IGIOP[IF] Speq surqiy 3S ,og!1 S ‘euney AUBYSLIC x x ace jydo9x9 poalenustoyip Areas 5 ese ai a E Speq muy uog odey oSor © jou seunry yIOX MON ‘sdnors A ueIURySIIQ, PUP UPIIdST) SUOJSOUNT VAQIN) apuRIN) {00g (WAI) HON AOD Sg eee Spoq IaATY Y10K (WREUSS)) DCUS BOA, speq Yyoorq neauuos auojspues odsery QfoZ gdser) Ulojsey Ul suOT}eUIO., Jo aIqQv Ty, 45 DEVONIAN OF GASPE: KINDLE iil STRATIGRAPHIC RELATIONS Ordovician Rocks.—The oldest rocks of this region are well exposed at Cape Rosier. The seashore at Cape Rosier light house exposes in shore cliffs and in the intertidal zone, black shales with bands of dark limestone extending seawards some 200 yards. Along the 214 miles of the shore separating the light house and the base of the cliffs which expose a contact between the Devonian limestones and the Quebec group shales, much of the pre-Devonian sediments are concealed. But disconnected outcrops of the old rocks south of Cape Rosier light are seen to include here purple, red and olive shales, the latter associated with light grey sandstones. The Gaspé limestones and shales rise in vertical cliffs some 700 feet above the black slate series which J. M. Clarke conceived to be separated from the older rocks near the shore by a great overthrust fault which brought the Lower Devonian in contact with the “Ordovician-Cambrian”’ (8). Concerning this same contact Logan in 1844 expressed the cautious opinion that “I have not been able to satisfy myself” whether or not the Gaspé limestones and calcareous shales re- pose on the older series. The faulting which can be demonstrated to have occurred in this area is of the horizontal offset type with very limited dis- placement. It is well illustrated about a half mile south of the Gaspé Limestone and Black shale contact south of Cape Rosier where a vertical dike extends some 60’ or more above high tide level. Two small horizontal faults have displaced the upper and lower portions of this dike in a northerly direction. The upper horizontal displacement is about 30 feet while the lower move- ment is just equal to the width of the dike,—about six feet. At the Quebec group and Devonian contact a fault plane nearly vertical trending at right angles to the shore line, is exposed just above high tide level cn the shore between the Devonian lime- stone and the Black shale (see pl. II, fig. 1). Here the horizon- tally grooved and slickensided limestone surface (left side) also indicates an offset fault of horizontal throw probably comparable with those at the dike. No evidence appears Ito exist for any 12 BULLETIN 82 46 great thrust fault. Frequent and abrupt changes in strike and dip make any close approximation of the total thickness uf the Quebec series im- possible in the vicinity of their eastern limit. ‘flic.e are however localities further west along the coast where the strike and dip are quite regular as at Cloridorme. The character of a part of these beds at Cape Rosier is indicated in the section below. Section at Rosier light house Dark to black hard fissile shale interbedded with bands of green shale and occasional thin bands of limestone. Inclusions of coarse limestone conglomerate from 1’ to 6’ thick or more oc- . cur in the midst of the black shale at intervals. 200 f. Hard limestone conglomerate of various sizea pebbles mostly 1” or less but many 3” or 7 LMOKE CTOSS fst he a eae st ee hae 4-5) e. Thin bedded limestone in regular strata 3” to yin ee ee ies Nk a as an eae en 6! de Darksandyarcen shale =altemnating |. sa aeunan 50’ Cran Gney limestones conelonlenate (=== seme Bei b. Grey, dark shale. Thin bedded limestone inter- bedded? 2: stoo a csiive tae ise okies eee eal eee 65/ a. Limestone conglomerate (to sea border low ticle) IVs le Aha eel iN nei Ace a eae eee 5 330 Logan estimated at Cape Rosier a thickness of “not much under goo feet” (p. 28). Other beds referred to by Logan near Cape Rosier (southeast) include strata comprising “red, purple, black and olive green shales” associated with light grey sandstone and thin layers of black bituminous limestone. It is proposed to use in this paper the name Cape Rosier beds for the rock series exposed at Cape Rosier and along the shore southeast of the lighthouse as far as the Devonian limestone cliffs. The name St. Lawrence Shale which Dr. Ruedemann (20) applied to these beds is not available because it was given to a limestone formation in Wisconsin by Winchell (39) in 1874. 4? DEVONIAN OF Gasph: KINDLE 13 A graptolite zone found north of the Cape Rosier lighthouse by C. H. Kindle which occurs in the section below yielded a faunule which has been studied by Dr. Rudolph Ruedemann. Section* in cliff face behind small sandy beach north of Cape Rosier lighthouse about 200 yards. Sandy slates, some black (top of cliff) 12’ Black shale LO.’ Dark shale band (graptolites) Be Limestone argill. fine textured Be Dark shales with limestone bands Ge Limestone (base at high tide level) Aa aN ON /\(Oj0i0> ee Ze! Dr. Ruedemann has furnished the following list of the grapto- lites found in this section: “The lot from Cape Rosier proved exceedingly interesting, especially from Locality C. H. K. No. 6, 150 yds. north of Cape Rosier lighthouse. This contained: Dictyonema approximatum sp. nov. D. pertextum sp. nov. Licnograptus elegans gen. nov., sp. nov. Dendrograptus fruticosus Hall Callograptus salteri (Hall) var. strictus nov. Tetragraptus similis (Hall), fragment. Leptobolus sp.” “This fauna indicates a Deepkill or Point Levis age for the fauna, probably lower Point Levis.” In the graptolite faunules submitted for study by Dr. Ruede- mann from four stations along the coast between the Cape Rosier light and Fox River, a distance of about 28 miles he recognizes two horizons, the older listed above, of Deepkill or Point Levis age and the younger of Normanskill age represented by the faunules listed below. ga. [East side of Griffin Cove Corynoides gracilis Lapw. Thamnograptus capillaris Hall Nemagraptus gracilis (Hall) * (By road 150 paces north of the road speed sign near lighthouse), 14 BULLETIN 82 48 N. gracilis var. exilis N. gracilis var, linearis Dicranogr. ramosus (Hall) Cryptogr. tricorms (Hall) Climacograptus modestus Rued. This is lower Normanskill gb. East of Gros Ruisseau 2 miles Didymogr. sagitticaulis Gurley Dicellogr. sextans var. exilis Lapw. Dicranogr. ramosus Hall Also Normanskill oc. East of Gros Ruisseau 14% miles Nemagr. gracilis (Hall) Dicranogr. gramosus (Hall) Dicranogr. furcatus (Hall) Diplogr. acutus Lapw. Climacogr. parvus (Hall) Lower Normanskill. gd. East of Gros Ruisseau 4 mile Corynoides gracilis Nemagr. gracilis Didymogr. sagitticaulis Dicellogr. ci. moffatensis D. sextans D. divaricatus Diplograptus euglyphus D. acutus Climacogr. bicornis C. modestus Also lower Normanskill in age Prof. Chas. Lapworth’s study (16) of the graptolites of this area lead him to recognize a Cape Rosier Zone, 7 Zone or Dictyonema sociale and Bryograptus’ of Calciferous age equiv- alent to the Tremadoc rocks of Great Britain and the Ceratopyge and Dictynema beds of Norway. The younger Canograptus Zone of Griffin Cove and Marsouin River he made the equiva- . lent of the Middle Llandeilo beds of Great Britain. Both Reude- mann and Lapworth thus agree in considering the graptolites of 49 DEVONIAN OF GASPE: KINDLE a5 Cape Rosier to represent a fauna older than those at Griffin Cove. Various views have been held concerning the correlation of the Normanskill graptolites (21). These appear to indicate that its faunas may have a range from Beekmantown to Trenton. The evidence now available indicates that the sediments comprising the Cape Rosier beds include horizons ranging from lowest to middle Ordovician age in the Cape Rosier district. The close and in some places complex folding of these shales and associ- ated limestones and sandy sediments must in different areas bring different horizons of the old rocks in contact with the Devonian sediments which rest on them. In the interior, Agnostus, sp. and representatives of one or more other Cambrian trilobite genera have been found by I. W. Jones in the Dartmouth River valley in these early sediments. Twenty miles southeast of Gaspé a middle or upper Cambrian formation was discovered and mapped a few years ago which holds a considerable Cambrian fauna. (C. H. Kindle, Fig. 8, Mem. 183, Geol. Surv. Can.) Basal Devonian Sediments.—The lowest of |the three Devon- ian formations which Dr. J. M. Clarke introduced to include the Devonian rocks below the Gaspé Sandstone ~he named the St. Albans beds, the type section being exposed in the cliffs of Mt. St. Albans (9) and near the northern end of the De- vonian cliffs south of the Ordovician rocks section about Cape Rosier. The basal 70’ of these beds, No. 1 of Logan’s section, he described as “grey limestones in layers of from six to eight inches thick which are separated by greenish calcareo-argill- aceous shale gradually increasing towards the upper part.” In the Fox River Road section the divisions b and a, below St. Albans beds appear to include beds lower than those known to Dr. Clarke in that formation. They are here given the name Griffon Cove River beds. ‘The section which follows is located about 13 miles northwest of the base of the Forillon Peninsula. It is exposed at the side of the Fox River—Gaspé highway and in the banks of the stream above the gorge cut in the Cape Rosier black shale and slates directly above the covered bridge. The other formation names previously used by Dr. Clarke are believed to approxim- 16 BULLETIN 82 50 ate the limits given them by him in the coast section: Fox River Road Section.—( Starting near top of south slope of Mountain ridge) Sta. 20 g¢ Grand Gréve Limestone (Top of section) Hard blue grey limestones decaying to soft PS Uirttaape@ C Kap etee tee ace, te ede ae ee 50+ 20 { Cape Bon Ami Beds Shaly calcareous beds largely covered, with MUMehOUSH tml Obite tall sete =e ? 20 e Blocky green to drab rather soft shale with Sonne IAG Or imeclchign Saas 60024 Fossils scarce. Jaonurus seen. 20 d Hard sandy shale with calcareous lenses, Sanalll Spanier alovwsaoeraye: 250 j= 20 c St. Albans Beds Coarse grey limestone in thin sheets inter- bedded below with sandy shale —_____. TO = Small brachiopods (Spirifer) and Favosites common. 20 b Griffon Cove River Beds Shales with two calcareous reefs of Stromatoporoids. Large ostracodes abund- ant in rather dark shale below lowest reef 150’ 20 a Sandy greenish shale with interpolated bands or limestone in) basal 20, =a AO’ Lowest limestone band with large ostracodes anol Comells, aorbhavdkeraye ee 10’ B Cape Rosier Beds A Black shale with occasional thin lenses Ol CcOnelomenrate ht 7.1 ell eke ee ee 300+ The strike of beds in A and B is nearly at right angles to that of beds in the younger series above, thus indicating 'the great unconformity separating the Cape Rosier beds from those which follow. The Griffon Cove River section is located about midway be- tween Fox River and the Forillon Peninsula. It is exposed along 51 DEVONIAN OF GASPE: KINDLE lz) Griffon Cove River about 414 miles from Griffon Bay and has considerable interest with reference to the lowest post-Ordovician faunules included in it. It includes in 35 pa crinoid genus ranging from West Virginia to Newfoundland previously un- known in Canadian sections. Like the Fox River Road section, it displays beds not exposed south of Cape Rosier at the contact of the St. Albans beds and the Ordovician shales of the Cape Rosier beds. Griffon Cove River Section—Vop of section 35 s_ begins some 400 yards below waterfall over limestones at road bridge. Griffon Cove River Beds Sta. 35 s Drab limestones softer than r and ascillaceausypatthy GCoyereds 2.2" 225 35 r Thin bedded magnesian limestone bands with black stems of plants (?) and calcareous bands with small brachiopods Forms short gorge 20’ wide ___---_. ie 20’ 35 q Limestone shale and covered —__ Oy 35 p Hard limestone bands with bulbous fossils (Scyphocrinus) ; 3 on surface of slab in place 2” in diameter with four DRO Sh eee en ee Se ee 2 o Red shale with bands of grey shale on = (plait Mica OmmemiSs Tita lALbGT )) ve se 60 35 n Conglomerate largely of quartz and SPE IG bOlle miBei DICS uy ie sata ee Es 15 35 m Thin bedded calcareous sandy shale with hard limestone and conglomerate [Sai eee a ee ee 2 35 1. Coarse conglomerate SiO? pebbles. 6” Str 6. 25° H.—D..28° S. W. 35 | Hard thin bedded magnesian lime- Stone Aes ie eee eee Be eaee ee eee Tribe. siopueodontimi, Caster, NOW ._..._..----.-----=-i-0cs-se<-sesonano--= LDIVeP TENA VCLES START To Cg RU Dae Re Oe ee ee ee ee (Geis), Tieelanqonrrl, tSlneere 1s(15) 9 Ye eee eee ree eee Genusm ohaleritares @as term ne wiser... 222) eke 2c ceaetecs=ct (Gresaubes) WY cclkeeuiniabiey, ORR a TAYE ee ee Re a ivew Do unvalimon ders tO jCOd ONGICS: ete ees 8 aoe ee reese Genus Protomegas.ropaia Caster, new ...._......-....---2css--ee002-=-+= Genus Merastrophiam Caster mew. .2-.2 22 .occ coe eee ceeenceesescce ee Gentism Cymosinop iam Caster, sme Wace 222 cecccedseensssesceceasenneenne GennseDichyostophian Gasten, Me wie. 99. eesces-ckrssceeensena ae The Douvillinoid Stropheodonts of Colombia Genus Megastrophia Caster, new (Species M. hopkinsi bo Sova op) GS O1 H C2 So DO oon Rmowww wut w Ww c= tS BULLETIN 83 Caster, new, p. 42; M. pygmea Caster, new, p. 45) -.......... Genus Cymostrophia Caster, new (Species C. schucherti Caster, new, p. 48; ?C. waringi Caster, new, p. 54; C. Ghielk@yil (CASNGIR, IEW 1D. BO) ccecestessseccecenecsancBao-baresnsoeOEncEaese Genus Dietyostrophia Caster, new (Species D. cooperi Caster, SANE 7) aR a ei a a a rece eee ee MMe Sere OMS GO po l1e O.Cl OTIC pee ee em Sn Genus Stropheodonta Hall, 1852, s.s. (Species 8. kozlowskii OhRier, Mew, Os OOS Sh, Sis 10s WO) ccteeeceec re eeetedentcccesece — ‘qnherijoyey’ IDonuhvaillinbawirmal: Ces TN a ceca ectboneseeccoascessose Malle ale £0str oy atime Cals tere 1 © Wi gestae Pee eae neonate Primitive Leptostrophids Mires ae erp lan aie SLOW pee ees elec een ey aoe Sec OLAS ee cere Genus Protoleptostrophia Caster, new The Nervose Leptostrophids Genus Nervostrophia Caster, new INH G yea all Caibe SINC TaViO See ees ee een ees oe Cec ok ne eee ee Genus Sulcattositn opp lata Caister pe 1'e vy eee eee en HE} cleats wll © 11's 70,6 sitar; spo Cl See etre ee en Genus Australostrophia Caster, new BIVe © Ogee be Cle ase osts 01s ti Oso Gl Syren te ee Genus Rhytistrophia Caster, new (Species R. caribbeana Caster; variety colombia Caster, new, p. 87) ST He PTET 69'S 15 © 0 HT CLS pe ane ete oh vee Genus Leptostrophia Hall and Clarke, s.s. __.....----e-ce-00- Subtamaly.. Siroph'onelllnmneesys Caister. srle wees sess eeeseeeneee eee Genus Amphistrophia Hall and Clarke, 1892, s.s. (inelud- ne StarOooln@yomanom “Ayreraaoneel, MON ee) oc ercecece ars ceesnnccen Genus Pholidostrophia Hall and Clarke, 1892 Genus Strophonella Hall, 1879, s.s. Genus Strophonelloides Caster, new Ghermnis Olnveromumapemie, | (Chasen, TEN a os cco acsarnceme ence Genus Strophonella Hall, 1879 (Species S. meridionalis Caster, new, p. 107; S. floweri Caster, new, p. 109) HamulysOxnthotetideemWaacenymll'S3 45cm ex Cy ae eessess eee ena Genus Schellwienella Thomas, 1910 (Species 8. goldringe Caster, new, p. 116; variety juvens Caster, new, p. 119) Some ‘‘ Austral’’ representatives of Schellwienella __..... Family Productide Gray, 1840 (Subfamily Productelline Stolmmuelaerey Ewayel IL erae WG) a nacegereee nt tee onee seeereSsosee Genus Productella Hall, 1867 (Species P. ef. spinulicosta 3 eH eat oyataee 22) tema ie ee spree ee See LS eee ud ese Mamualya'Chonetides call andl Clarke ws 9:2 geiasteeesee nets eniee eens Genus Eodevonaria Breger, 1906 (Species EH. imperialis 104 105 CoLOMBIAN DEVONIAN FAUNA: CASTER Caster, new, p. 122; variety parva Caster, new, p. 126; variety transversa Caster, new, p. 128; species E. reedi (CHISKiaIES TIEN) JO IPAS), ee a ee eee Genus Chonetes Fischer de Waldheim, 1830 (Species ?C. ef. stiibeli Ulrich, p. 131; C. aff. billingsi Clarke, gens, p. BETO iy Cee rn REN Se te Ne ie Se Se cane a ces Comparisonot 5" Austral’? Chonetids 2.202222. ah eeccents se Genus Chonostrophia Hall and Clarke, 1892 (Species C. HOGI AS LC Tops C.Wa) meres e-wh ce Menuen Ec ee Sven eh Ele a tS. Oxrdens helotrenmatcmelbseech Gry eon ere ee ee ee ete Superfamily Atrypacea Schuchert and LeVene (Family Atrypi- de Gill, 1871, Subfamily Atrypine Waagen, 1883) _.. Genus Atrypa Dalman, 1828 (Species A. harrisi Caster, new, p. 140; variety nasuta Caster, new, p. 142) Family Celospiride Hall and Clarke, 1895 Genus Vatulina, call tsG0) (Species Vij sp.) Genus Anoplotheea Sandberger, 1855 (Species A. ?silvetii (TRALEE Rs Coe tein eI Scent mk ee Se rr Superfamily Spiriferacea Waagen, 1883 (Family Spiriferide King, 1846, Subfamily ?Reticulariine Waagen, 1883) __.. Subfamily Phricodothyrine Caster, neW _._.......222-.:-c----.------5-+--- Genus Elytha Fredericks, 1918 (Species E. colombiana Cas- {ULM DAVEN 75 Py OE ec LAU CD) rg ee Oe OE eee nee Subfamily Spiriferine Schuchert, 1913 (The ‘‘Spirifer divari- OLMIS se EAR GIN Cmegl O)e 01D) wen aces sears a Meee a Genus ‘‘Spirifer’’ (Species ‘‘S.’’ kingi Caster, new) ........ Genus Acrospirifer Helmbrecht and Wedekind, 1923 (Species AS Olsson @astonien eye spelio Gites tn oad. ---- e ee Genus Australospirifer Caster, new (Species A. ef. antare- ticus variety 1 Caster, p. 162; variety 2 Caster, p. 163) Genus Brachyspirifer Wedekind, 1926 (Species B. palmere (GYD ATED) | skeet Sis RR pe eee Genus Paraspirifer Wedekind, 1926 (Species P., sp.) -....... Genus Spinocyrtia Fredericks, 1916 (Species ?S. cf. valen- Tastee (CUR Ier ete) A pose ELSI N) oo ne es ae a nce Superfamily Rostrospiracea Schuchert and LeVene (Family Meristellide Hall and Clarke, 1892, Subfamily Meris- qureulllminngees Ww eneneaeraly TINS })\" Saale Se ee ee eer sere Genus Meristella Hall, 1860 (Species M. wheeleri Caster, OVEN a eS ee Genus Pentagonia Cozzens, 1846 (Species P. gemmisuleata BSG REL TENET Pel Dioe 112) | ei oe ee ee Superfamily Terebratulacea Waagen, 1883 (Family Megan- (Svea iy: NV Neer ILSKS 127) ie ea Sa a eee Genus Meganteris Suess, 1855 (Species M. australis Caster, AEN)» hoe Starch bate nc te Meters Ee eae a or 6 BULLETIN 835 ACE Tbae SE GSN) Oe oP RRS LAN Oy oR aR oem Nr eres 2 @reuTTy UE Ofb Oe CHIT aa SO ater tek ae oe ec ee ll cee ec Nn ee AO ae\ oneconaye) UE) iiss 0)5 eh iene a none eka py) ee Meer aee Se ae AMUR ne ae eres PG TID Vel BAIS PO Siete oes oe eee Cae RC Lr as oe RS tine Se te AN JO FOC YO GLE SS oe TE AN Se 5 EE ee en cen ne Se Pele cy dave sixes eo ers ee ee ek) a aie Cypricardimiapets suloin den tal Wiers\o Or cleeessesseeswens = smneneaee PAC aIIG UL @TO.E CLETUS [seem © cl: 1a mee nt en VATA OCC UEIM, Ds TB) COIS WEE aaa ccc secece ce ceesenecnceeneese= et oP terimecare Caster yur Ss ;ccsc. sees ose eee ee ene (ESET S(O) ONO YO ez anaes 00 WARES EA gee ea eS i ee 1 BA OP AOYs Ny ae ance aie wp SL EOE V8 OE eo a aN a ee y 5X AU] OVO YOY? cree RR ed NSP Fs aie I a A ee ee ane A eae) ee Da. nat CYS rsate NEKO KGL path ah a el Oc ate NUS ea cee a eal ae NaN SNe i) Rea 0) ON ez Harrap nen alten NON a cP eee TARE Ce ce Ory Re ce ool TEANBKOOIOE) Cit, Senlhwere, ICO vA ST ee hates secon ueee ?Dalmanites ef. patacamayaénsis Kozlowski _................. 2 Civ ASUS HAS ay eee kas WE AE EE 0 eee su PET OMLATONODUS SOs Vest eee ae Oren rie Gus eee ORS alee oes ee Notes on two new papers describing Devonian faunas of the South- erm, "EVemisphere: 2 use ae oe aI ae ee TRS ik Li ora ao Tyee A eS Ie ee ees Sa Plates 106 A DEVONIAN FAUNA FROM COLOMBIA* By KENNETH E. CASTER University of Cincinnati jdsysd eo VCdl The deseription of the more striking elements of the first Devonian fauna found in the State of Colombia, South America, is the objective of this paper. Olsson and Caster briefly discuss the stratigraphic relationships of the Colombian Devonian in the first part of the paper. Caster, in the second part of the paper, treats the systematic paleontology and discusses the probable age of the fauna. The assemblage is evaluated as essentially the equivalent of North American early Onondagan faunas, with some Oriskan- ian derivatives of unusual aspect. The fauna proves to have a more striking ‘‘boreal’’ than ‘‘austral’’ flavor, and opens the problem of the isolation of South America during the early Devonian. Forty-nine species and varieties, chiefly brachiopods, assigned to thirty-seven genera, are recognized and illustrated. Of these, twenty-three species and five var- ieties are described as new to science. Five new genera (plus nine new genera not represented in the Colombian faunule), one new family, three new subfamilies and three new tribes are proposed in the Brachiopoda. There are several new generic records for South America. A partial re- classification of the crenulate-hinged Strophomenacea, principally as known in the Western Hemisphere, is outlined. ACKNOWLEDGMENTS To Mr. Axel A. Olsson and Dr. Parke A. Dickey goes credit for the original discovery of the Colombian Devonian strata. Their original collection, and so far as known the only collection yet made, is the basis for this study. The kind cooperation of these gentlemen and of the International Petroleum Company by whom they were employed has made the present investigation possible. The results are now presented for publication with the permis- sion of Dr. O. B. Hopkins, Director, and Mr. O. C. Wheeler, Chief Geologist of that company. I am especially grateful to Mr. Olsson for turning this material over to me for study. Mr. Olsson has generously waived his rights of co-authorship through dis- covery and no little preliminary preparation and study of the fauna. His interest and assistance during the course of the work have been of great benefit. * Published with the permission of Dr. O. B. Hopkins, Director, and Mr. O. C. Wheeler, Chief Geologist, of the International Petroleum Com- pany, Toronto. 8 BULLETIN 83 108 I am fortunate in having had the opportunity to work and examine collections during the course of the investigation at the University of Cincinnati Museum, the Paleontological Research Institution in Ithaca, New York, the New York State Museum in Albany, and the United States National Museum. The facili- ties and collections in Ithaca were made available through the courtesy of Dr. K. V. W. Palmer and Prof, G. D2 Marrs im Albany through Dr. Chas. C. Adams, Director of the State Mu- seum, and Dr. Winifred Goldring, Assistant State Paleontologist ; and in Washington through Dr. R. S. Bassler, Head Curator of Geology, and Dr. G. A. Cooper, Assistant Curator of Paleont- ology. Dr. Cooper has been exceedingly generous in his co- Operation during the study, and in helpful criticism of the manu- script. The opinions and conclusions expressed in the paper are, however, entirely the responsibility of the writer. Publications not available in Cincinnati were generously loaned or contributed by the Paleontological Research Institution, Dr. Rudolph Ruedemann, Dr. Thore Halle of Sweden, Dr. Robin S. Allan of New Zealand, Dr. Ramon Koztowski of Poland, Dr. Pierre Pruvost of France, Drs. Etienne Asselberghs and Eugéne Maillieux of Belgium, and Dr. R. Mendez-Alzola of Uruguay. Photostats of several inaccessible papers were made through the courtesy of Dr. Pearl G. Sheldon of Ithaca, New York. I am indebted to Professor G. D. Harris and the Board of Trustees of the Paleontological Research Institution for facilities of publication, The defrayal of the cost of the collotype plates has been met by the Faber Publication Fund for Paleontology at the University of Cincinnati Museum. Much of the photography for the paper has been creditably done by Mr. Stewart Jones, stu- dent in geology at the University of Cincinnati. The constant ccoperation of Anneliese S. Caster has facilitated the preparation of the manuscript and the seeing of the paper through the press. All of the Colombian faunal material on which this study is based is on deposit at the Paleontological Research Institution. KENNETH E. CASTER University of Cincinnati November 5, 1938 109 COLOMBIAN DEVONIAN FAUNA: CASTER cy) PAR | GENERAL CONSIDERATIONS By AxEL A. OLSSON AND KENNETH FE. CASTER OCCURRENCE During the winter of 1935, while engaged in a geological recon- naissance of the Cordillera Oriental, or Eastern Andean Range of northeastern Colombia, Axel A. Olsson and Paul Dickey dis- covered the Devonian faunule described in this report. A _ pre- liminary notice of the occurrence and content of the faunule was presented by Olsson and Caster, 1936 (1937), before the Paleont- ological Society. The fossils here described were obtained at the north side of the small village of Floresta, long. 72°53’ W., lat. 5°51’ N., on the automobile road between Santa Rosa and Cor- rales in the western part of the Departmento de Boyaca, Colombia. Geographically this is a little east of north from the town of Sogamosa, the terminus of the railroad from Bogota. It has long been suspected that Paleozoic rocks occur in the Eastern Andes but the discovery of fossils to prove the surmise is of relatively recent occurrence. Marine Carboniferous strata are now known to be widely distributed in Colombia, especially in the Gacheta and Villavicencio districts to the east and southeast of Bogota, although as yet they have not been critically studied. Some fossils believed to belong to the Lower Devonian were col- lected in Quebrada Honda, north of Villavicencio, from loose boulders by R: Scheitbe, 1917 (1933) (described by W. E. Schmidt, 1933), and by R. E. King (according to Schuchert, 1935, p. 673) from a nearby locality. The discovery at Floresta by Olsson and Dickey is the first record, so far as we are aware, of the finding im situ of Devonian strata in the State of Colombia, although Devonian rocks are well known in the Sierra de Perija of western Venezuela (Weisbord, 1926; Liddle, 1928; etc.). The Eastern Andean Range has long been known to be com- posed principally of Cretaceous strata which rest on an older series called the Giron or “Old Red” series. It had been hitherto supposed that the Giron in this region rests directly upon a meta- morphic and igneous complex of gneisses, mica schists and diorite 10 BULLETIN 83 110 granites. In western Boyaca a wedge of Devonian strata ap- parently intervenes. These strata, for which the name Floresta is proposed below, consist of rather soft, yellowish to cream colored shales with some thoroughly indurated layers. The buff shales may originally have contained considerable calcium caibonate, for they have the appearance of being a residual deposit from a highly calcareous argillite. The fossils are preserved solely as external or internal molds which show structural details in fine perfection. The present study has been carried out principally by the use of plasticine casts. The name Floresta series is intro- duced for thie Devonian sequence between the Giron and the igneous-n.etamorphic complex. It seems clear that the [I'loresta sequence underlies the Giron, but opportunity for detailed field study of the relationships has not yet been available. For a short Cistance toward Santa Rosa from the Floresta occurrence expcsures are absent, then follow barren yellowish shales which are directly overlain by the harder, slaty red beds of the Giron series. In the opposite direction, toward Corrales, the first ex- posures met with are deeply weathered crystallines. These rela- tionships seem clearly to show that the Floresta series lies be- tween the typical red beds of the Giron and the basement complex, but whether they are transitional with the “Old Reds” or uncon- formably underlie them has not been established. The Cachira series of western Venezuela, usually considered as lower Middle Devonian in age, is extensively developed in the Sierra de Perija. According to Liddle, 1928, the Cachira is comprised of a thick sequence of sandstones, shales and lime- stones, and apparently occupies a similar stratigraphic position in relation to the “Old Red” and the crystallines as does the Flor- esta series. It seems quite clear from the faunal comparisons that the Cachira and Floresta series are in part at least contem- poraneous, although they have only a few species in common. The Cachira fauna and stratigraphy are very imperfectly known as yet, however, and certainly there is a striking facies difference in the fossiliferous horizons of the two series. The Venezuelan material described by Weisbord came principally from a massive limestone replete with corals. Molluscs, brachiopods and bryo- zoans are also common in his fauna, whereas in Colombia, 111 CoLOMBIAN DEVONIAN FAUNA: CASTER zal brachiopods and bryozoans are exceedingly abundant, but corals are extremely rare, and molluscs nearly equally sparse. The two faunas seem to be more alike in trilobite content. This aspect the two faunas appear also to share equally with the Bolivian Devonian, described by Koztowski, 1923. Further study of larger collections may show this similarity to be a preliminary illusion. The abundant bryozoan fauna of the Cachira and Floresta series is an unusual feature for South American Devonian strata. This phase of the Colombian fauna is now being studied by Dr, A. H. McNair of Dartmouth College. PART IT THE FAUNA By KENNETH E. CASTER AGE AND RELATIONSHIPS The Floresta fauna described in this paper is only a small por- tion of what seems clearly will be very extensive biota when more careful and detailed collecting is possible. The small amount of material on hand, however, shows so many interest- ing features that it has seemed worthy of description despite its inadequacy for a complete picture. Several years have already elapsed since the original collection was recovered, and at least three unsuccessful attempts have been made to secure additional material by colleagues in.Colombia primarily on other assign- ments. The reader will do well to bear in mind the reconnais- sance nature of the present paleontologic report and withhold final evaluation of the discovery until a more complete report is pos- sible. In the following list of species occurring in the Floresta series of Colombia, the symbols in the adjacent columns refer to the key, below. The species bearing an asterisk are new generic stocks in South America. I APPARENT PLACE IN THE NORTH AMERICAN DEVONIAN SYSTEM, BASED ON SFECIES AFFINITIES 1. Helderbergian series: North American Lower Devonian. 2. Oriskanian series: upper Lower Devonian*. (Grande Greve * North American section, after Clarke and Schuchert, 1899. 12 BULLETIN 83 112 facies of principal importance). 3. Ulsterian series (especially lower Onondagan) : lower Middle Devonian. 4. Erian (Hamilton) series: upper Middle Devonian. In the faunal list, column I, when referring to the key above, the italicized numbers indicate closest resemblances. II REGIONAL DISTRIBUTION OF CONGENERS IN SOUTH AMERICA AND ENVIRONS 1. Amazonas region of Brazil (Hartt, Rathbun, Clarke, Katzer, etc.). 2. Bolivia (d’Orbigny, Ulrich, Knod, Koztowski, etc.). 3. Argentina (Clarke, Thomas, etc.). 4. Uruguay (Mendez-Alzola, etc.). 5. Venezuela (Weisbord). 6. Falklands (Morris and Sharpe, Clarke, etc.). III PRINCIPAL DISTRIBUTION OF THE SPECIES GROUPS A. “Austral” (Distribution principally south of the Caribbean and Mediterranean). B. “Boreal” (Distribution principally North America and Eu- rope). C. “Cosmopolitan” (Known at least from most regions having Devonian faunas of comparable age). THE FLORESTA FAUNA JUL UE Pholidops florestee Caster Dre STAG B Lepteena boyaca Caster Renee Ce Megastrophia hopkinsi Caster* 34 2 B M. pygmza Caster* aes i B Cymostrophia schucherti Caster* I-3 ? B C. waringi Caster* I-3 ? B C. dickeyi Caster* Toons B Dictyostrophia cooperi Caster* 34? B Stropheodonta koztowskii Caster 3-4 2°? Ce 1B) ?Stropheodonta, sp. ra Tag ? Rhytistrophia caribbeana, var. colombia Gastery yy T=si5 AGE Strophonella meridionalis Caster* 2 P B 1B} COLOMBIAN DEVONIAN FAUNA: CASTER Strophonella floweri Caster* Schellwienella goldringee Caster S. goldringe, var. juvens Caster Productella cf. spinulicosta Hall Eodevonaria imperialis Caster FE. imperialis, var. parva Caster [Ié. imperialis, var. transversa Caster ?Chonetes cf. sttbeli Ulrich ?Chonetes cf. billingsi Clarke, gens. Chonostrophia knodi Caster* Atrypa harrisi Caster A. harrisi, var. nasuta Caster Meganteris australis Caster* Meristella wheeleri Caster Pentagonia gemmisulcata Caster* Elytha colombiana Caster* “Spirifer” kingi Caster* Acrospirifer olssoni Caster Australospirifer cf. antarcticus, var. Pere antarcticus, vat. 2) Brachyspirifer palmere Caster Paraspirifer, sp. ?Spinocyrtia cf. valenteana Hartt Vitulina, sp. Anoplotheca cf. silvetii (Ulrich) ?Camarotoechia, sp. ?Cryptonella, sp. ?Derbyiana, sp. Cypricardinia cf. subindenta Weisbord ?Aviculopecten, sp. A. PAviculopectcn, sp. LP. ePterinea, 0.,.sp. ?Fenestella venezuelensis Weisbord Phacops cf. salteri Koztowski ?Dalmanites cf. patacamayaénsis Kozlowski ?Cyphaspis, sp. *’Homalonotus, sp. bd td U ech esl louie a @ nie) leo ql @l les! ASBC? rrannwt ee) 14 BULLETIN 83 114 DISCUSSION The tabulation brings out the predominently “boreal” flavor of the Colombian fauna, rather than an “‘austral” aspect which one might expect in view of virtually all previously described South American faunas. The Venezuelan fauna of Weisbord, 1926, is the unique exception among described faunas, unless the Ama- zonas fauna of Hartt and Rathbun, 1874, proves similarly allied. Their identifications and comparisons strongly suggest this. The faunas of Brazil, described by Clarke, 1913, seem to be more closely related to what we ordinarily think of as the typical “aus- tral” faunas of Uruguay, the Argentine, and the Falkland Islands. The Bolivian faunas, which are perhaps best known of all in South America, lend a curious suggestion of antipodal admixture, the “boreal” aspect of which seems to increase northward. In northern South America the Devonian faunas thus far reported are almost wholly of a “boreal” stamp. Without discussing in detail the relative age relations of the Floresta Devonian in terms of the North American or European column, it appears probable that in Colombia and Venezuela we have a southward continua- tion of the early Middle Devonian faunas of the Appalachian geosyncline. At any rate, there seems to be good evidence of seaway communication between the continents when the Venezue- lan and Colombian strata were formed. That this seaway ex- tended across the Llanos region of today and around the Guiana shield into the northern and northwestern part at least of the Amazonas basin is also likely. From the strong admixture of southern and northern elements in Venezuela and Colombia, and decreasing importance of “boreal” species and genera southward, it would appear that open seaways extended between most, if not all, the known Devonian areas in South America, and that dis- tance principally acted as a barrier between the two foci of faunal distribution. These two foci apparently did not exist in Ordovi- cian times, when truly cosmopolitan faunas existed in the southern and central Andean region of South America, as brought out by the cosmopolitan graptolite faunas of the Andes. In Silurian times what may have been happening in the “austral” region of the Western Hemisphere is practically unknown. The so-called 115 CoLOMBIAN DEVONIAN FAUNA: CASTER 15 Silurian beds in Argentina described by Clarke are possibly De- vonian, as far as can be judged from the fauna described and illustrated. This is also the verbal opinion of colleagues who have examined the fauna in the field. The Argentine fauna may be semewhat older than those of Brazil, Bolivia and Colombia, but ret more than an epoch of the Devonian seems at most to inter- vene. It is the “austral” purity of the Argentine fauna that has made correlation difficult. The faunal statistics would suggest that near the beginning of Oriskanian time in the North, in the Western Hemisphere at least, occurred an extremely important diastrophic event, which opened the western seaways from the Gaspé Peninsula in the north to the South Atlantic for comming- ling of previously more or less separated faunas. The maximum spread of the seaways in the South presumably came in Onon- dagan time (as did it not also in the North’), when most “boreal’’ forms reached the southern continent. The northern fauna seems to have been a dominant one, however, for to my knowledge, no typically southern genera or species reached the North American continent, whereas Venezuela and Colombia were virtual “boreal’”’ outposts. If further study brings substantiation to these paleo- geographic speculations, it is not impossible that the present highly unsatisfactory intersystemic boundary between the Silurian and Devonian must be re-examined in the interest of trying to reconcile a major diastrophic break of systemic importance to the insignificant place of separation it now holds between present divisions of the extended Lower Devonian (Clarke and Schuchert 1899). From the tabulation it is seen that the present interpretation of the Colombian (and Venezuelan) fauna would align it essentially with the early Onondagan of the North. Scheibe, 1917-1933, Schmidt, 1933, and Schuchert, 1935, have thought of the fragments of Devonian rocks hitherto found in Colombia as Lower Devonian in age. The determination was based on the fragmental crinoid fauna described by Schmidt, and was recognized as only a tem- porary age assignment. The present fauna is preeminently early Onondagan in aspect, but also has strong similarities to that of the limestone facies of the Oriskanian so well developed in the Gaspé Peninsula of Quebec. It may be that in the south there 16 BULLETIN 83 116 yvas a longer carry-over of Oriskanian elements than in the north. There are, however, no elements in the present fauna that are especially reminiscent of the Schoharie fauna of the north. That there may be a very close affinity between the northern Andean Devonian faunas and those of the western United States is sug- gested by the few published reports, and even more so by unde- scribed material from the west which I have had the opportunity to examine. A somewhat similar carry-over seems to have oc- curred in the west also. As yet, however, in the absence of de- tailed modern studies (published) not much more satisfactory correlation can be ventured between the presumable Oriskanian- Onondagan strata of the western United States and those of the Appalachian region, than between the Colombian strata and the typical Oriskanian-Onondagan. SAS IMOAIEMUC, IPANL BON WOILOG NY BRACHIOPODA Subclass GASTROCAULIA Thomson ? Order NEOTREMATA Beecher ? Superfamily CRANIACEA Waagen ? Family CRANIIDZE Gray, 1840 Genus PHOLIDOPS Hall, 1859, or LINGULAPHOLIS Schuchert, 1913 Pholidops (or Lingulaphslis) filorestz Caster, n. sp. Plate 1, figs. 1-3 Shell large for either Pholidops or Lingulapholis. Outline of both valves subcircular ; apex submarginal and only very slightly produced posteriorly. Length and maximum width subequal ; greatest width is a little anterior of the middle. Ventral valve of low convexity; dorsal valve planate or slightly concave centrally. Both valves show a relatively wide peripheral area of contact which in one specimen (5454) 1s developed into a rounded convex peripheral flange in the dorsal valve, (see fig. 3). Ventral valve shows a weak subventral cementation scar (figs. 1,2). Growth was holoperipheral in early ontogenetic stages, but became hemiperi- pheral (or possibly mixoperipheral) before half grown. Both 117 COLOMBIAN DEVONIAN FauNA: CASTER 17 valves show a tendency for a narrow posterior plate or “false area” to develop between the valves as in Lingulapholis. Varices relatively prominent, imbricating and becoming fascicularly lamel- late in the Iter half of the shells. Postlateral slopes show faint sug- gestions of radi. Musculature of ventral valves not known. Muscle attachn-erts of the dorsal valve are very prominent; located sub- centrally cn either a callus platform, or on a deep central invagin- aticn of the corsal shell. The appearance of the mold suggests a subcircular platform on which the bilobate anterior adductor im- pressiorvs are most prominent. The posterior adductor impres- sions are extended outside the central prominent zone as faint postlateral lobate scars. The median impression (diductor scar?) between the anterior adductor lobes is very distinct. Dimenstons.—Lenegth of types is 7 mm. and maximum width 6.5 mm, Discussion.—Schuchert, 1913 (p. 296), has questioned the po- sition in the Brachiopoda of the genus Pholidops, and has pointed out that the genus (and Lingulapholis as well) has fully as many atremate as neotremate characteristics. The inferred relationship to the Craniacea and Crantide, largely hypothetical as yet, is postulated on general similarity in form. These matters need net especially concern us at this time. So far as known, this is the first clear-cut record of the genus Pholidops (or Lingulapholis) in the Southern Hemisphere Devon- ian. Thomas, 1905 (p. 258), lists a problematical fossil from the Argentine Devonian as Pholidops?, but did not describe or illustrate it, and I have no way of knowing the bearing of this record on our species. The affinities of this fossil seem to lie with forms described from the Lower and lower Middle Devonian in North America. The generic assignment would be definitely Pholidops Hall, s. s. were there not suggestions of a pseudoarea such as Schuchert, 1913 (p. 206, pl. 53, figs. 14-19), described for Lingulapholis terminalis (Hall) from the Oriskanian of Mary- land. The Colombian specimens do not recall the genotype, how- ever, neatly so much as they recall P. areolata Hall, 1867 (p. 31, pl. 3, figs. 4, 5; Hall and Clarke, 1892, pl. 4i, figs. 25, 26), from the Schoharie grit of New York (which Dunbar, 1919, p. 86, 18 BULLETIN 83 118 lists also from the Camden Chert) especially in the shape and proportions of tne muscular platform. If, however, I am correct in judging the tiny areas shown on plate 1, figs. 1, 2, as cementa- tion sites, then it is possible that the similarity 1s more apparent than real, for the comparable markings in Hall’s specimens are assigned to the ventral valve. Hall, 1867, and Hall and Clarke, 1892 (p. 155), were under the impression that this genus is un- attached, and always expressed doubt as to the identification of the two valves. Schucheit, 1913 (p. 295), pointed out that Pholidops was cemented, and apparently did not have a functional pedicle. It appears that the Colombian specimens were attached in the Pholidops manner, but very meagerly. They seemingly also had a functional pedicle, if a faint pseudoarea, as in the free genus Lingulapholis, is an acceptable criterion for judging the presence of a pedicle. The well developed peripheral flange of the dorsal valve in the Cclombian species easily sets them apart from North American forms. P. arenaiia Hall, 1867 (p. 413, pl. 3, fig. 10; Hall and Clarke, 1892, pl. 4i, fig. 24), of the Oriskanian is smaller and with less clearly differentiated muscular areas, and P. tunuda Schuchert, 1913, from the Maryland Oriskanian (Shriver Chert) is far too linguloid to warrant detailed comparison. Lingulapholis termin- alis (Hall) differs as outlined above. P. bellula Walcott, 1884 (p. 113, pl. 2, fig. 6), from the Lower Devonian of the. Eureka District is much smaller and has a distinct U-shaped muscle seat. His ?P. quadrangularis is so doubtfully congeneric with our forms that comparison is unnecessary. /. ovata Hall, 1859 (p. Ago, pl. 103B, fig. 7; Hall and Clarke, 1892, p. 157, pl. 41, figs. 22, 23), from the lower Helderberg (Coeymans) has a more quad- rangular muscle platform, and more prominent and extensiform posterior adductor lobes. The form from the Grande Greve limestone ccmpared with P. ovata by Clarke, 1908 (p. 212, pl. 47, figs. 10, 11), recalls the South American material somewhat more than the topotype material in the apparently total absence of postlateral adductor impressions. The form referred to P. ovata by Weller, 1903 (p. 226, pl. 20, figs. 27-29), shows a periph- eral flange, but is a much smaller fossil. The P. ovata of Schu- 119 COLOMBIAN DEVONIAN FAUNA: CASTER 19 chert, 1913, from the Helderberg of Maryland has far too lobate muscle impressions to be confused either with Hall’s original material or the South American material. It is not impossible that when better material has been found of the fossil referred to Lingula, sp. by Reed, 1903 (p. 167, pl. 20, fig. 3), from the Bokkeveld beds of South Africa that it will be assigned to Pholidops rather than to Glossina (Reed, 1925, p. 35). The general contour only suggests this. Types.—Holotype: Pal. Res. Inst. No. 5454; paratype: No. 5454A. This material is on deposit at the Paleontological Re- search Institution, Ithaca, New York, and likewise all the rest of the material on which this study is based. Order PROTREMATA Beecher Suborder STROPHOMENOIDEA Oepik Superfamily STROPHOMENACEA Schuchert Family RAFINESQUINIDZ Caster, new With the elevation of the strophomenoids to subordinal rank, the long-awaited reorganization of this important branch of the Brachiopoda seems to be at hand. At this place only those phases of the superfamily are considered which the Colombian fauna may help illustrate. It is here proposed that the group formerly known as the subfamily Rafinesquinine Schuchert, 1893, be elevated to family rank.. The new family Rafinesquinide ap- parently contains several subfamilies, one of which will be re- stricted Rafinesquininz, centering around the genus Rafinesquina. It is highly probable that the genus Leptena should be considered the nucleus of another subfamily. The present study will not watrant any proposals in this connection, however. Genus LEPTENA Dalman, 1828 Genoleetotype—Productus rugosa WUisinger. Ordovician. Leptzna boyaca Caster, n. sp. Plate 1, figs. 4-13; Plate 10, figs. 1-3 Large, concavo-convex strophomenoid having the general characteristics of typical representatives of “Leptena rhomboi 20 BULLETIN 83 120 dalis,’ but showing what appear to be specific differences: length to hinge width usually as 2 to 3; hinge line usually auriculate; both valves in adulthood prominently geniculate, convexly in the ventral and concavely in the dorsal. Postgeniculate surface, 7. e., “visceral discs,” of both valves slightly convex and prominently rugose, with usually 14 or 15 rugze in adult shells. These corru- gations are irregular and asymmetrical, the front slope of each being gentler than the rear. The amplitude of the corrugations increases regularly toward the front. The geniculate condition apparently appears only in adulthood. Height of geniculate “trail” measured on the ventral valve is usually about half of the leneth of the rugose surface of the ventral valve, and slightly less than half the length of the corresponding area when measured on the dorsal one. Surface radially multistriate ; strize relatively straight and regular on planate surfaces, but wavering and crowded on geniculate area. The striz are fine and thread-like, and are separated by rounded interspaces about twice as wide as the diameter of a single stria. The radii increases by intercalation (appearing as bifurcation when viewed in the external molds) and usually originate on the crest of the corrugations on the posterior surfaces of the shell, but arise very irregularly on the up-bent zone. Internally, the “trail” is partially dissociated from the body cavity of the shell by a low sub-angular callus septum, or “dia- phragm” in the ventral valve immediately anterior of the last corrugation. (See Dunbar and Condra, 1932, p. 178, for nomen- clature of analogous dorsal structure in the Productide.) No evidence of such a “diaphragm” has been seen in the dorsal valve, but may be present, in as much as most “Leptena rhomboidalis” specimens show such a structure. The external characteristics of the cardinal areas are not known; some intimations of these are, however, shown by inter- nal molds. The interiors of adult shells show evidence of con- siderable secondary calcification. The surface corrugations are only obscurely represented on internal molds, and the entire inner surface, especially of the dorsal valve, is papillose or granular. The papilla are drop-like with the tapering end directed toward 121 COLOMBIAN DEVONIAN Fauna: CASTER 21 the beak, and show a general longitudinal orientation. In both valves the internal papilla are most concentrated and less elongate in the immediate region of the muscle platforms. Toward the front of the shell and in the auricular zones they tend to be aligned so that they somewhat resemble the surface pustules of Reticularia. The internal characteristics are shown on plates 1 and to. The ventral muscle platform is well developed, and sharply outlined by an elevated peripheral extension of the dental lamellz. The only interruption of this lamellar wall is in the front and middle. The platform is usually transverse, as in 5469 (plate 1, fig. 5), but may be subcircular, as in 5468 (plate 1, fig. 4). It is chiefly occupied by the flabellate diductor scars which show radial stri- ations. Mesially the diductor scars are separated but not enclosed by an elevated anteriorly expanding ridge to which the adductors were attached. The postlateral borders of the scar are delimited by narrow elevated carinze which become evanescent at about the anterior third where the whole adductor scar is slightly expanded. At the place of maximum expansion the adductor scars are less elevated than in the middle. For virtually its entire length the adductor ridge bears a narrow center carina or knife-edge which separates the right and left scars. The median ridge stops abruptly at the anterior edge of the muscle platform, but is not delimited by the wall of dental lamellz as is the diductor area on either side. The teeth are very large, and widely divergent. The hinge line is essentially straight. In the dorsal valve the musculature is equally striking. (See plate 1, fig. 7; plate 10, figs. 1-3). As in the ventral valve, the attachment seats are elevated on a callosity from the interior of the shell, but are not so well defined by the dental lamellz as they are in the ventral valve. The cardinal process is bipartite, and extends only slightly, if at all, beyond the cardinal line as strong diverging carine. This can be seen on the plate 10, figs. 2, 3. The dental sockets are relatively large to accommodate the large opposing teeth. The crural platform is well developed posteriorly, where it is elevated on either side considerably above 22 BULLETIN 83 122 the hind periphery of the muscle platform. Forward extensions of the crural platform outline the narrowly flabellate posterior adductors in much the same manner as the dental lamellee out- line the ventral diductor scars. These extensions reach further to the front as weak, recurving and incurving borders of the nar- row and anteriorly attenuate forward adductor scars. From the base of each part of the cardinal process prominent angular ridges extend toward the front. These are considerably elevated above the level- of the muscle platform, and converge toward the middle at about the contact of the anterior and posterior adductor scars. They increase in elevation forwardly, where, united, they continue as a low septum recessed in a depression between the adductor scars. This septum is evanescent in the deepest part of the recess, which occurs at about the mid-zone of the anterior adductors, but increases in prominence in the front quarter of the platform. Near the anterior border of the platform the septum is well elevated and spade-like, and extends for some distance over the pallial region in front of the muscle-scar structures. In the rear part of the shell, between the extensions of the cardinal process before they unite, there is a low median septum on the narrow elevated platform which their outward fusion creates. This median carina is conspicuous to the point of evanescence of the median structure near the middle of the muscle platform. The dorsal muscle platform is much narrower, and more elon- gate than the ventral one. The sides are subparallel, converg- ing only slightly anteriorly. The posterior adductor scars are irregularly hexagonal in outline. They are bounded postlaterally by the vertical edge of the crural platform; postmesially and me- sially by the sloping edges of the cardinal process extensions ; anteromesially by the recessed edge of the anterior muscle scars ; and anterolaterally by an incurving of the peripheral crural exten- sion at the most elevated part of the muscle platform where there is a vertical elevation from the inner shell surface; and laterally by the crural extensions which here are subpazallel. The eatire posterior scar rises from the floor of the valve toward the front. Delicate arborescent ridges cover the postericr impressions. In each of these impressions there are two zones. The part lying between the forward apex of the crural platform and the mesial 1p COLOMBIAN DEVONIAN FAUNA: CASTER 29 convergence of the cardinal process extensions is slightly more elevated than the rest of the scar. The front portion of the scars is bordered by a flange of crural extension. The anterior adduc- tors are of the same general shape as the posterior, but are some- what narrower, and more produced toward the front, which gives them an equal sided, subtriangular outline. If the most forward extension of the rear adductors is taken as the apex of the tri- angle, the side extending from this point to a point a short distance in back of the convergence of the procéss extensions is subequal to the slightly arcuate anterolateral side. The bases of the tri- angles diverge from the posterior apices and the intermediate zone is excavated. The surface of the anterior scars is on a lower plane than the posterior ones and is undulatory throughout, the posterior part being slightly higher than the anterior, and central- ly concave. This creates the appearance of a flange adjacent to the mesial fossa. The attenuated crural extensions disappear toward the front and do not extend quite so far forward as does the mesial carina. The diductor fossa, while excavate through- out, is deepest toward the front, where the septum is most pro- duced. The entire surface of the median platform rises toward the front, so that the converging anterior boundaries stand out in bold relief from the pallial region. There is no evidence of an internal concentric carina or “diaphragm” in the dorsal valve at the point of geniculation as in some species of this genus. Dimensions.—The dimensions in millimeters of a representative suite are as follows: 5468 5469 5470* 5471 5472* 5473 5474 5475 Width hinge ime; 90) 32 45 40 40) 16 24 Length hinge 35 35 28 33 25 ine. 9 17 Length ‘‘trails’’ wines AKG 5 14 Length “*visceral dise’’ 30 31 26 25 25 ine. 9 17 Length muscle sear 15 16 19 8 Width muscle sear 14 23 ila 12 All of the above specimens are either casts or molds as fol- lows: 5468, 5469 and 5473 are internal ventral molds; 5470 is an internal dorsal mold; 5471 and 5475 are external ventral molds ; 5472 is an external dorsal mold, and 5474 is an external cast, *Dorsal valves 24 BULLETIN 83 124 valve probably dorsal. Discussion.—Of the generic assignment of the Colombian ma- terial there can be no question; the pattern is too similar to the genotype, Leptena rhomboidalis. As always, when a new spe- cies of Leptena is proposed, there immediately is entailed the search for tangible differences by which the new form may be differentiated from the many forms found at so many horizons over the world which have been assigned to “Leptena rhomboid- alis’. First, the writer is of the impression, shared by Grabau, 1931 (p. 20), when describing new Devonian Leptena, that the many forms assigned to this species are quite likely homeo- morphs. Second, with such inadequate information on facies variation in the Southern Hemisphere, it seems unwise to pro- pose subspecific names at this time. The criteria for specific differentiation in the genus lie, it would seem, principally in the minutiz of the musculature, size and general contour. South Arerican representatives of the genus, either listed as “L. rhombow alis” or under other names, are very rare. Frag- mental material from the Devonian of Rincon de Alonso in Uruguay has been provisionally listed as Leptena, sp. by Mendez- Alzola, 1934 (p. 32), but may be corrugated Leptostrophias, or possibly the fcssil described as Stropheodonta argentina Thomas, 1905 (p. 261), but assigned to Leptena by Clarke, 1913 (pp. 290, 340-341), after examination of the material. (See below). Leptostrophia caribbeana Weisbord, 19260 (p. 10), from Vene- zuela, which he considered close to the Argentine material, is, however, a true leptostrophid. The Venezuelan species, which is here referred to a new genus, is also present in our fauna, but has no beaiing in the present comparison. The “austral” association of leptenids and corrugated leptostrophids has a parallel in the Oriskanian (Grande Greve limestone) of the Gaspé Peninsula in the north. In immaturity Lep- tena looks very much like Leptostrophia oriskama. This was commented on by Clarke, 1908 (p. 184), in his description of the Gaspé fauna. While I am unconvinced as yet of the wis- dom of Clarke’s pronouncement on the Argentine fossils, and am still receptive to the idea that the material may be stropheodontid, 125 COLOMBIAN DEVONIAN FAUNA: CASTER o5 the differences in appearance should be mentioned. Thomas’s fossils are far less rugose, smaller, and if his description and illustrations can be relied on, have virtually none of the typical lepteenid muscle characteristics. They do show faint corruga- tions of the shell, which Clarke said were minimized by Thomas, and are apparently slightly geniculate. No other reports, to my knowledge, carry identifications of Leptena from the South American Devonian. The Colombian material is very similar to the Helderbergian, Oriskanian and Onondagan forms of Lepte@na described from North America. Especially reminiscent is L. rhomboidalis, var. ventricosa (Hall) of the Oriskanian. This form, as described by Hall (Strophomena rugosa, var. ventricosa Hall), 1859 (p. 417), Hall and Clarke, 1892 (pl. 15 A), and Schuchert, 1913 (p. 308), from New York and Maryland, is presumably specifically dis- tinct and may well be taken for the nonce as typifying the lep- teenid development in the Oriskanian. The Colombian fossils share the abnormally large size of Hall’s species, but show ephe- bic geniculation not present in the North American material. There are other important differences, as a comparison of the plates will show, especially in the internal characters. Hall’s, 1859, drawing and Schuchert’s, 1913, photograph of the same speci- men, when compared with L. boyaca, show very clearly a more prominent and massive cardinal process in the dorsal valve, a less strongly delimited muscle platform, shallower median de- pression, and more striking and continuous median septum or carina. The Hall species has a dorsal callosity or “diaphragm”, but no “trail”. The South American material at hand shows no evidence of a diaphragmal callosity. The ventral interior of the Oriskany form bears a nearly circular muscle scar, the halves of which almost enclose a median depressed zone of lenticular shape. In the Colombian specimens, this central zone is slight- ly expanded and open in front. The Oriskanian material ap- parently does not have a median carina in the midst of the ventral process. There seem to have been at least two strains of lepteenids in the Lower Devonian. Those of the Helderbergian are usually smaller and recall Silurian and even Ordovician representatives, 26 BULLETIN 83 126 whereas the Oriskanian and Onondagan forms, as Nettelroth, 1889 (p. 150), and others have pointed out, are consistently larger, more flabellate and ventricose than the previous forms. Dunbar, 1919 (p. 86), illustrated Leptena (cf.) rhomboidalis from the Camden chert of Tennessee. His specimens show L. ventricosa characteristics and also Helderbergian attributes. Leptena boyaca is one of the commonest fossils in the Colom- bian fauna, and shows considerable variation, but all specimens bear sufficient characters in common to warrant the present as- sumption that we are dealing with a single lepteenid strain, which is in all likelihood closely related to L. ventricosa of the North American Oriskanian. Types.—Holotype: Pal. Res. Inst. No. 5470; paratypes: Nos. 5468, 54609, 5471-5474, 5471A, 5474A. Family STROPHECDONTIDZ Caster, new The crenulate-hinged Strophomenacea are the subject of a more extensive study now in progress by the writer. In the final report on the investigation, the major classification of those brachiopods will be discussed in some degree of fullness. It seems wise to digress here sufficiently to outline the general scheme of classification of the lower categories, and point out some of the criteria of value in defining the more striking generic lines. This seems necessary in order better to bring out the structural and stratigraphic correlations of the Colombian fauna. It appears that the entire group of crenulate-hinged Stropho- menacea bears characters of structural value adequate to war- rant recognition as a separate family of the Strophomenacea. The name STROPHEODONTIDA is hereby proposed for this new family centering around the genus Stropheodonta to which most crenulate-hinged brachiopods were at one time or another re- ferred. The possession of a crenulated hinge in these rafines- quinoid shells, coupled with a community of quite similar inter- nal structures, seems to be of family value. It appears after a rather careful study of the better part of the entire group as developed in the Western Hemisphere, that in this stock at least, the character of resupination is not so important as former classi- 127 COLOMBIAN DEVONIAN FAUNA: CASTER Li) ~ fications would imply. Schuchert and LeVene, 1929, for ex- ample, dissociated the strophonellids from the stropheodontids mainly on resupination it would seem, and referred the former to the Oithotetine, with which they seem to show, omitting re- versed convexity, no major classificatory correlation. It seems to express relationships much better to place both groups in a common family and recognize the resupination as principally a subfamily, or even less significant characteristic in this stock. In the stropheodontids Douvillina and Douvillinella, the latter being resupinate, the character is apparently not of more than generic value. The part which wholesale homeomorphy plays in the Prachiopoda daily grows more obvious, but to what ex- tent differently derived lines can almost precisely ‘“‘mimic”’ each cther thas in each case to be thoroughly proved. It seems more probable in this case that both lines were derived from the rafinesquinids and soon after inception, resupination was estab- lished in one as a dominant trait. For this basal cleavage of the family Stropheodontidee two subfamilies are proposed in line with the characteristics suggested above: the Stropheodontine and the Strophonelline. Within the Stropheodontinze three great groups are recog- nizable, as Hall and Clarke, 1892, understood long ago when proposing subgeneric names for the all-inclusive genus Stropheo- donta. These groups are here designated tribes of the subfam- ily and take their names from the respective genera which they recall: Stropheodontim, Leptostrophiini, and Douvillinint. The first tribe, as here considered, embraces three types of shells for which separate tribal groupings may eventually prove necessary. The groups are here listed as the “brachyprionids”, the “douvil- linoid stropheodontids” and the “true stropheodontids”. Sev- eral new generic groups and species assemblages are also pres- ent in the St-ophonelline. Keys and abbreviated discussions of these matters follow: Key to the Subfamilies and Tribes of the Stropheodontide ’ A. Rafinesquinoid shells which are coneavo-convex, plano-convex, or rarely bi-convex and have a partially or comple'ely crenulated hinge at some stage of their ontogeny. STROPHEODONTINAE, new. 28 BULLETIN 83 128 B. Adult shells concavo-convex or subconcavo-convex. C. True braceplates (Stiitzplatten) present in the dorsal valve, shells usually small, and muscle scars usually strongly encircled loyaelanvell anid posits sass sees eee Tribe Dowvillinini, new. CC. True braceplates absent in the dorsal valve, although false braceplates may be well developed for muscle attachment; shells of moderate to large size in most species, and muscle sears ordinarily mot encireled in the douvillinoid manner. ___..___....... EE ESED Ne Site Ra LI MSA A = lee Coin cine ERE Tribe Stropheodontini, new. BB. Adult shells plano-convex or subplano-convex; no braceplates in GOS all vellivc Spamieteeoes tant ere Tribe Leptostrophiim, new. AA. Strophomenoid shells which are convexo-planate (resupinate) and have a partially or completely crenulated hinge at some stage of their development; braceplates not known in the dorsal valve. _........ STROPHONELLINAE, new. Subfamily STROPHEODONTINAE Caster, new The following genera principally represent the tribes of the Stropheodontine as recognized at the present time: Stropheodontini The Brachyprionids Brachyprion Shaler. Strophomena leda Billings, genotype. Silurian. Mclearnia Caster, new. Brachyprion mertoni MeLearn, genotype. Silurian. Shaleria Caster, new. Strophomena gilpeni Dawson, genotype. Silurian. The Douvillinoid Stropheodontids Protomegastrophia Caster, new. Leptena profunda, Hall, genotype. Silurian. Megastrophia Caster, new. Stropheodonta concava Hall, genotype. Middle Devonian. Dictyostrophia Caster, new. Dictyostrophia cooperi Caster, new, genotype. Middle Devonian. Cymostrophia Caster, new. Leptena stephani Barrande, genotype. Middle Devonian F2. The True Stropheodontids Stropheodonta Hall, s.s. Strophomena demissa Conrad, genotype. Middle Devonian. Douvillinini Dowvrillina Oehlert. Strophomena dutertrei Murchison, genotype. Frasnian F2 Douvillinella Spriesterbach. Douvillina filifer Schmidt, genotype. Middle Devonian. Group of Strophomena inequistriata Conrad, auct. Middle Devonian. Group of Stropheodonta cayuta Hall. Upper Devonian. Group of Stropheodonta arcuata Hall. Upper Devonian. 129 COLOMBIAN DEVONIAN FAUNA: CASTER bo io) Leptostrophuni Primitive Leptostrophids Group of ‘‘Leptostrophia planulata’’ (Hall), Manlius and Helderbergian. Group of ‘‘Stropheodonta’’ bipartita Hall, Manlius and Helderbergian. The Perplana Group Protoleptostrophia Caster, new. Strophomena blainvillii Billings, genotype. Oriskanian. (Conrad), auct. Middle Devonian. ? Group of ‘‘Leptostrophia perplana’ The Corrugated Leptostrophids Rhytistrophia Caster, new. Stropheodonta beckti Hall, genotype. Helderbergian-Erian. The Nervose Leptostrophids Nervostrophia Caster, new. Stropheodonta nervosa Hall, genotype. HKrian-Senecan. (?Chemung). Group of Leptostrophia junia (Hall). Middle Devonian. The Sulcate Nervosz Sulcatostrophia Caster, new. Leptostrophia camerata Fenton and Fenton, genotype. Upper Devonian. Linguate forms and other forms of Suleate Nervosx in the Eifel. Edentulous Leptostrophids Australostrophia Caster, new. Leptostrophia ?? mesembria Clarke, genotype. Onondagan, Brazil. True Leptostrophids Leptostrophia Hall and Clarke, s.s. Strophomena magnifica Hall, genotype. Oriskanian. Leptostrophia explanata (Sowerby) - Subgeneric strain of Leptostrophia. Coblenzian. The subfamily Stropheodontine is characterized by a general rafinesquinoid contour in most of the species. This varies rela- tively slightly so far as the outline goes, but there is considerable variation in the detail of relative convexity and concavity. The surface is commonly ornamented with fine elevated costellee which ordinarily are arranged in alternations of several fine radii be- tween more prominent costellz. Concentric surface ornament, excluding simple varices, is rare in the group as a whole, although apparently reaches generic proportions in isolated lines. Inter- nally virtually all genera and species have a strong pair of ventral dental lamellz (plus paradental lamelle in some cases), which vary in length and direction. In some forms accessory calcar- eous muscular “diaphragms” separate the adductors from the diductors. A v-shaped posterior ventral “‘process” of greater or 30 BULLETIN 83 130 less development and in scme cases a strong ventral median sep- tum are generally present, but vary greatly in detail. In the dorsal valve the cardinal process is usually well developed, and may be either sessile or pedunculate; the strength and general plan of the parts of the process vary considerably in different genera, Crural plates are well developed in virtually all forms, but their length and direction vary a great deal. The presence or absence of a median dorsal septum and anterior adductor plates or braceplates (Stiitzplatten) are also important generic criteria. The ventral “‘process” in most forms fits between the prongs of the dorsal process, thus creating an effective articulation which is reminiscent of the cyclodont condition in the pelecypods. ‘The true hinge teeth and sockets are in general obsolescent, but oc- casionally are rather well developed. Tribe Stropheodontini Caster, new The brachiopods falling in the tribe Stropheodontini conform most closely to the original plans of the genus Stropheodconta. It was especially the groups of Stropheodonta denussa and Stro- pheodonta conceva that Hall apparently had in mind when the crenulated strophomenoids were differentiated. It is highly probable that in the completed study of the stropheodonts, how- ever, that the “concava” group (Megastrophia) will be elevated to the rank of tribe. The protean brachyprionids and the close congeners of Protomegastrophia, all in the Silurian, will also be separated into one or two new tribes. The information, gleaned principally from North American faunas, suggests these steps. It seems best to let the matter rest until European species can also be included with some degree of certainty. A Key to the Divisions and Principal Genera of the Stropheodontini. A. Hinge either partially or wholly crenulated, but usually incompletely so; surface usually marked by an alternating pattern of radii which are not especially fasciculate; partial or complete encirclement of the muscle scars of one or both valves relatively common; strong varices of growth uncommon. B. Ventral median septum well developed; shells usually of small size; delthyria primitively gaping; hinge crenulations restricted to very few on either side of the delthyria, usually located on a tiny expanded zone of the hinge; shells thin _........... The Brachyprionids. 131 COLOMBIAN DEVONIAN FAauNA: CASTER 31 _C. Ventral median septum undivided anteriorly. D. Ventral muscle sears flabellate, expanded, and bordered post-laterally by what appear to be elongate continuations of the dental lamella; hinge crenulations very few, restrict- ed to a small triangular zone adjacent to the delthyria; process in dorsal valve very weak _... Brachyprion Shaler, s.s. DD. ventral muscle scars narrow, elongate, rather than trans- versely flabellate; scars rectangular in outline; crenulations few, located on a narrow, tear-shaped expansion of the hinge plate adjacent to the delthyria; ventral muscle scars near- ly surrounded by lamellar extensions..Mclearnia Caster, new. CC. Ventral median septum divided anteriorly in the plectambonitid manner: in some species (not in the genotype) the forks of the septum join the paradental lamelle completely to encirele the MUS CLEP SCAT So ete ean OU ae pele Mets t Shaleria Caster, new. BB. Ventral median septum obsolescent or absent; shells usually of med- ium to very large size; delthyria in adulthood usually partially or com- pletely closed, but may be secondarily opened by abrasion or other factors; hinge crenulations usually do not extend for more than half the length of the hinge, but may, in forms having strongly differen- tially wrinkled shells, nearly fill the hinge; shell usually relatively thick and may be wrinkled longitudinally either regularly or differ- entially ; cardinal process usually very prominent. _.......................--..--- ee SS a ee Nees ees Saleh iE ds The Douvillinoid Stropheodonts. E. Shells large and thick; anterior dorsal diductor sears not strongly scoriaceous nor especially elevated or elongate, although of large size; hinge crenulations extend from one- third to slightly less than one-half the width of the hinge; ventral muscle sears large, flabellate, and only posteriorly delimited by lamellar extensions; delthyria may be open, partially closed by lateral ingrowths which resemble paradeltidia, or by an internal deltidial callosity; delthy- rial zone usually an equal sided triangle; surface usually GO Aue seecteees okt eo Protomegastrophia Caster, new. KE. Shells large, but may be relatively thin; anterior dorsal diductor sears are either elongated, elevated and roughened areas, which in some cases become scoriaceous plates of at- tachment (pseudo-braceplates), or elongate submerged pits with roughened surfaces separated by a broad median septum or callus. F. Shells large, relatively thick; ventricose; alternating radial ornamentation with faint varices, no corrugations or wrinkles; anterior dorsal diduector sears are elevated, but not much produced beyond the adduec‘or sears; hinge erenulations extend for about half or slightly more than half the hinge width; ventral muscles are large, flabel- late and nearly, if not completely, encircled by lamellar extensions; dorsal adductor scars usually subtrigonal; only anterior dorsal median septum, if any, present; del- thyria closed by deltidia, but always very narrow, usual- ly lnear and sometimes inconspicuous; cardinal process LOMACEOUS ee eww. SA Ae Megastrophia Caster, new. FF. Shells large, relatively thin; usually not especially ven- tricose; alternating ornamentation with faint growth lines in some specimens; shells uniformly differentially corrugated or wrinkled in a seersucker manner; anterior (Je) to BULLETIN 83 132 dorsal diduetor sears are much elongated as scoriaceous subparallel ridges in front of the adductor scars. These are the pseudo-braceplates. Hinge crenulations extend for nearly the full width of the shell; strong median dorsal septum arises from a callus at base of prominent dorsal process; anterior portion of shell often genticu- lated, leaving posterior portion of both valves somewhat flattened to recall a visceral dis¢. _.o oo... cece eee FIFE. Shells large, relatively thin; usually not especially ven- tricose; alteinating radii with very strong concentric elevated lines forming a striking dictyate surface pat- tern; radial subangular plications of the shell follow the principal radii even onto the early portions of the shell. Anterior dorsal diductor sears are located in depressed, elongate pits which are roughened internally. The pits are separated by a median callus or septum, probably corresponding to the anterior median septum.. Ventral muscle scars are not strongly delimited, and are of me- dium size. Dorsal adductor scars are very small, and Suomen. Dictyostrophia Caster, new. AA. Hinge in adult shells holocrenulate, or nearly so; shells of medium size, regular, without any striking folds or corrugations; surface radii coarse and usually strongly fasciculate and not truly alternat- ing in pattern; lamellose varices common in several forms; ventral muscle sears relatively large, flabellate, and not conspicuously limited by dental lamelle, which are usually obscure; paradental lamelle not known; ventral median septum usually frail, obsolescent or absent; dorsal process relatively strong, and usually sessile. 0.222200 DS ERENCE N YB ROSS Se sie sneer Nee eu es Stropheodonta Hall, s.s. The Brachyprionids Only our inability to judge as yet the part played by homeo- morphy in this early group of stropheodontids prevents assign- ing them to a separate tribe or possibly splitting them into two tribes. It seems likely that all lines of the later crenulate-hinged brachiopods had antecedents within this essentially Silurian stock, or “amalgam”. The genera here outlined are only a very small part of the latent genera typified by the widely varying species of “Brachyprion”. In the following generic discussions the char- acteristics of the genotype species have been principally consid- ered. In making these comparisons the collections of the U. S. National Museum, New York State Museum, and University of Cincinnati Museum have been of inestimable value. In view of the absence of true brachyprionids in South American strata, only a skeleton discussion in here presented. 133 COLOMBIAN DEVONIAN FAUNA: CASTER 33 Brachyprion Shaler, 1865 In addition to the characters indicated in the key above, the genotype, Strophomena leca Pillings, as represented from Jum- pers, Anticosti, in the U. S. National Museum (94397), shows a sessile cardinal process which rises from a low postumbonal platform. The prongs of the process are subterete and project very slightly to overhang the hinge and notothyrium. On either side of the process socket, ridges divaricate at a high angle, leav- ing a rather prominent dental socket between them and the hinge crenulations. Medially the posterior platform extends toward the front as a low rounded callus. In the ventral valve it ap- pears that the true Cental plates are elongate and constitute the marginal limits of the muscle scars. None of the specimens of the genotype has shown evidence of the paradental plates found in Mclearma and Shaleria, nor any evidence of a strong median ventral septum such as we find in those genera. The cotypes of Brachyprion newsomensis Foerste, 1919, (U. S. Nat. Mus. 87032) and Brachyprion plana Foerste, 1909, (U. S. Nat. Mus. 84914) from the Ohio Silurian appear to be assignable to this genus in the strict sense. The genus seems to occur exclusively in the Middle and Upper Silurian of the Northern Hemisphere, but may range into the Helderbergian. Shaleria Caster, n. gen. Diagnostic in this new genus is the character of the ephebic plates which lie immediately imsi’e the elongate dental plates and extend beyond them anteriorly as delimitations of the ventral muscle scars. These accessory plates are here termed paradental plates. These are the plates which apparently fuse indistinguish- ably with the true dental plates in Mclearnia and join the rami of the bifid median septum completely to encircle the muscle scars. In Stropheodonta gilpeni Dawson, the genotype, the den- tal plates themselves are elongate, nearly vertical to the hinge and nearly subequal in length to the paradental structures. The paradental plates are outwardly deflected toward the front in a broad are which is essentially concentric with the arcuate branch- es of the median septum. This is of course a variable feature, 34 BULLETIN 83 134 and hardly of constant generic value. Examples of Shaleria gilpeni from the Stonehouse formation on McPherson’s Brook, Arisaig, N. S., in the U. S. National Museum collection (36942) show the nature of the paradental structures very well. It ap- pears that Strophomena ornatella Salter ( e.g. Davidson, 1874, pl. 43, figs. 16-20) is a Shaleria, although the bifid median septum and paradental plates are anchylosed to surround completely the muscle zone. This may prove to be a distinct generic develop- ment when there is opportunity to study Salter’s species. David- son’s figures (e.g. idem, fig. 19b) suggest the presence of over- lapping dental and paradental plates in this species. The types of Brachyprion shaleri Williams, 1913, (U. S. National Museum 58952) show in some specimens a faint tendency for the median septum to split distally. Here also the dental plates are very long, but no sign of paradental overlap has been observed. The Williams material, however, shows additional dorsal features, such as strong postlateral muscle delimitations, prominent cardin- al teeth and sockets. These may, when coupled with the struc- tures mentioned above, prove of generic value. I fail to see any very close comparison between Williams’ species and the resupinate Strophonella striata (Hall), 1843, a comparison which Williams stressed. The genus Shaleria is known only from the Upper Silurian (Middle?) but may range into the Helderberg- ian. Mclearmia Caster, n. gen. Of considerable value in the classification of the stropheodon- tids is the nature of the ventral muscle scars. The size and shape of the scars and degree of delimitation by a peripheral carina or callus are the important criteria. The principal features of generic importance are brought out in the preceding key. It appears likely that the specimens from the Ardenne Devonian of Belgium identified by Asselberghs, 1930, as Stropheodonta orna- tella (Salter) may be a holocrenulate congener of Mclearnia. The bearing of Salter’s typical material has been discussed above under Shaleria. Koztowski’s, 1929, Brachyprion subinterstrialis and var. serentensis from the Polish Silurian may belong to Mc- learnia. Several specimens of Koztlowski’s species were exam- 135 CoLOMBIAN DEVONIAN FAUNA: CASTER 35 ined in the U. S. National Museum (84320-84322), but certain diagnostic characters are concealed in this material. Barrande’s Leptena costatula, 1847, 1879, also seems to have the internal features of the present genus. This is especially true of the ma- terial identified as Barrande’s species by Kozlowski, 1929 (p. 100). If these conclusions are correct, it is possible that Mc- learnia is the direct antecedent of Stropheodonta demissa and allies. The known range of the genus is Upper Silurian, but holocrenulate derivatives probably also occur in the Helder- bergian. The Douvillinoid Stropheodontids Four genera among the stropheodontids are here likened to the Douvillinini, but it is presupposed that the similarities are largely homeomorphic. The genus Protomegastrophia (Leptena profunda Hall) presents few of the characters recalling the dou- villinids, but does appear in other respects to be antecedent to the genus Megastrophia (Lower? and Middle Devonian) which shows incipient douvillinoid characters. The genera Cymostro- phia and Dictyostrophia show these homeomorphic aspects well developed. The principal douvillinoid features of the group oc- cur in the dorsal valve, where are found scoriaceous anterior diductor lamellz which are grossly comparable to the braceplates in the Douvillinini, with which they may be homologous. The scoriaceous plates in the present group differ from those in the Douvillinini principally in not having a median apex which is much the highest point on the douvillinid plates. There is no sign of incurvature of the plates toward the median line to form a cone or tube as they do in the Douvillinini. Large, and strong- ly encircled ventral diductor scars are much like those usually found in the Douvillinini. The shells of the douvillinoid strophe- odonts also recall on a gigantic scale the general shield shape of the true douvillinids. At first the entire group, excepting Proto- megastrophia and doubtfully Megastrophia, was assigned to the Douvillinini, but further study seems to point to the present ar- rangement of the classification. The range of the stropheodontid genera showing strong douvillinoid traits appears to be Oriskan- ian through Erian. This is of course antecedent to the most characteristic development of the Douvillinini in the late Middle and Upper Devonian. 36 BULLETIN 83 136 Protomegastrophia Caster, n. gen. The genotype, Leptena profunda Hall, 1852, from the Middle Silurian (best development in the Waldron) has in_ receit writings been ordinarily assigned to the genus Brachyprion. So generalized is the structure of the species group that it could still be assigned nearly as readily to the brachyprionid group of genera as to this douvillinoid assortment. Typical material of the geno- type has been examined in the fine collections of Waldron ma- terial at the University of Cincinnati, the U. S. National Museum, and at Albany, New York. “Leptena’ profunda is one of the largest of a considerable number of large Silurian crenulate- hinged brachiopods. The shell material is exceptionally thick. The ventral valve is extremely ventricose in most instances, and the dorsal valve correspondingly concave. The crenulations ex- tend further along the hinge than in the three genera of brachy- prionids described above. The ventral muscle seats are large, flabellate, striate, and in all respects very similar to those of the genus Megastrophia, lacking only very strong lamellar margina- tion. It seems that virtually all of the traits of the Devonian genus are anticipated in subdued style by the Silurian giant. The curiously spatulate cardinal process of some species of Megas- trophia (undescribed as yet, but usually referred to Stropheo- donta concava, s. 1.) is anticipated in Protomegastrophia. Lilke- wise, the weak ventral process, which is of little articulatory function, is found. Examples of Protomegastrophia (species not described) in the University of Cincinnati Museum, from the Racine formation, show rather well the subequal sided del- thyrium and notothyrium, each closed by a convex plate or callus in adulthood. The notothyrium usually has the appearance of being closed by a callus deposit, possibly associated with a con- vex chilidium. The deltidium appears to be a convex plate in the apex of the delthyrium and is excavate toward the commis- sure plane, which it attains only on the margin of the delthyrium. In some cases it seems as though paradeltidial plates, with a short convex deltidium between them, are present in the apex of the delthyrium. The Racine specimens show ventral diductors of the size and proportion seen in Megastroplia, which similarly completely surround the lenticular adductors. In shells assigned 137 COLOMBIAN DEVONIAN FAUNA: CASTER 3 ~l to the genotype, Hall, 1852, described the structure which we have called the ‘‘ventral process” as a “projecting, grooved and bidentate process in the ventral valve through which the pedicle passes”. The delthyrial foramina are usually gaping in imma- turity, but closed in maturity as previously explained. Oblique crenulations extend less than half the hinge width and diverge from the beak. In many respects of proportion and structural detail, the Silurian species foreshadows Megastrophia hemispher- ica of the Onondagan. Stropheodonta niagarensis Winchell and Marcy, if not conspecific with the genotype, certainly belongs in the new genus, and likewise, Stropheodonta convexa Prouty of the Maryland Silurian. The genus is known from the Niagaran, and may possibly occur in the Upper Silurian. No true repre- sentatives are anticipated in the Helderbergian, although Megas- trophia will probably occur there. Megastrophia Caster, n. gen. The genus Megastrophia is taken as the prototype of the group here called the Douvillinoid Stropheodontids. The principal struc- tural features of the genus are brought out in the key and forego- ing discussion of the supposedly antecedent genus in the Silurian. In studying this genus the excellent collections of Stropheodonta concava Hall, genotype, and allies in the U. S. National Museum, the New York State Museum, and the University of Cincinnati Museum have been of great value, and the characters outlined are based on these collections. The ventricosity, great size and thickness of the adult shell, incomplete hinge crenulation, en- circlement of the large, heart-shaped ventral adductor scar, es- sentially linear delthyrial area, and general ponderous irregular- ity of the ephebic shell, all seem to point to phylogerontism, Yet the genus ranges from Oriskanian through Hamiltonian time. The cardinal process in this genus is usually spatulate, and broad- ly pedunculate ; at all times ponderous. In the genotype of Megastrophia the posterior diductors of the dorsal valve are attached to elevated, anteriorly expanded and relatively narrow platforms which bear linear ridges and nodules. ’ These platforms originate just in front of the cardinal process and terminate near the front margin of the elongate adductors. The posterior adductor carinz are usually separated by a rela- 38 BULLETIN 83 138 tively deep fossa. In front of the adductor scars and cuneiform posterior diductors, the anterior diductors are attached to rough- ened or scoriaceous, elongate-ovate elevated zones which usually do not rise so high above the floor of the valve as the posterior muscle platform. Between these roughened scars there is an an- terior median septum in ephebic shells. Gerontism may exag- gerate the septum. This septum usually does not reach the rear platiorm and extends forward slightly beyond the anterior di- ductor scars. In “austral’ representatives, at least, of Cymostro- phia, described below, the median septum extends from the base of the cardinal process as a very prominent tapering structure. Here also, the anterior diductor scars are much extended and elevated so that they appear like scoriaceous lamelle which are excavate above, but have a superficial resemblance to the brace- plates (Sttitzplatten) in the Douvillinini. They do not show any of the detailed structure of the braceplates, however, and there can be no question of their being principally places of muscle attachment. These false braceplates of the M/egastrophia allies are termed pseudo-braceplates. In Dictyostrophia they are only slightly developed and usually appear as elongate scoriace- ous excavations rather than elevated lamelle. The surface orna- ment of Megastrophia is made up of alternating costellz which are crossed by concentric elevated lines, ordinarily finer than the finest radial ornament. The result is a fine dictyate pattern. The concentric lines are usually very inconspicuous, and often cannot be distinguished without a lens. This type of ornament is great- ly exaggerated in Dictyostrophia. Most North American representatives of the genus, without much critical examination, have been assigned to Stropheodonta concava or closely allied species. There are apparently a great many valid undescribed species thus concealed. Stropheodonta murchison. (dArchaic and de Verneuil) of the Lower Emsien of Belgium (U.S. National Museum 87244-9) shows a slight de- velopment of the pseudo-braceplates, the while sharing many telling features with Dictyostrophia, yet will probably be referable to a new genus. Stropheodonta herculea Drevermann of the Siegen Devonian in Germany also has weak pseudo-braceplates, but its more continuous dorsal median septum recalls the genus Cymos- 139 CoLOMBIAN DEVONIAN FAUNA: CASTER 39 trophia. Two species of Megastrophia are described from the Colombian faunule, below. Cymostrophia Caster, n. gen. This new genus with the genotype Leptena stephani Barrande from the Devonian F2 fauna of Konieprus, Bohemia, will include a large number of differentially corrugated shells having the gen- eral aspect of Stropheodonta corrugatella Davidson in Great Britain, Stropheodonta patersoni Hall (and authors), Stropheo- donta reticulata Stainbrook, etc., in North America. Among the characters brought out by the preceding structural key and following description of “austral’’ species, certain ones appear to be of primary generic value. These would include, without selective arrangement as to importance: large size of the adult shells which are concavo-convex, but not ordinarily extremely ventricose ; shell material which is relatively thin, especially so for shells as large as these; ornament consisting of strong rounded elevated costelle with intercalated finer costellz, which thus cre- ate an alternating pattern; concentrically corrugated shell in the intermediate zone of fine costelle. The corrugations are inter- rupted by the coarser radii, thus creating a surface which looks like the fabric seersucker. This makes the recognition of the genus easy from only a fragment of the shell material. So far as can be judged, this character is of genetic significance, and does not seem to have appeared in divergent lines. The ventral di- ductor scars are large, flabellate, and not strongly delimited. They enclose subfusiform adductor scars. In the dorsal valve the car- dinal process is relatively stout, and sessile. It has at its base an elevated median septum which seems to have acted as a buttress. There are radial secondary buttresses also, presumably repre- senting a split socket plate, or perhaps socket plates and crural plates. There are also relatively inconspicuous pseudo-brace- plates which are upwardly excavate and coarsely scoriaceous for muscle attachment. The hinge is nearly completely crenulated. The genus Cymostrophia was probably derived from or is pos- sibly a homeomorphic recurrence parallel to the differentially cor- rugated rafinesquinids in the Ordovician. Holtedahl, 1916 (pl. 3), e. g. has illustrated from the Kristiania region several simi- 40 BULLETIN 83 140 larly corrugated rafinesquinids: Rafinesquina? schnudti Jagel, R. munsteri Holtedahl, and R.? ringerikensis Holtedahl. The, presumably represent a distinct generic line in the rafinesquinids, but might be the antecedents of the stropheodontids here consid- ered. Willard’s genus Ptychoglyptus, 1928, was proposed for somewhat more regularly corrugated rafinesquinids from the Chazyan, which may also be related to this development. — The genotype of Cymostrophia is so well illustrated by Bar- rande, 1879 (pl. 40, figs. 10-30; pl. 55, figs. vi, 1-9), that very little additional information can be added. The F2 fauna of Konieprus bears many brachiopods which recall the lower Mid- dle Devonian of North America, The representatives of the genotype which I have examined (U. S. National Museum 53453) came from the type locality in Bohemia. Stropheodonta paterson Hall of the North American Onondagan is apparently closely allied to the species, and gives a very satisfactory picture of the genus here. Hall’s species, like most North American ones of this generic bearing, shows less closely spaced corruga- tions than the Bohemian form, and in this respect is reminiscent of the South American species described below. In none of the species known tome do the differential corrugations extend in adulthood much beyond the zone of greatest gibbosity. The corrugation is principally a feature of the zone that in other genera might be termed the visceral disc. Barrande shows the nature of the corrugation extremely well in his original illustra- tions. Apparently the strong mucronation exhibited by the geno- type is unknown in the Americas. Three Colombian species of Cymostrophia are described below. Dictyostrophia Caster, n. gen. Like the other douvillinoid stropheodontids, this genus is char- acterized by large to gigantic size, and relatively ventricose shells which are concavo-convex. The shell material, even in matur- ity is relatively thin. The surface is marked by radial coste and intervening costelle, usually four or five in number. There are also concentric elevated cords which are subequal to the series of costelle and cross them to form a very striking grid pattern 141 COLOMBIAN DEVONIAN FAUNA: CASTER 41 resembling cord netting. The shell is subangularly corrugated along the lines of the principal coste, thus creating an unusual combination of surface detail. The concentric lines are not or- dinarily present on the earliest third of the shells, but the sub- angular corrugations extend to the very apex in most cases. The dorsal valve has a flattened “visceral disc” peripherad of which the shell is rather sharply geniculated. There may be developed internally a low callous diaphragm at the line of up-bending in the dorsal valve, although not so prominently as in Megastrophia. The ventral valve has moderately large and flabellate ventral muscle scars which are relatively faint and indifferently encircled by lamellar extensions. The dorsal valve has remarkably small recessed and subovate adductor scars which are separated by a weak median septum extending from the base of a frail cardinal process. The posterior diductors are obscure, but were appar- ently attached to narrow submerged zones on either side of the median septum and between the small adductor scars. The an- terior diductor scars are deeply excavated, elongate muscle pits which have a scoriaceous lining; between the pitsis alow callous ridge, or possibly anterior median septum. The submerged pits are thought to be homologues of the excavate surfaces of the pseudo-braceplates in Megastrophia and Cymostrophia. In “boreal” faunas Stropheodonta concava, sensu lato, will probably contain the relatives of this genus best developed, as far as known in the Colombian Devonian, although rumors are afoot of fine examples in the Nevada Devonian section. As shown under Megastrophia, above, the Stropheodonta concava group has both radial and concentric ornament, but in a lesser degree. The interiors of the two groups in the strict sense, are however, quite different. The European Stropheodonta murchisoni (d’Archaic and de Verneuil) of the Belgian Emsien deposits has angular surface plications very similar to Dictyostrophia, but does not, to my knowledge, show a reticulated surface pattern. The dorsal in- terior of the Belgian material does recall that of Dictyostrophia to a certain degree, but they differ in ventral characters, espe- cially in the extraordinarily strong development of a much ele- vated and broad muscle platform and median septum in the lat- 49 BULLETIN 83 142 ter. As mentioned before, the Belgian species is probably typi- cal of a generic strain, but appears to belong to this same group of stropheodontids. The genotype, Dictyostrophia cooperi Cas- ter, from Colombia is described below, THE DOUVILLINOID STROPHEODONTIDS OF COLOMBIA Genus Megastrophia Caster, n. gen. Genotype.—Stropheodonta concava Hall. Middle Devonian. Megastrophia hopkinsi Caster, n. sp. Plate 3, fig. 19; Plate 4, figs. 1-2 Ventral valve very large, ventricose or inflated centrally ; hinge line and length subequal; anterior outline shield-shape to sub- circular; hinge line broadly produced, and terminally rounded; anterolateral periphery slightly expanded; anterior produced slightly ; posterior two-thirds of shell essentially planate, rising toward the front; most produced on mesial fold; anterior part subgeniculate ; mesial swell or fold is obsolescent in the largest specimens. From small fragments of the surface characters on the larger mold, the shell is seen to be coarsely punctose and mul- tistriate. The striae appear to be of an alternating sort, with the primary ones rising like fine threads, subangularly from ex- cavate interspaces of about two or three times the width of a primary costella. The interspaces are finely striate, there being a variable number of striz in each interspace, but from 6 to 12 have been counted. Concentric ornament, if present, is of ex- treme delicacy. Palintrope relatively high, apparently externally smooth, and crenulate on the commissure surface in the mesial part for a little more than half the hinge length; mesial height of the palintrope about 3 mm.; centrally (on either side of the delthyrial opening which is not known) the crenulations of the hinge are continued externally for about 1 mm. on the external palintrope surface, and taper off to nothing where the hinge cren- ulations themselves disappear. The demarcation between the smooth palintrope surface and the crenulate zone is very sharp. Subumbonally the shell of the palintrope is about 2.5 mm. in thickness, and decreases toward the hinge apices. The ventral interior is dominated by elongate, flabellate di- ductor muscle scars having an angle of about 65°, and is very prominently bordered both on the sides and in front by an ele- vated extension of the dental plates. The muscle scars are in- 143 COLOMBIAN DEVONIAN FAUNA: CASTER 43 cised in a callus which extends across the entire posterior part of the shell. A broad “moat” is excavated around the enwalled muscle scars. It is deepest laterally and becomes evanescent antero-mesially. The “moat” is coarsely granulose and ridged throughout. The diductor scars completely encircle the lentic- ular adductor scars which are raised on an obscure mesial eleva- tion and separated by a weak median septum. The greatest ra- dial length of the diductor impressions is slightly more than their greatest combined width; the mesial length is to the radial length about as 3 is to 4. The lateral boundaries of the scars are very much elevated above the floor of the valve. They are highest toward the rear where the muscle seat is slightly excavate, and of about the same height above the floor of the valve anteriorly, but are there filled to the crest by secondary calcification within the muscle seat. The extended dental lamellz rise from the floor obliquely and pronouncedly overhang the muscle seat on the sides. In front they are nearly vertical, where recurved mesial- ly the bordering carinz stand lowest as cordate delimiting ridges. Here they unite with a median anterior boss from which the ob- scure septum extends for a short distance into the pallial zone. Unfortunately the subumbonal characters are not preserved, but it appears on specimen 5406 that there is a slight cavity for the reception of the cardinal process. On this specimen are also pre- served under the beak the front ends of the posteriorly converg- ing crests of the “ventral process,” which very likely met at the palintrope wall. The adductor muscle scars occupy a relatively large central lenticular area which completely separates the di- ductor muscles. Posteriorly the adductors are recessed below the general level of the diductor seats, and are bordered by low, anteriorly diverging ridges which become evanescent at about the zone of maximum width of the adductors. A low, irregular continuation of these ridges separates the adductors in front from the diductor seats. Along the median line the adductor scars rise on the slope of an obscure septum, which appears to separate them by an evanescent carina extending forward beyond the muscle markings. The front portion of the adductor scars rises very steeply from the general level of the muscle seats onto 44 BULLETIN 83 144 a prominent callus which marks the place where the right and left diductors and the attenuated adductors meet. Both the ad- ductor and diductor scars are radially striate, and the adductor scars at least seem to have been arborescently marked. The whole extramuscular area of the shell interior is ridged and stri- ated with fine pustulosities. The inner corrugations and pustules are strongest on the postlateral flanks of the muscle platform, The internal striz and ridges appear to be negative reflections of the surface markings. In specimen 5406 the striz of the middle zone appear to converge just anterior of the edge of the visceral surface or disc. They appear to be much finer here than else- where on the interior. Dimensions.— 5406 5453 Hinge width 55 mm. 80 mm. Length of shell 55 mm. 70 mm. Maximum he:ght of shell 25 mm. 25 mm. Length visceral dise 40 mm. 45 mm. Max. width visceral disc 37 mm. 46 mm. Max. radial length diductors 22 mm. 33 mm. (left) 40 mm. (right) Mesial length muscles 20 mm. 25 mm. Max. width adductors 13 mm. 15 mm. Discussion.—‘ Austral” comparisons are restricted to the present fauna, for no stropheodont in the Southern Hemisphere, known to me, approaches the Colombian material in size or general char- acteristics. The wider affinities appear to be with “boreal” Amer- ican forms and more distantly with those of the European De- vonian. The present species is much the largest stropheodont in the Colombian fauna. The outline is more subcircular than for any of the other four representatives in the fauna. The internal ventral characters of the other species are unfortunately not known, but the surface markings are so clearly differentiable that few other structures are necessary for quick recognition. Dictyostrophia cooperi is angularly plicate or fasciculate and has very pronounced secondary costellz and varices creating a regu- lar upraised reticulation on the surface which makes even a small fragment recognizable. Cymostrophia schucherti has radii of two sizes, the smaller being far more numerous than the pri- mary, and the whole surface is finely wrinkled after the manner 145 CoLOMBIAN DEVONIAN FAUNA: CASTER 45 of the “boreal” Stropheodonta patersonit Hall or the European S. corrugatella Davidson, but without concentric elevated varices. Cymostrophia dickeyi is another Colombian concavo-convex stropheodont of large size. It is smaller and more transverse than M. hopkinsi and has curious duplex primary radii in the mesial region and is finely cancellate. The interior of the dorsal valve of C. dickeyi has two curious subparallel carinz extending frontward from the beak, and the cardinalia are more delicate in this species than in Dictyostrophia coopert. Megastrophia pygmea is much smaller, but otherwise quite similar to the pres- ent species. It may be a juvenile form of this giant. In the “boreal” Devonian, Megastrophia hopkinsi seems to be most like Megastrophia concava (Hall) and M. hemispherica (Hall). The shape, ventricosity and ornamentation are all very similar, but the strongly bordered muscles of the pedicle valve in M. hopkinsi have no North American duplicate to my knowl- edge. The size of the ventral muscle scars is comparable to cer- tain forms of MW. concava in the Onondagan of New York (N. Y. State Museum; U. S. National Museum), but the muscle angle of our material is much smaller, the muscles themselves more divergent anteriorly and more deeply re-entrant anteromesially. Specimen No. 5453 in our collection recalls “Stropheodonta con- cava” illustrated by Kindle, tgor (pl. 6, fig. 1), from the Jeffer- sonville limestone of Indiana, This appears to be a common species in the Colombian fauna, but is known as yet only from internal molds of ventral valves. No dorsal valves of this magnitude, ornamentation, or general contour have been found. Types.—Holotype: Pal. Res. Inst. No. 5406; paratypes: Nos. 5453 and 5453A. Megastrophia pygmza Caster, n. sp. Plate 3, figs. 16-18; Plate 5, figs. 1-4; Plate 6, figs. 11-13; Plate 8, figs. 13-14 Ventral valve convex shield-shaped with subequal length and width; hinge straight and slightly produced at extremities ; um- bonal area most expanded, but entire mesial zone is broadly ele- vated into an undefined fold. The outline is regularly rounded 46 BULLETIN 83 146 ellipsoidally ; postlaterally it is slightly constricted in front of the hinge extensions. Toward the front the shell shows signs of gen- iculation, and is peripherally slightly scalloped. The surface is radially finely striate in an alternating manner, there usually being two or three fine striz between the coarser ones. New striz of both types originate interstitially. No tendency toward corruga- tions or fascicular arrangement of the striz has been observed. ~ Concentric lines are not visible on the specimens at hand. The palintrope is flat and slightly apsacline. It is relatively high mesi- ally and slopes off toward the extremities, and is vertically striat- ed. The hinge is apparently crenulated for nearly the entire length. On either side of the ventral delthyrial region there are two low hinge teeth. Dental lamelle extend around relatively large, elon- gate, and collectively heart-shaped diductor muscle scars, thus very clearly delimiting these structures. The diductor scars near- ly completely surround the mesial adductors which are situated on elongate, spindle-shaped elevations separated by a mesial fur- row. Just outside these adductor elevations and overlapping their front ends are two ridges which originate in the diductor zone and extend forward into the pallial region for a distance about equal to the length of the adductors. Out of the diductor scars arise two posteriorly converging angular carinee which meet before at- taining the inner palintrope (deltidial) wall to which the resultant confluent median process is attached. This is the so-called “ven- tral process”. The posterior terminus of the “process” protrudes slightly above the hinge line and appears to enter the hinge zone of the dorsal valve in a position that would be just behind the car- dinal process or between the prongs of this structure if its branches are sessile. The inner wall of the palintrope on either side of the median process is slightly concave as though to receive the prongs of the dorsal process. The deltidium appears not to reach the commissure plane, thus leaving a slight open space for admission of the cardinal process. The palintrope wall overhangs the inter- ior of the ventral valve and is excavate beneath. The pallial zone about the muscle scars is slightly granulose. The remainder of 147 CoLOMBIAN DEVONIAN FAUNA: CASTER 47 the shell interior appears to be smooth or only very. slightly striate. The following measurements are based on the holotype: Dimensions.— Hinge width 20 mm. Median Icngth shell 19 mm. Maximum width ventral museles 9 mm. Radial length ventral museles 6.5 mm. Length adductors 5mm. Discussion.—Megastrophia pygme@a may prove upon further study to be either a variant or ontogenetic stage of the preceding species. The structural features are highly distinctive, however, and are certainly of specific value so far as the material at hand will indicate. The absence of any shells intermediate in size be- tween the gigantic and probably gerontic Megastrophia hopkinsi and this small shell is, of course, hard to explain in terms of ontogeny. Until more can be learned of the life habits of these organisms, this condition cannot be properly evaluated. Too fre- quently such an association is taken to mean genetic distinctness, when it may only indicate an unusual adolescence. Perhaps the best contrast in internal structures is brought out by the natural size figure of a plasticine cast of Megastrophia hopkinsi and an enlarged one of the present species shown on plate 3, figs. 19 and 17. The surface details of the small species are very different from those of the large one, as even casual scrutiny will show. A comparision of plate 5, fig. 1, with the internal mold upon which some of the surface features are im- pressed (plate 4, fig. 1) demonstrates this very well. Plate 8, fig. 14, also shows surface details of the small species. There is nothing in “austral” faunas with which this species might be compared since the entire group is new to the Southern Hemisphere. In North American faunas “Brachyprion” schu- chertana Clarke, 1909 (e.g. pl. 9, fig. 10), from the Dalhousie beds, has somewhat similar ornament, and is comparable in size. It also shows the same irregularity of anterior outline that is found in the Colombian species. Clarke’s species is as little known as the Colombian, and further comparisons are really of small value. Types.—Holotype: Pal. Res. Inst. No. 5398-5398A, external and internal molds of a ventral shell; paratypes: Nos. 53985, C. D, and 5405A. 48 BULLETIN 83 148 Genus Cymostrophia Caster, n. gen. Genotype.—Leptena stephani Barrande. Devonian F2 (Coblenzian) Bohemia. Cymostrophia schucherti Caster, n. sp. Plate 1, figs. 14-17; Plate 2, figs. 1, 4-6; Plate 3, fig. 2; Plate 6, figs. 7-10; Plate 10, figs. 4-6 Large shells, strongly concavo-convex; hinge line greatest ~ width of shell and slightly produced in acute, subacute or rounded auricular extensions; width somewhat greater than length; sides constricted immediately in front of the hinge and regularly round- ed toward the front of the shell which is also regular. The valves appear to be very closely appressed in life, and therefore have nearly opposite contour. Posterior seven-eighths, or thereabouts, of the shell is very nearly planar, (1.e. only slightly convex in ventral valve and slightly concave in dorsal one) anterior fraction of shell is abruptly but arcuately geniculated, and may be even very slightly constricted in life. Most shells as now recovered are considerably constricted peripherally. Some shells appear to have been more regularly rounded throughout and may have been even slightly ventricose. Other shells are slightly produced anteromesially. There are shallow sinuses in the ventral valve extending from the margin of the shell at the lateral constrictions toward the beak, which thus separate the hinge zone from the rest of the shell. The hinge zone is regularly convex in the ventral valve, and rises slightly above the general contour of the shell. The palintropes are very narrow, and detailed features are not known. Both valves are broadly regular in contour, but are differ- entially corrugated in a manner recalling the fabric seersucker. (See plate 1, figs. 14-17.) This shell structure is brought out by the interruption of relatively regular concentric corrugations by strongly elevated primary radii. The corrugations are propor- tionally much closer spaced on the earlier portions of the shells than on ephebic portions, and may be completely absent on the geniculated zone, although usually not. In some shells the cor- rugations are very indistinct, and in others differentially so. The corrugations, whatever their degree, usually affect the whole shell. 149 COLOMBIAN DEVONIAN FAuNA: CASTER 49 External ornamentation consists of alternating radii (e.g. p. I, fig. 15; pl. 6, ig. 8) composed of strong elevated coste. About 14 principal coste are present on the earliest part of the beak, and increase by intercalation. New coste appear to arise each as a fine stria in the midst of a narrow furrow. Approximately 28 of the primary coste constitute the chief ornament over most of the shells. Toward the zone of geniculation this number is approxi- mately doubled, and on the geniculated zone the number may be again doubled. The primary radii increase in strength toward the front of the shell, but very gradually. This means that the initial set is always very slightly stronger and more prominent than the later appearing radii. Petween the primary ribs in slightly de- pressed interspaces are 14 or 15 fine elevated striz which appear also to increase by intercalation, but apparently do not become primary coste toward the front. The striae seem to double in number somewhat 1n advance of the intercalation of new costz, so that as soon as a new costa is visible, it has the usual number of striz separating it from adjacent coste. In some shells (e.g. 5450, pl. 2, fig. 5) the fine striz are not well developed, and the intercalation of primary coste is much more rapid, and their final number much greater than average. Very fine concentric striz are sometimes preserved as shown on plate 6, fig. 15. Usually they show best on aberrant examples in which the fine radial strize are indistinct. They can be detected in some specimens with a lens as tiny, discontinuous acicular extensions from the sides of the microstriz. The interiors of both valves are negatively striate, i.e. incised furrows correspond to the principal coste on the surface. The finer radial ornamentation does not leave an impression on the interior. The differential corrugations of the shell show on the interior, though less distinctly than on the exterior, due to a slight amount of internal secondary calcification. The central region of the shell is granulose or punctose, especially adjacent to the muscle platforms. The hinge is essentially holocrenulate, except for the mesial zone. The crenulations are borne on a tri- 50 BULLETIN 83 150 angular zone on the floor of the dorsal valve and on a triangular thickening of the ventral palintrope. In the dorsal valve the cren- ulations cease abruptly at the crural plates which are strongly de- veloped on the posteriorly upbent surface. (See plate 6, figs. 9, 10.) In the ventral valve the crenulations cease at the hinge teeth. The denticles decrease in size toward the hinge apices, and are essentially normal to the hinge line throughout. The intericr of the ventral valve is dominated by large, flabel- late diductor muscle scars which completely enclose the adductors. The musculature reaches at least half way to the anterior margin, and sometimes a little further. The diductor scars are recessed behind and are strongly delimited postlaterally by broad callous ridges rather than extensions of the dental lamellae. From the line of their greatest width forward and around the front margin they are indistinctly outlined, and in some forms merge incon- spicuously with the floor of the valve. The surface of the scars is marked by radial lines, some of which extend forward beyond the general confines of the scars. Anteriorly the diductor scars are separated by a low median septum which seems to be an ex- tension of the forwardly tapering lenticular adductor platform. The adductor muscles are attached to a subfusiform arbores- cently ridged zone which is located mesially in the hind third of the shell. Like the diductor scars, the adductors are depressed below the general level of the adjoining shell (in this case the diductor scars) posteriorly, and rise anteriorly on a platform which stands at a slight elevation above the diductor level. On the midline the rear adductor scars are separated by a deeply in- cised fossa which widens posteriorly, whereas the anterior ad- ductors appear to be separated by a very slight mesial carina which extends forward as the septum between the front of the diductor scars. The arborescent zone seems to give way to an irregular surface beneath the anterior adductors. Postlaterally the adductecr scars are bounded by broad swells of the diductor floor, and immediately in back by forwardly diverging carine apparently originating at the base of the ventral “process”. Of the last structure, little is shown in our material, but apparently it is present, well developed as an articulatory structure fitting between the prongs of the ventrally introduced dorsal process. 151 CoLOMBIAN DEVONIAN FauNA: CASTER 5) The hinge teeth are short, stubbed, but apparently not serving any articulatory function, since no corresponding sockets have been noted in the dorsal valve. These structures are shown by plate 2, fig. 6. The interior of the dorsal valve is dominated not by muscula- ture, but by the ruggedly developed cardinal process, its appur- tenances and scoriaceous braceplates. The general features of this valve are shown by plate 6, figs. 9g and 10. There it will be seen that the cardinal process is very strong, composed of two essentially terete subparallel prongs which rise from the posterior callus, and overhang the hinge. They are buttressed behind by lamellar extensions against the posterior surface of the valve, thus creating a cavity between the prongs which is a little wider than the width of a single prong. This cavity was presumably oc- cupied in part at least by a “process” extension from the ventral valve. Anteriorly the prongs of the process are supported by a thornlike carina or rounded septum which is subequal to the width of the interprong space at the place of origin and tapers to beyond the mid-point of the shell. Two additional pairs of brace carinz radiate from the cardinal process; a short pair originates along side the median septum at the base of the prongs, and extends for about one-fourth the length of the septum, while a second pair originates at the side of each prong and divergently radiates toward the front, making, if projected, about a 20° angle with the median septum. Subparallel to the median septum, diverging slightly from it in front on either side for its anterior three-fourths are two very much roughened ridges which originate as low crests, rise gradually to stand relatively high near their mid-length and taper near the front, where they extend slightly beyond the median septum. These appear to be homeomorphs of the brace- plates (Sttitzplatten) which probably served as muscle attach- ment sites and may also have had the same brace function as the homeomorphic structures in the Douvillinini, 7.e. they kept the shells far enough apart to make life possible for the very much depressed animal. They have the appearance of being a scori- aceous callous deposit, of rather irregular surface and contour. Further comparison with the douvillinoid structure is hardly profitable. The crural plates are well developed as_ laterally 52 BULLETIN 83 152 diverging ridges on the posteriorly upbent wall of the shell. They probably rested against the ventral hinge teeth as a phase of the articulation. Dimensions.— 5426* 5451 5449 5426A* Estimated hinge width 60mm. 60 mm. 66 mm. 60 mm. Median length shell 41 mm. 54 mm. 55 mm. 41 mm. Length visceral disc 37mm. est. 38mm. 45 mm. o7 mm. Width shell at lateral constriction 55 mm. 50 mm. 62 mm. 55 mm. Median length ventral diductors 29 mm. 24 mm. Radial length ventral diductors 33 mm. 28 mm. Width ventral diductor scars 35 mm. 27 mm. Length ventral adductor scars 21 mm. 20 mm. Width ventral adductor scars 8 mm. 5mm. Distance between hinge teeth 15+ mm. 15 mm. Length dorsal median septum 22 mm. Maximum width process 6 mm. Between prongs of process 2.5 mm. Between crural plates 12 mm. Length pseudo-braceplates 16 mm. Discussion.—The shells in this genus, although very large, are exceedingly thin; the primary coste appear to be delicate radial corrugations of the shell, whereas the intermediate striz are ap- parently formed by additional calcification. This character of the ornament may prove to be a generic feature. Hall and Clark, 1892 (p. 286), thought of Stropheodonta as without, or at most with obsolescent, cardinal teeth and dental plates. Certainly this Colombian species could hardly be so classified. Plate 10, figs. 4 and 5, shows a ventral mold and cast which exhibit a more numerous type of surface radii. In this form there is less contrast in the strength of primary and secondary radii. It also shows more delicate differential corrugation than any other Colombian material. In this form also the hinge is slightly extended and there is a suggestion of a median fold on the * Holotype 153 COLOMBIAN DEVONIAN FAUNA: CASTER 5: ventral valve. It may, when better known, constitute a distinct variant of the main species. Cymostrophia schucherti may be distinguished from the various stropheodonts in the Colombian or other South American deposits by the differentially corrugated and strongly concavo-convex shells; by the strong dorsal process and douvillinoid pseudo- braceplates ; by the nearly holocrenulated condition of the hinge which it probably shares only with ’Cymostrophia waringi, be- low ; and by the large flabellate ventral diductor scars which are incompletely delimited. The closest resemblance (possibly homeomorphic) to the Col- ombian species seems to be the northern Stropheodonta patersoni Hall of authors in America and Stropheodonta corrugatella (Dav- idson) or Stropheodonta stephani (Barrande) in Europe. That a great many species as yet undescribed are still lurking under these names in literature seems very clear. Stropheodonta reticulata Stainbrook, 1938, is a species which has previously been concealed in this manner. Stainbrook’s species appears to be assignable to Cymostrophia. S. patersoni Walcott, 1884, may be conspecific with Stainbrook’s species, or, more likely, may be still another representative of the genus. At a distance it appears that Stroph- eodonta corrugatella Gosselet, Barrois, etc., 1920, of the Liévin Lower Devonian may be an early representative of the genus, which is probably specifically distinct. The relationship of Stro- pheodonta inequiradiata Hall is discussed under ?Cymostrophia waringi, below. But it should be noted here, that virtually all “boreal” species assigned to S. patersoni or S. inequiradiata, and to a limited extent to S. corrugatella, have recorded representa- tives which show rather well developed intercostal reticulations, which our Colombian material does not show. The internal characteristics of the dorsal valve of the reticulated Colombian species, assigned below to a new genus Dictyostrophia, are very different from Cymostrophia. It seems not unlikely, however, that in many cases representatives of these two stocks have been lumped together under a common name. Types.—Holotype: Pal. Res. Inst. No. 5426-5426A (outer and 54 BULLETIN 83 154 inner mold of dorsal valve) ; paratypes: Nos. 5427, 5428, 5449, 5450, 5451, 5451A, and 5460A. ?Cymostrophia waringi Caster, n. sp. Plate 2, figs. 18, 14; Plate 3, fig. 1 Shells large, concavo-convex, transversely scutelliform; hinge straight, apices somewhat produced in rounded auricular exten- sions, separated from the main body of the shell by lateral con- strictions ; periphery in front of the constrictions regularly round- ed, and usually not centrally produced. On the anterior lateral and front margins the shells are slightly geniculated. Contour of the shells regular. The exterior surface bears radial ornament comprised of alternating costellz and fine strize, which run very regularly from apex to margin, and show virtually no undulation or curvature. The radii arise as rounded ridges on the surface, and whether they are solid lines of accessory calci- fication, or miniature corrugations 1s not known. ‘Their number is increased by intercalation. They very slowly increase in size toward the margin. The spaces between the principal costellee are essentially flat, or only very slightly excavated, and are orna- mented with very much finer micro-ornamentation. This con- sists of low-lying convex radii which usually number from 8 to 10 between the costelle, and increase also by intercalation. These features are brought out by the illustrations. (See plate 3, fig. 1.) Crossing the finer striz, but interrupted by the costelle are ex- ceedingly fine and closely spaced concentric striz which are also elevated. This creates a very fine grid pattern over the better preserved interspaces of the shell. Apparently, however, not all regions of every shell had this fine reticulation, and some individ- uals show it not at all. The ventral palintrope is apsacline. De- tails of the hinge crenulation and shell interiors are not known. Dimensions.— 5448 5448B Hinge width 64 mm. 60 mm. Median length shell 42 mm. 40+ mm. Width shell at lateral constriction 50 mm. 56 mm. Discussion.—Obviously, it is impossible to determine on the basis of the material at hand, the precise generic relationships cf the present shells which seem clearly specifically distinct from 155 COLOMBIAN DEVONIAN FAUNA: CASTER On or the other Colombian stropheodonts. Almost exclusively on the basis of form they are tentatively assigned with some doubt to Cymostrophia, The internal features will be diagnostic. The surface ornament differs from the genotype and C. schucherti, above, in the absence of any sign of differential corrugation, and in having fewer striz between costella. The shells differ from Cymostrophia dickeyi in being less transverse, more inflated, and more pronouncedly concavo-convex, and also in having no sign of corrugation. With Dictyostrophia cooperi, this species shares intercostal reticulation, but the contrast is much the same as be- tween a fine muslin and a monks cloth, so much coarser is this feature in Dictyostrophia, which moreover has_ fasciculation of the radial elements, and subangular to angular crests to the fascicles in most cases. When the internal features of the present form are known it may prove to be more closely allied to Dicty- ostrophia than now appears. (On the surface of the holotype of this species is attached the external mold of what appears to be a new species of auloporid coral. The details of this cnidarian are too poorly preserved to make an analysis worthwhile at this time. It is an important bit of ecological data, however, since it is one of the only two ceelenterates known as yet from the Colombian deposit, the other being the calicinal imprint of a small tetracoral. I have tended to view development of Cymostrophia schucherti, ?C. waringi, C. dickeyi and possibly Dictyostrophia cooperi in South America as a counterpart of the yet unsatisfactorily classi- fied group in the Schoharie and Onondaga of North America which have ordinarily been assigned to Stropheodonta patersoni, S. inequiradiata and variants, wherein occur most of the external variations recorded for the Colombian faunule. The curious douvillinoid internal features, especially of the dorsal valve, sug- gest isomorphism on a grand scale. Types.—Holotype: Pal. Res. Inst. No. 5448; paratype: No. 5448B. 56 BULLETIN 83 156 Cymostrophia dickeyi Caster, n. sp. Plate 3, fig. 20; Plate 4, figs. 3-8; Plate 8, fig. 15 : Shell of medium to relatively large size, subplanoconvex or concavo-convex ; only moderately inflated at most; usually both valves relatively flat; hinge line straight, greatest width of shell; slightly produced at extremities in rounded processes which are separated from the main body of the shell by lateral constrictions. Length usually about two-thirds the hinge width; anterior margin transversely rounded, and usually somewhat irregular. Anterior part of ventral valve slightly upbent, and dorsal valve correspond- ingly deformed, for the valves lie in close proximity throughout. The surface is regularly striate. Relatively coarse cord-like coste separated by 9 or Io finer costelle cover the entire surface in the stropheodont manner. The coste are recessed in a fur- row for their entire length, or until the extreme geniculation zone is reached. The intercostal zone is, on the other hand, convex upward. This gives the surface a curiously inverted aspect when compared with other stropheodonts in the fauna. The principal coste originate as fine costelle in the bottom of grooves on the apical portion of the shell. The interspaces, even on the early part of the shell, are finely costellate. New principal coste orig- inate in an intercalary manner out of furrows that develop in the midst of an intercostal space. The new costz appear all at essen- tially the same stage of growth over the shell, and concomitantly with them new costellz appear in an intercalary manner also, so that almost everywhere the number of intercostal costelle is essentially the same. This regularity of surface sculpture is a very characteristic feature of this species. On the upbent portion of the shell, at least on the dorsal valve, the coste rise out of the bounding grooves in which they lie elsewhere on the shell, and constitute the cordate crest of fascicular bundles of costellz. In other words, the relative convexity of the coste and costellz is reversed on this zone. Here also the costelle are irregular and wavering, which is in marked contrast to their regularity and straightness elsewhere. Over the entire surface fine varices of growth appear as delicate elevated striz. They are so faint that they are invisible to the unaided eye and nowhere give a cancel- 157 COLOMBIAN DEVONIAN FAUNA: CASTER ~ Or late effect. The hinge margin of the dorsal valve, upper surface, is curiously concave, even at the beak, and the upper edge of the dorsal palintrope is elevated above the surface as a carina with rounded keel. What the nature of the palintrope surface may be is not known. Nature of the delthyrium is also not known. The interior of the dorsal valve carries unusual features which seem to merit separate generic recognition for this fossil and congeners. These characteristics have been outlined under the foregoing generic analysis. The shell is exceedingly thin, and the surface ornamentation is seen from the internal mold to be truly corrugated, for the interior of the shell carries the negative form of even the finest surface radii nearly as distinctly as the external mold does the positive. The muscle markings are small, inconspicuous, and with poor delimitation. There is apparently no accessory calcification under the beak in the form of a muscle seat, for the surface radii are discernible over essentially the whole muscle area, but the postlateral region about the muscle scars is finely granulose. The two prominent subparallel “braceplate” ridges extend for- ward from the bases of the pillars of the cardinal process. They are granulose on their crests and likewise the zone between them is granulose. This median zone is divided by a septum of less elevation than the “braceplates”, but subequal to them in length. It is fused posteriorly to a slightly elevated callus which surrounds the bases of the pillars of the process and the termini of the paral- lel ridges. The central triad of ridges reaches about one-third of the distance toward the front margin. The cardinal process rises perpendicularly from the floor of the valve and posterior callus. It consists of two subparallel elongate pillars, triangular in cross section, and excavate posteriorly, which slightly overhang the hinge. The excavate terminus of each pillar is slightly crenu- lated. The space between the pillars is about one and a half times as wide as a single pillar. The adductor muscle scars are very faintly elevated from the floor of the valve. They are well defined on their postlateral margins by widely divergent ridges which extend outward from the bases of the cardinal process to the anterolateral corners of the scars. They are elsewhere sans 58 BULLETIN 83 158 limitation. The postlateral margins of the scars are slightly exca- vate. The front margin of the scars is not well defined, but in some shells has a truncated appearance and in most grades into the granulated pallial zone. The inner edge of the adductor scars is distinct as far as the junction with the cardinal callus, and is essentially a straight line. The surface of the scars is radially ridged. Sectors which are nearly as large as the scars them- selves separate the outer muscle scars from the parallel carine. These sectors are striate and granulose after the same manner as the scars and may also have been the scene of muscle attach- ment. In back of the adductor scars, and slightly less divergent than they, two well developed subangular crural ridges appear to originate at the hinge and continue forward to the anterior side of the postlateral reentrant in the adductor margin, where the ridge appears to fuse with the carina at the margin of the scars. The shallow posterior depression between the bounding carina and the crural ridges may represent a socket zone. Hinge crenu- lation occurs only on the central third of the hinge. The crenu- lations are weak and closely spaced under the beak, and disappear by gradual diminution laterally. The palintrope does not rise internally above the floor of the valve, and the crenulation sockets are inset below the general valve level. Externally the palintrope has been forced upward on the posterior margin as a vertical ridge. There is, therefore, no inner palintrope surface, and no posterior cavity. The dorsal valve was strictly opercular. Dimensions.—The following measurements were taken from two external dorsal molds and one internal dorsal mold. All are incomplete, and the dimensions are approximations. 5444 5437-5438 Width hinge 60 mm. 50 mm. Length of shell 35 mm. 32 mm. Visceral length 30 mm. Length median septum 13 mm.. Length parallel carine 12 mm. Radial length adductors 6 mm. Maximum distance across muscles 10 mm. Width cardinal process 1.75 mm. Total length hinge ecrenulations 17 mm. Crural angle 120° Cardinal boundary angle 125° Angle inner adductor boundaries 65° Angle pseudo-braceplates ice 159 CoLOMBIAN DEVONIAN FAUNA: CASTER 59 Discussion.—This species differs from ?C. waringi in being much more leptostrophid in its shell relations and in having a much weaker cardinal process, an imperfectly crenulated hinge, and obsolescent differential corrugation. The strongly developed dorsal adductor scars are also very characteristic in this species, whereas they are not clearly defined in previous species, above. The surface markings in the present species are highly charac- teristic. In the delimitation of the costelle by adjacent furrows we may have an example of arrested development, for in Cymos- trophia schucherti it was pointed out that the principal costellz originate in the midst of furrows, but soon all signs of the initial groove are lost. Some individuals in C. dickeyi show a tendency for the costelle to form crests of subangular corrugations of the shell in much the same manner as seen in Dictyostrophia. It would appear in the few cases observed in C. dickeyi that the phenomenon might be attributed to lateral compression during fossilization rather than initial corrugation. In no cases have any concentric varices been noted in typical examples of this species. In 5438A, a dorsal mold, the usual frail trident is shown and also slight differential corrugations. The obviously incom- plete and exceedingly weak hinge crenulation also points to the possibility of arrested development, as does likewise the delicacy of the cardinal process. The affinities with “boreal” forms here again probably lie in the Stropheodonta patersoni—S. inequira- diata complex. With the “boreal” douvillinoid Stropheodonta inequiradiata (Conrad) as interpreted by Hall, 1867 (e.g., p. 106, pl. 18, figs. 2a-2k), there is considerable similarity in the structures of the dorsal valves. This last “boreal” species is as- signed as genotype of a new genus in the forthcoming study of the Stropheodontide mentioned above. The details of the brace- plates, musculature, ornament, and size preclude the possibility of the Colombian species being assigned to this genus. Homeo- morphy is the probable explanation of the resemblance. Types.—Holotype: Pal. Res. Inst. No. 5437-5438, internal and external dorsal molds of one shell; paratypes: No. 5444, an ex- ternal dorsal mold; 5438A, an external and internal dorsal mold; also 5446, not illustrated. 60 BULLETIN 83 160 Genus Dictyostrophia Caster, n. gen. Genotype.—Dictyostrophia cooperi Caster, n. sp. Colombian Devonian. The generic analysis is given above. Dictyostrophia cooperi Caster, n. sp. Plate 2, figs. 7-12; Plate 4, fig. 9; Plate 6, figs. 1-6 Shells concavo-convex, transverse; hinge width greater than maximum length of shell; hinge termini auriculate, and rounded; anterior third of dorsal valve pronouncedly geniculate; the peri- pheral upbending continuing to near the hinge extension on both sides where a very sharp demarcation occurs in the form of a lateral constriction of the shells. The “visceral disc” is appar- ently somewhat more depressed at its front margin than else- where on the shell. Surface of the dorsal and ventral valves prominently and an- gularly plicated posteriorly, and possibly on the venter through- out (5455). Each angular corrugation is surmounted by a cord- ate costa. The angularity of the corrugations gives way on the earliest part of the shell to mere rounded fascicles; when about half grown the angular corrugations become most prominent, and in the ephebic condition the addition of many intercalary coste, each at the apex of a smaller angular corrugation, makes the pli- cation less apparent. On the apical portion of the shell there is a curious reversal of corrugation: the principal coste, which number between 16 and 18, appear first as very faint ridges in apical incised striz or grooves between which there are convexi- ties covered with very fine radial costella. About 5 mm. in front of the apex the faint ridges rise out of their grooves. Each ridge is for a short distance bounded on either side by a shallow con- tinuation of the groove as delimiting furrows. The ridges soon attain such elevation that the bounding furrows are mere shallow concavities at the base of each costa, on the angular slope which descends to the middle of each intercostal zone. The intercostal zones are angularly excavated except at the places where costze originate. The entire intercostal surface is crossed by fine costel- le. The grooves from which the principal costelle emerge con- tinue to the terminus of the beak, and likewise the very fine inter- costal striz, some of which reach the size of secondary costz at later stages of shell growth. It appears, therefore, that a large 161 COLOMBIAN DEVONIAN FAUNA: CASTER 61 part of the radial ornamentation of the later shell is represented in miniature on the earliest preserved portion of the shell. So far as can be deduced from the material at hand, the only radial elements to appear in later growth stages are intercostal striz. These are not numerous. Progressive strengthening of the al- ready present structures appears to have been the rule. The strengthening proceeded differentially, however. The striz lying in essentially the middle of the intercostal spaces each time re- ceived the strengthening and this rather abruptly, essentially synchronously in all interspaces, and at periodic stages of shell growth, rather than constantly and variably. The effect is that ef periodic intercalation. Apparently until the stage of growth was attained when a costella was to be enlarged into a costa, it maintained a rather constant size after the first few millimeters of growth. After enlargement began it continued gradually to increase in size. Costellae were much more rapidly advanced into coste from the zone of geniculation outward than on the area of the “visceral disc”. On the geniculated region of the shell so many of the costella have been enlarged into small coste that the angularity of the paucicostate juvenile zone gives way to a low- angled multifasciculate condition, where only a few costelle exist between coste of various sizes. On the extreme front mar- gin of adult shells there is a uniformly multicostate ornamenta- tion (of variably sized coste, however), without corrugations or fascicles. The costellz are not so straight as the coste. They usually show a uniform irregularly wavering tendency. Over the whole post-embryonic shell there are found strongly ridged varices which are most prominently developed in the spaces between cost, but are represented on the coste themselves as slightly elevated knobs and bosses of regular spacing. In the interspaces the varices are subequal to the costellee and of about equal spacing, although somewhat more irregular. The inter- sections of the varices and costellze are faintly nodose. The whole dictyate effect is rather similar to a loosely woven textile, such as cheese cloth. The prominence of the reticulation appears to vary slightly with individuals, and may be greater on the ventral valve than on the dorsal. . 62 7 BULLETIN 83 162 The nature of the palintrope, dentition and crenulation of the hinge are not known. The dorsal interior shows certain highly interesting characteristics. The whole interior of the shell is coarsely corrugated or radially striated, the depressions being along the raised lines of the surface. The whole surface is finely pustulose or granular, and very strongly so in the posterior part of the shell. There is an internal callus “diaphragm” at the line of upbending of the shell. This callus mass is thickest and of greatest length near the central part of the shell. In back of the diaphragmal callus the shell is concentrically excavated and strongly ridged. Subumbonally the muscle scars are embedded in a broad callus, the sloping sides of which are coarsely and abun- dantly pustulose. The actual places of dorsal muscle attachment occupy a very small area in the postcentral part of the shell. The adductor scars are correspondingly small, elongate, ellipsoidal areas which are separated by a median carina. The scars are well delimited postlaterally by callous ridges which diverge from the base of the cardinal process. The scars are most elevated above the floor of the valve toward the front, where they are delimited by a pre- cipitous slope. A reentrant mesial zone between the anterior extension of the muscle seats is deeply excavated and mesially divided by a septum which appears to be the discontinuous exten- sion of the median septum. The anterior septum extends to the beginning of the diaphragmal callosity. The sides of the exca- vated zone are slightly elevated above the general slope of the muscle callus. The zone occupies an ellipsodial area which ends with the anterior median septum at the diaphragmal callosity. This excavation may represent the attachment seat of the anterior adductors. The posterior pair was presumably attached along the median septum. The cardinal process appears to be pedunculate and to rise directly from the floor of the valve. The prongs of the process are laterally divergent. Crural characters are not preserved on the material. Dimensions—The following measurements were taken from the holotype, an internal and external mold of a nearly perfectly preserved ventral valve (5449A and 5445). 163 COLOMBIAN DEVONIAN FAUNA: CASTER 63 Width hinge 70 mm. Length of shell 36 mm. Hinge extension beyond visceral dise 17 mm. Median length visceral dise 29 mm. Width muscle scars 13 mm. Radial length adductor sear 8mm. Length posterior median septum 7 mm. Total length median septum 14 mm. Width eardinal process 12 mm. Width between sockets 12 mm. Width anterior median depression 5mm. Discussion.—Without going into the generic features of this interesting reticulated species in any detail, it must be noted that certain gaps in our knowledge, most notable of which is lack of information on the hinge crenulations, sockets and process, make absolute assignment hazardous. The general contour of the shell, obvious concavo-convexity and unequal striz bespeak northern forms usually assigned to Stropheodonta concava Hall (e.g. Hall and Clarke, 1892, pl. 14, fig. 19), especially as developed it. the North American Onondagan and Hamilton. Some of the struc- tural features of this species have already been discussed in pre- ceding pages. In the “boreal” species the very similar, albeit smaller, diaphragmal callosity is present; also narrow, ellipsoidal posterior adductor scars, and recessed anterior adductor scars, both separated by a median septum. Similarly in S. concava the process is broadly bi-pronged and the dental sockets relatively close together at the base of the process. The visceral disc por- tion of the shell of S. concava is coarsely ridged as in the Colom- bian species. Externally the profiles of S. concava (e.g. Hall’s 1867 illustration, pl. 14, fig. 2a) and Dictyostrophia cooperi are almost identical. Significant in the comparison are the alate hinge extensions. The surface markings especially point to the relationship between the two species, for both show similar sur- face corrugations, but the “boreal” form, although having closely crowded concentric striz, does not ordinarily have a reticulated appearance. This last feature seems to be a Colombian specializa- tion within the group, or a quite independent development in a stock which more or less paralleled the S. concava developments in other respects. It is not unlikely, however, that this line has a North American expression in such small forms as the one listed by Hall, 1867 (pl. 12, fig. 11), as Stropheodonta patersoni? from 64. BULLETIN 83 164 the Lower and lower Middle Devonian of New York State. The specimen illustrated in this connection does not have the well- wrinkled aspect of typical S. patersoni, and probably represents a distinct specific development. The surface coste are angularly elevated as in the Colombian species, and the entire surface is finely cancellate after the general manner of our specimens. Walcott, 1884 (p. 120, pl. 2, fig. 11a), described and illustrated a reticulated specimen from the Lower Devonian of Combs Peak, Nevada, assigned to Stropheodonta inequiradiata Hall of the New York Helderberg strata. Walcott’s specimen has surface reticu- lations which are very reminiscent of the Colombian material, but are inadequately described for detailed comparison. The types of Walcott’s material are not especially helpful, but certainly a distinct species was involved. It is an interesting occurrence parallel to that of the eastern American Helderbergian, where forms assigned to S. inequiradiata and S. patersoni apparently intergrade. The Colombian fauna seems to echo this same asso- ciation. Before closer comparisons can be made, a restudy of the North American Lower Devonian stropheodont fauna is essential. Types.—Holotype: Pal. Res. Inst. No. 5445 and 5449A, inter- nal and external molds of one individual; paratypes: No. 5455, a partial external mold of a ventral valve; 5445A. THE TRUE STROPHEODONTIDS The crenulate-hinged brachiopods having the general char- acteristics of Stropheodonta demissa (Conrad) are an entity apart from the rest of stropheodontids. The group appears to embrace structural lines that exceed the bounds of a single genus. It may develop that this aggregate will constitute the Stropheodontini in full, when a thorough revisory study is carried out. The genus Stropheodonta, as used below, in the strict meaning of the geno- type, will apparently not cover the entire group of species now de- scribed or describable from the Lower into the Upper Devonian, and at one time or another for the most part referred to the geno- type. Whatever may be the outcome of the true stropheodontids, only the genus in the strictest sense has any immediate bearing. 165 COLOMBIAN DEVONIAN FAUNA: CASTER 65 Genus STROPHEODONTA Hall, 1852, s.s. Genotype.—Strophomcna demissa Conrad. Middle Devonian. The genus Stropheodonta in the restricted sense of this paper is typified by the genotype Strophomena demissa Conrad. It rep- resents an end development apparently out of an as yet undeter- mined brachyprionid stock. (See discussion of Mclearnia, above. ) Stropheodonta, s.s., is in the adult stage characterized by a holo- crenulate hinge. It seems likely that all species of true Stropheo- donta, like the genotype, possess no foramina in the palintropes and usually no evidence of even deltidial plates. The genotype and its relatives have always an exceptionally high dorsal palin- trope. The ventral muscle scars spread flabellately without car- inate delimitation. The dental lamellae are weak or obsolescent. The ventral bifid process is well developed and fits between the divergent prongs of the dorsal cardinal process which rest in cavi- ties or even pits under the ventral beak. A posterior median septum or carina separates the elongate diductor muscles of the ventral valve. The process of the dorsal valve is sessile, the prongs rising from the subumbonal callus as parallel pillars which are posteriorly directed. The adductor scars are reniform and imperfectly delimited by callus deposits. A median septum ex- tends forward from the posterior callus to about the middle of the shell. The crural plates are obsolescent and apparently were never functional in maturity in Stropheodonta, s.s. The ado- lescent and ephebic portions of the genotype externally bear sub- equal alternating radii which arise principally by bifurcation, but occasionally by intercalation. Costellze are apparently not present in the group. The alternating radii are made up of mature and newly formed ones rather than consistently fine and coarse radii. In the genotype and related species the earlier part of the shell may be angularly corrugated or coarsely costate. Warthin and Cooper, 1938, have shown that the strata bearing Stropheodonta demissa, s.s., do not occur west of New York State, and Cooper, 1938 (personal communication), has reported it quite unlikely that any of the specimens in the many reports of this species else- 66 BULLETIN 83 166 where are truly conspecific with Stropheodonta demissa. The use of the species name in the past has been virtually generic, and it is highly improbable that reports of the species in more distant places can be otherwise interpreted. Stropheodonta, s.s., appears to be found only in the Middle and lower Upper Devonian, the world over. The first definitely known representatives in South America occur in the Colombian fauna. Stropheodonta kozlowskii Caster, n. sp. Plate 3, figs. 3-15; Plate 5, figs. 14-19 This is a small to medium sized, symmetrical, concavo-convex brachiopod with a straight hinge line, produced extremities, and slightly greater width than length. The ventral valve is mesially most gibbous along an anteriorly expanding zone which has the aspect of an undefined fold (in some individuals very clearly defined, however) of which there is a counterpart in the dorsal valve in the form of an undefined sinus or mesial depression. The outline of the shells is essentially shield-shape; the periphery is slightly constricted just in front of the auriculate hinge extensions and thence is regularly curved toward the front. The front mar- gin is subcircular, and only slightly produced mesially at the shal- low sinus and fold. The greatest height of the ventral valve opercular in function ; the beak is very low, and rises only slightly occurs at about the mid-point of the shell. The dorsal valve is above the hinge line as the only convexity of the shell. Over much of the surface the dorsal valve is essentially flat, but bears a shallow mesial sinus-like depression without conspicuous delimit- ations. At the front margin the shell is slightly up-flexed, thus indicating the recessed nature of the valve. The characteristics of the hinge margin are not known; likewise the internal features. The surface markings constitute the chief character for recog- nition of this species in the Colombian fauna. These consist of many elevated, rounded or subangular radii arranged in fascicles, each with a common origin out of a series of original radu on the beak which merge into a single mesial swell at the extreme pos- terior part of the beak. The radii increase in threes, by trifidy ; the mesial part of each radius is much larger than the newly arisen lateral branches which split off on either side at the same time. 167 CoLOMBIAN DEVONIAN FAUNA: CASTER 67 At first the newly arisen branches are finer than the parent radius, but within a distance of about one-third the total length of the adult shell the lateral branches attain the same size as the parent, whereupon they give rise each to a pair of new lateral radii, but the “parent” radius in each case appears never to branch more than once. The rate of subdivision of the radit was much more ra- pid on the earliest part of the shell, and appears to have decreased as the individual matured, thus on most shells it appears that there are 15 principal radii all originating in equal development on the beak, yet in some specimens it can be seen that these are really two fascicles of seven each with a radius of equal strength in the middle. The radii appear to be of an alternating type only in those zones of growth where new ones have recently branched off. Definitely none of the shells has shown an intercalary origin of the radii; the barest approach to this comes where a varix appears to interrupt the juncture of the tiny nascent radii and the parent radius, but in all cases the paired nascent radii converge notice- ably and for some distance lie on the slope of the parent radius. Having given rise to two lateral radii, the mesial or parental radii become the most elevated crests of fascicles in the ventral valve. In the dorsal valve this condition is not so prominent, and in fact, may well be reversed. Certainly the mesial swell of the ventral valve and mesial depression of the dorsal precisely conform to this scheme. In the dorsal valve there is a tendency for the pri- mary or parental radii to give rise to essentially full-fledged lat- eral ones of essentially the same strength as the mesial one. This gives a curious tritonlike aspect to the ornamentation on the more mature parts of the shell. Branching on all radii appears to have occurred at essentially the same moment at regular inter- vals in the growth of the shell. The interspaces between radii are consistently narrower than the radii, usually being about half as wide, and are ordinarily essentially flat-floored. Concentric varices occur with no regularity on the surface. These are most prominent on the dorsal valve in our material. Concentric orna- mental lines of great fineness are present over most of the shell surface, but can be seen with a lens best on the postlateral areas. They have most consistent development on the flanks of the ele- vated radii. The surface never has a cancellate aspect. 68 BULLETIN 83 168 Dimensions.— 5429* 5430 5431 Width hinge 26 mim. 26 mm. 28 mim. Length of shell 18 mm. 26 mm. 28 mm. Anterior width fold or sinus 10 mm. 9 mm. 8 mm. Discussion.—In the absence of internal features and palintrope structures, it is admittedly impossible definitely to assign this in- teresting species to genus, but the preponderance of characters seems to lie with the true stropheodonts. Of these, the “boreal” genotype, Stropheodonta demissa (Conrad) bears the closest re- semblance, so far as I can ascertain. Especially reminiscent are some of the Schoharie and Onondaga forms illustrated by Hall, 1867, in the first satisfactory diagnosis of the genotype. It is in- teresting, however, that none of Hall’s material showed the can- cellate surface and subtuberculated striae which Conrad, 1842, described for the species. The Colombian material is compared with the species as recognized by Hall, 1867, rather than with Conrad’s original material, which may possibly not be conspecific. In size and contour, S. kozlowsku and the early Devonian repre- sentatives of S. demissa (Hall, 1867, pl. 11, figs. 14-16) are very similar, but in surface markings the Colombian species appears to be unique in the trifid nature of the coste and regularity of pattern. Hall’s S. demissa is especially characterized externally by posteriorly prominent radu, originating as nine or ten on the beak and increasing very rapidly by both bifurcation and intercalation toward the front where the surface is covered with very fine sub- equal radii. In some variants the radii remain essentially coarse throughout and in strength compare with the Colombian material. But in general, compared with “boreal” S. demissa, even the fine strie of the Colombian species would appear coarse. No speci- mens of S. demussa, to my knowledge, show trifid radial increase. Until better material from Colombia is obtainable, which may in- dicate other relationships, S. koztowskii is considered as lying close to the genotype line of the stropheodonts, but showing highly individualistic features. Apparently no “austral” material has been described which closely resembles S. kozlowsku in detail. It is likely that the con- * Dorsal valve of 5430. 169 COLOMBIAN DEVONIAN FAUNA: CASTER 69 geners will turn up in material generally listed or illustrated un- der “Orthotetes’, “Schuchertella’ or even “Orthis’, for in gen- eral contour the species is very similar to many forms so identi- fied in Brazil, Bolivia and South Africa, (e.g. Knod, 1908, pl. Bug, 7). Strophomena haferi Katzer, 1903, the types of which have been examined in the New York State Museum (8422/1), 1s pre- sumably a Stropheodonta, s.s., as Clarke, 1913 (p. 291), suggest- ed. The lectotype has a few lines in the external mold of the palintrope which strongly suggest a crenulated hinge. The ven- tral valve only is known of Katzer’s species, and the contour 1s similar to Stropheodonta demissa (Conrad). The surface, how- ever, as impressed on the lectotype, is of a strongly alternating sort, and certainly cannot be closely compared with S. demussa, s. s. The musculature of the ventral interior is also quite dif- ferent from Stropheodonta, s. s. Instead of being flabellate and divergent as in Stropheodonta, s. s., the muscle area is a depressed, narrow, elongate, lenticular zone which reaches far beyond the middle of the valve, and bears mesially a low septum. This in- dividuality, not found in any of the stropheodonts known to me, probably will eventually lead to a separate generic designation for Katzer’s species. The surface radii of S. koztowsku are of much the same appear- ance as of Schellwienella agassizi and S. sulivami. It was with the former species that I first identified the material, but closer scrutiny showed a reversal ( 2e¢. normal state) of convexity and the existence of an obscure sinus and fold in the present species. It showed as well a very different origin of the radial ornament. This species and Schellwienella agassizi have sufficient external similarity to suggest parallel develop- ment, and in the absence of S. agassizi and presence of the usual- ly associated S. swlivani, it seems not unlikely that Stropheodonta koztowskii may well have held in Colombia the ecologic niche usu- ally occupied in the “austral” Devonian by Hartt’s species. Fur- ther exploitation of the Colombian deposit will undoubtedly clarify this point. Hartt’s, 1874, description indicated that sev- eral of his fragmental paratypes of Streptorhynchus agassizi had extended hinge margins, although Clarke, 1913, in reillustrating 70 BuLLETIN 83 170 the species showed no specimens with this outline. Close scru- tiny of Hartt’s material may yield this present form among the paratypes. The trifid increase of radial ornament and small size of the species serve adequately to distinguish S. koztowskit from any of the other stropheodonts described in this report. Types.—Holotype: Pal. ‘Res. Inst. No. 5406A ; paratypes: Nos. 5429-5431, 5431A-E. ?Stropheodonta, sp. Plate 10, figs. 7, 8 A single specimen of an internal dorsal mold of a stropheodont shell is at hand which exhibits features not observed in other Colombian material, and so far as known, not hitherto recorded in ‘“‘austral” faunas. The illustrations bring out the very prom- inent median boss of the shell which occupies the position of an anterior median septum, but may involve a depression of the shell itself, rather than being a mere secondary deposit. The frail sessile cardinal process is shown by the plasticine cast, and likewise the recessed and small muscle scars. The shell may repre- sent a deformed, or even pathologic specimen belonging, possibly, to one of our species described above, but the closely packed radial ornament is not known for any of the already described forms. Illustrated specimen.—Pal. Res. Inst. No. 5448A. Tribe Douvillinini Caster, new The true douvillinids are unknown from the Southern Hemis- phere, and there is little point in discussing them in any detail at this time. Studies now under way indicate that the genus Douvillina, s. s., as determined by the genotype, Dowzillina du- tertrei (Murchison) from the middle Frasnian of Europe, is not known in North America, although it now appears that the prin- cipal development of the tribe occurred on that continent. It furthermore appears that several distinct lines of douvillinid evo- lution are represented in North America, and await generic designation. Several of these are indicated in the preceding list ; others are now in manuscript. The genus Douwwvillina, s. s., ap- parently represents a highly specialized end-product rather than the prototypic line. Material from Moscow (Erian, Muddle lyk COLOMBIAN DEVONIAN FAUNA: CASTER (fal Devonian) referred by Hall to Stropheodonta inequistriata (Con- rad), in Volume 4 of the Paleontology of New York, shows some of the specialized traits of true Douwvillina, but the mass of subse- quent material assigned to the Conrad species from the Hamilton and Tully is quite different in character. Principally, they fail to show anterior closure of the median dorsal pit which Oehlert’s genus apparently requires. Beginning in the Onondagan (possibly in the Oriskanian) small stropheodontids appear in which occur prominent dental plates bearing anterior elongations to border the diductor scars. The junction of the true dental plates and the prolongations (pos- sibly paradental plates) is considerably elevated to form a median apex to each “dental plate’. These apices are often incurved toward the middle of the valve and sometimes almost touch the apices of a secondary pair of lamellz which form along the con- tact of the diductor and adductor scars, and have a midway ele- vation very similar to that occurring in the dental plates and their extensions. The secondary plates between the two sets of muscle scars are here termed the muscular diaphragms. ‘Their apices curve outward, 7. e. laterally, toward the apices of the dental lamellz, and may in some cases join them at their apices to form calcareous rings. On the median line there is also a bladelike median septum which separated the adductor muscles. In de- rivative and gerontic forms the muscular diaphragms tend to dis- appear, or become concealed with secondary calcification which mantles them and the median septum as well. This forms an elevated mesial callus ridge probably for adductor attachment. The derived condition seems to be of generic worth. Derivative forms of this ilk first appear in the Tully limestone (uppermost Erian or basal Senecan), where they have usually been assigned to Stropheodonta inequistriata (Conrad). The dorsal valves show equally striking features, the develop- ment of which correlates very satisfactorily with those of the ventral valve. The evolution of the so-called braceplates or Sttitzplatten in the dorsal valve seems to be one of the principal criteria for generic differentiation in the tribe. These plates originate in the middle zone at the base of the cardinal process and on either side of the elongate but low median septum or plat- form which also abuts the cardinal process. From the very be- 72 BULLETIN 83 2 ginning the plates are differentiable into a posterior and an an- terior half. The posterior portion consistently lies lower, and may in some cases be represented only by a slight roughening of the muscle scars adjacent to the median septum. ‘The anterior halves, on the other hand, are ordinarily built up by granular callus material, and rise very abruptly with a posteriorly concave slope from the hinder halves. The anterior portions of the plates rise highest near their contact with the back portions, and slope off gradually toward the front of the shell. Their upper surfaces are usually roughened slightly, or even grooved in the earlier species. In the dorsal valve, as in the ventral, there is a tendency for the highest elevations on the plates to bend toward each other over the median septum, to which they sometimes fuse, but over which they more often become contiguous, sometimes to form a ring, but more frequently to form a posteriorly tapering calcare- ous half cone which lies on the septum toward the rear. The ex- tensions of the anterior plates usually curve toward the center, but may diverge greatly. In Dowwvillina, s. s., the plates apparently are joined anteriorly by an arcuate callus deposit which thus makes a subcircular pit between them. This is a highly special- ized growth. In most of the earlier species the median dorsal septum reaches forward as far as or beyond the plates, but in later forms the septum stops near the origin of the anterior plates. In a few species a secondary anterior septum appears between the forward prongs, and in Douwvillina, s. s., this forward septum is fused to the arcuate callus and extends to near the front mar- gin of the shell. Contrary to the views expressed by Spriesterbach, 1925, this study of the douvillinid braceplates has lead me to view them with Hall and Clarke, 1892, as scenes of muscle attachment. The hinder half of each plate is the place of attachment of the pos- terior diductors, and the much elevated portion the place of at- tachment of the anterior diductors. That they may also have served some such brace function as Schmidt, 1912, and Spriester- bach, 1925, postulated, is not improbable. The condition of the braceplates seems to be a major generic feature. Ephebic shells, and not gerontic, however, must be examined in order to make ba | 173 COLOMBIAN DEVONIAN FAUNA: CASTER use of almost any of these internal douvillinoid structures, for in the old age condition, much absorption and secondary callus deposition occurs throughout the tribe. Resupination is also ap- parently of generic value in this group. One complication in the study of these shells is the great similarity of the contour and surface details of quite distinct lines of development. Exceed- ingly critical study of the so-called species groups is essential to a clarification of the problem. External homeomorphy, or an extreme case of external conservatism seems to be the rule in the tribe. This running survey indicates the essential generic criteria which seem to be of greatest value in studying and reassigning the douvillinids. It also indicates the nature of the structures in this group which have been mentioned in discussing the homeo- morphic development of douvillinoid structures among some of the stropheodontids above. Tribe Leptostrophini Caster, new The tribe as here recognized is essentially equivalent to the genus Leptostrophia of authors. That the genus was a very plas- tic one, subject to a great number of variations, has been recog- nized from its inception. Williams and Breger, 1916 (p. 26), Clarke, 1913 (p. 286), and others, have shown some of the inno- vations that occurred in the genus, and intimated a little of the part homeomorphy may have played in the derivation of similarly appearing lines of descent. Many of these structural differences appear to be of generic or subgeneric value, and certainly have great stratigraphic value as well. The classification here outlined will be dealt with more fully in an extended work now in prog- ress on the crenulate-hinged Strophomenacea. Only the most striking and important generic lines in the Leptostrophiini are here mentioned. The tribe Leptostrophiini is characterized by the well known set of traits by which the original division Leptostrophia, s. lL, was noted as a group distinct from the typical stropheodon- tids. Of these, the plano-convex shell is the most striking fea- ture, but the presence of a strong, sessile cardinal process at- 74. BULLETIN 83 174 tached to a trifid posterior callus, enormous flabellate ventral diductors, and especially of ventral subumbonal process pits, and fascicular diductor muscles with secondary leaflike septa “muscle diaphragms” between them, all bespeak this as a line apart from the common company of stropheodonts. A Key to the Divisions and Principal Genera of the Leptostrophum A. Surface ornament of the alternating type; consisting of principal costelle which wax and wane to create a subspinose appearance in some cases. (The Nervosa condition.) B. Shells planate, or subplanate; no fold or sinus; shells thin; external ornamentation easily discernable on the interior. C. Shells of moderate size with distinctly nervose external ornament, and no marked tendency for different ornamentation pattern on the StwiOm valves miami ts sae eae Nervostrophia Caster, new. CC. Shells of huge size, very flat; nervose condition obscure; dorsal and ventral valve with slightly different ornament _ WWW... ee ata) nea a ania sh cea NN Vieni ga eae SER RS ““Leptostrophia junia’’ group. BB. Shells concavo-convex, but otherwise with the characteristics of the Leptostrophiini; fold and sinus broadly developed; shells thick and lamiell ateu sa mn UR Lae mcuiae Anon ela Sulcatostrophia Caster, new. AA. Surface ornament consisting of costelle of essentially the same size, and never conspicuously of an alternating type. D. Shells very large, plano-convex; pedicle groove well developed in ventral valve, spondyloid; prongs of cardinal process biram- ous or bifoliate; process cavity in ventral valve obseure........ radio ARIE UR eT Aaa Tak Leptostrophia Hall and Clarke, s. s. DD. Size usually medium to small for the group; pedicle groove usually absent; if present, very small and not spondyloid; ecar- dinal process usually simple; well developed process cavities in the ventral valve. EK. Shells markedly corrugated, either regularly or differentially ; ventral process usually obscure; ventral muscles usually rela- tively narrow; cardinal process attached to a broad trian- gular platform resembling inner palintrope wall; external dorsal palintrope linear ............ Rhytistrophia Caster, new. EE. Shells either not corrugated, or if so, obscurely, and without the characters listed in EH, above. F. Hinge at least partially crenulated.... eee primitive Leptostrophids and Protoleptostrophia Caster, new. FF. Hinge edentulous, or with only a few nodes near the ex- tremity which are apparently not comparable with crenu- NEGCLOMIS ee eNMe NL ec sullen taeda Australostrophia Caster, new. PRIMITIVE LEPTOSTROPHIDS This is not the occasion to enter into a discussion of the ances- try of the leptostrophids; it does appear, however, that they are first clearly differentiable in the late Silurian. Certainly they were 175 COLOMBIAN DEVONIAN FAUNA: CASTER a | or well along in the Manlius (uppermost Silurian, American col- umn) and offer no especial difficulties in identification. The Silurian forms appear to have true dental teeth (nodes) and den- tal buttresses, which were poorly developed and are as a conse- quence only indifferently known in the Helderbergian and true Devonian forms. (See Protoleptostrophia, below.) The group was presumably derived independently of the other Stropheo- dontids from the brachyprionids, and has apparently therefore only a broad early relationship with the Stropheodontini. The hinge of the Leptostrophiini has in no case been found to be fully dentic- ulate to the degree attained in the true stropheodonts. Le ptos- trophia planulata (Hall), s. l., from the Manlius and Helderber- gian, and Stropheodonta bipartita Hall from the Manlius appar- ently lie near the ancestral stock of the leptostrophid line, and will, when better known, probably be recognized as the nucleus of a protean genus, more primitive even than Protoleptostrophia, below. THE PERPLANA GROUP Genus PROTOLEPTOSTROPHIA Caster, n. gen. The group of species, both described and in manuscript, which center around Leptostrophia perplana (Conrad) and Leptostro- phia blainvillii (Billings) constitute a very highly characteristic branch of the Leptostrophiini, and one which is easily detected either by external or internal structures. The shells of most spe- cies in this group are only subplano-subconvex, and in many cases are sub-biplanate. In other instances, however, the dorsal valve is slightly concave. The surface of both valves is typically orna- mented with closely crowded rounded costelle of subequal width and height which are usually from two to three times as wide as the intervening concavely subangular spaces. The radii increase both by bifurcation and intercalation in the genotype, but in some species seem to increase entirely by intercalation. The radii are slightly irregular above, with a tendency in the genotype to be- come obscurely nodose, but in certain Middle Devonian forms become prominently nodose, or even spinose. (This has possibly attained the prominence of a generic character in the Middle De- 76 BULLETIN 83 176 vonian.) Concentric varices are exceedingly fine, and cover the radii and interspaces. Lamellose growth lines are also irregu- larly developed, and obscure concentric corrugations are a com- mon inheritance of the group, which probably may appear at any time. They are, however, never so strikingly developed as in the L. beckii group (Rhytistrophia). Some shells show very prom- inent puncte in the interspaces between radii. Internally the ventral valves of the genotype show low, callus- like nodes under and abutting the hingeplate on either side of the delthyrial zone. These have the position and apparent func- tion of dental braces or obsolescent lamelle. The dental nodes do not attain the commissure plane, and can have had no articu- latory function. Moreover, they are not visible in many speci- mens, but material from the Gaspé sandstone in the U. S. Na- tional Museum referred to the genotype shows them very distinct- ly. Centrad of the inferred true dental nodes lie elongate, sub- angular ridges which bound the large flabellate diductor scars. Preparations in the U. S. National Museum by Beecher of re- lated species in the Hamilton (perplana complex) show that these ridges are made up of globules, and even hollow spheres of calcareous callus material. They are not dental plates, since they lie inside the dental nodes, and overlap them for a short dis- tance. They seem rather to be paradental plates or their homo- logues, already discussed under the Douvillinini. In virtually all other groups of the Leptostrophiini the dental nodes are not pres- ent, and were it not for this contrary information, the paradental plates might justifiably be construed as modified dental plates. The delthyrium is narrowly triangular, and open only at the hinge on either side of the terminus of the median septum which abuts the deltidium for its full height. The median septum is divisible into a subumbonal portion which is elongate, lozenge shape, and an anterior portion which may be tenuously connected with or quite disjunct from the posterior portion. The anterior septum is blade-like, tapering off to extreme delicacy at each end. It separates the adductor muscle scars and in some cases extends forward also to separate the anterior portion of the diductor scars. The posterior septum is really made up of the fusion of Wy (ce COLOMBIAN DEVONIAN FAauNA: CASTER 77 and callus filling between the anteriorly and downwardly (i. e. down from the posterior callus and inner palintrope wall toward the floor of the valve) divergent prongs of the ventral process and two posteriorly and upwardly divergent lamellee which bound the pedicle pit anteriorly. This is inferred from other, less spe- cialized stropheodonts in which these prongs apparently posterior- ly bounded the lozenge-shaped pit in which the pedicle muscles were principally attached. These component parts are often much concealed by secondary calcification so that the posterior septal region is merely an elevated platform resembling a spear- head at the terminus of the bladelike true median septum. The posterior septum sometimes bears an obscure median dimple which is all that remains of the so-called pedicle pit. In other genera, these posterior parts create a structure around the base of the pedicle which might be compared (analogously) with either a syrinx or pseudospondylium. Postlaterad of the prongs of the ventral process and subumbonally placed are elongate, ovate pits for the reception of the cardinal process. For these the name process pits is suggested. In the present group the pits often reach to the palintrope wall or may be excavated into it. Shells have been seen in which the cardinal process had worn holes through the ventral palintrope on either side of the delthyrial zone. The diductor muscles were apparently split into bundles, and low radial bladelike septa separated these bundles on the scar. The diductors may reach nearly to the front margin of the shell, and are without definite anterior limitations. The adductor muscles are usually obscurely attached to a relatively large lentic- ular area on either side of the median septum, They extend from the posterior septum for usually about two-thirds the median length of the diductors. The interiors of both valves are coarsely pustulose. | my In the dorsal valve the cardinal process is seen to be composed of sessile prongs which rise from the posterior callus and over- hang the hinge. Each prong is tear-shaped in cross section, the narrow part directed toward the hinge. No evidence of hinge sockets is known in the group. Extending forward from the rear platform is a tapering median septum which is usually rather short. The posterior platform is continued around the ovate ad- 78 BULLETIN 83 178 ductor scars as a broad callus into which they are posteriorly re- cessed. The posterior diductors were attached on either side of the median septum, in elongate areas, sometimes slightly de- pressed. The anterior diductors were attached mesially in an ovate zone at the end of the cardinal septum, but their seat of at- tachment is usually most obscure and is detected only in extreme gerontism. The posterior callus is very coarsely pustulose. The hinge is nearly completely crenulated in the genotype, but less so in several allied species. The Leptostrophia perplana (Conrad) complex, including vir- tually all of the species placed under synonymy by Hall, 1867, and Williams and Breger, 1916, as well as many yet undescribed, will constitute the bulk of the present group of leptostrophids. At least two additional genera are apparently yet to be described in the group from the North American fauna. The group is prob- ably represented in the Bolivian Devonian, and will undoubtedly turn up in the Colombian fauna when it has been more thorough- ly collected. The genus Protoleptostrophia, s. s. is apparently limited to the Oriskanian, but may extend in each direction for a short space of time. The perplana complex is particularly well developed in the Middle Devonian and probably does not extend into the Upper Devonian, or at most, not beyond the true Che- mung. THE NERVOSE LEPTOSTROPHIDS The nervose leptostrophids are a very distinct Middle and Up- per Devonian (American column) development: They may be derived from the early perplana stock as Hall, Hall and Clarke, Williams and Breger, and others have inferred by their reference of virtually all subplanate leptostrophids to “Stropheodonta- per- plana’. The nervose forms share a characteristic alternating type of surface ornament in which the main radii wax and wane in their course from beak to periphery, giving a very irregular and even elongate-nodose radial effect. Between these incom- plete main costellz are several much finer radii which are rela- tively regular in strength, but usually are wavering in direction for their entire length. The number of fine radii between the principal costelle varies considerably, but seems to be always 179 COLOMBIAN DEVONIAN FAUNA: CASTER 79 greater than that of costelle. This is a very different type of surface detail from that seen in the perplana group, with which they do share very similar shape and convexity of the valves. Internally the nervose forms ordinarily have the ventral pro- cess very strongly developed, so that it often gives the effect of a small secondary set of lamellae comparable to the paradental ones which are also very well developed. The paradental plates often fail to reach the palintrope wall by a considerable distance. The ninge is incompletely crenulated in all of the known representa- tives of this group. In the dorsal valve the characters are rather similar to those in Protoleptostro phia. Genus NERYVOSTROPHIA Caster, n. gen. This genus, based on Stropheodonta nervosa, Hall, is charac- terized by exceedingly thin shells which are subplanate, the ven- tral one usually being only very slightly more convex than the dorsal one, which in some cases is correspondingly very slight- ly concave. The surface ornament in the genotype usually gives a first impression of hirsuteness, for the principal costelle are conspicuously discontinuous, and irregular in height where de- veloped. The ornament is really comprised of delicate radial cor- rugations of the shell which usually are reflected in reverse on the interior. New radii originate by intercalation. Between the principal radii are usually from four to six very fine undulatory radii which also originate by intercalation, but are essentially con- tinuous. The hinge is crenulated for about two-thirds of the width, the crenulations usually diverging from the median line toward the front of the shell. The delthyrium is closed by a con- vex deltidium which may be secondarily opened at the commis- sure plane for the reception of the cardinal process. Internally the ventral valve usually shows prominent paradental lamellz which ordinarily fail to reach the palintrope wall. The ventral process is strongly developed, but the anterior boundary of the pedicle pit is usually obscure, but may be well developed as a specific feature. The median septum varies considerably in de- gree of development. The ventral muscle scars are usually not 80 BULLETIN 83 180 well defined. They are flabellate areas, which in certain Upper Devonian species of large size, are well differentiated by lateral bordures of callous material. The adductors were attached in the Upper Devonian shells to elongate median subparallel plates or platforms which are separated either by a median septum, or by a sharply angular median fossa, which sometimes has a relict septum in the bottom. In the genotype the adductor scars are ordinarily not well shown. In the dorsal valve there are no striking differences from Pro- toleptostrophia which might be termed of generic value. The Senecan epoch and Chemung age of the Chautauquan were the heydays of Nervostrophia in North America, although the stock is well represented in the Erian also. Certain shells from the Bolivian Devonian suggest its presence in the “austral” realm but it is unknown as yet in Colombia. Leptostrophia junia (Hall) of the Erian is a specialized con- gener of Nervostrophia which attained truly gigantic size for so thin-shell and planate a brachiopod. In this form the nervose characteristic is not so striking, although readily detectable. There is a tendency for the intermediate radi to disappear on the ven- tral valve in L. junia while they become coarser on the dorsal valve, and form with relatively regular primary costelle a very striking and characteristic surface detail. Fine concentric varices are also occasionally present, and may cause the radii to appear slightly lamellate. This feature is also seen in certain species of Sulcatostrophia, below. Internally L. junia also shows in ex- aggerated manner most of the features which characterize Nervo- strophia. Especially striking is the nodose ventral process which recalls the strong process development 1n otherwise not very sim- ilar L. esplanata (Sowerby) of the lower Coblenzian, (which is presumably a true Leptostrophia, or assignable to a closely re- lated subgenus). The hinge of L. junia is only about half crenu- lated; the palintrope nearly linear. It seems at this stage of in- vestigation that L. juwnia represents a distinct generic offshoot of the Erian Nervostrophia line. 181 COLOMBIAN DEVONIAN FAUNA: CASTER 81 THE SULCATE NERVOSAE While it is not improbable that the development of a shallow fold and sinus may have happened in several distinct lines of the leptostrophids, the group here under discussion seems quite def- initely to be derivative from the nervose line. The principal dis- tinctions are a strongly arcuate or ventricose shell of heavy com- position, with a lamellose exterior in adulthood, and the presence of a broad fold and sinus. Internally the normal leptostrophid features, while present, are blanketed by much secondary calcifi- cation, at least in the genotype. Genus SULCATOSTROPHIA Caster, n. gen. The genotype Leptostrophia camerata Fenton and Fenton, 1924, from the Hackberry stage of the Upper Devonian is a highly specialized shell of diminuitive size for the leptostrophids. Leptostrophia calvint (Miller) [= L. quadrata (Calvin) ] from the Independence shale is also assignable to this genus (or an antecedent subgenus) and appears to be the ancester of the geno- type.* The allied form, Leptostrophia rockfordensis Fenton and Fenton, also from the Hackberry, is somewhat larger, but also well illustrates the characteristics of the new genus. The ventral valve of the genotype is strongly ventricose, or inflated. In some cases it is anteriorly geniculate. The hinge in the genotype is shorter than the greatest width of the shell which comes at about the anterior half or two-thirds. Mesially the ventral valve bears a shallow sinus which in the genotype often becomes very pro- nounced. The dorsal valve is strongly concave. This is of course unusual for the Leptostrophiini. In fact, this character excludes the shells from the leptostrophids as originally defined. The dorsal valve fits very closely upon the inner surface of the ventral valve, and bears a median fold which is the counterpart of the ventral sinus. The surface of the shells clearly bespeaks * Tam indebted to Mr. Merrill A. Stainbrook of Texas Technological College, for his generosity expressed in a letter of October 24, 1938, in reply to one of my own outlining my proposals concerning Sulcatostrophia. Quotation from his letter follows: ‘‘Since it will be a year at least be- fore I shall have anything ready on the Independence fauna it seems best for you to go ahead with your genus since it would be more appropriate to include it in your classification. I gladly relinquish what rights, if any, I have in the genus to you.’’ 82 BULLETIN 83 182 affinity with the nervose leptostrophids. There is, however, a difference of ornament on the two valves, which in some cases is extremely striking. The nervose characteristics come out well on the ventral valve where the principal radii are much elevated, and wax and wane very noticeably, The finer radi are undula- tory in their course. Very often though, the intermediate radii are obscure in the ventral valve, or may be differentially developed on a single valve. In the dorsal valve the radii are very regular and of the alternating type. Although the shells of L. camerata are of marked difference in size, contour and relationship, the ornament, very curiously, recalls that of Leptostrophia junia of the Hamilton. The shell substance is very thick, and toward the front becomes characteristically foliate or lamellose. Often the dorsal valve is abbreviated anteriorly so that the interior of the ventral valve shows externally from the dorsal side. The inter- iors of both valves receive a great deal of secondary calcification, but this is probably an ecologic feature, since it occurs in many Hackberry shells. The dorsal valve usually bears a strong peri- pheral callus. The internal features are reminiscent of the con- dition already described for Nervostrophia, but tend to show also the effects of excess calcification. The Siulcatostrophia development seems to be restricted to the calcareous shale facies of the Upper Devonian in the interior of the North American continent, but apparently embraces more specific diversification than the already described species would indicate. The stock is not known to me elsewhere. There are, however, other sulcate leptostrophids, such as curious linguate forms which occur in the Eifel (Calceola beds) of Germany. They probably represent a distinct genus. This is true also of the in- distinctly sulcate Leptostrophia assella gens of Europe. The Chemung group of New York contains inflated leptostrophids, as yet undescribed, which recall Suwlcatostrophia in the condition of the valves. They retain however the thin shells and character- istic ornament of the true nervosa group and bear no sign of median sulcation, The Chemung forms show internal features that mark them as probably meriting subgeneric assignment. 183 CoLOMBIAN DEVONIAN FAUNA: CASTER 83 EDENTULOUS LEPTOSTROPHIDS Genus AUSTRALOSTROPHIA Caster, n. gen. Clarke, 1913 (p. 286), described an abundant leptostrophid fauna from the Amazonas Devonian of Brazil under the name Leptostrophia?? mesembria. This species superficially resembles L. tardifi, as Clarke pointed out, but amazingly enough does not possess hinge crenulations in adulthood, although it may show signs of them in early ontogenetic stages. The Brazilian develop- ment is certainly of generic importance, as Clarke himself intim- ated by his generic interrogation. For the genus the name Aus- tralostrophia is proposed, with Clarke’s species as the genotype. Clarke has shown the presence of tiny chonetid-like nodes near the extremities of the hinge line in the shell, which may be looked upon as one of the generic features. His original diagnosis is so complete that there is no point in abstracting his discussion here. The Colombian fauna shows, however, that his conclusion that all “austral” leptostrophids were of the Australostrophia type is, in the light of discoveries in northern South America, not satisfac- tory to-day. A few observations on the paratypes of Clarke’s, 1913, Brazil- ian Devonian material of Australostrophia mesembria in the New York State Museum follow: Clarke’s figures show the outlines accurately and his fig. 4o shows surface detail rather well, although it exaggerates unneces- sarily the nodes on the varices, which are very prominent, but not nearly so abundant as his figures would lead one to suppose. Figure 40 also shows the varices disproportionately wide. The ornament is, however, very highly characteristic and quite unlike any “boreal” leptostrophid known to me. The hinge is, as Clarke insists, wholly without crenulations. Clarke’s figures 39 and 40 represent very well the ventral in- terior of this genus. Characteristic of the genotype is the prom- inent pair of hinge teeth, 2.5 mm. long, which diverge at approx- imately 90°. The large flabellate muscle scars are delimited by low ridges which extend from the hinge teeth, but are less elevat- 84 BULLETIN 83 184 ed above the floor of the valve than the teeth. The anterior bor- der of the muscles is usually not so distinct as Clarke’s illustra- tions would indicate. A strong median septum separates the diductor scars. Ina nearly perfect internal mold (paratype) hav- ing the general proportions of Clarke’s figure 41, the septum is 13 mm. long, and extends posteriorly to the apex of the delthyr- ial zone, where it appears to split slightly. The septum is tri- angular in cross section and sharp above. In the dorsal interior, as Clarke showed on his plate 22 (figs. 33-30), the structures are much reduced. There is a frail, sessile cardinal process which probably performed little if any articula- tory function, since it is barely elevated above the hinge. The crural plates are short, low and diverge at about the same angle as the ventral hinge teeth. They are separated from the hinge by relatively deep dental sockets. The adductor muscles are shown on some of the paratypes in the New York State Museum to have been attached to small flabellate depressed zones about 3 mm. in length. These are located on either side of a postmedian low smooth platform which extends anteriorly in a trifid manner. The posterior undivided part of the platform is about 2 mm. in length and the extensions of about the same length. Each prong, middle and two lateral, is split anteriorly or at least medially excavated. This creates a very curious structure which may have served as attachment place of the adductors and also perhaps as incipient braceplates which were possibly homeomorphs of the Stutzplatten of the Douvillinini. Clarke’s figure 35 suggests faintly the two lateral branches of this median structure. The lateral prongs of the median process are usually slightly more elevated above the interior of the valve than the posterior portion, and may stand out, as in Clarke’s figure 35, as dissociated knobs which are medially excavate. ‘ Discussion of “austral” leptostrophids.—As indicated above, the differences between the Amazonian fossils described by Clarke as Leptostrophia?? mesembria and any Colombian-Venezuelan materials as yet known are of generic importance. Chief external difference lies in the absence of corrugations and presence of no- dose strize and concentric varices in Clarke’s material. Clarke, 1913 185 CoLOMBIAN DEVONIAN FAUNA: CASTER 85 (p. 289), reviewed the other “‘austral’’ species having the aspect of Leptostrophia, and concluded that crenulate hinges were up to that time unknown in South America. This is a view which in the light of Weisbord’s, 1926, discoveries in Venezuela and the present abundant development of the group in Colombia, will no longer hold. It may develop, however, as Clarke supposed, that many of the specimens assigned to Leptostrophia, Stropheodonta, and Strophomena will turn out to be representatives of the retro- grade and edentulous stropheodont stock, here assigned to Aus- tralostrophia. Others are certainly true stropheodontids and are discussed under that heading in this paper. Stropheodonta argen- tina Thomas, 1905, may be a Leptena as Clarke, 1913, indicated, but I see no reason to urge that it necessarily per se requires Sil- urian age determination for the Argentine deposits. This view is fortified by the presence of a lepteenid in the present Devonian fauna. Apparently the corrugated Leptostrophias (/hytistrophia) of the northern Andes are an unique occurrence for South Am- erica. They are, moreover, apparently unknown elsewhere in the whole “austral” realm. Orthis concinna Morris and Sharpe, 1846, from the Falkland Islands has the aspect of a holocrenulate Leptostrophia, as Clarke, 1913 (p. 285), wrote, and as Reed, 1903 (p. 169), agreed, the while astutely pointing out its close affinities with the “boreal” Leptostrophia perplana gens. Morris and Sharpe’s material seems to belong in the Protoleptostrophia grouping, and likewise much of Reed’s Bokkeveld material. Clarke, 1900 (p. 87), also identified this gens (L. perplana) from the Amazonas Devonian of Maecurt and Cuaru and did not specifically rescind the identi- fication in his, 1913 (p. 289), pronouncement, cited above. Corru- gations of the Rhytistrophia type are to be expected in the “‘aus- tral” representatives of Protoleptostrophia, in keeping with boreal developments, but appear as yet not to have been recorded. I do not share Clarke’s views on the close relationship between L. con- cinna (Morris and Sharpe) and L. magnifica-L. tardifi of the north. The forms which Clarke considers Morris and Sharpe’s spe- cies from the Devonian of San Juan, Argentina, in which the 86 BULLETIN 83 186 crenulations are restricted to the umbonal zone in maturity, are probably not congeneric. They seem to belong rather with the New Zealand form described by Allan, 1935 (p. 12), from the Reefton beds as L. reeftonensis. These are to my mind more likely brachyprionids than leptostrophids. The subumbonal, or at least much restricted, denticles as well as the shallow sinus and fold belong rather with a Brachyprion of a Douvillina relative than with Leptostrophia. Corrugations such as those on L. reefton- ensis are not present on shells from the Falklands or Argentina, to my knowledge. I see no striking points of similarity between the New Zealand material and L. magnifica and congeners, which Allan stressed. I have, however, seen none of the Reefton ma- terial at first hand. THE CORRUGATED LEPTOSTROPHIDS Leptostrophid shells of lepteenoid aspect are a characteristic part of North American Oriskanian and Onondagan faunas. They extend downward into the Helderbergian (home of the geno- type), but are apparently not known in Erian or later faunas. The stock is derived from the perplana group, which also shows in a sporadic and incomplete degree this tendency toward corru- gation of the shells. It is even possible that Conrad’s original spe- cimen of Strophomena perplana from the Onondagan, which he described as having obscure corrugations, was an immature in- dividual of the form later described by Hall as Stropheodonta beckiui, but in the absence of the types, it seems wisest to follow Hall’s lead in the recognition of L. becki as a distinct species, inasmuch as his material came from the Helderbergian, with which the Onondagan has few, if any, species in common. Genus RHYTISTROPHIA Caster, n. gen. Genotype.—Stropheodonta beckiit Hall. Helderbergian. The new genus Rhytistrophia is here proposed for the De- vonian (Helderbergian to Erian) leptostrophids having regularly and prominently corrugated shells in the manner of Stropheodonta becku Wall, the genotype (Helderbergian). It is not un= likely that the genus may descend into the Manlius and may range upward into the lower Hamilton faunas of some areas. The sur- 187 COLOMBIAN DEVONIAN FAUNA: CASTER 87 face details of this genus are very similar to the more conserva- tive member of the perplana group, above; 1.e. they have regu- lar rounded costella which are closely crowded and much wider than the interspaces. The radii are slightly irregular, or even obscurely nodose. Concentric varices are present, but usually not prominent. Strong lamelle of growth are also occasionally present. The corrugations of the shell usually extend from hinge margin on one side to the other, but may be discontinuous. Thev are usually symmetrical folds. Material in the U. S. National Museum from the Linden form- ation (Helderbergian) at Perryville, Tennessee, of an undescribed species closely allied to the genotype, shows the internal features of the genus very well. The hinge is incompletely crenulated, and in the Helderbergian forms, the crenulations usually reach less than half the width. The ventral interior shows the presence of the ventral process, but in the genotype only obscurely. In the undescribed species the whole posterior area is amalgamated into an elongate lozenge-shape platform. The paradental plates are well developed and nearly attain the palintrope. The angle of the plates is apparently of specific value. In the dorsal valve the cardinal process shows each prong to be bipartite, after the man- ner of true Leptostrophia. This characteristic is only faintly shown in the Colombian material. (See pl. 5, fig. 12.) The car- dinal process is attached to an elongate triangular callus that ex- tends along the palintrope for about the same distance as the hinge crenulations. The prongs of the process are usually tri- angular in cross section, and behind them is a callus which closes the notothyrium. The dorsal median septum is short, broad and attached to the posterior platform. The adductor muscles are bounded by roughtened callus ridges. The exposed dorsal palin- trope is essentially linear. Additional structural features of the genus are brought out un- der the description of the Venezuelan-Colombian species and Col- ombian variant, below. Rhytistrophia caribbeana, var. colombia Caster, n. var. Plate 5, figs. 5-13; Plate 6, fig. 14; Plate 8, figs. 5-12; Plate 11, fig. 9 One of the most abundant brachiopods in the Colombian De- 88 BULLETIN 83 188 vonian fauna is a typically corrugated stropheodont of general leptostrophid characteristics which seems assignable to the same species as the fragmental shell described by Weisbord, 1926, as Leptostrophia caribbeana from the Devonian of Venezuela. An abundance of material showing both external and internal charac- teristics makes it now possible to give a more thorough analysis of this interesting fossil, and offers the basis for designating the Colombian shells a distinct variety of Weisbord’s species. Shell large, thin, and prominently corrugated; semielliptical, semicircular or even on rare occasions subquadrate in outline. Hinge straight, sometimes slightly produced, usually is greatest width of shell; hinge commissure denticulate for entire width. Shells are ordinarily biconvex, or planoconvex. The ventral valve is, however, never ventricose or inflated, and the dorsal valve shows no evidence of ever being concave, unless to a very slight degree peripherally, although usually somewhat flatter than the ventral valve. Narrow, but definitely not linear, interareas are in both valves; that of the ventral being apsacline and somewhat higher than that of the dorsal, which is anacline. Delthyrium broadly triangular and apparently closed by a flattened plate. Surface covered with regular threadlike elevated striz separ- ated from each other by rounded furrows of about equal width. The striz are consistently wavy or undulatory, and show slight irregularities in thickness. They increase by intercalation rather than by bifurcation and each stria is posteriorly, near the source of origin, much more wavy than in its peripheral development. Concentric varices are very obscure over the main body of the shell, but may be detected on the postlateral zones. Even here they play no conspicuous part in the surface ornamentation, En- tire surface rugose due to leptenoid concentric corrugations of some irregularity. There are usually from 10 to 14 prominent corrugations in an adult shell, but in some cases there are fewer. The radii tend to be best developed in the zones between corru- gation crests. Corrugations increase in prominence anteriorly. The internal features of the ventral valve, as judged from sey- eral internal molds, are dominated by large flabelliform and re- cessed diductor scars which are delimited postlaterally by pustu- lose calli, the portions of which adjacent to the muscle scars are 189 COLOMBIAN DEVONIAN FAUNA: CASTER 89 elevated as strong ridges. These ridges are apparently formed as callosities in adult shells about the extensions of the dental plates. Posteriorly the plates are extended onto the inner palin- trope surface. The diductor scars extend anteriorly for more than half the length of the shell, and sometimes for as much as two-thirds the length; the front margin of each scar is lobate, but the number of lobes varies. The scars are without sharp anterior limits. Each diductor scar is radially striate throughout and an- teriorly ridged; sometimes in gerontic individuals there are three or four prominent carine or muscle diaphragms, separating the anterior lobes of each muscle seat. Posteriorly the diductor re- cess is continued onto the palintrope as rather sharply defined somewhat circular depressions underneath the fragile deltidium. These subdeltidial pits were probably recipients of the prongs of the cardinal process. Mesially the flabelliform diductors are sep- arated by a well elevated septum and toward the rear by an elong- ate, fusiform expansion of the adductor scars. The adductor scars are elevated on a slight callosity above the plane of the diductors, in adult shells, and are faintly defined, especially later- ally, in immature shells. In adult shells the diductor scars are postlaterally bordered by forwardly diverging extensions of the dental ridges which end in front at or near the place of greatest width of the scars. The dental ridges converge behind and con- tinue on the inner palintrope surface (deltidial surface) where in some specimens they apparently meet on the commissure plane. In either case, they stand up from the hinge plane as articula- tory knobs which fit into sockets of the dorsal valve on either side of the cardinal process. These are probably the homologues of a ventral “process” described by Hall, 1867 (p. 78), for Stropheo- donta demissa, genotype of Stropheodonta. All ventral valves at hand show these internal dental buttresses. The posteriorly bor- dering carinee may converge toward the rear at the palintrope angle and therefore be accessory structures, possibly the scene of attachment of the posterior adductors and of separate origin from the true dental plates. Specimen 5149 (pl. 11, fig. 9) indicates this possibility on the internal mold. In no case is a conspicu- ous pedicle pit formed by the dental plates. The adductor scars 90 BULLETIN 83 190 are divided by a central deep narrow slash which extends to the front margin of the adductor zone. On either side of the slash are posteriorly diverging ridges which meet at its end to form the median septum between the diductor scars. The septum has its greatest elevation near the middle of the shell in gerontic forms and becomes evanescent before the front margin of the diductor zone is attained. In ephebic individuals the median septum is a structure of the posterior third of the shell. In some individuals the adductor zone occupies most of the posterior, or umbonal re- cess. The postlateral portions of the shell interior are thickened by callosities and bear many coarse granulations which decrease in size anteriorly. The shell corrugations show clearly on the interior, and likewise the surface radu, especially in the peripheral zone. Ventral valves are, however, much thicker than the dorsal ones and therefore the surface structures tend to show through less distinctly. Several shells bear prominent discontinuous radial grooves or scratch-like markings on the interior which are most deeply incised on the parts which correspond to the crests of the surface corrugations. The palintrope is thickest toward the axis and the hinge crenulations which extend across the entire commis- sure surface are best developed in this zone where they are longi- tudinal ridges. Toward the apices of the hinge the denticles become mere pointed elevations. Fach crenulation is extended on the palintrope exterior as an elevated carina, only slightly less prom- inent than the commissure denticulations, thus giving the cardinal area, of the ventral valve at least, a vertically striated exterior. The dorsal interior shows equally characteristic markings, the most prominent of which are the adductor muscle seats and strong bipartite cardinal process. The muscle scars are narrower than in the ventral valve, and more pronouncedly delimited postlateral- ly by pustulose callosities which extend from the base of the car- dinal process anterolaterally for about three-fifths the length of the shell in this direction. The contact of the adductor seat and the callosities is essentially vertical and shows a difference of elevation usually exceeding a millimeter. The callosities begin to disappear toward the front of the shell near the anterolateral terminus of the muscle seats, and the postlaterally do not attain the hinge zone. The general plane of the adductor scars rises above the floor of the valve toward the front. The posterior adductors 191 COLOMBIAN DEVONIAN FAUNA: CASTER 91 are about twice as large as the anterior ones and lie as elongate, outwardly encircling arcs which are covered with faint arbores- cent markings. The zone adjacent to the anterior adductors is elevated in an elongate swell. Toward the rear the posterior adductor seats are circularly bordered by the meeting of the ex- planate bases of the anteromesially converging process extensions and the bases of the delimiting callosities. The anterior adductor scars lie along side the prominent mesial ridge formed by the union of the convergent process extensions. They too are cres- centic in outline and extend forward as attenuated lines on either side of the mesial carina a little beyond the posterior adductor ° scars. They are also covered with arborescent longitudinal grooves and ridges. The boundary between the two pairs of adductor scars is relatively sharp, the anterior ones being on a slightly lower plane (i.e. more recessed) than the posterior ones. The cardinal process consists of two strong and only slightly diverging pillars which rise nearly vertically from the posterior subumbonal callosity and are recurved behind so that the two termini over- hang the palintrope and extend a slight distance into the umbonal cavity of the ventral valve, where they may rest against the pos- terior wall. The pillars of the process are braced as long as they are in the dorsal valve by narrow posterior buttresses between them and the inner surface of the palintrope, apparently forming inner thickenings along the notothyrial area. Basally the two prongs of the process are separated by an excavated area of about the same width as each pillar. Low, broadly rounded, descending extensions from the base of each pillar converge a short distance in front of the process to form a low, rounded mesial ridge which separates the adductor scars. At about the same radial distance forward as the vertical bounding ridges extend, the central ridge is bulbously expanded in the center of each anterior adductor arc, and abruptly narrowed toward the front where it continues as a sessile acicular extension several millimeters beyond the muscle zone to the edge of the pustulose pallial region. No true teeth have been observed in the ventral valve. The median stropheodontid ventral “process” is well developed. The two small knobs rest in shallow sockets between the two 92 BULLETIN 83 192 prongs of the cardinal process, rather than on either side, for the knobs are so close together, and so often fused into a miniature process-like overhang in the ventral valve, that they must needs articulate between the widely spaced pillars of the process. There are, however, postlateral depressions at the base of each process pillar which certainly have the aspect of sockets, though no coun- terpart structures have been seen in the ventral valves. ~ The posterior three-fifths of the shell is internally pustulose and essentially without evidence of surface corrugations, presumably due to a mantle of secondary calcification in the pallial region. The peripheral zone of the shell shows internally the prominent surface corrugations and also the surface radi very clearly, which is in contrast to the ephebic condition of the ventral valve. The hinge surface rises but slightly above the inner surface of the valve, although the palintrope is externally about 3 mm. high at the beak. The umbonal cavity is filled by pallial deposition in the adult stage. The denticular fossettes are the negatives of the denticles in the ventral valve. Dorsal palintropeare without vertical strize in all specimens observed. : As shown on plate 8, fig. 7, the corrugations of the shell in this species begin at an early ontogenetic stage. Wherever observed, the nascent corrugations are less regular than they are in their later state. Plate 8, figs. 8-11, shows the early stages of the massive median septum and posteriorly conjoined strong crural plates, which in the ephebic stage are largely obscured by a great amount of secondary callus material. At this early stage also, the hinge extremities seem nearly always to be rounded. The hinge crenulations are much coarser and fewer in number, al- though extending the full hinge length, than in adulthood. While it is presumable that these small shells represent a juvenal stage, they may in truth be a distinct species or variant. The very strength of the crural process in shells so small strongly suggests maturity. The absence of shells of intermediate size also forti- fies this surmise. For the nonce, however, I consider them all as one species. 193 CoLOMBIAN DEVONIAN FAUNA: CASTER iio) (vt) Dimensions.—The following dimensions in millimeters are tak- en from internal and external molds. 5417 5418* 5419 5420* 5421+ 5436 Length shell 21 32 29 26 25 35 Width shell 39 44 25 36 46 40 Umbonal height shell 1.5 2.5 2 Length median septum 10 15 14 8 Width muscle area 13 20 10 18 Length adductor sears 6 5.5 6 Width adductor sears 4.5 3 4.5 Width cardinal process 4.5 Height cardinal process 2 Diameter visceral zone 16 No. strong corrugations of the shell 10 3S 6+ 14 7-8 15 Discussion.—Internal characters in these shells seem to be of greatest generic value. Of these, the accessory carinz or ridges which delimit the ventral adductor muscles posteriorly are important. These may have been occupied by the posterior set of adductor muscles. Also important are the strongly developed mesial “ventral process” plates of which the “accessory” ridges probably are the extensions. The apparent- ly degenerate condition of the teeth, prominent development of the dental plates and crowding of the “accessory” plates on the inter- nal palintrope wall to form a set of parallel, or upwardly converg- ing “teeth” which overhang the hinge like a miniature, mesially split, ventral process, is rather unusual. The presence of lateral pit-areas on either side of the process in the dorsal valve, which are apparently not occupied by teeth from the ventral valve, and of a dorsal subumbonal pit into which the median “ventral pro- cess” fits is also characteristic. The “boreal” expression of the L. beckii gens does not seem to possess these characteristics. The “austral” representatives do not have a tiny, subumbonal pedicle pit which is ore of the indicial characters of the genotype. In the Colombian material the dental extensions do not converge so obtusely as to appear like a transverse bar connecting the bases of the lamellze as they do in L. becku. The elongate lozenge- shape area formed by the intersection of the opposedly divergent “accessory” carinee and low ?dentel ridges which unite to form * External mold. + Dorsal valve. 94 BULLETIN 83 194 the mesial ridge may be the homologue of the pedicle pit in the “boreal” forms. The posterior buttresses of the cardinal process between the pillars and dorsal palintrope wall also appear to be unique features of the South American material. Clarke’s species Leptostrophia?? mesembria (genotype of Australostrophia) has a somewhat similar cardinal buttress and dorsal musculature, but is edentulous and without the corrugations of the present form. The Colombian representatives of R. caribbeana (Weisbord) appear to have fewer, more widely spaced prominent corrugations than topotypic material. The Venezuelan type is a fragment, however, of uncertain orientation and shows only external features. This specimen has about 16 prominent ridges whereas our material usually has about 12 to 14. His specimen indicates a more circular outline than any of our specimens show; a more nearly equal length and width. Weisbord reports that the striz of the Venezuelan speci- men increase by bifurcation, but the Colombian material shows definite intercalation instead. This may have been inaccurate observation, for many of the earlier described species having inter- calation of new striz were reported to increase by bifurcation. Examination of Weisbord’s type substantiates this surmise. The differences existant between the Colombian-Venezuelan material and the “boreal” Rhytistrophia becku have already been brought out in the discussion. That the two groups are of com- mon descent seems rather clear. The closest resemblance seems to occur between our shells and Helderbergian forms in the Appalachian province. R. becki, var. tennesseensis (Dunbar) 1920 (p. 129,), from the Birdsong shale of Tennessee has a much larger, more circular shell than the South American fossils. It has moreover many more undulations on the surface and these are far more irregular and discontinuous than on any of the “austral” material. Types.—Holotype: Pal. Res. Inst. No. 5421, an internal mold ef the dorsal valve, but of equal importance for generic charac- ters are the paraytpes 5417, 5419, and 5436, which are ventral inner molds. Additional paratypes: Nos. 5417A, 5418, 5420, 5421A-C. 195 COLOMBIAN DEVONIAN FAUNA: CASTER 95 TRUE LEPTOSTROPHIDS The genus Leptostrophia, s.s., must be taken as the basis of comparison for the tribe Leptostrophiini, and therefore establish- es what is meant by “true leptostrophids.” It is perhaps unfor- tunate that so striking and obviously phylogerontic a species as Leptostrophia magnifica should have been selected (Schuchert, 1897) as the genotype of the Hall and Clarke genus, for by so doing, the more protean forms, such as the protoleptostrophids, are atypical. It seems to me that Leptostrophia, s.s., as determ- ined by the genotype and congeners in the Oriskanian and homo- taxial equivalents elsewhere is a byproduct rather than a phylo- genetic stage in the history of the later species. Genus LEPTOSTROPHIA Hall and Clarke, s.s. As determined by the genotype, Strophomena magnifica Hall, from the Oriskanian, Leptostrophia is seen to be a very special- ized and stratigraphically restricted group of shells, most of which are of gigantic size. The collections in the New York State Museum, U. S. National Museum, and University of Cin- cinnati Museum are the source of material here presented on the genotype of Leptostrophia. The shells are large, and plano-convex, but in some cases the dorsal valve may be slightly concave. The shell material is rela- tively thin for the size of the shell. The surface is ornamented with closely spaced rounded radii which increase principally by implantation. There are ordinarily present also relatively prom- inent concentric varices which are strongly developed in the in- terspaces between radii, but usually are not visible on the crests of the radii themselves. This interrupted reticulation is not vis- ible on all shells, and may prove to be a specific feature, especial- ly of the genotype in the Oriskanian. The palintropes are very well developed in both valves, and are vertically cordate, the cords being the edges of plates whose ends on the commissure plane form the hinge crenulations in the ventral valve, and pillars between sockets in the dorsal valve. The palintrope wall is rel- atively thick. Internally the ventral valve exhibits some of the most striking generic features. The delthyrium gives the impression of being 96 BULLETIN 83 196 agape, but it is apparently always closed by a deltidium which is extremely concave, and over the external surface of which the pedicle possibly functioned as in a trough. The concave deltid- ium extends internally for some little distance in front of the palintrope, and by the deposition of secondary lime beneath, cre- ates the impression of a sort of “pseudospondylium” in the ventral valve. This grooved delthyrial filling is mesially ridged as though it were composed by the lateral outgrowth of paradeltidial ma- terial rather than by the formation of a single plate. It is also marked with concentric varices. The pedicle groove-plate ap- parently covers the ventral process. This presumably partially contributes to the lime filling beneath the plate which creates the pseudospondyloid appearance. The hinge is usually crenu- lated for nearly its full width. Although the cardinal process is a ponderous structure in the dorsal valve, the secondary calcifi- cation in the subumbonal portion of the shell fills the process cav- ities which are so well developed in the protoleptostrophids. The articulation of the two valves is very slight, as the almost uni- versal dissociation of the fossil valves would indicate. There are neither hinge teeth nor dorsal sockets, although there are carinz which may correspond to socket ridges in the dorsal valve. The diductor muscles are enormous, flabellate, and may reach to near the front margin of the shell. The muscles were apparently separated into strands, for the scars have well developed radial septa (“muscle diaphragms”) subequal to their length. The scars are bounded laterally by callus ridges, which occasionally rise quite high above the surrounding callus deposit. These struc- tures are apparently paradental ridges as in the other leptostro- phids. The front margin of the muscles is without definite delim- itation. The adductor muscles were attached to elongate lenticu- lar zones cn either side of the median septum. In some cases the anterior adductors are slightly more elevated than the posterior. The median septum is usually present, but seldom well developed. It extends from near the front margin of the pedicle groove to near the front edge of the diductors. The dorsal interior is similarly highly specialized. The process is a ponderous device which rises as subparallel columns from 197 COLOMBIAN DEVONIAN FAUNA: CASTER 97 the posterior callus. The prongs are elongate trigonal in cross section and apparently grooved above. They barely overhang the hinge. On either side of each prong of the process is a sub- sidiary ridge which in some cases is very prominent. These give the process a curious tetrapartite appearance in gerontic forms. The subsidiary plates are probably specialized, albeit non- functional, socket plates. The notothyrium is closed by a subangu- lar convex callus, or chilidium, which appears to fit, at least at the apex, into the concave deltidium. The chilidial callus ex- tends within the shell as a short acute growth between the prongs of the process in some shells. The posterior callus is subtriangu- lar. Buttresses reach from the front of the process to encircle the adductor scars postlaterally. The adductor scars are elong- ate-ovate depressions which diverge only very slightly. Between them lie, on either side of the median septum (a variable struc- ture), the posterior diductor scars which are relatively enormous, elongate roughened zones which in some shells are almost twice the length of the adductors, and may nearly attain the front mar- gin of the shell. In the genotype there are additional anterior adductors in the form of irregular roughened mounds just in front of the adductor scars, and not reaching so far forward as the posterior scars. This condition is variable in allied species. The median septum in the genotype is very well developed as a blade-like carina separating the middle portion of the posterior diductors, but not attaining the posterior platform. In allied forms, the median septum is strictly leptostrophid as in most of the tribe where it originates at the posterior platform, and ex- tends as a blunt and stubbed spine about as far forward as the adductor scars. Most adult shells of the genotype show a well developed peripheral callus on the inside of the dorsal valve. This is in some cases a very striking ridge. There are apparently several species lurking under the name of the genotype in eastern American deposits of Oriskanian age. Leptostrophia magniventra (Hall) is a ventricose relative of the genotype which shares most of the essential generic characters. Leptostrophia explanata (Sowerby) of the lower Emsien appar- ently occupies the same environmental niche, but is a separate generic strain, albeit not distantly removed from Leptostrophia, 98 BULLETIN 83 198 s. s. No South American representatives of this genus in the strict sense are known as yet. Certain fragments in the Colom- bian faunule offer a basis for expecting its eventual discovery there. This group will be taken up in more detail in a subsequent paper. Subfamily STROPHONELLIN Caster, new The subfamily Strophonelline has been proposed, above, for most of those Stropheocontide which are resupinate in their shell orientation. The following genera and species-groups are clearly recognizable : Amphistrophia Hall and Clarke, 1892, s. s. Strophomena striata Hall, genotype. Niagaran. ?Pholidostrophia Hall and Clarke, 1892.* Stropheodonta nacrea Hall, genotype. Devonian. Strophonelloides Caster, new. Stropheodonta reversa Hall, genotype. Upper Devonian. Chemungia Caster, new. Stropheodonta celata Hall, genotype. Chemung, Upper Devonian. Strophonella Hall, 1879, s. s. Stropheodonta? semifasciata Hall, geno- type. Upper Silurian. Principal Species-Groups in Strophonella, s.s. A. American and European differentially corrugated forms, e.g. S. jamesont Reed. Silurian. B. Regularly corrugated strophonellids, e.g. S. leavenworthana (Hall), Helderbergian. C. American and European forms of the S. williamsi Kindle and Breger type. Silurian. D. Group having the surface ornament of S. continens Clarke, gens. Principally Oriskanian. HK. Group having subangular and fascicular plice, e.g. 8S. ampla (Hall). Helderbergian-Ulsterian. It now seems very probable that at least three tribes are repre- sented in this list of genera. The pholidostrophids, if they are genetically related at all to the strophonellids with which they are customarily considered, certainly constitute a separate line of evolution from the other generic stocks, and would merit at least tribal distinction. Pholidostrophia is not resupinate, and shows other non-Strophonellid characteristics. My present inclination, subject to further evidence before crystalizing, however, is that the genus Pholidostrophia as now used really embraces several generic groups, and should be given subfamily ranking on a par with the Stropheodontinz and Strophonelline. Aside from men- * Not resupinate. 199 CoLOMBiAN DEVONIAN FAUNA: CASTER 99 tion in the key below, the pholidostrophids will not be discussed at this time. The amphistrophids are apparently a tribal group, possibly including several undescribed genera, which occur prin- cipally in the Silurian. They seem to be quite apart from, albeit possibly ancestral to, the main line of strophonellid development. The third and largest group would center around the genus Strophonella and would include as a nucleus the last eight items on the foregoing list. Possibly the two holocrenulate genera would also constitute a tribe. Definite proposals in line with these intimations must await opportunity to study more fully American and especially European collections. A structural key to the genera and groups recognized at this time follows: Key to the Principal Genera and Species-Groups of the Strophonelline A. Shells nacreous, without radial ornament; not resupinate.... 2... Pholidostrophia Hall and Clarke. AA. Shells not nacreous; with radial crnament; resupinate. B. Hinges only partially crenulated. C. Hinge crenulations restricted to a small triangular plate adjacent to the partially open delthyrium; ventral adduct- or muscle sears elongate ovate, and ordinarily not complete- ly encircled by paradental plates. Shells of relatively small SIZ CR Rumer eane nec ee rar meee Amphistrophia Hall and Clarke (inel. Strophoprion Twenhofel). CC. Hinge crenulations occur on elongate plates which usually reach for nearly half the width of the hinge or even a lit- tle further; delthyrium closed, or secondarily opened; ven- tral muscles flabellate, and completely encircled by the paradental plates and callus ridges. Shells usually of large size. D. Surface of shell relatively regular; radii fine and of the alternating type, in usual proportion of 3 or 4 to 1; hinge less than half crenulated Strophonella Hall, s.s. DD. Surface of shell variable, may be regular; radial orna- ment variable; hinge half or more than half ecrenu- lated, but not holocrenulate. EK. Alternating type of radii; surface differentially corrugated in a ‘‘seersucker’’ manner____....___..... Group of S. jamesoni Reed. HE. Regular, non-alternating type of surface radii (or at least usually of this type) ; surface regu- larly, concentrically corrugated in a leptenoid manner sroup of S. leavenworthana (Hall) EEE. Surface not corrugated either differentially or concentrically to any appreciable degree. 100 BULLETIN 83 200 F. Surface ornament comprised of very wide- spaced principal ecostelle between which oceur large numbers of fine costelle. No regular concentric VariceS_./.....-.-22- Group of S. williamsi Kindle and Breger. FF. Surface ornament fasciculate, or sub- fasciculate, not conspicuously of the al ternating type. Concentric varices may be well developed. G. Surface ornament subfasciculate ; concentric varices present and in closely spaced series. Shells of mod- erate size, not pronouncedly trans- verse in most cases GG. Surface ornament strikingly fasicicu- late, or schuchertelloid; shells tend- ing to huge size, and transverseness pee ee Group of S. ampla (Hall). BB. Hinge holocrenulate or nearly so. H. Shell substance thick, ponder- ous, lamellose, only faintly sul- cate; radii fasciculate; shells of MOG CrAbeyES!7 Cu ea Strophonelloides Caster, new. HH. Shell substance normal, not thick, lamellose or ponderous; shells strongly suleate; radii simple, regular and _ equally Spaced) size lar ce meee Chemungia Caster, new. Genus AMPHISTROPHIA Hall and Clarke, 1892, s.s. (including STROPHOPRION Twenhofel, 1914). The characteristics of this genus, determined by typical exam- ples of the genotype Strophomena striata Hall, from the Niagar- an in the U. S. National Museum, New York State Museum, and the University of Cincinnati Museum, are very different from those of many of the later species usually assigned to this genus. The shells of all the species I have examined which seem definite- ly assignable to this genus, s.s., are small to medium size. They are moderately resupinate, but may be pronouncedly so, being in some related species even strikingly geniculate. The surface radii are very fine and alternately costellate, with several differ- ent types of detail included under the genotype. The hinge is only very slightly crenulated on triangular plates on either side 201 CoLOMBIAN DEVONIAN FAUNA: CASTER 101 of the delthyrial area, The crenulation plates can be seen very distinctly to overlie the palintrope wall, and are separated from it by a stria or groove. This plate rises above the hinge in all observed specimens. The delthyrium is primitively open, or only partially closed at the apex by an outwardly convex callus. The interior of the ventral valve is without cardinal teeth, unless it should develop that the crenulation plates are themselves the homologues of such structures. There extend from near the inner palintrope wall for the full length of the di- ductor muscles outer bounding carinz of callus origin. These are apparently paradental plates such as occur in the Stropheodon- tine. In the genotype the diductor muscles are not delimited in front, but in closely related species they may have a very faint anterior ridge setting them off. The ventral process is obscure in the genotype, but a very delicate spearhead terminus to the well developed median septum probably represents the homologue of this structure in the stropheodonts. The median septum while simple in S. striata, forks anteriorly in related species (possibly a generic character). The dorsal valve bears a weak cardinal process, and a faint posterior muscle platform. The genus Amphistrophia has ovate, not flabellate, ventral muscle scars. They are not delimited by plates or callus. The hinge in this genus, as well as Strophonella, is incompletely cren- ulated, as is brought out below, but in Amphistrophia the crenula- tions are limited to a very few on a triangular plate on either side of the delthyrium, whereas in Strophonella they extend for a greater distance along the hinge, on a more evanescent plate. The genus Strophoprion Twenhofel, 1914, is apparently almost precisely a synonym of Hall’s and Clarke’s Amphistrophia, as Holtedahl, 1916, and McLearn, 1924, have pointed out. The genus was seemingly introduced without examination of the genotype of Amphistrophia and with the picture in mind of many of the later forms usually assigned, apparently quite erroneously, to that genus. Amphistrophia was proposed by Hall and Clarke, 1892 (p. 292): “Should it be considered useful to recognize the incipi- ent and progressive features of the species S. striata and probably S. patenta toward a full manifestation of generic characters, and distinguish them from Strophonella in its more mature condition 102 BuLLETIN 83 202 of development, the term Amphistrophia may prove expressive of their apparent double relationships as shown in the young and mature shells.” The type species was further described by them as “a reversed Brachyprion, bearing precisely the same relation to Strophonella in its fuller development as that group does to Stropheodonta.” Twenhofel’s brief analysis of his genus is almost precisely that of Hall and Clarke for Amphistrophia. Furthermore, Twenhofel’s genotype, Strophomena geniculatum Shaler is very closely allied, in my estimation after examining specimens in the U. S. National Museum from the Anticosti fauna, to Stropho- mena striata Hall, genotype of Amphistrophia. It is highly ques- tionable if there are any supplementary features in Shaler’s spe- cies on which a separate genus might be construed. Actually, of course, the genus Strophoprion is also at least philosophically a synonymn for Strophonella, s.s., since here again is found a par- tially crenulated hinge, rather than a holocrenulate one such as Twenhofel postulated. I am not in accord with Holtedahl’s opin- ion, 1916 (p. 64), that the absence of an anterior bounding ridge in front of the ventral diductor muscles is of no generic import- ance, for it seems to be of as much importance in the Strophonel- lids as in the Stropheodontids, where several genera seem justi- fied chiefly on this criterion. Apparently the closing of the anterior diductor scars in the Amphistrophids came principally in the lat- est Silurian in North America. Currently however, no additional generic groups are being proposed in the Amphistrophids al- though several seem undoubtedly to be present, both in North America and Europe. Genus PHOLIDOSTROPHIA Hall and Clarke, 1892 The genus Pholidostrophia, based on Stropheodonta nacrea Hall, as mentioned above, presumably is a very distinct develop- ment of the crenulated Strophomenacea and quite possibly de- serves a separate family or subfamily ranking. That several generic lines are represented in these curious shells seems patent after only a cursory examination of the group. The current use of the generic name is far too broad, it seems, and even the speci- mens assigned to the genotype possibly embrace more than one genus, not to mention several species. 203 COLOMBIAN DEVONIAN FAUNA: CASTER 103 Genus STROPHONELLA Hall, 1879, s.s. The genus Strophonella Hall, as determined by the Upper Silurian genotype, Stropheodonta? semifasciata Hall, is of quite different aspect from that commonly presented by species assigned to it. This brief discussion of the genus is based on genotypic material from the Silurian, principally Waldron, in the collections of the U. S. National Museum, New York State Museum, and University of Cincinnati Museum. It is found that the shells truly assignable to the genus are for the most part of large adult size, imperfect crenulation (usually not more than half of the hinge length) and with alternating costelle (geno- type). In fact, complete crenulation of the hinge presumably did not occur in the strophonellids to any marked degree until Upper Devonian times. (See Chemungia and Strophonelloides, below.) The ventral interior of the genotype shows very large, flabellate, albeit posteriorly pedunculate, diductor scars which are bordered their entire margin by callus ridges. These are presumably in part at least paradental. The median septum is present, but not ordinarily prominent. The adductor muscles occupy a large len- ticular medial zone and are defined quite distinctly. The hinge crenulations number 25 or 30 or even more, and are attached to a plate superposed on the palintrope surface, apparently much as in Amphistrophia, but it does not so conspicuously overhang the hinge. The delthyrium is closed by a convex plate or callus, and no hinge teeth are known. shells usually fail to retain this character and external casts show only the marks of concentric lamelle. It seems clear, however, that this fim- briate feature is to be accredited as a normal character of S. antarcticus and the combination of radii, spinules and lamella which we have shown to be present in several of the austral Spirifers is so highly distinctive that it is not known to me to occur in boreal species. : All these facts indicate a common initial stock for the austral Spirifers 160 BULLETIN 83 260 which is unlike in sculpture any boreal stock, and the resultant mature or specific expressions seem to be the outcome of intensive and retrograde developmen. under conditions of geographic isolation. (p. 264) Clarke was obviously much impressed with the uniqueness of this group of “austral” spiriferoids, to which he assigned all his Brazilian material, and to which he implied that most previously described South American forms belonged. He was at a loss to make “boreal” comparisons, for the condition seemed to him not duplicated elsewhere. While admitting the unusual features of this development, I am neither impressed with the all-em- brasive ubiquitousness of the group in “austral” faunas, nor of its utter dissociation from northern assemblages. Certainly in Co- lombia, as shown in this paper, many of the standard northern types of spiriferoids are present, and it is likely that this admix- ture of the elements from the antipodes is not wholly unique in this rew faunule. Certainly many of the spiriferoids described from Brazil by Hartt and Rathbun, 1874, show none of the Aus- tralospirifer features, and even some of the specimens illustrated by Clarke are highly questionable. Of the better known spiriferoids from South America, the fol- lowing species seems to belong to Australospirifer: Spirifer ant- arcticus Morris and Sharpe, 1846; idem, Kayser, 1897; idem, Clarke, 1913 (pars) ; Spirifer hawkinsi Morris and Sharpe, 1846; idem, ‘Clarke, 1913; Spirifer Rayserianus Clarke, 1913 (pars) ; Spirifer theringt Kayser; idem, Clarke, 1913; ?Spirifer parana Clarke, 1913; ?Spirifer contrarius Clarke, 1913; ?Spirifer lauro- sodreanus Kayser, 1897; ?Spirifer chuquisaca Ulrich, 1892, ete. All of these forms apparently show the structural features of the genus, although varying considerably in outline, number and type of plicee, and degree of septation. Spirifer antarcticus Morris and Sharpe, 1846, has been an enigma ever since it was described, but if we may rely on the clarification which Clarke, 1913, gave the species as represented in the Falklands, it certainly has the sur- face details of the present genus, but has only the very faintest median dorsal septum (if any) and a relatively distinct ventral median septum. The form of the shells as reported by Clarke is exceedingly variable, and may be too inclusive. Spirifer haw- kinst and Spirifer orbignyi Morris and Sharpe are of the same 261 COLOMBIAN DEVONIAN FAUNA: CASTER 161 gereral stamp and both possess a median dorsal septum as was originally shown (although the orientation of the valves was re- versed in the plate analysis), and may possess a ventral one al- so. In Spirifer kayserianus Clarke the dorsal septum is very prominent, and the ventral one obscure. It is in this species that the ornamentation which I think of as crucial to the present genus is best developed (or at least most fully discussed). In Spirifer iheringt Iayser, especially as illustrated and discussed by Clarke, 1913, the radial ornament is very completely developed as it ap- pears also to be in Spirifer antarcticus, s.s. In S. theringi the median dorsal and ventral septa are best developed. SS pirifer parana, S. contrarius and S. lawro-socreanus, on internal features and the absence of contrary information on the exterior, appear to belong in the present genus. The Spirifer buarqueanus Hartt (in Rathbun) complex, as interpreted by Derby (in Clarke, 1913, p. 241), may also belong in this genus. Clarke’s Spirifer katzeri, however, as well as Hartt’s Spirifer pedroanus and the large spe- cimen referred to S. buargiucanus by Katzer seem better refer- able to Brachyspirifer than to the present genus. Clarke pre- sented no convincing evidence that they share the curious surface details of Australospirifer. The surface texture of the present genus seems to be very much akin to the condition found in some of the forms which Hall and Clarke, 1893, assigned to the Osteolati. Clarke, 1913, re- ferred all of his Brazilian spiriferoids to the Radiati on the basis of fine radial ornament, but they could in no sense be considered multiplicate (one diagnostic feature) although they are multiradi- ete. Furthermore, the presence of fine radii derived from elong- ate pustules of the fimbriate type recalls rather closely the condi- tion in S. marcyi Hall. (See Williams, 1913, p. 54.) It is curi- ous too, that Clarke, while pointing out that in some of the “aus- tral” species the pustules or tiny spines are located on radii for part of the surface (precisely the description given by Hall and Clarke for their Osteolati), should still have pronounced them Radiati, The enormous delthyrial zone which Morris and Sharpe first described and illustrated in their S. antarcticus seems to carry through the entire group, and certainly fits the Osteolati 162 BULLETIN 83 262 picture very well. The intimation of a rudimentary syringeal plate in Spivifer hawkinsi, in Morris’ and Sharpe’s original analysis and Clarke’s reanalysis, as well as a suggestion of this develop- ment in several of the “pre-syringothyrid” southern species (or variants), recalls Spirifer marcyi in the North which is a division of the Osteolati. From the boreal “pre-syringothyrids” (see Will- iams, 1913) Australospirifer differs principally in the recapitula- tory aspect of the surface ornament, and in the presence of un- usually strong concentric features in some representatives. It looks now as though this “austral” Onondagan (or earlier) development may very well be the ancestral stock out of which’ the very typical “boreal” Middle Devonian Osteolati might have been derived. It still appears as though these primitive Osteolati developed a series of homeomorphs of the bimbriata unispinei pauciplicates with which, in Colombia at any rate, they inter- mingled, and of course may possibly have interbred. Australo- spurifer may possibly be the ancestral stock of Spinocyrtia. Two forms have been found in the Colombian fauna which seem quite definitely assignable to Auwustralospirifer. Neither, unfortunately, is adequately known to make specific comparisons very reliable. Australospirifer cf. antarcticus, var. 1. Plate 12, figs. 7, 8; Plate 13, figs. 21, 22 ’Spirifer antarcticus Morris and Sharpe, Geol. Soc. London, Proe., vol. 2, 1846, p. 276, plate 11, figs. 2a, 2b. Spirifer antarcticus Morris and Sharpe. Clarke, Mon. Serv. Geol. y Mineral. do Brasil, 1913, p. 258, pl. 18 (pars). Several fragments of spiriferoid shell are at hand from the Co- lombian faunule which show many of the features usually assigned to Spirifer antarcticus, although hardly conforming to that species in the strict sense of the original designation or illustration. I have in mind such shells as Clarke, 1913 (plate 18, fig. 3), illus- trated from Jaguariahyva as this species, the specific identification of which I feel may well be an open question. The lamelle, even on these small shells, are exceedingly strongly developed. There are usually eight or nine subangular plice on each side, which are separated by contrastingly regularly concave interspaces. 263 CoLOMBIAN DEVONIAN FAuNA: CASTER 163 There is no evidence of pustules on the early part of the shells, but then none is extremely well preserved. Over most of the shell the continuous fine lineations are well preserved, and in some instances can even be seen on the inner molds. The inter- nal mold shows the presence of a pair of septumlike dental lam- ella which extend anteriorly for nearly one-third the length of the valve, They embrace not more than one plica on either side of the median sinus. A well marked median septum is also pres- ent in the ventral valve. The characters of the dorsal valve are not known. It seems futile to attempt further comparisons on such scant information, but the surface details coupled with the internal structure certainly point to shells hitherto referred to the Morris and Sharpe species, but which in all likelihood will warrant a separate name when better known. Illustrated specimens.—Pal. Res. Inst. Nos. 5408, 5408A-C. ?Australospirifer cf. antarcticus, var. 2. Plate, fe, 21: Plate 13, figs. 19, 20. Cf. Spirifer antarcticus Morris and Sharpe. Clarke, Mon. Serv. Geol. y Mineral. do Brasil, 1913, pl. 18 (pars). Several fragmental specimens are at hand which have the con- tour, general plical arrangement, and internal ventral details of Falkland Island forms attributed to Spirifer antarcticus by Clarke, 1913. Without judging the accuracy of Clarke’s por- trayal of the species, present comparisons are made with such forms as those which he diagnosed as the Morris and Sharpe spe- cies on his plate 18, figs. 15, 16. In the Colombian shells certainly, differences are adequate to warrant specific separation when bet- ter preserved material is at hand. This variety is distinguished by slightly produced hinge, moderate inflation, very high and ex- ceedingly thickened area, large delthyrial zone, prominent and regularly rounded ventral sinus of which the bounding plice are much the strongest on the shell. On either side of the sinus there are 10 plus 2 plice, which is a somewhat larger number than average for. this species as portrayed by Clarke. Each plica is subangular on the sides, but rounded above; the interspaces are regularly concave, and slightly wider than the plice. Crossing 164 BuLLETIN 83 264. the plicee are lamellose varices which are especially prominent toward the front of the shells. None of the shells is well enough preserved for finer surface markings to show, although sugges- tions of radial strie can be detected. The internal mold of the ventral valve shows a deep rostral cavity which is separated by a median septum. The hinge teeth are short and rugged, and their supports are continued as encircling bands nearly around the perimeter of the rostral cavity. The suggestion of a delthyr- ial plate (syringothyroid syringeal plate) shows on the inner mold, and recalls the impression recorded for Spirifer hawkinsi Morris and Sharpe, and many illustrations of Spirifer antarcticus. As in the preceding, it seems likely that the present form may well be quite distinct from the species, s.s., but the material will not warrant the description of a new species. Some of the speci- mens attributed to Spirifer pedroanus Hartt by Rathbun, 1874, may fall here, but the comparisons are fraught with risk in view of the probability that most of the types of the Hartt species are Brachyspirifers. Illustrated specimens.—Pal. Res. Inst. Nos. 5471B, 5407D, 3407G, all internal and external ventral molds in a poor state of preservation. Genus BRACHYSPIRIFER Wedekind, 1926 Genotype.—Spirifer carinatus. Middle Devonian. Brachyspirifer palmere Caster, n. sp. Plate 10, figs. 12, 13; Plate 12, figs. 5, 6 Shells of medium size, transverse; hinge line greatest width of shell; extremities acute; outline of shell essentially transverse- ly triangular; sides from hinge extremities to median zone essen- tially a straight line; sinus produced anteriorly to a slight degree ; valves moderately inflated; cardinal area moderately high; fold and sinus nonplicated, and very prominent; each side bears 17 or 18 plications which rise as rounded ridges above the common surface, and are separated by narrow subangular interspaces about one-half as wide as the plice. The interspaces- adjacent to the fold and the plicaze adjacent to the sinus are somewhat more prominent than these features elsewhere on the shell, and thus 265 CoLOMBIAN DEVONIAN FAUNA: CASTER 165 very effectively delimit the mesial regions. The entire sur- face is covered with sublamellate varices which tend to become pronouncedly lamellate on the extremities. They stand out best in the interspaces between plice, but can be traced over the ribs as well. None of the specimens at hand shows any signs of sur- face puncte, pustules or lineations in addition to the varices. The dorsal internal mold shows very characteristic muscula- ture: from a slight rostral filling or callus in the apex a well de- veloped median septum extends anteriorly for about one-fourth the length of the shell and acts as a wall between two elongate oval, rather deeply recessed, muscle cavities which also extend for about one-fourth the length of the shell. These are very striking spiriferoid features. The crural plates leave the palin- trope at a very low angle, and are not especially prominent. The internal features of the ventral valve are not known. ‘The di- mensions are brought out by the illustrations. Discussion.—In the South American Devonian the closest form-ally of the present genus appears to be the Brazilian Spirifer pedroanus Hartt (in Rathbun), 1874, (=Spirifer katzeri Clarke, 1913, fide Derby in Clarke, 1913, p. 242) which has the same general outline and contour, but apparently fewer ribs (10-16) on a side, more angular median sinus and fold, and apparently much less conspicuous median dorsal septum and recessed dorsal muscle seats. The comparison can be made, of course, only on the proviso that Clarke was mistaken in assuming a radiate or pustulose surface for his species. He certainly did not succeed in showing such a structural surface in his magnified surface de- tail (idem, pl. 21, fig. 4), and mentions nothing in the text about the surface specifically, for he was merely proposing a new name for a mistaken identification of Katzer. Elsewhere in the report of course, Clarke pointed out that to his knowledge all of the Brazilian spirifers possessed the surface details here assigned to the new genus Australospirifer. Neither the account of Spirifer pedroanus in Rathbun, 1874, nor any subsequent specific detail would indicate that the Brazilian form, any more than Brachy- spirifer palmere, shares the fimbriate-pustulose-striate-lamellate surficial features of the great group of southern spirifers. It is 166 BuLuEtTIn 83 266 my impression that the present Colombian species and Hartt’s Brazilian form are congeneric. The Venezuelan specimens assigned by Weisbord, 1926 (pl. 5, fig. 8), to Hartt’s species similarly do not show the “austral” type of spiriferoid surface, nor does Weisbord’s Spirifer audacu- lus, var. gulianus. In the northern faunas the closest relative or parallel appears to be the Spirifer audaculus (S. medialis)-Spirifer macronotus line of the Hamilton. It is here that we especially find the same general outline and contour duplicated many times, and the same sharply chiselled plicee and lamelle of growth without fimbriz or pustules. It is here also, especially in the Spirifer macronotus line (e.g. Hall, 1867, pl. 38A, fig. 20), that we find very similar dorsal musculature and septation developed. The South Ameri- can expression appears to be prenuncial of this stock in the North American Middle Devonian, and shows in subdued form the traits to be elaborated in the group later on. Size and gibbosity, increase in height of cardinal area, etc. are phases emphasized in the Hamilton faunas. It is very likely that Koztowski’s, 1923 (pl. 9, figs. 31, a, b), Bolivian form referred to Spirifer, aff. audaculus belongs to the same prenuncial stock as the Colom- bian form, but appears to show specific differences, among which the very apparent flattening and broadening of the plice, infla- tion of both valves, development of rounded extremities and in- spicuous dorsal musculature are important. The Bolivian form would suggest more the true S. awdaculus stock in these respects than would the Colombian material. Weisbord’s, 1926, Spirifer audaculus, var. zulianus has received much the same comparisons as the present species, and may in truth be very closely allied. The Venezuelan form, however, appears to possess a much high- er palintrope (e.g. Weisbord, pl. 5, figs. 1, 2), than the Colom- bian form, with apparently many more plicze on a side ( Weisbord reports 20 to 30) and no evidence of concentric varices or lamelle, although Weisbord suggests their presence. His types show them very faintly. The interior of the dorsal valve (Weisbord’s plate 5, fig. 8) shows much the same type of recessed muscle scars reported for Brachyspirifer palmere, although he does not 267 COLOMBIAN DEVONIAN FAUNA: CASTER 167 show a median septum in his illustrations. It is important thus to establish quite certainly the presence of this highly typical Middle Devonian spiriferoid of the North in association with early Middle Devonian fossils. The suggestion of direction of migration seems inescapable. If one were to judge by the Hartt, 1878 (p. 25), record only and overlook Clarke’s, 1913, opinion on the surface details of all Brazilian spiriferoids, it seems very probable that Hartt’s Spirifer duodenarius (?) Hall is closely allied with if not the same as Brachyspirifer palmere. Types.—Holotype: Pal. Res. Inst. No. 5407, external and in- ternal dorsal molds of an individual; paratypes: 5407A, an ex- ternal ventral mold; 5407B, 5407C, 5407E, 5407F, partial exter- nal dorsal and ventral molds which show surface details very clearly. Genus PARASPIRIFER Wedekind, 1926 Genotype.—Spirifer culirijugatus. Middle Devonian. Paraspirifer, sp. Plate 12, fig. 9 One gigantic internal ventral mold, which is unfortunately im- perfect and poorly preserved, has the general aspects of the inter- ior of Spirifer acuminata and congeners in the Onondagan and Coblenzian. The internal mold bears no record of surface plicze or more delicate structures, and unfortunately is damaged at the crucial postumbonal region. For the nonce, its meaning and relationships are problematical, but it seems worth recording since I am unaware of similar forms in southern faunas. S‘pirifer theringit Clarke approaches this form in size, but seems always to show evidence of surface plice, and certainly has a much less profound median sinus in the ventral valve. S. iheringi appears to have a proportionately smaller muscular zone, and much less rugged dental plates and hinge teeth. Illustrated specimen.—Pal. Res. Inst. No. 54106. Genus SPINOCYRTIA Fredericks, 1916 Genotype.—Delthyris granulosa Conrad, 1839. Middle Devonian. One imperfect internal ventral mold from the Colombian fauna is identical with the spiriferoid described in Rathbun, 1874, as 168 BULLETIN 83 268 Spirifer valenteana Hartt, (cf. Rathbun, pl. 8, fig. 11). This specimen, as also the original Rathbun one from Brazil, shows very distinctly the broad posterior rostral cavity of the Spirifer granulosus- Spirifer marcy line. The dental plates are exceeding- ly strong and extend anteriorly from the inner palintrope as sub- parallel walls. Between them, and attached to the apical portion of the deltidial zone is the imprint of a syringeal plate, which also characterizes this branch of the Osteolati of Hall and Clarke, 1893. It appears that Rathbun’s material from Brazil, and also the pres- ent Colombian discovery, are assignable to Fredericks’s genus Spinocyrtia on the basis of internal comparisons with the genotype, even though the surficial features of both the Brazilian and Col- ombian materials are unknown, and for Fredericks at least, these are the diagnostic criteria. ?Spinocyrtia cf. valenteana (Hartt) Plate 13, fig. 27 Cf. Spirifer valenteana Hartt, in Rathbun, Buffalo Soc. Nat. Sci., Bull. vol. 2, 1874, p. 241, pl. 8, fig. 11. Hartt’s analysis in Rathbun’s paper agrees with our material in nearly every respect, but this may well be due to the incomplete information on the materials from both Brazil and Colombia. Hartt’s description (in Rathbun), 1874, follows: Test above medium size, ventricose, thick, trilobed in outline and slightly transverse, with the greatest width along the hinge line. Ventral valve very convex, most elevated between the beak and middle. Cardinal angles de- pressed, with the cardinal margins concave. Beak probably large and curving over a rather constricted area. The margin of the valve is dis- tinetly trilobed, caused by the extension forward of the broad mesial sinus beyond the general margin of the valve; leaving the cardinal extremity on one side at nearly a right angle, it curves regularly inward for more than one-half the whole length of the valve and one-fifth the width, when it gradually bends outward, forming a shallow reéntrant curve before reaching the forward projection of the sinus, around which it extends in an elliptical curve. The distance across, from the center of one reéntrant curve to the other, is about twice the length of the prolongation of the sinus beyond the general margin of the valve. Mesial sinus very broad and shallow, regularly rounded in the bottom, and with its margins unde- fined; width of sinus nearly one-half the width of the valve, the whole an- terior lobe of the valve being occupied by it; in the case it is nearly as broad near the beak as at the front. The surface of the valve curves regu- larly and quite strongly from the beak to the front margin; from each side it curves rapidly upward for about one-fourth the width, and then de- scends gradually to form the sinus, which is very slightly and regularly coneave. The dental plates, as indicated by the moulds, were very high and thick behind, thinning out gradually as they advance. They are wide- ly separated, the distance between them being nearly one-third the width bo for) Wo) COLOMBIAN DEVONIAN FAUNA: CASTER 169 of the valve, and they extend forward, parallel with each other, for two- thirds the length of the valve. Between the dental plates in the mould are indistinct impressions of muscular markings, consisting of an ovate, slightly depressed space, rounded behind, where it is immediately enclosed by the dental plates, and gradually narrowing to a point anteriorly, not extending as far forward as the dental plates. This impression seems, however, too limited to include all the muscular markings of the ventral valve. The illustration on plate 13, fig. 27, will indicate how far I am justified in applying Hartt’s analysis to the Colombian speci- men. Illustrated specimen.—Pal Res. Inst. No. 5443B. Superfamily ROSTROSPIRACEA Schuchert and LeVene Family MERISTELLIDAE Hall and Clarke, 1892 Subfamily MERISTELLINAE Waagen, 1883 Genus MERISTELLA Hall, 1860 Genolectotype.— (Hall and Clarke, 1892) Merista levis (Hall) (Atry- pa levis Vanuxem, 1842). Middle Devonian. This is one of the rarer genera of brachiopods in the Devonian faunas of the southern continents, and is also rare in the Colom- bian fauna where it is represented by a large and striking species. ’ Meristella wheeleri Caster, n. sp. Plate 12, figs. 14, 15 Shell large, length and breadth subequal; maximum width at about middle of shell; ventral valve with a profound, broad and subangular sinus which is produced anteriorly and upwardly in a linguate manner. The surface of the shell is smooth, with- out even recognizable growth lines, but a varix or two is present near the front and many rather prominent ones are present on the lateral portions of the shell, where also faint radii and a suggestion of puncte can be distinguished. Beak slightly incurved, probably perforated; area constricted and tere- bratuloid. Interior of ventral valve is dominated by a large rostral cavity and recessed muscle seat. The cavity and muscle scars are bor- dered by very strong, dental ridges which posteriorly buttress large and powerful cardinal teeth. The muscle platform is radial- ly striated and also bears concentric ridges. The surface of the interior, outside the postmedian zone, is indistinctly ridged and dimpled with vascular markings. These features are brought out by the illustrations. Dorsal features not known. 170 BULLETIN 83 270 Dimensions.— 5461 5462 Hinge width 15 mm. 18 mm. Maximum width shell 33 mm. 40 mm. Median length shell 32 mm. 34 mm. Length to maximum width shell 17 mm. 25 mm. Length rostral cavity and muscle seat 18 mm. 22 mm. Width rostral cavity and muscle seat 11 mm. 12.5 mm. All of these measurements are based on internal molds of ven- tral valves. Discussion.—The generic assignment of this species is based wholly upon the features of the internal mold of the ventral valve, but these so closely conform to the general scheme of Meristella that there can be little doubt of the identification. In specific details this Colombian form differs so markedly in every respect from all other forms hitherto described or illustrated from the southern Devonian that close comparisons are really impossible. This is very nearly as true when northern faunas are considered. Ulrich, 1892, described a presumably genuine Meristella from the Bolivian Devonian as Meristella riskowskyi which compares rather well with the average North American forms attributed to Meristella nasuta Hall. Ulrich made his comparisons with that species while also mentioning the somewhat less similar Meris- tella hoskinsi Hall of the American Hamilton. The shell of the present species differs from Ulrich’s in being larger, much more transverse and less globose. Especially striking in the Colom- bian species externally is the profound ventral sinus and corre- sponding dorsal fold. The internal features of Ulrich’s shells are not known. It is doubtful if the form listed by Thomas, 1905, from the Argentine Devonian as Meristella, sp.? Kayser is really referable to this genus, Certainly it has no obvious affinity to the present form. Clarke, 1913 (p. 346), also questioned the Kayser-Thomas generic determination. The Brazilian form of Meristella first recorded as Amphigenia by Derby, but later de- scribed by ‘Clarke (and Derby?) 1913 (p. 264), as Meristella septata is far larger than the present species, much more planate, and distinctly subcircular in outline. It is also without fold or sinus. The musculature, while that of Meristella, does not spe- al COLOMBIAN DEVONIAN FAUNA: CASTER 171 cifically recall the present form. In the North American Devonian faunas there are several spe- cies with which broad comparisons are justified, though in no case is there evidence of especially close relationship. Meristella princeps Hall, (e.g. from the Coeymans) is nearly as large and does have a relatively deep sinus and corresponding fold. It is, however, not so transverse, the sinus is by no means so conspicu- ous, and the musculature is less recessed and more restricted in area. Meristella lata Hall of the Oriskany (and Grande Gréve) is comparable in size, but is without more than a suggestion of median fold and sinus. Meristella nasuta Hall has variants in the Schoharie and Onondaga (e.g. Hall, 1867, pl. 48, fig. 5, 6) which compare with Meristella wheeleri in size, but are again without striking fold and sinus. The muscular differences are obvious if the illustrations are compared. It is likely that some of the variants within the species-group described from the Grande Gréve limestone of the Gaspé Peninsula, as Meristella champlaint Clarke, 1908 (e.g. pl. 30), may approach the Colom- bian form in size and general aspect. If any northern forms were to be chosen as essential equivalents, this last one would be my selection. Types.—Holotype: Pal. Res. Inst. No. 5461-5461A, internal and external ventral molds of one shell; paratype: No. 5462. Genus PENTAGONIA Cozzens, 1846 Genotype.—Atrypa unisulcata Conrad, 1841 (—Pentagonia peersi Coz- zens, 1846). Onondagan. The genus Pentagonia shows characteristic meristelloid inter- nal features with slight modification of the hinge plate. The ex- ternal features are very strikingly characteristic: they include a pentagonal or hexagonal outline, and a very broad, somewhat flattened ventral sinus which is subangularly delimited from the iateral portions. In the dorsal valve there is a prominent median sulcus which usually bears a mesial groove, or subsidiary sinus. On each side of the central duplicate fold is typically a single prominent subangular corrugation near the posterior margin. In the Hamilton forms the lateral sulcus tends to be duplicate. The surface is prominently marked by concentric strie. 2, BULLETIN 83 272, Pentagonia gemmisulcata Caster, n. sp. Plate 10, figs. 16, 17 The external dorsal mold of a single specimen of Pentagonia has been found in the Colombian material. From this fragmen- tal evidence, however, specific differences from the North Ameri- can genotype and allies are clearly distinguishable. The size of the shell is large; outline pentagonal. The valve bears mesially a very prominent duplicate sulcus, the groove be- tween the two portions being very deep, but rounded in the bot- tom. The sulcar region stands up in sharp relief from the com- mon contour of the valve. Postlaterally on each side is a single rounded, but subangular, corrugation which parallels the hinge margin. The zone between the lateral plicee and the sulcus is broadly convex, suggesting an incipient corrugation in that area, and giving the shell a gibbous appearance. The surface is marked with regular elevated varices, which tend to become fasciculate or lamellate toward the front, on the flanks of the sulcus and in the intervening zone between the sulcus and the lateral folds. The shell is meristelloid in external texture. Illustrations are natural Size. Discussion.—There is nothing resembling this shell known from the South American fauna, or for that matter, as far as I am aware, from the entire “austral’”’ province. Structural similari- ties are very close, however, between Pentagonia gemmisulcata and the North American genotype. The resemblance is less strik- ing between the Colombian species and the duplicate forms or- dinarily found in the Hamilton of the North. In none of the “boreal” material which I have been able to study in Albany or Washington, has the median sinus in the dorsal fold been near- ly as deeply developed as in the Colombian specimens. The rule in the north seems to be for an almost universal disappearance of the median sinus on the fold toward the front of the shell. The lateral folds seem to be consistently less prominent and much shorter. They usually also fail to attain the margin of the valve. The resemblance seems to be closest to the Onondagan representa- tives of the genus in the North. Type.—Holotype: Pal. Res. Inst. No. 5410A. 273 COLOMBIAN DEVONIAN FAuNA: CASTER 173 Superfamily TEREBRATULACEA Waagen, 1883 Family MEGANTERIDAE Waagen, 1882 Genus MEGANTERIS Suess, 1855 Genotype.—Terebratula archiaci de Verneuil, 1850 (non Megalanteris suesst Drevermann, 1902).* Devonian. The characteristics of this genus have been discussed at some length by Hall and Clarke, 1892, and the known structures of the present species seem to fulfill every requirement for its inclusion in the genus Meganteris as currently employed. Meganteris australis Caster, n. sp. Plate 9, figs. 25, 26; Plate 13, figs. 9-15 Shells large, moderately biconvex; outline elongate ovate, or even transversely ovate; greatest width usually about one-third the length in front of the hinge line; ventral valve somewhat more convex than dorsal, but not inflated or especially gibbous ; beak slightly incurved; apparently bearing a tiny pedicle open- ing at apex; cardinal area terebratuloid; commissure essentially a plane, definitely so in front; anterior peripheral zone of both valves apparently consistently somewhat flattened. Surface of both valves is megascopically smocth, but microscopically very finely and evenly punctose, the puncte being irregularly concen- trically arranged and rather evenly spaced. On the outer half or third of the adult shell there are ordinarily present two or three very prominent equispaced sublamellose varices that have the appearance of being periods of growth cessation. They create a steplike anterior surface to the shell. The edges of the lamel- lze sometimes bear very distinct radial lines which are also pres- ent in series on the margin of both shells in adult specimens. The interior of the ventral valve shows in reverse the imprint of the strong anterior varices, and bears an anteriorly expanding subumbonal cavity for muscular attachment. The cavity is faint- ly delimited by dental lamellae which become evanescent near the middle of the shell. The teeth are strong, triangular and over- hang the hinge to fit in opposite sockets. These features, as * Allan, 1935, selected Drevermann’s species as genolectotype. Opinion 65 of the International Rules of Zodlogical Nomenclature, also mentioned by Allan, 1935 (p. 23), stands in the way of this selection. To establish the Hifelian species, which Suess misidentified, as genolectotype, thereby fulfilling Suess’s intentions, would apparently require action by the In- ‘ernational Commission. 174 BULLETIN 83 ey 274 portrayed on the inner mold, are shown on plate 13, fig. 11. The interior of the dorsal valve shows similarly the reverse imprint of the varices, but is dominated by the rugged process and affiliated callus. The process stands up in bold relief from the apical zone of the shell which it completely obscures internal- ly. It apparently rises from a callus platform, from. which an- teriorly protrude, a few millimeters from the floor of the valve, two subparallel conical plates. The cale ribbon or loop that may have been present is not preserved in any of the ma- terial. The process is cleft behind and terminally. Each apophy- sis is slightly produced to over-hang the hinge. The terminus of each is faintly scored as if interlocking with some structural ele- ment in the ventral umbonal cavity. The musculature of the ventral and dorsal valves is shown by the illustrations. Dimensions.— 5424 5425 5422 5423 5422A Hinge width 18 mm. 20 mm. 20 mm. est.24 mm. 7 mm. Median length shell 38 mm. 36 mm. 35. mm. 25 mm. 13 mm. Maximum width shell 32 mm. 31 mm. 40 mm. 30 mm. 14 mm. Discussion.—To my mind there can be no question of the ge- neric affinity of this Colombian species. The specific resem- blances lie principally with the “boreal” forms from the Oriskan- ian, usually attributed to MWeganteris ovalis (Hall). To this same species-group Meganteris swesst (Drevermann) (e.g. Megan- teris archiact (deVerneuil) of the Eifelian in Hall and Clarke, 1892, p. 281) apparently belongs; possibly also Meganteris thunet Clarke, 1908, and in all likelihood Meganteris neozelanica Allan, 1935, from the Keefton Devonian of New Zealand, which recalls strongly Clarke’s Meganteris diobolaris. Meganteris ovalis (Hall), 1859, is a much smaller form, and bears radial ornament in the true Rensseleria tradition, albeit obscurely. The forms of this species illustrated by Hall and Clarke, 1892, may not be entirely conspecific, but the illustrated forms in this work, (figs. 19, 20, 21), recall very closely the posteriorly grooved and apically bifid ponderous cardinal process of Weganteris australis, although present on a smaller shell. Many of the smooth shells of Meganteris thunei Clarke, 1908, from the Grande Greve lime- 275 COLOMBIAN DEVONIAN FAUNA: CASTER Te stone of Gaspé, recall in outline, size and contour, the Colombian form, but appear consistently to have a much frailer cardinal pro- cess and hinge plate (e.g. Clarke, 1908, pl. 27). The immature specimens attributed to this species from the Colombian deposit recall in outline and contour certain forms from the Oriskanian described by Clarke, 1909, as Meganteris diobolaris which is a smaller and more rotund form, with smaller cardinalia. The New Zealand species, which is the only other representative of the genus known thus far from the so-called ‘“‘austral” faunas, ap- pears to be as highly variable in contour and outline as the present species. It seems, however, to be consistently smaller, and in general, more ovate than the Colombian specimens. The dorsal muscle zone appears to be proportionately narrower, and the pro- cess of dissimilar shape. That they belong to one rather closely allied stock, seemingly of world-wide distribution in late Lower or early Middle Devonian time, appears very clear. It is possible that some of the many Rensseleria reported from South Africa and South America may prove upon better acquaintance to be Meganteris. Most show prominent plications, however, and seem therefore to be at least specifically distinct from this nearly glab- rous form. Types.—Holotype: Pal. Res. Inst. Nos. 5425, 5422, internal and external molds of a dorsal valve; paratypes: Nos. 5424, an internal ventral mold, 5423, an external dorsal mold, and 5422A, external mold of an immature shell, orientation not known. Incerte sedis ?Camarotechia, sp. Plate 7, fig. 21 A single fragment of an angularly plicated shell may be refer- able to Camarotechia or a somewhat similarly plicated genus, but in the absence of any intimation of internal structures, or really critical external ones, further identification is at this time impos- sible. As far as the fragment at hand will indicate, the shell is probably new to the South American Devonian, but might prove to be congeneric with certain forms usually referred to Leptocalia flabellites in “‘austral” faunas. Illustrated specimen.—Pal. Res. Inst. No. 5427A. 176 BULLETIN 83 276 ?Cryptonella, sp. Plate 13, figs. 24, 25 One tiny external dorsal impression of a terebratuloid shell is at hand, which is too poorly preserved to warrant a guess at its identity. The known features are shown by the illustrations, which are natural size. Illustrated specimen.—Pal. Res. Inst. No. 5463A. ?Derbyina, sp. Plate 11, figs. 15-17 One small shell in the collection shows on the internal ventral mold the strong dental plates of Derbyina. The shell is subangu- larly corrugated in the rhynchonelloid manner and is also marked by irregular growth lines. The corrugations show a tendency to become slightly inflated between the varices, thus creating a cur- ious nodose appearance which, as far as I am aware, is unique in the South American Devonian. The specimen has been compared with Derbyina jamesiana (Hartt) in the New York State Museum collection as identified by J. M. Clarke from the Devonian of Rio Maecuru, Brazil, which would appear to be the most closely resembling form as yet de- scribed. The Colombian shell is much narrower, with a more trigonal outline. The Brazilian form does not have nodose plice. Illustrated specimen.—Pal. Res. Inst. No. 5485. APPENDIX The fauna of the Colombian Devonian which has come to light thus far is, as we have seen, quite amazingly rich in brachiopods of unusual stamp, but decidedly depauperate in most of the types of remains one ordinarily associates with the South American Devonian strata. Thus we see but the most meager fragments of Mollusca. Only one unidentifiable tiny gastropod was found after the most careful search. Half a dozen pectenoid shells and one small Cypricardimia complete the molluscan picture. There are no signs of either the conulariids or the cephalopods. The molluscan scraps thus far retrieved are illustrated on the plates and briefly discussed below. Bryozoans are rare items in South American Devonian faunas, but are exceedingly abundant and varied in the Colombian de- AEC COLOMBIAN DEVONIAN FAUNA: CASTER il7/ “I posits as they are in the Venezuelan, Here only two fragments are shown, and these quite uncritically, for it is hoped that this unusual phase of the fauna may be separately presented. The aspect of the Bryozoa is even more strikingly reminiscent of the early Onondagan faunas of the North, however, and will probably be of great value in establishing the relative age of the northern Andean Devonian, Ostracods are also very abundant in much of the material, but their poor state of preservation has discouraged me from under- taking their study. Better material will certainly make a very interesting study. Crinoid columnals are common in the ma- terial at hand, although no terminal portions have been found. From the varied shapes of the columnals it seems safe to pre- dict a numerous fauna when thorough collecting at Floresta is possible. The calicinal imprint of a single tiny tetracoral has been found in our material. This seems probably to be Cyatho- phyllum bchvianum Woztowski so far as the poorly preserved fragment would warrant identification. The facies does not ap- pear to have been one especially conducive to coralline growth, however, and no duplication of the fauna found in the rather pure limestone conditions which prevailed during part of the De- vonian in Venezuela is anticipated. Trilobite fragments are more abundant than any of the other lesser items in the fauna, but even here the material is extreme- ly fragmental and the preservation is none too good. As far as can be judged from the collections, no new forms are involved. This is unusual in view of the large variety of new brachiopods. PELECYPODA Cypricardinia cf. subindenta Weisbord Plate 11, figs. 6-8; Plate 14, figs. 19-21 Cf. Cypricardinia subindenta Weisbord, Bull. Amer. Pal., vol. 11, No. 46, 1926, p. 26, pl. 6, fig. 6. Several small fragments of pelecypod shell impressions have been found which suggest very much the form described by Weis- bord, 1926, as Cypricardinia subindenta of the Venezuelan De- vonian. The Colombian material shows the delicate chevron 178 BULLETIN 83 278 like markings of Conrad’s Cypricardinia indenta which Weisbord could not detect on the Venezuelan form, although surmised might be present on better preserved material. As can be seen from the present figures, the material will hardly warrant de- tailed description or wide comparisons. It does apear, how- ever, that this pelecypod is in all likelihood a relative of the North American Middle Devonian species. Illustrated material.—Pal. Res, Inst. Nos. 5479, 5479A. ?Aviculopecten, sp. A Plate 12, fig. 16; Plate 13, figs. 17, 18 Several small fragments of a rather large monomyarian shell having the surface ornamentation of the aviculopectenoids is at hand. In view of the fact that none shows the hinge zone or mus- culature, generic placement seems hazardous. The surface sculp- ture of the commonest form is well shown by the illustrations. This type of shell is not known to me previously from the South American Devonian, but is nearly duplicated many times by northern species. The most striking features of the surface orna- ment are rounded, nodose, alternating radii which arise by im- plantation and are crossed by varices which are elevated in the interspaces but incised on the radii, thus creating the nodose ex- pression on the latter. Studied specimens.—Pal. Res. Inst. Nos. 5414A, 5414. ?Aviculopecten, sp. B Plate 18, fig. 16 Another pectenoid shell, apparently of somewhat larger size, and of quite different sculptural pattern is also known from frag- ments in the Colombian deposit. Unfortunately, the shell is too poorly preserved to warrant description. It is characterized by rather irregularly spaced radii of variable size, some of which are rounded, but others are definitely angular or subangular. The varices of growth are less prominent than on the form mentioned above, and it is not impossible that when better known it will prove to be the same as or closely allied to Weisbord’s Aviculo- pecten yeakeli (1926, pl. 6, fig. 3), from Venezuela, which is also quite imperfectly known, but does appear to have subangular radii, at least on the internal mold. Studied specimens.—Pal. Res. Inst. Nos. 5412, 5413, 5415. 279 COLOMBIAN DEVONIAN FAUNA: CASTER 179 ?Pterinea, n. sp. Plates fies. 13. 14 One tiny pterineoid shell is known from an imperfect external mold. While the surface pattern suggests the fossil attributed to ?Aviculopecten, sp. A, above, the large extensiform wing on this left valve suggests rather Pterinea, or some closely allied genus. The known features are sufficiently well shown by the illustrations to make further analysis unnecessary. The shell is undoubtedly immature, and until better known, does not offer the necessary diagnostic features for specific description. Illustrated specimen.—Pal. Res. Inst. No. 5412A. GASTROPODA Sp. illustrated IE, WB 4 ane, AAG In the entire collection from Colombia only one tiny crushed fragment of a gastropod shell has been recovered. What its rela- tionships may be I have not been able to determine, but illustrate it as a rarity in the faunule which promises to be of considerable interest in new collections from Colombia. Illustrated specimen.—Pal. Res. Inst. No. 5484. BRYOZOA ; As already indicated the Colombian faunule is exceedingly rich in bryozoans, especially the fenestelloids, many of which are ap- parently new to science. Fortunately many of the zoaria are pre- served sufficiently well in the residual clay to make specific de- scription possible in some cases. In view of the advisibility of having these interesting remains studied by a specialist* in the group, only two fragments of the commonest types are shown in this paper as a gesture toward a better appreciation of the nature of the fauna. The specimen illustrated on plate 14, fig. 22 (5480A) is possibly the form described by Weisbord, 1926, as Fenestella venezuelensis. Certainly close counterparts of all of Weisbord’s forms and many more are present in Colombia. Plate 14, fig. 24 (5480B) of this paper shows another type of fenestell- oid which is exceedingly common. The conical fillings of this * This study is being undertaken by Dr. A. H. MeNair, 1938. 180 BULLETIN 83 280 type of zoarium are very abundant in the collection. Figure 23, plate 14, (5481) is probably the external mold of a zoarium which is rather unusual. It was molds of this sort that Kozlowski, 1923 (p. Iot, pl. 4, figs. 19, 20, 20A), described as “Corps prob- lematique” from the Bolivian Devonian. They are abundant in our collection, and may prove to be well enough preserved in some cases to make more complete description possible. ANTHOZOA The calicinal imprint of a single tiny tetracoral, about 3 mm. in diameter, occurs in the Colombian collection. It is too poorly preserved to warrant even tentative identification (unless it be Cyathophyllum bolivianum Woztowski) but is of considerable ecological importance. The external mold of a single auloporid was found attached to a stropheodontid shell, and has been mentioned above. OSTRACODA Many of the clay blocks from Floresta show impressions of great numbers of Ostracoda. They have seemed, however, too poorly preserved to warrant description at this time, especially in view of the probability of securing much larger and more valu- able collections in the future if the material is gathered with an eye to the rarer fauna. It might be mentioned that the few forms identified in our collection recalled rather strongly the ostracod fauna of the Camden chert of Tennessee, now in process of description of Dr. R. S. Bassler, (personal communication, July 1938). TRILOBITA When better material of the Colombian Devonian sediments is discovered, it will undoubtedly be very rich in trilobite materiai that will merit detailed study. The collection has yielded a great number of external and internal impressions of tests of all sizes, but for the most part in exasperatingly poor state of preservation. There are apparently four different generic groups represented in 281 COLOMBIAN DEVONIAN FAUNA: CASTER 181 our fragments, and these are illustrated on plate 14. The fauna appears to be closely allied in its trilobite content with that which Koztowski, 1923, described from Bolivia. Phacops cf. salteri Kozlowski Pilate 4 ess leo 7— iG Cf. Phacops salteri Kozlowski (new name), Ann. de pal., Tome 12, 1923, p- 54, pl. 6, figs. 1-6. (Synonymy here given.) Relatively commonplace in the Floresta faunule are fragments of phacopid trilobites having the general aspect of Phacops salteri Koztowski of the Bolivian Devonian. A representative suite of the fragments is illustrated on plate 14. The material retrieved thus far in Colombia will not warrant detailed comparisons or augmentation of Kozlowski’s thorough analysis of the Bolivian fossils. The specimens illustrated by figures 14 and 15 may be- long to quite another type of trilobite. Illustrated specimens.—Pal. Res. Inst. Nos. 5405, 5452, 5456, 5458, 5458A, 5458B, 5476, 5478. ?Dalmanites cf. patacamayaénsis Kozlowski Plate 14, figs. 3-6 Cf. Dalmanites patacamayaensis Kozlowski, Ann. de Pal. Tome 12, 1923, DesG, ple 2s hie 2. Several pygidial casts and molds in the Floresta fauna recall Dalmanites patacamayaénsis Kozlowski from Bolivia, but in the absence of more data, even the generic identification is hazardous. The Colombian pygidia, while superficially resembling the Pata- camaya trilobite, are externally papillose and bear a row of median bosses on the axis which the Bolivian species apparently does not possess. This feature is well shown by the internal view shown on plate 14, figures 3 and 4. It has been suggested that the speci- emn illustrated in plate 14, fig. 5, recalls the proetid genus Deche- nella more than Dalmanites. Illustrated specimens.—Pal. Res. Inst. Nos. 5457, 5477, 5477A. ?Cyphaspis, sp. Plate 14, fig. 17 A single imperfect cranidium bears a superficial resemblance to the genus Cyphaspis, and whets the curiosity to see more and better preserved material from the Colombian faunule. Illustrated specimen.—Pal. Res. Inst. No. 5480. ?Homalonotus, sp. Plate. 14, fig. 18 A single relatively broad, imperfect pygidium may belong to 182 BULLETIN 83 282 an homalonotid trilobite, but only new material can establish the nature of the intriguing fragment. Illustrated specimen.—Pal. Res. Inst. No. 5465A. NOTES ON TWO NEW PAPERS DESCRIBING DEVONIAN FAUNAS OF Ele SOUTHERN HEMISPHERE While this paper was in press two recent monographs dealing with Devonian faunas of the Southern Hemisphere have come to my attention. The first, by Mendez-Alzola, 1938, (June) reached me in December, 1938. This report covers in a rather detailed manner the Devonian faunas of Uruguay previously reported on by Alzola, 1934. None of the Uruguayan species appears to have an especially close relationship to the Colombian assemblage. Dr. Alzola appears to have very fine supplimentary data on the well- known ‘‘austral” faunas of Argentina, the Falkland Islands, the Parana area of Brazil and Bolivia. As might be suspected, the affinities are closest to faunas described from the immediately ad- jacent areas. The ubiquitous “Schuchertella agassizi” (Hartt), Chonetes falklandicus Morris and Sharpe, “Leptoceelia flabellites” (Conrad), “Spirifer theringv’ Kayser, etc., in the Brachiopoda, and a variety of pelecypods, some of which are new, but most of which are identified as Clarke’s or Reed’s “austral” species, illus- trate the “orthodoxy” of the Uruguayan fauna. The trilobites apparently show certain specific and varietal dissimilarities from the well-known southern species, but augment the ‘“‘austral” flavor of the assemblage. The Uruguayan fauna may be looked upon as typically southern, and without any conspicuous northern contamination. The Alzola report enlarges our picture of the typical South American Devonian fauna and extends the distri- bution of the fauna considerably. In the current (December, 1938) number of the Quarterly Journal of the Geological Society of London, Dr. Jack Shirley publishes a paper on the Lower Devonian fauna of the Baton River beds of New Zealand. This paper adds considerably to 283 COLOMBIAN DEVONIAN FAUNA: CASTER 183 our knowledge of the interesting New Zealand Devonian first gained through Allan’s valuable paper on the Reefton fauna in 1935. The Baton River fauna has been described almost exclus- ively in terms of European (and subordinately North American) Lower Devonian species. This brings out very strikingly the “boreal” or cosmopolitan aspect of the fauna. The Baton River material, as illustrated, is even more deformed and fragmental than the South American material in this paper, and is apparently only very sparcely known. Without in any way disparaging the excellent work of Dr. Shirley, it does appear that the assignments are to be taken as only generically definitive in several instances. Apparently facies differences at least in part are responsible for the non-appearance of most of Allan’s species. I noted in a rather hasty reading of the paper only two species reported in common. Several of the cosmopolitan species( genera) described and illustrated are beyond doubt generically allied to fossils de- scribed from Colombia in the present paper. Shirley’s Stropheodonta stephani (Barrande) is indubitably closely allied to Barrande’s species, and is therefore certainly assignable to the new genus Cymostrophia of which Barrande’s species is the genotype. As brought out in the foregoing text, this genus is very well represented in the Colombian fauna. His Leptostrophia explanata (Sowerby) will bear comparison with Rhytistrophia caribbeana (Weisbord) from the Cachira series of Venezuela and the variety colombia from the Colombian Floresta series. The New Zealand specimens are concentrically deformed, but do not show the regular corrugations of the genotype, Stro- pheodonta becku Hall, or Colombian-Venezuelan specimens, but may prove to be allied to this stock. There are other compari- sons to be made, but this will illustrate the bearing of the Baton River fauna on that of this paper. As indicated above, the Baton River fauna has greater similar- ity to European-Asiatic and North American faunas than to most hitherto described faunas of South America. I would not now wish to stress any very close relationship between New Zealand and South America in the Devonian, but in tracing the the dis- tribution of the cosmopolitan “boreal” faunas too little considera- 184 BULLETIN 83 284 tion has been given to the work of Hartt and Rathbun, 1875, 1878, on the Para Devonian of Brazil and especially to the Cachira fauna of Weisbord, 1926, in Venezuela. The Colombian fauna of the present paper gives further data on these northern South Am- erican faunas, and likewise has important bearing on world distri- bution of the “boreal” faunas. Evidence is at hand, therefore, to show quite certainly that Venezuela and Colombia were boreal out- posts in early Devonian times. This necessarily introduces the question of how completely the “austral” province was isolated at this time. All who have studied southern Devonian faunas con- cur in the opinion that there must have been an isolated center of evolution for the fauna. This was presumably South Atlantic in general location. It seems probable that this province was isolated in some manner to the north, east, south and west, except for rela- tively narrow places of contact. It now appears that one of the principal regions of contact with “boreal” faunas was through what is now the Andean area of South America. Clarke, Knod, and to a less degree Koztowski, concurred on the isolation ot South American faunas. Kozlowski’s Bolivan faunas carried in- timations of intermingling, and the present paper on Colombia illustrates an advanced stage of the admixture further north. lye daresay: wthat. there ) was s muche they y samemmisola. tion between “‘austral’” and ‘‘boreal” faunas in Devonian times as we see today on the West Coast of South America where the tropical fauna of the Panamic province meets the “austral” cold water fauna of the Humbolt Current. If the paleogeographers of the future, using present criteria, were to examine the fossil evidence now in process of formation on the continental shelf of the East Coast of the United States they would probably require at least a long barrier peninsula if not an “isthmian link” running off to the northeast in the vicinity of Cape Hatteras, where the New England and Gulf Stream faunas meet so abruptly. Addi- tional study brings out, however, considerable intermingling of certain elements of both faunas. This sort of condition seems better to fit the faunal relations of the Devonian of South America, now that intermingling is quite well established. £85 COLOMBIAN DEVONIAN FAUNA: CASTER 185 BIBLIOGRAPHY LIST OF WORKS CITED Allan, R. S. The Fauna of the Reefton Beds, (Devonian) New Zealand; with notes on Lower Devonian animal communities in re- lation to the base of the Devonian; New Zealand Geol. Surv., Pal. Bull. 14, 72 pp., 1 tab., 1 map, 5 pls., 1935. Asselbergs, E. Déseription des Faunas Marines du Gedinnien de 1’Ar- denne; Mus. Royal d’Hist. Nat. de Belg., Mem., No. 41, 73 pp-, 6 pls., 1930. Barrande, J. Systeme Silurien du Centre de la Bohéme, vol. 5, text and plates, 1879. Beecher, C. E. The origin and significance of spines. A study in evolu- tion; Amer. Journ. Sci., (4) vol. 6, pp. 1-20; 125-136; 249- 268 3329-359, 1898. Billings, K. -aleozoic Fossils. Vol. 1, containing descriptions and figures of new or little known species of organic remains from the Silurian rocks; Geol. Surv. of Canada, 426 pp., 401 figs., 1865. Biological Society of Washington. International Rules of Zodlogical No- menelature; Biol. Soe. Wash., Proe., vol. 39, pp. 75-104, 1926. Breger, C. L. On Eodevonaria, a new sub-genus of Chonetes; Amer. Journ. Sei. (4), vol. 22, pp. 534-536, 1906. Clarke, J. M. Molluscos Devonianos do Estado do Para; Arehiv. do Mus. Nae. do Rio de Janerio, vol. 10, pp. 49-174, pl. 1-8, 1899. The Oriskany Fauna of Becraft Mountain, Columbia Coun- ty, New York; N. Y. State Mus., Mem. 3, vol. 3, 128 pp., 9 pls. 1900. Early Devonic History of New York and Eastern North America; N. Y. State Mus., Mem. 9, Part 1, 252 pp., maps and text figures, 48 pls., 1908. Early Devonie History of New York and Eastern North America; N.Y. State Mus., Mem. 9, pt. 2, 250 pp., maps and text figs., 34 pls., 1909. Fosseis Devonianos do Parani; Serv. Geol. e Mineral. do Brasil, Mon. 1, 353 pp., 27 pls., 1913. Clarke, J. M., and Schuehert, C. The nomenclature of the New York Series of Geological formations; Science (n. s.), vol. 10, pp. 874-878, 1899. Conrad, T. A. Observations on the Silurian and Devonian systems of the United States, with the description of new organic remains ; Aead. Nat. Sci. Phila., Journ., vol. 8, pt. 1, pp. 228-280, pls. 12-17, 1842. Davidson, T. British Fossil Brachiopoda; Mon. Paleontog. Soc. 1864- 1880. 186 d’Orbigny, A. Dunbar, C. O. BULLETIN 83 286 Voyage dans 1’Amérique Méridionale, vol. III, Paléontol- ogie, 1842. Stratigraphy and correlation of the Devonian of western Tennessee; Tenn. State Geol. Surv., Bull. 21, 127 pp., 11 pls. and figs., 1919. New species of Devonian fossils from western Tennessee ; Conn. Acad. Arts and Sci., Trans. vol. 23, pp. 109-158, pl. 1-5, 1920. Dunbar, C. O. and Condra, G. E. Brachiopoda of the Pennsylvania Sys- Fenton, C. L. Fenton, C. L., Fischer, P. Fcerste, A. F. George, T. N. tem in Nebraska; Neb. Geol. Surv., Bull. 5 (2), 377 pp., 25 figs., 43 pls., 1932. Studies of evolution in the genus Spirifer; Publ. of the Wagner Free Inst. of Sci., vol. 2, 436 pp., 204 figs., 50 pls., 1931. and Fenton, M. A. The stratigraphy and fauna of the Hackberry Stage of the Upper Devonian; Mus. Geol., Univ., Mich., Contrib., vol. 1, 260 pp., 9 figs. 45 pls., 1924. Manuel de Conchyliologie et de Paléontologie Conchyli- ologique ou Histoire Naturelle des Mollusques Vivants et Fossiles suivi d’un Appendice sur les Brachiopodes par par D. P. Oehlert; 1369 pp., 1158 figs., 23 pls., Paris, 1887. Silurian fossils from the Kokomo, West Union, and Alger horizons of Indiana, Ohio and Kentucky; Cinti. Soc. Nat. Hist., Journ., vol. 21, pp. 1-41, 1909. Notes on Silurian Fossils from Ohio and other Central States; Ohio Journ. Sci., vol. 17, pp. 187-204, 233-267, 1917. The British reticulate Carboniferous Spiriferide; Quart Journ. Geol. Soc. (London), vol. 88, pp. 516-575, pls. 31- 35, 1932. (Bibliography very complete.) Gos:elet, J., Barrois, Ch., Leriche, M., Crepin, A., Pruvost, P., and Du- Grabau, A. W. Hall, J. bois, G. Déseription de la faune Siluro-Devonienne de Liévin; Soc. Géol. du Nord., Mém., Tome 6, Fase. 2, pp. 70- 225, pls. 10-17, 1920. Devonian Brachiopoda of China. I. Devonian Brachiopoda from Yunnan and other districts of South China; Text. Paleontologica Sinica, ser. B, vol. 3, fase. 3, 545 pp., 67 figs., 188 tables, 1931; Idem, Plates, pp. 546-752, 54 pls., 1933. Geology of New York. Part IV, comprising the survey of the fourth geological district; 683 pp., tables and figs., 1843. Deseription of the organie remains of the lower Middle Di- vision of the New York System; Nat. Hist. N. Y., Pal., vol. 2, 362 pp., 1852. Paleontology: Volume 3, Containing descriptions and figures of the organic remains of the Lower Helderberg Group and the Oriskany Sandstone; Nat. Hist. N. Y., Geol. Surv. N. Y., Text, 532 pp., 1859; Part II, 120 pls., 1861. Paleontology: Volume 4, Part 1, Containing descriptions and figures of the fossil Brachiopoda of the Upper Helder- berg, Hamilton and Chemung Groups. Nat. Hist. N. Y., Geol. Surv. N. Y., 428 pp., 63 pls., 1867. 287 COLOMBIAN DEVONIAN FAUNA: CASTER 187 Hall, J., and Clarke, J. M. Paleontology: Volume 8. An introduction to Hartt, C.F. the study of the genera of Paleozoic Brachiopoda; Part 1, Geol. Surv. N. Y., 367 pp., 2 tab., 20 pls., 1892; Part 2, 394 pp., pls. 21-84, 1893. Contributions to the geology and physical geography of the Lower Amazonas; Buffalo Soc. Nat. Sci., Bull., vol. 2, pp. 236-271, 1874. (in Rathbun, R.) The Devonian Brachiopods of the Pro- vince of Para, Brazil; Boston Soc. Nat. Hist., Proe., vol. 20, pp. 14-39, 1878. Hartt, C. F., and Rathbun, R. Morgan Expedition of 1870-71: on the De- Helmbreeht, W. Holtedahl, O. Katzer, I. Katzer, F. Kayser, E. Kindle, Ei. M. vonian Trilobites and Mollusks of Ereré, Provinee of Para, Brazil; Lyceum of Nat. Hist. N. Y., Ann., vol. 11, pp. 110- 127, 1875. , and Wedekind, R. Versuch einer biostratigraphichen Gliederung der Siegener Schichten auf Grund von Rens- selerien und Spiriferen; Gliickauf. 59, Nr. 41, pp. 949-953, 1923. The Strophomenide of the Kristiania Region; Vidensk. Skr. I. Math. Nature. K1., Nr. 12, pp. 115, 16 pls., 1916. Das Amazonas-Devon und seine Beziehungen zu den an- deren Devongebieten der Erde; Sitzungsberichte d. k6nigl. bohmischen Gesell. der Wissensch. Math.-naturwiss. Classe, VOL SOT Ppa t-o0s ee ple SOT Grundziige der Geologie des Unteren Amazonasgebietes (des Staates Para in Brasilien) ; 129, pp., 15 pls., text fig., maps. Leipzig, 1903 (Max. Weg). Beitrige zur Kenntnis einiger paleozoise..er Faunen Siid- Amerikas; Zeitschr. d. Deutsch. Geol. Gesell. vol. 49, pp. 274-317, pls. 7-12, 1897. Alguns Fosseis Paleozoicos do Estado do Parana; Revisto do Museo Paulista, vol. 4, pp. 301-311, pls. 1, 2, 1900. The Devonian and Lower Carboniferous faunas of south- ern Indiana and central Kentucky; Bull. Amer. Pal., vol. 3, No. 12, 111 pp., 1899. The Devonian fossils and stratigraphy of Indiana; Ind. Dept. Geol., Nat. Res., Ann. Rep. 25 (1900), pp. 529-758; 773-775, 1901. The Onondaga fauna of the Allegheny region; U. S. Geol. Surv., Bull. 508, 144 pp., 13 pls., 1912. Kindle, HE. M., and Breger, C. L. Paleontology of the Niagara of northern Kknod, R. Kozlowski, R. Lake, 12 Indiana; Ind. Dep. Geol. and Nat. Res., Ann. Rep. 28, pp. 397-486, 1904. Devonische Faunen Boliviens; Beitr. zur Geol. u. Pal. von Stid-Amerika, pt. 14, Neues Jahrb. Miner. Geol. Pal., Beil. Bd. 25, pp. 493-600, pls. 21-31, 1908. Fauna Devonienne de Bolivia; Ann. de Pal. Tome 12, 112 DD ie Lose LOM se NOS. Les Brachiopodes Gothlandiens de la Podolie Polonaise ; Paleontologia Polonica, Tome 1, 254 pp., 95 figs., 1 map, 12 pls., 1929. The Trilobites of the Bokkeveld beds; South Africa Mus., 188 BULLETIN 83 . 288 Ann., vol. 4, pp. 201-220, pls. 24-28, 1904. Lambert, R., and Mendez: Alzola, R. Un nuevo yacimiento fosilifero devou- ico en el Departmento de Durazno; Inst. Geol. del Uruguay, Bol. 24, pp. 169-174, 1 pl., 1938. Licharew, B. Ann. Soc. Pal. Russie, vol. 7, pp. 117-137, 1928-1929, (1930), Liddle, R. A. The Geology of Venezuela and Trinidad; 552 pp., 83 pls., 22 figs., J. P. McCowan, Ft. Worth, Texas, 1928. MecLearn, F. H. Daleoutolony of the Silurian rocks of Arisaig, Nova Scotia; Canada Geol. Surv., Mem. 137, 179 pp., 30 pls., 1924. Maillieux, E. Le Couvénien de 1’Ardenne et ses Faunes; Mus. Roy. d’Hist. Nat. de Belg., Mém. 83, 1938. Mendez-Alzola, R. Contribucion al Conocimiento de la Fauna Devoniea de Ricon de Alonso; Inst. de Geol. y Perfor. del Uruguay, Bol., No. 21, pp. 21-55, pls. 3, 4, 1934. Fosiles Devonicos del Uruguay; Inst. Geol. del Uruguay, Bol. 24, pp. 1-115, 15 pls., 1938. Morris, J., and Sharpe, D. Description of eight species of brachiopod shells from the Falkland Islands; Geol. Soe. London, Proe., vol. 2, pp. 274-278, pls. 10, 11), 1846: Nettleroth, H. Kentucky fossil shells, a monograph of the fossil shells of the Silurian and Devonian rocks of Kentucky; Ky. Geol. Surv., Frankfort, Ky., 245 pp., 36 pls., 1889. Olsson, A. A., and Caster, K. E. Devonian iossils from Colombia, Souih America, (abstract), Proc. Geol. Sec. Amer., 1936, p. 369, NO STE Paeckelmann, W. Versuen einer zusammemfassenden Systematik der Spir- iferide King; Neues Jahrb. ftir Miner. ete. Beil.-Bd. 67, Abt. B, pp. 164, 1O3ar Rathbun, R. On tke Devonian Brachiopoda of Ereré, Browinee of Para, Brazil; Buffalo Soc. Nat. Sci., Bull., vol. 2, pp. 236- 271. pls. 8-10, 1874. The Devonian Bracziopoda of the Province of Para, Bra- zil; Boston Soc. Nat. £ci., Proc., vol. 20, pp. 14-39, 1878 Raymond, P. E. The Developmental Cnanges in some common D_vonian Brachiopods; Amer. Journ. Sci., vol. 17 (4), pp. 279-300, pls. 12-18, 1904 Reed, F.R.C. Brachiopoda fiom the Bokkeveld beds; South Africa Mus., Ann., vol. 4, pp. 165-197, pls. 20-23, 1903. Some new fos:ils from the Bokkeveld Beds; South African Geol. Meg., vol. 3, pp. 301-316, pls. 16, 17, 1906. New Fossils from the Bokkeveld Beds; South African Mus., Ann., vol. 4, pp. 381-405, pls. 47, 48, 1908. Revision of the fauna of the Bokkeveld Beds; South Af- frican Mus., Ann., vol. 22, pp. 27-225, pls. 4-11, 19205. Scheibe, Robert Informe acerca de las investigaciones en la regién de Tocaima La Virginia y Girardot; Colombia Min. Indust., Bib. Dep. Minas y Petrol., Compilacion de los estudios geologicos officales en Colombia, 1917-1933, vol. 1, pp. 42-51, 1933 289 COLOMBIAN DEVONIAN FAUNA: CASTER 189 Schenk, E. T., and MeMasters, J. H. Procedure in taxonomy including a reprint of the International Rules of Zoédlogical Nomen- clature with summaries of Lee ae pee to the present date; Stanford University Press, 193 Schmidt, W. HE. Cultrijugatus Zone unil Sitoree aaron stidlich der Attendorn-Hisper Doppelmulde; Jahrb. kénigl. Preuss. Geol. Landes., Bd. 33, pt. 2, pp. 265-318, pls. 22, 23, 1912. Schuchert, C., Synopsis of American fossil Brachiopoda, ineluding bibli- ography and synonymy; U. 8. Geol. Surv., Bull. 87, 1893. Historical Geology of the Antillean-Caribbean Region; 811 pp., 16 pls., 107 figs. John Wiley and Sons, New York, 1935. Schuchert, C., and Cooper, G. A. Brachiopod genera of the suborders Orthoidea and Pentameroidea; Peabody Mus. Nat. Hist., Mem., vol. 4, pt. 1, 270 pp., 36 figs., 29 pls., 1932. Schuchert, C., and LeVene, C. M. Brachiopoda, Generum et Genotyporum Index et Bibliographia; Fossilium Cataogus I: Animalia, Pars 42, 140 pp., 1929. Schuchert, C., (and Maynard, T. P.) Systematic paleontology ef the Low- er Devonian rocks of Maryland: Brachiopoda; Md. Geol. Surv., Lower Devonian, pp. 290-449, 1913 Schwartz, E.H.L. South African Paleozoic Fossils; Albany Museum, Records, vol. 1, pp. 347-404, pls. 6-10, 1916. Scupin, H. Spiriteren Deutschlands; Paleont. Abhandl. N.F., Bd. 4, latte Gy JOO: Shirley, Jack Some aspects of the Siluro-Devonian Boundary Problem; Geol. Mag., vol. 75, pp. 353-362, 1938. The fauna of the Baton River beds( Devonian), New Zea- land; Quart. Jour. Geol. Soc. London, vol. 94, pt. 4, pp. 459-506, pls. 40-44, 1938. Spriesterbach, J. Die Oberkoblenzschichten des Berzischen Landes und Sauerlandes; Jahrb. d. koénigl. Preuss. Geol. Landes., Bd. 45, p. 432, vl. 6; 1925. Stainbrook, M.A. Atrypa and Stropheodonta from the Cedar Valley Beds of Iowa; Journ..Pal., vol. 12, pp. 229-256, pls. 30-35, 1938. Stewart, G.A. Fauna of the Silica Shale of Lucas County; Geol. Surv. Ohio, (4), Bull. $2, 76 pp., 5 pls., 1 map, 1927. Thomas, I. Neue Beit-iige zur Kenntniss der devonischen Fauna Ar- gentiniens; Zeitsch. d. Deutsch. Geol. Gesell., 1905, pp. 233-£90, pls. 11-14, 3 figs., 1905. Twenhofel, W.H. Tie Anticosti Island Faunas; Canada Geol. Surv., Mus. Bull. 3, (Geol. sez. 19), 38 pp:; 1 pl., 1914. Ulrich, A. Paleczoise ec Versteimerungen aus Bolivien; Beitr. zur Geol. u. Pal. von Siidamerika, pt. 1, 110 pp., 5 pls., 1892. Van Cleave, H.J. An index ts the International Rules of Zodlogical Nomenelature; Amer. Microscop. Soe., Trans., vol. 52, pp. 322-325, 1938 Waleott, C. D. Paleontology of the Eureka District; U. 8. Geol. Survey, Mon. 8, 298 pp., 24, pls., 1884. 190 BULLETIN 83 290 Warthin, A. S. Jr., and Cooper, G. A. New formation names in the Michi- Weisbord, N. EH. Weller, 8. Williams, H.S8. Williams, H. 8., gan Devonian; Wash. Acad. Sci., Journ., vol. 25, No. 12, pp. 524-526, 1935. Venezuelan Devonian Fossils; Bull. Amer. Pal., vol. 11, No. 46, pp. 220-272 (1-52), pls. 35-41 (1-7), 1926. The Paleozoic Faunas; Geol. Surv. N. J., Rep. on Pal., vol. 3, 462 pp., 53 pls., 1903. The Mississippian Brachiopoda of the Mississippi Valley Basin; Ill. State Geol. Surv., Mon. 1, text: 508 pp., 30 figs., plates, 186 pp., 83 pls., 1914. Recurrent Tropidoleptus zones of the Upper Devonian in New York; U. 8. Geol. Surv., Prof. Paper 79, 103 pp., 18 figs., 6 pls., 1913. and Breger, C. L. The Fauna of the Chapman Sandstone of Maine, including descriptions of some related species from the Moose River Sandstone; U. 8S. Geol. Survey, Prof. Paper 89, 347 pp., 2 figs., 27 pls., 1916. 291 CoLOMBIAN DEVONIAN FAUNA: CASTER 191 BXPLANATION OF PLATES Photographs made by the author with the aid of Mr. Stewart Jones, student of Geology at the University of Cincinnati. (Plates furnished ready for insertion by the University of Cincinnati Museum through the Faber Fund for Paleontology) PLAT EI. (VIL) All specimens illustrated on this and the succeeding plates are from the Devonian strata of Floresta, Department of Boyaca, northeastern Col- ombia, South America. 14. Cymostrophia schucherti Caster, n. sp. —_ Soh et, OR) et AS Dorsal external mold. See also plate 6, figs. 7-10. Holotype. Pal. Res. Inst. No. 5426;-x 1 15. Cymostrophia schucherti Caster, n. sp. —...._. 2 ee et WAS Enlargement of the surface of the holotype specimen (fig. 14). Pal. Res. Inst. No. 5426; X83 16. Cymostrophia schucherti Caster, n. sp. E _ 48 Enlargement of the surface of a dorsal Sastar Peta on ech concentric varices are unusually well developed. Paratype. Pal. Res. Inst. No. 5427; X 3 ieeecymostrophia schucherti Caster, n. sp. —2...... — 2 48 Enlargement of the well corrugated aeewee eeued to the fabric seersucker. For additional ilusteations of this species see also plate 2, figs. 1, 4-6; pl. 3, fig. 2; pl. 6, figs. 7-10, 15, and pl. 10, figs. 4-6. Paratype. Pal. Res. Inst. No. 5428; xX 3 192 BULLETIN 83 292 EXPLANATION OF PLATE 1 (7) Figure Page i; Pholidops floreste” Caster, n. sp. 3 ee 16 Internal mold. Paratype. Pal. Res. Inst. No. 5454A; x 1 2; “Pholidops) florest2: (Casters n> ‘Spey Eee 16 Enlargement of internal mold figured above. Paratype. Pal. Res. Inst. No. 5454A; X 38 Pholidops) florestz! Caster, n. sp. 0.2. eee 16 Internal mold showing musculature. Holotype. Pal. Res. Inst. No. 5454; xX 3 40 Leptenay boyacay (Caster, ny Spy 5 aa EE ee 19 Ventral internal aaelldl, somewhat compressed laterally. Para- type. Pal. Res. Inst. No. 5468; x 1 5. -septenay boyacal | Casitename ssi yeas eee ele 18) Ventral internal mold of an undeformed specimen, showing the strongly delimited diductor scars and papillose interior of the shell. Paratype. Pal. Res. Inst. No. 5469; x 1. ive) C. Leptzna boyaca Caster, n. sp. — ce 2 19 Side view of same specimen seen in Ve, Be ishowing impres- sion of the diaphragmal ridge and the degree of shell genic- ulation. Paratype. Pal. Res. Inst. No. 5469; x 1 Weptena boyaca Caster nt (sp, 2 ee eee 19 Dorsal internal mold showing the imprint of the strongly elevated subumbonal platform and muscle seats. See also plate 10, fig. 1. Holotype. Pal. Res. Inst. No. 5470; xX 1 85= Leptena iboyaca (Casters in), Spit] ee ee ee 19 Ventral external mold showing usual auricular tendency of the hinge extremities and wavering irregularity of the radial ornament on the geniculated surface. Paratype. Pal. Res. Inst. No. 5471; XxX 1 9. -Weptwena’ boyaca Caster. mn. ‘sp. 22.2 eae Eee 19 Dorsal external mold. Paratype. Pal. Res. Inst. No. 5472; x 1 10. Leptena boyaca Caster, ny isp.t: 2. ee ee 19 Enlargement of specimen shown in fig. 9, bringing out surface details. Paratype. Pal. Res. Inst. No. 5472; x 2.5 Jd eptena boyaca ‘Caster: ni isp. 2. ee eee 19 Dorsal external mold of immature individual showing early establishment of the: peltate outline. Paratype. Pal. Res. Inst. No. 5474; xX 1 12. Leptzena boyaca Caster, n. sp. ea ol de Need oe eer ae 19 Enlargement of specimen in ae 11, to show details of the narrow interspaces and wide coste at this stage which is in contrast to adult ornament as shown in figure 10. Paratype. Pal. Res. Inst. No. 5474; X 1.75 1B. lLeptoenamibeyacas (Casters cree (Spo yy eee cae eae 19 Umbonal portion of a dorsal external mold showing the trans- ition of ornament from immature to adult condition. For additional illustrations of this species see plate 10, figs. 1-3. Paratype. Pal. Res. Inst. No. 5475; x 1 Continued on previous page PLATE 7, VOL. 24 BuLuL. AMER. PALEONT. No. 83, Pu. 1 PLATE II (VIII) ZGymostrophiaawaringin@asteky ne sSP. se - 54 Enlargement of a portion of the ventral external mold of the holotype. See also pl. 3, fig. 1. Pal. Res. Inst. No. 5448; x 25 ACGVMOStLOPh ia Wariner ws GaSbery MS. ae 54 Still greater enlargement of the surface of the holotype, show- inx very fine reticulation. Pal. Res. Inst. No. 5448; x 3 194 BULLETIN 83 294 EXPLANATION OF PLATE 2 (8) Figure Page 1. Cymostrophia schucherti Caster, n. sp. 48 Ventral external mold of umbonal area showing curious asym- metrical tendency sometimes occurring in this species. See also plate 3, fig. 2, for enlargement of this specimen. Paratype. Pal. Res. Inst. No. 5466A; X 1 2,3. Strophonella floweri Caster, n. sp. — 109 See also plate 10, figs. 9-11. Holotype. Pal. Res. Inst. No. 5486; XxX 1 4, Cymostrophia schucherti Caster, n. sp. 48 Ventral internal mold which has been slightly compressed laterally. Paratype. Pal. Res. Inst. No. 5449; x 1 5. Cymostrophia schucherti Caster, n. sp. 48 Fragment of the external mold of specimen 5449 (fig. 4). See also plate 6, fig. 15, showing concentric varices. Paratype. Pal. Res. Inst. No. 5450; x 1 ; 6. Cymostrophia schucherti Caster, n. sp. 48 Ventral internal mold showing very well the characteristics of the musculature. The differential rippling of the surface has left its imprint on the anterior part of the mold. For additional illustrations of this species see pl. 1, figs. 14-17; pl. 3, fig. 2; pl. 6, figs. 7-10, 15; pl. 10, figs. 4-6. Paratype. Pal. Res. Inst. No. 5451; xX 1 7. Dictyostrophia cooperi Caster, n. sp. 60 Dorsal external mold of the holotype. See also pl. 4, fig. 9 and pl. 6, fig. 1. Pal. Res. Inst. No. 5449A; x 1 8. Dictyostrophia cooperi Caster, n. sp. 9. 60 Enlargement of the umbonal features of the external mold of the holotype, showing the absence of concentric varices, alter- nation of radii and slight convexity of the mid-zone of the radii between the principal radii. Holotype. Pal. Res. Inst. No. 5449A; X 1.5 9. Dictyostrophia cooperi Caster, n. sp. 60 Still greater enlargement of the umbonal portion of the ex- ternal mold of the holotype to show character of the radii. Holotype. Pal. Res. Inst. No. 5449A; X 3 10. Dictyostrophia cooperi Caster, n. sp. 60 Enlargement similar to fig. 9, above, of the anterior portion (geniculated zone) of the external mold of the holotype, show- ing the reticulation occurring there. Holotype. Pal. Res. Inst. No. 5449A; X 3. 11. Dictyostrophia cooperi Caster, n. sp. 60 Portion of the ventral external mold of an individual that may have suffered slight lateral compression, although to no great extent. See also pl. 6, figs. 5, 6. Paratype. Pal. Res. Inst. No. 5455; xX 1 12. Dictyostrophia cooperi Caster, n. sp. — 60 Enlargement of the specimen seen in fig. 11, to show retic- ular detail. For additional illustrations of this species see pl. 4, fig. 9; pl. 6, figs. 1-6. Paratype. Pal. Res. Inst. No. BYIIS <3 Continued on previous page No. 83, Pu. 2 BULL. AMER. PALEONT. PLATE 8, VOL, 24 ene Se S558: 14, 15. 16-18. 1S), 20. 21. PLATE III (IX) Stropheodonta kozlowskii Caster, n. sp. = ees Plasticine cast and original dorsal internal mold showing un- usually well the internal features of the genus and species. Holotype. Pal. Res. Inst. No. 5406A; xX 1 Megastrophia pygmza Caster, n. sp...» ss Plasticine casts of the ventral interior of the holotype speci- men. Figs. 17, 18 are two views of the same cast showing the secondarily opened delthyrium, median deltidial buttress, and incomplete hinge crenulation, as well as details of the muscu- lature. See also pl. 5, figs. 1, 4; pl. 6, figs. 11-18; pl. 8, figs. 13, 14. Holotype. Pal. Res. Inst. No. 5398. Fig. i, < Anoplotheca cf. silvetii (Ulrich) —. External ventral mold and plasticine cast showing known fea- tures of this unique specimen in the Colombian collection. Pal. Res. Inst. No. 5460C. Fig. 25, < 1; fig. 26, X 3 7Australospirifer cf. antarcticus var. 2 — 2 External ventral mold (reversed illumination). For addi- tional illustrations of this species see pl. 13, figs. 19, 20. Pal. Res. Inst. No. 5407G; xX 1 Acrospirifer olssoni Caster, n. sp. od PL See Internal ventral mold showing median Septum. For addition- al illustrations of this species see pl. 9, fig. 24; pl. 10, figs. 15, 18; pl. 11, figs. 10-12; pl. 12, figs: 10-13. Paratype. Pal. Res, Inst. No. 5447, Bic. 28) x ie fic, 295 x 2 Vitulina, sp. ADS eS cae ETS EIS SS, Be al nh en: Exte-neal molds of a 1 unique e pustulose ‘shell of unknown ‘rela- tionships. See also pl. 13, fig. 23. Pal. Res. Inst. No. 5460. IBiiee, BH, SK GIS ite, Bul, SK 8} 140 143 163 156 142 204 BULLETIN 83 304 EXPLANATION OF PLATE 7 (13) Figure Page 1. Schellwienella goldringz Caster, n. sp. — ~~ 116 Internal dorsal mold, showing the short dental plates, low median septum, and flabellate muscle scars. See also pl. 8, fig. 2. Holotype. Pal. Res. Inst. No. 5463; xX 1 2. Schellwienella goldringe Caster, n. sp. 116 Fragmental external ventral mold. Paratype. Pal. Res. Inst. No. 5466; X 1 3. Schellwienella goldringz Caster, n. sp. ee 116 External ventral mold. See pl. 8, fig. 1, for plasticine cast. Paratype. Pal. Res. Inst. No. 5467; x 1 4, Eodevonaria reedi Caster, n. sp. —__--_____--___ 129 External dorsal mold. For additional illustration of this species see pl. 9, figs. 1-2. Paratype. Pal. Res. Inst. No. 5403; xX 1 : 5,6. Eodevonaria imperialis, var. transversa Caster, n. var. 128 External dorsal molds showing the character of the surface ornament. See also pl. 11, figs. 18-20. Paratype. Pal. Res. Inst. No. 5404.° Fig. 5, X 1; fig. 6, x 2 7,8. Chonetes aff. billingsi Clarke, gens. 2 Sates Fragmental external ventral mold and ‘plasticine ‘cast. Pal. Res. Inst. No. 5460B. Fig. 7, x 1; fig. 8, X 3 9,10. Eodevonaria imperialis. var. parva Caster, n. var. — — 126 Plasticine cast of a dorsal internal aaolld showing the medion! septum, strong hinge crenulations. and weak cardinal plates. Paratype. Pal. Res. Inst. No. 5484B. Fig. 9, X 1; fig. 10, X 2 11,12. Eodevonaria imperialis Caster, n. sp. 122 Exterlan dorsal mold showing the postlateral fasciculation of the varices, slight median fold, and peripheral flange. Paratype. Pals Res, Inst) Nos 54345 Mick 1 Iie, INS, S< ile sees NG, SK 2 17,18. Eecdevonaria imperialis Caster, n. sp. —. 122 Internal dorsal mold, showing the median septum, and large, flabellate muscle impressions. For additional illustration of this species see pl. 9, fig. 3. Paratype. Pal. Res. Inst. No. yHBB}, | Ihe, A SK INS salen, AS SK aL ts) 19,20. Productella cf. spinulicosta Hall 121 Internal ventral mold with some of the surface features superimposed. Only specimen found in the collection. Pal. Res. Inst. No. 5459. Fig. 19, x 1; fig. 20, x 3 Pee Camaxrotor chia ‘Spi eit, ee a 75) Pal. Res. Inst. No. 5427A: ) BlythaycolombiananC@aster;, may )Sypsg 146 Internal ventral mold of the holotype. See figs. 1 and 2, above. Pal. Res. Inst. No. 5489A; X 1 5 Gnueblythancolompbianay Caister. rien po pe 146 Internal ventral mold of an immature individual. Note the impression of the external spines on the internal mold. Para- type. Pal. Res. Inst. No. 5441. Fig. 5, & 1; fig. 6, XK 2.5 ft ~Elytha xcolombiana (Caster; ni sp) 2 146 Internal dorsal mold of an immature individual. Paratype. Pal. Res. Inst. No. 5442; xX 1 82) bby tha colombianay Casters ii) sos essen ease aes eS 146 Fragment of an external ventral mold, showing the median depression at the base of each spine which corresponds on the exterior to the position of the internal median dia- phragm. Paratype. Pal. Res. Inst. No. 5448; x 1 oe eMevanteris australis (Caster, npsp, eee 173 External dorsal mold showing the prominent lamelle, and the faint radii toward the front of each lamella. See interior of same individual, fig. 12, below. Holotype. Pal. Res. Inst. No. DAZ 2 ol: 10. Meganteris australis’ Caster, n. sp. 22 SS eee 173 External ventral (?) mold of an immature individual. Note faintness of the radii on the varices. Paratype. Pal. Res. IDNs INO BARBS —S< Al 11. Meganteris australis Caster, n. sp. 173 Internal ventral mold of an individual of about the same size as the holotype. Note short but strong hinge teeth, well developed lameliw, and obscure muscle scars. Paratype. Pal. Res. Inst. No. 5424; x 1 12°14, Mesanteris) australis ‘Caster, ns sp.) 2 2 Eee 173 Internal dorsal mold of the holotype. Note impression of bi- partite cardinal process, rugged crura, and recessed muscle seats. Fig. 13 shows impression of the cardinal area, and fig. 14 is a plasticine cast of the same. See also fig. 9, above, for external characteristics. Holotype. Pal. Res. Inst. No. 5425;
    LRRETACHSL RSLS Sioa ee The base of a zoarium, probably the same “species - as : fig. 22, above. Pal. Res. Inst. No. 5480B; xX 1 180 ITS) 218 BULLETIN 83 318 EXPLANATION OF PLATE 14 (20) Figure Page 12h a cops cis s al teri ao z/l OiyS Kees 181 Fragment of an external mold showing the highly character- istic glabellar swelling and papillose surface of the species. IFES INOS; Iba INO, BARA, Iie, il SK Be ine, Bo SK Al 3-6. ?Dalmanites cf. patacamayaénsis Kozlowski _._ 181 External and internal molds of the pygidium “o8 a Werilopee which recalls the Bolivian form described by Kozlowski. The median spines may prove to be a specific character, how- ever. Figs. 3, and 4, an external mold; fig. 5, an internal mold showing the marginal flange very well. Pal. Res. Inst. Nos. oATTA and 5477, respectively. “Figs 38) xX 3); fie, 4° xX Te iies By ><. il (ee wehacops) clan Salt erie Keo 21) osyy;s Ki geese een 181 Fragment of an external dorsal mold of part of the thorax and pygidium. The circular hole is a cavity left by the weathering of a crinoid stem. Pal. Res. Inst. No. 5405; &« 1 8. Phacops cf. salteri Kozlowski —_ 2 elon Internal dorsal mold of a portion of. “ane Wiorae "Pail, Res. Inst. No. 5458; Xx 1 DOS ehacopsiict.salteri Xozlowski Sees 181 Portion of an external mold and plasticine cast of the cepha- lon of an average sized adult. Pal. Res. Inst. No. 5458A; X 1 122, iPhacops) ci salteritc Kozlowski 22 220 2 ee eee 181 External cephalic mold of a young individual. Pal. Res. IONS Us INO. YH, Agito, “aL DK als shee, IB. SK B 3s ehacopsm ct salteritykozlows ki ee 181 Plasticine cast of the head mold illustrated as figs. 11 and WAS SKB Asse eh acopsmchss sal teri 2 keo Zi oj wis clea nee 181 An internal cephalic mold presumably belonging to this species, but showing exceptionally widely spaced eyelobes. Pally INES, Ibngied! INO, HDG, lies U4 Se alg iver Il, OK BS 116.9, Bhacops (ct. ‘salterioiozlowski 225). ee 181 Internal mold of a portion of the mentum of this trilobite. Pal. Res. Inst. No. 5478; X 1 ie 2C yphaspisy.bSip.. ees se ee eRe ee) oe Ae Ne Te ed ee 181 A portion of the cranidium of an unknown trilobite which shows some resemblance to the form illustrated by Kozlow- ski, 1923, pl. 4, fig. 11, from the Bolivian Devonian. Pal. Res. Imisit., Nios 5438037 x 18: 2homalonotuss (spy ee te ee ee 181 Internal mold of the pygidium of an unknown trilobite. Pal. Res. Inst. No. 5456A; xX 1 19-21. Cypricardinia cf. subindenta Weisbord — eee WaT External mold and squeezes of a fragment of shell presum- ably closely related to Weisbord’s Venezuelan species. Note the chevron markings on the enlargement of the squeeze, Continued on previous page PLATE 20, VOL. 24 Buu. AMER. PALEONT, No. 83, Pu. 14 AMERICAN PALEONTOLOGY © aS sh i VOT XXIV NOE 84°& 85 IrHaca, N. Y. TE SP OA: BULLETINS OF AMERICAN PALEONTOLOGY Volume 24 Number 84 Notes on Cypraea Heilprini Dall and Cypraea Chilona Dall with New Species from the Pliocene of Costa Rica By WILLIAM MARCUS INGRAM April 75, 1939 PALEONTOLOGICAL RESEARCH INSTITUTION Ithaca, New York Ue Ss Ay NOTES ON CYPRAEA HEILPRINI DALL AND CYPRAEA CHILONA DALL WITH NEW SPECIES FROM THE PLIOCENE OF COSTA RICA* By WILLIAM Marcus INGRAM Cornell University NOTES ON CYPRAEA HEILPRINI DALL AND CYPRAEA-~CHILONA DALL During a recent study of fossil Cypreide in the United States National Museum the writer had opportunity to examine type material of Cyprea heilprint Dall and Cyprea chilona Dall. A comparison of the former species with the illustration accom- panying Dall’s original description showed that the figure of the ventral view was idealized, and does not represent the true char- acter of the ventral surface of the shell. Illustrations of the holo- type specimen of C. heilprini Dall are included here to aid one in identifying this species when a comparison with the holotype is impossible. (Figs. 1, 2.) Dall’s original description of heilprini is also included here with the writer’s comments about the figure of the ventral view. Cyprea heilprini Dall Plate dy fies. 1 2 Cyprea heilprini Dall, 1890, Trans. Wagner Free Inst. Sci., vol. 3, p. 166, pl. 11, figs. 2, 2a. This species is best described by comparison with C. pinguis. It dif- fers from the latter in being more cylindrical; in its somewhat straighter and proportionally narrower aperture, which is also less curved at the posterior commissure; in being less elevated in proportion to its length and having the posterior slope of its dome less abrupt; in being more attenuated laterally at either end, and in the greater production of the extreme ends of the base; lastly, it appears to average smaller than pinguis, and none of the specimens indicate any such Aricia-like basal * Greatful acknowledgment is extended to Dr. Paul Bartsch, Dr. Harald Rehder, and Mr. F. 8S. MacNeil of the United States National Museum for courtesies offered the writer. Appreciation for advice rendered is due Dr. C. W. Merriam of the Department of Paleontology of Cornell Univer- sity. The photographs were obtained through the courtesy of the United States National Museum. The writer wishes to thank Professor G. D. Harris and Dr. K. V. W. Palmer of the Paleontological Research Institu- tion, Ithaca, New York. bo 4 BULLETIN 84 32 callus as pinguis assumes in its fullest stage of development. The speci- mens show nothing of the spire; the teeth are strong but small, and not extended (even faintly) across the basal callus; there are twenty-two on the right lip and about eighteen on the left, which latter are less prom- inent. The largest specimen measures 26.5 mm. long, 17.0 mm. wide and 15.0 mm. high. The smallest was only 21.5 mm. long, but of about the same proportions.—(Dall, 1890). Dall’s illustration of the ventral view of heilprini gives the im- pression that the specimen is broader in proportion to its length than it really is. His figure shows no tooth bordering the colum- ella incisure anteriorly; in reality there is a prominent tooth an- terior to the incisure, and the incisure is deeper than his figure portrays. The photograph of the ventral view shows that the anterior canal lips are produced ventrally, the columellar lip be- ing especially pointed. The original figure does not indicate this condition of the anterior canal lips, and shows the columellar lip to be blunt. Holotype.—Numbered 114103 in the United States National Museum, Washington, D. C. Type locality—Ten Mile Creek, one mile west of Bailey’s Ferry, on the Chipola River, Calhoun County, Florida. Lower Miocene. Cyprea chilona Dall Plate 1, figs. 3, 4 Cyprea chilona Dall, 1900, Trans. Wagner Free Inst. &ci., vol. ILI, p. 1195, pl. 39, figs. 1, 3. Shell rotund, heavy, anterior canal very slightly produced; base angled at outer margins on both columellar and outer lip sides ; anterior canal straight, about 4 mm. broad at point of max- imum width; posterior canal curved, about 5 mm. broad at point of maximum width; the strange curvature of the aperture at the center is possibly due to a malformation of the shell in this re- gion; anteriorly the aperture is straight, posteriorly it is curved to the left; the aperture is about 6 mm. broad at the point of maximum width; teeth heavy, rounded, indefinite toward the posterior end of the base; incisures between the teeth broad; no shell color is preserved. This description is based upon the lecto- type. Length 42.25 mm.; width 37 mm. Dall (1900) described the species Cyprea chilona from an il- 323 CHIPOLA AND Costa Rica CypRaHAS: INGRAM 5 - lustration. In order to facilitate further identification of this spe- cies the writer has included here a brief description based on the lectotype. Figures of the lectotype designated by the writer (figs. 3, 4) are also included. The lectotype in the United States National Museum is without doubt the specimen in dorsal view figured by Dall for specific distinction. A syntype figured by Dall in ventral view is numbered 164928 in the above museum. Lectotype-—Numbered 498388 in the United States National Museum, Washington, D. C. Type locality—Chipola Beds, Alum Bluff, Florida. Lower Miocene. NEW “CY Phe shiVvONIC@sT A RICA: The Cyprea reported here are from the Pliocene of Costa Rica. They were collected by Dr. W. P. Woodring in 1917 from a railroad cut two and one-half miles outside of the town of Limon, Costa Rica. They are now housed in the United States National Museum, locality No. 8461. Cyprza bartschi, n. sp. Platelets conor Shell ovate-subdepressed ; from point of greatest width, 17 mm., shell narrows to 5 mm. anteriorly, and to 6 mm. posteriorly ; canals produced; dorsal convexity slopes gradually toward an- terior canal, and abruptly toward posterior canal, forming near- ly a right angle; spire obscured; rounded depression in shell just to left of spire; lateral extremities marked by a raised line formed by the angled shell base ; base convex ; posterior canal noticeably curved to the left, anterior canal slightly curved to the left; aper- ture narrower posteriorly than anteriorly; teeth strong; colum- ellar teeth extend but slightly on columella into aperture ; colum- ellar teeth arranged in a fairly straight line along base, and ex- tend from 2 to 3 mm. over base; 4 posterior columellar teeth ex- tend further on base than the rest, 3 of them extending over the columellar side of the posterior canal; outer lip teeth are ar- ranged about evenly in their extent over the base, central few being shorter than the rest; teeth rounded; interstices between teeth broad and concave; most anterior few of outer lip ani 6 BULLETIN 84 324 columellar teeth extend over the anterior canal lips. Dimensions.—Length 25.75 mm.; width 17.00 mm.; height 12.75 mm. Cyfrea bartschi is apparently related to the lower Miocene fossil, Cyprea raymondrobertsi Pilsbry (1921) of Santo Domin- go. It differs from it in having the columellar lip of the pos- terior canal more produced; also the columellar teeth in bartschi extend as raise! ridges over the columellar lip projection of the posterior canal. The anterior canal in raymondrobertsi is straight and in bartschi it 1s curved to the left. The teeth in the former species do not extend over the lips of the anterior canal as they do in the latter species. The anterior canal lips are straight in a dorso-ventral direction in Pilsbry’s species while they are angled in bartschi. The aperture is narrower, and the base is more angled in bartschu. This species is named for Dr, Paul Bartsch, Curator of Mol- lusca and Cenozoic Invertebrates, of the United States National Museum. Holotype.—Housed in the United States National Museum, Washington, D. C. Type locality.—Morin Hill, railroad cut 2% miles outside of the town of Limon, Costa Rica. Pliocene. Cypra cinerea, var. morinis, n. var. Pl. 1, figs. 8, 9 Shell cylindrically-oblong; canals surrounded dorsally by an impression; posterior and anterior canals but slightly produced; spire obscured, depressed; base convex ; teeth finer on columellar side of aperture than on outer lip side; columellar teeth longer than outer lip teeth; aperture narrow, curved toward the left posteriorly, nearly straight anteriorly; aperture about twice as broad anteriorly as posteriorly. The enamel is preserved. The shell color is a uniform dirty- brown dorsally, fading to a greyish-white on the shell base. Some of the interstices between the columellar teeth are colored brown. This variety differs from Cyprea cinerea Gmelin in possessing an elongate shell, resembling that of Cyprea isabella Linnzeus. The columellar and outer lip teeth are finer. Although the shell 325 CHIPOLA AND Costa Rica CyPprmAS: INGRAM ~] color was described above this may not be the original colora- tion, for it is not unlikely that some color distortion has taken place. However, the color of the interstices between the colum- ellar teeth seems to have been well preserved. Dimensions.—Length 28 mm.; breadth 16 mm.; height 13.50 mm. Holotype.-—Housed in the United States National Museum, Washington, D. C. Type locahty—Morin Hill, railroad cut 21% miles outside of the town of Limon, Costa Rica; locality No. 8461. Pliocene. BIBLIOGRAPHY Dall, W. H. 1890. Contributions to the Tertiary Fauna of Florida with Especial Reference to the Miocene Silex-beds of Tampa and the Pliocene Beds of the Caloosahatchie River. Transactions of the Wagner Free Institute of Science. vol. 3, p. 166, pl. 11, figs. 2, 2a. 1900. Tertiary Fauna of Florida. Transactions of the Wagner Free Institute of Science. vol. 3, p. 1195, pl. 39, figs. 1, 3. Pilsbry, H. A. 1921. Revision of W. M. Gabb’s Tertiary Mollusea of Santo Domingo. Proceedings of the Academy of Natural Sciences, vol. LX XIII, p- 365, pl. 30, figs. 1, 2. 3. 8 BULLETIN 84 326 EXPLANATION OF PLATE 1 (21) Figure Page 1.2.9 Cyprea-heilprini(Dall 22... ee eee 3 Lower Miocene, Florida; holotype, No. 114103 in U. S. N. M.; about natural size. 3, 43° Cy prea ‘chilona Dall 22. Ce ee 4 Lower Miocene, Florida; lectotype, No. 498388 in U. S. N. M.; about natural size. 5-7. -€y prea: bartschi, n,) spi.) 2 ee eee 5 Pliocene, Costa Rica; holotype. Housed in U. S. N.-M.; about natural size. 8,9. Cyprea cinera, var. morinis, n. var. 6 Pliocene, Costa Rica; holotype. Housed in U. S. N. M.; about natural size. No. 84, Pu. 1 BuLuL. AMER. PALEONT. PLATE 21, VOL. 24 4 z aa Saez BULLETINS OF AMERICAN PALEONTOLOGY Volume 24 ave w Number 85 New Fossil Cypraeidae from the Miocene of the Dominican Republic and Panama, with a Survey of the Miocene Species of the Dominican Republic By WILLIAM MARCUS INGRAM April 15, 1939 PALEONTOLOGICAL RESEARCH INSTITUTION Ithaca, New York 1, Ss Ae NEW FOSSIL CYPRAEIDAE FROM THE MIOCENE OF THE DOMINICAN REPUBLIC AND PANAMA, WITH A SURVEY OF DHE. MIOCENE SPECIES OF THE DOMINICAN REPUBLIC By 7 Wittiam Marcus INGRAM Cornell University The family Cypreide is well represented in numbers of fossil species occuring in Santo Domingo. The following species have been reported from the Miocene: Cyprea campbelliana Pilsbry, Cyprea cinerea Gmelin, Cyprea dominicensis Gabb, Cypraca henekem Sowerby, Cyprea isabella Linneus, Cyprea raymon- robertst Pilsbry, Cyprea spurca Linnaeus, Cyprea .spurcoides Gabb, Cyprea noueli Maury, and Nuclearia gabbiana (Guppy). This paper adds one new species and one new variety of Cypreide to the Miocene fauna of Santo Domingo, and one new species to the Miocene fauna of Panama. Grateful acknowledgment is due Dr. Paul Bartsch, Dr. Harald Rehder, and Mr. F. S. Mac Neil of the United States National Museum for allowing the writer to examine these species that are housed in the National Museum and are described here. DESERIPRION OF SPECIES Cyprza gurabonis, n. sp. Pip iaetia swoon Shell light, ovate; dorsal surface rounded; anterior and pos- terior canal extremities slightly produced; outer lip portion of anterior canal extremity slightly longer than columellar lip por- tion ; outer lip portion of posterior canal extremity produced about twice that of columellar lip portion; lateral shell margins rounded gradually into the base; outer lip angled, leav- ing a ridge on outer lip lateral surface; aperture filled with matrix so that the teeth are not visible ; both anterior and posterior canals curve toward the left of the shell; anterior portion of shell slopes 4 Bubuetin 85 330 gradually to anterior production of canal; posterior portion of shell slopes abruptly to production of posterior canal. Color wanting. Named for Gurabo formation. This species in the general bulbous character of the dorsal surface of the shell superficially resembles this character in Cyprea semen Cooke, from the lower Miocene of Anguilla. This bulbous character is not well shown in the figures included here. The species is quite distinct from others reported from Santo Domingo. Dimensions.—Length 17 mm.; width 11 mm.; height 9.25 mm. Holotype —Numbered 483461 in the United States National Museum, Washington, D. C. Type locality—United States National Museum station 8737, District of Monte Cristi, Gurabo River about 5 miles above Gurabo Adentro at base of coral limestone and above conglomer- ate. Gurabo formation, middle Miocene. Cyprza merriami, n. sp. Plate 1, figs. 10, 11 Shell heavy with a high dorsum; anterior and posterior canals produced, deeply notched; posterior canal notch 9 mm., anterior canal notch about 6 mm.; anterior canal covered by a shelf about 5 mm. broad on the dorsal surface; anterior canal bounded later- ally by flanges which are quite prominent; spire obscured, with a depression to the right; base but slightly convex on columellar side, convexity exists in center, the base becoming flattened an- teriorly and posteriorly; outer lip side of base slightly convex throughout its length; teeth especially prominent on outer lip; these teeth broad, rounded, interstices between them rounded; anterior 7 columellar teeth prominent, other columellar teeth very indistinct ; interstices between columellar teeth broad; columellar teeth extend but slightly on columella; teeth on both sides con- fined to lips surrounding aperture. Named for Dr. C. W. Merriam of the Department of Paleon- - tology of Cornell University. 331 WEsT INDIAN MIocENE CyPpR#AS: INGRAM or This is the largest species of fossil cowry recorded from Pana- ma, and exceeds in size and general shell bulk the large Cyprea willcoxt Dall from Florida. The shell is extremely heavy, and bulky, and superficially resembles the general outline of a living Cyprea arabica Linnzus. Dimensions.—Length 74.90 mm. ; width 50 mm. ; height 40 mm. Holotype-—Housed in the United States National Museum, Washington, D. C. Type locality—North shore of Nancy’s Cay, Panama. Mio- cene? Cyprea henekeni var. potreronis, n. var. Pies oe Shell obovate; two nodules on dorsal surface with a deep pit between them, nearest to the nodule on the left of the shell; pos- terior shell portion extremely thick, heavy; anterior canal much produced, narrow, flanged; flanges depressed in center, leaving a prominent raised ridge around their margins; posterior canal bounded dorsally by shell extensions which are heavy and ex- tend about 6 mm. out from canal; teeth confined to extreme ven- tral margins of aperture; teeth on columella side of aperture nar- rower than those on outer lip side; incisures between columellar teeth broader than these of outer lip teeth; anterior part of aper- ture broad, and fairly straight; posterior part of aperture curved to the left; shell color brown on both dorsal and ventral surfaces ; teeth and columella both colored brown. Named for Potrero, Santo Domingo. This variety differs from the species in possessing a deep pit between the nodules on the dorsal surface. The posterior canal notch is deeply cut, and the anterior canal is extremely narrow. The shell flanges around the anterior canal are much broader than in the species, henekeni, and these are bounded by a raised ridge that is prominent. The posterior bulk of the shell is heavier, the bulk being slightly produced in this region. The combina- tion of these characters will readily separate this variety from the species, C. henekent Sowerby. 6 BULLETIN 85 332 Dimensions.—Length 61 mm.; breadth 48 mm.; height 31.25 mm. Holotype-—Housed in the United States National Museum, Washington, D. C. Type locality.—Bluff on the right bank of Rio Amina at ford near Potrero, Provincia de Santiago, Santo Domingo. Gurabo formation, middle Miocene. MIOCENE-SPECIES OF THE DOMINICAN REPUBEIC In reviewing the literature of the fossil Cypreide of the Ameri- cas and the West Indies during a recent study, I have found no one work that contains a complete summary of the fossil species found in the Miocene of the Dominican Republic. In formulating a general bibliography of the fossil Cypreide of the Americas and the West Indies much time was consumed in tracing scattered literature pertinent to the subject; with this in mind original descriptions concerning Miocene species of Santo Domingo are included here. By carefully checking the collections of the United States Na- tional Museum, Academy of Natural Sciences, Paleontological Research Institution, and the Paleontological collection of Cornell University several type specimens have been located. The num- bers of certain of these are now cited for the first time, and should facilitate the location of such specimens for one desirous of examining them. The writer wishes to thank Dr. C. W. Merriam for advice rendered. Acknowledgment is made to the following institu- tions and persons who have aided in this study: Dr. C. W. Merriam for examination of specimens in the Paleontology collec- tion of Cornell University; Dr. Paul Bartsch, Dr. Harald Rehder, and Mr. F. S. Mac Neil for examination of material in the United States National Museum; Professor G. D. Harris and Dr. K. V. W. Palmer for examination of material in the Paleontological Research Institution; and Dr. B. F. Howell for examination of material in the Academy of Natural Sciences. Be) West INDIAN MIocENE Cypr@AS: INGRAM “I SPECTES “LISTED Cyprza campbelliana Pilsbry Cyprea campbelliana Pilsbry, 1921, Proceedings of the Aeademy of Nat- ural Sciences, vol. 73, pp. 365, 366, pl. 30, figs. 9, 10. The shell is oblong-oval, but slightly produced at the ends, moderately ealloused laterally, the callouses dappled with rather small dark spots; dorsal outline evenly arched, spire concealed. Outer lip having 24 teeth. Inner lip with 20 short teeth not running inwards as in C. cinerea. In the lower part of the columella an inner series of 5 short teeth may be seen. Length 30.3, lateral diameter 18.7, dorso-ventral diameter 15 mm. It is less convex than C. cinerea, the posterior slope of the dorsal out- line less abrupt. Moreover, the markings of the lateral callouses seem to be of a different character. The teeth are far less numerous than in C. dominicensis.—(Pilsbry, 1921) Holotype-—Numbered 3000 in the Academy of Natural Sciences, Philadelphia, Pennsylvania. Type locality.—Santo Domingo. Miocene.1 Cyprza cinerea Gmelin Cyprea cinera Gmelin, 1791, Systema Nature, 13th ed., p. 3402. Cyprea cinerea Gmelin, Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 364. Pilsbry (1921) reports one specimen from the Miocene of Santo Domingo. The present day distributional range of this species is in the waters of the West Indies, Florida, and South and Central Amer- ica. Cyprza dominicensis Gabb Cyprea dominicensis Gabb, 1873, Transactions of the American Philo- sophical Society, vol. 15, new series, p. 236. Cyprea dominicensis Gabb, Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 364. Shell very similar to C. lurida in form, sides sub-parallel, anterior end tapering more than the posterior, base slightly flattened; inner lip flexous in advance; teeth small, very numerous and not extended over base. This shell is closely allied to C. lurida and C. pulchra, but differs from both in that its teeth are small, regular, uniform, and end abruptly along a straight line. The last character at once separates it from the latter, while the size of the ecrenulations equally distinguish it from the former. The largest specimen is 1.5 inch long—(Gabb, 1873) The type has the form of C. lurida, being a little more produced at both ends. There are 36 teeth on the outer lip, 29 on the inner. Length 39.5, lateral diam. 23.2, dorso-ventral diam. 19 mm.—(Pilsbry, 1921) Maury (1917) lists a Cyprea dominicensis ? Gabb, from Rio 1, Pilsbry, 1921, ‘‘ With two or three exceptions, none of the labels bore any identification of locality or horizon further than ‘Santo Domingo’.’’ 8 BULLETIN 85 Gurabo at Los Quemados, Santo Domingo. Middle Miocene. A comparison of the holotype of this species to that of C. camp- bellina Pilsbry included here show the general shell shape of the two species to be similar. However, differences in the canals, apertures and teeth readily separate the two. Holotype—Numbered 3003 in the Academy of Natural Sciences, Philadelphia, Pennsylvania. Type locality.—Santo Domingo. Miocene.’ Cyprea henekeni Sowerby Plate 1 fiz. 3 Cyprea henikeri Sowerby, 1850, Quarterly Journal of the Geological So- ciety of London, vol. 6, p. 45, pl. 9, fig. 3. Cyprea henekent Sowerby, Gabb, 1873, Transactions of the American Philosophical Society, vol. 15, p. 285 Cyprea henekenit Sowerby, Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 365. Cyprea henekeni Sowerby, Maury, 1917, Bulletins of American Paleon- tology, vol. 5, No. 29, p. 114, pl. 19, fig. 4. Testa obovata, ventricosa, inflata, levis, dorso postice irregulariter tuber- eulifero, lateribus, pareecipue sinistro, obsolete granosis; extreminatibus, postica brevissimaé, antica, subproducta, apertura angusta, marginibus dentatis, dentibus paucis, magnis, rotundatis, canali brevissimo, reflexo. This species bears a general resemblance to Cyprea Mus and several others, which occasionally have irregular tubercles on the posterior part of the back; it may however be easily distinguished from all such by the dentition of both edges of the aperture, the teeth in this species, though not numerous, being large and prominent.—(Sowerby, 1850) Pilsbry (1921) states in comparing C. henekeni to C. mus, “This species resembles the recent C. mus, and has parallel varia- tions, both having smooth and bicornute or bituberculate forms. In C. henekeni the tuberculate form predominates, and the tuber- cles are larger, being thus more specialized than the modern race of the same stock.” There seems to be confusion in the literature concerning the proper spelling of the specific name of this species. Sowerby (1850) in the original description spelled the specific name henikeri. Sowerby in applying this name misinterpreted the collectors’ names as J. S. Heniker instead of J. S. Heneken. Sub- sequent writers have corrected Sowerby’s misinterpretation of the above name. Holotype.—? 2 See footnote No. 1. 335 West INDIAN MIOCENE CyPReAS: INGRAM 9 Type locality—Santo Domingo. Miocene. Maury (1917) lists this species from Cercado de Mao, Rio Gurabo, and Rio Cana at Caimito. Middle Miocene. Cyprza isabella Linneus Cyprea isabella Linneus, 1758, Systema Nature, 10th ed., p. 722. Cyprea patrespatrie Maury, 1917, Bulletins of American Paleontology, vol. 5, No. 29, pt. 1, p. 116, pl. 19, fig. 10. Cyprea isabella patrespatria Maury, Woodring, 1928, Carnegie Institu- tion of Washington, Publication No. 385, p. 317, pl. 21, fig. 9. Cyprea isabella Linneus, Gabb, 1873, Transactions of the American Philosophical Society, vol. 15, new series. ‘Cyprea isabella Linneus has been reported from the Miocene of Santo Domingo by Pilsbry (1921), Gabb (1873), and Maury (1917). The latter’s locality for this species is Cercado de Mao, Bluff1 Gurabo formation, middle Miocene, Dominican Republic. In reducing Cyprea patrespatrie Maury to synonymy with Cyprea isabella Linneus Pilsbry (1921) states: “Two speci- mens, which present no characters differing from the recent shells. The larger one closely resembles a recent C. isabella mexicana Stearns which we compared.” The living representative of C. isabella Linnzus in North America is Cyprea isabella mexicana Stearns (1893) from the Gulf of California. The holotype of this species is numbered 46581 in the United States National Museum from Tres Maries, Gulf of California. Additional localities based on material in the above museum are: Cape St. Lucas, Lower California, and Clarion Island, West Coast of Mexico. In the living state Cyprea isabella Linneus has a wide Pacific distribution, being found abundantly in the Fiji Islands, Tuamotu Islands, Hawaiian Islands, Samoan Islands, etc. Its distribu- tion has been recorded by Ingram (1937a), (1937b), (1938), (1939) from the following Islands of the Pacific; Hawaiian Islands, Palmyra, Washington, Fanning, and Christmas Islands, Guam Island in the Mariana Islands, and from American Samoa. Cyprza noueli Maury Platet 1s) fics 5 Cyprea noueli Maury, 1917, Bulletins of American Paleontology, vol. 5, No. 29, pt. 1, pp. 114, 115, pl. 19, fig. 5. Shell large, exceedingly globose, showing the apex of the spire, aper- 10 BULLETIN 85 336 ture somewhat curved; inner lip with about 16 rather weak teeth, the strongest anterior; outer lip with about 20 teeth, also strongest anteriorly and fading out posteriorly; upper margin of outer lip projecting; surface of shell entirely smooth, with traces on the back of a color pattern of white spots of varying size on a dark ground. Length of shell 60, width 44, thickness 39 mm.—(Maury, 1917) The closest relative of this species is Cyprea henekent Sowerby. It is readily separated from henekeni, however, by the bulbous character of the shell and the absence of the two tubercles on the dorsal surface. Holotype—Numbered 36984 in the Invertebrate Paleontology collection of Cornell University, Ithaca, New York. Type locality Expedition locality No. 16, Bluff 1, Cercado de Mao, Santo Domingo, middle Miocene. Cyprza raymondrobertsi Pilsbry Cyprea raymondrobertst Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 365, pl. 30, figs. 1, 2. 3. The shell is subovate in basal outline, solid, with the sides heavily eal- loused, laterally angular, the callus rising high, especially on the left side, and covering the spire, which is marked by a shallow irregular pit. Ends but little produced. The outer lp has 19 strong teeth, narrower than their intervals. Inner lip with 17 teeth. The base is rather strongly convex. Length 27, lateral diameter 19.2, dorso-ventral diameter 14.9 mm. Somewhat like C. arabicula Lam. by its angular sides, the aperture as in C. albuginosa Mawe.—(Pilsbry 1921) A similar form called by Pilsbry (1921) Cyprea raymond- robertst bowdenensis occurs in the middle Miocene of Bowden, Jamaica, It seems to the writer that this species is closer to C. spurca L. than to the above species mentioned by Pilsbry. Holotype.—Numbered 3995 in the Academy of Natural Sciences, Philadelphia, Pennsylvania. Type locality—Santo Domingo. Miocene.* Cyprea spurea Linneus Plate 1, fig. 2 Cyprea spurca Linneus, 1758, Systema Naturae, ed. 10, p. 724. Cyprea spurca Linneus, Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 365. Cyprea spurca Linneus, Gabb, 1873, Proceedings of the American Philo- sophical Society, vol. 15, new series, p. 235. Cyprea spurca Linneus, Maury, 1917, Bulletins of American Paleon- tology, vol. 5, No. 29, pt. 1, p. 115, pl. 19, fig. 6. Maury (1917) reports a definite locality for this species, hav- ing taken it in the fossil state from Bluff 1, Cercado de Mao; Zone 3 See footnote 1. j 337 West INDIAN MIocENE CyprmAS: INGRAM 11 1, Rio Cano at Caimito, Santo Domingo, middle Miocene. Maury (1921) states that this species is rare. In the living state this species has a varied distributional range. It has been reported from the North African Coast, South and Central America, Florida, and the West Indies. It is most abundant at the present time in the West Indies. Cyprza spurcoides Gabb Plate 1, fig. 4 Cyprea spurcoides Gabb, 1873, Transactions of the American Philo- sophical Society, vol. 15, new series, p. 235. Cyprea spurcoides Gabb, Maury, 1917, Bulletins of American Paleon- tology, volo o,No: 29, pt. 1, ps Lilo pl 195 fies. 7, 8, 9: Cyprea spurcoides Gabb, Pilsbry, 1921, Proceedings of the Academy of Natural Sciences, vol. 73, p. 365, pl. 30, figs. 4, 5. Shell similar in form to C. spurca, but somewhat broader and more nar- rowed in advance. Callus broad, convex below and _ slightly expanded laterally, not crenulated above as C. spurca. Crenulations of both lips well defined, more numerous on the inner than on the outer lip. Color pattern mottled irregularly. In eolor and size, this is not unlike C. bicallosa, but it differs from it in form and in the absence of the two callosities. It wants entirely the marginal pittings of C. spurca, which it approaches nearest in form.— (Gabb, 1873). Pilsbry (1921) states, “Gabb has compared it with C. spurca, but of recent species it seems to us closest to large examples of C. annulus in shape.” Holotype-—Numbered 2999 in the Academy of Natural Sciences, Philadelphia,. Pennsylvania. Type locality —Santo Domingo. Miocene. Maury’s (1917) lo- cality, Bluff 1, Cercado de Mao; Zone D, Rio Gurabo at Los Quemados, would indicate a middle Miocene occurence. Nuclearia gabbiana (Guppy) Plate 1, fig. 1 Cyprea gabbiana Guppy, 1876, Quarterly Journal of the Geologleal So- ciety of London, vol. 32, pp. 528-529, pl. 19, fig. 10. Pustularia nucleus (Linneus), Gabb, 1873, Transactions of the Amer- ican Philosophical Society, vol. 15, p. 236 (not C. nucleus Linn.) Pustularia gabbiana (Guppy), Pilsbry, 1921, Proceedings of the Acad- emy of Natural Sciences, vol. 73, p. 366. Cyprea gabbiana Guppy, Maury, 1917, Bulletins of American Paleon- tology, vol. 5, No. 29, pt. 1, pl. 19, fig. 12. Oval-elongate, rostrated at both ends, superiorly covered with large shin- ing tubercles which are almost circular upon the back, but become elong- ate and have a tendency to run into ribs near the thickened regularly grooved lip, whose dentations are continuous with the ribs on the outside. A dorsal groove separates the back into two nearly equal halves. The tubercles are larger than those of C. nucleus. 12 BULLETIN 85 338 The cowry for which I propose the above name has hitherto been con- sidered by me to be C. pustulata, and has been identified by Gabb as C. nucleus. I think it may be regarded as intermediate between those two spe- cies and it presents, I think, some characters which, combined with its distance in time and space from its nearest congeners, may warrant a pro- visional specific name.—(Guppy, 1876) Pilsbry (1921) presents conclusive evidence to show that this species is distinct from Cyprea nucleus Linneeus, “While this species has a general resemblance to P. nucleus L., it differs by the following characters: the raised transverse. lines which net the tubercles together are more numerous and conspicuous; the tubercles along the lateral margins are larger; the transverse ridges of the base alternate in size, but the smaller ones terminate at the margin of the aperture in teeth equal to those terminating the larger ridges; the teeth of the columellar side do not extend entirely within the aperture, but end on a sort of projecting ledge, inward from which a latticed-granulose sculpture is seen. Final- ly, the aperture curves more to the left at the upper end. Length TS eAMEON2O: 2) tania Holotype.—? Type locality.—Haiti. Miocene. Maury (1917) reports this species from Zone D, Rio Gurabo at Los Quemados, middle Miocene. BIBLIOGRAPHY Gabb, William M. 1873. On the Topography and Geology of Santo Domingos. Trans. of Amer. Phil. Soe., vol. 15, new series (read 1872), pp. 49-259. Gmelin, J. F. 1791. Linneeus’s Systema Nature, 13th ed. p. 3402. Guppy, R. J. L. : 1876. On Miocene Fessils of Haiti. Quart. Journ. Geol. Soe. London, vol. 32, pp. 528-29, pl. 19, fig. 10. Ingram, W. M. 1937a. The Family Cypreide in the Hawaiian Islands. The Nautilus, vol. 50, No. 3, pp. 77-82. 1937b. Cypreide from Christmas, Palmyra, Washington, and Fanning Islands. The Nautilus, vol. 51, No. 1, pp. 1-3. 1938. Cypreide from Guam. The Nautilus, vol. 52, No. 1, pp. 5-7. 1939. Cypreide from American Samoa with Notes on Species from Palmyra Island. The Nautilus, vol. 52, No. 3, pp. 103-105. 339 West INDIAN MIOCENE CyPR&AS: INGRAM 13 Linnzus, Caroli 1758. Systema Nature, Tomus 1, 10th ed. Maury, Carlotta Joaquina 1917. Santo Domingo Type Sections and Fossils. Bulletins of Amer- ican Paleontology, vol. 5, No. 29, pt. 1, pp. 1-251, pt. 2, pp. 1-43. Pilsbry, Henry A. 1921. Revision of W. M. Gabb’s Tertiary Mollusea of Santo Domingo. Proceedings of the Academy of Natural Sciences, vol. 73, pp. 305-435. Sowerby, G. B. = 1850. Descriptions of New Species of Fossil Shells Found by J. 8. Heniker Esq. Quart. Journ. Geol. Soc. London, vol. 6, p. 45, pl. Oe fies 3: Stearns, R. F. C. 1893. On rare or Little Known Mollusks from the West Coast of North and South America, with Descriptions of New Species. Proceedings of the United States National Museum, vol. 16, pp. 341-352. Woodring, W. P. 1928. Miocene Mollusks from Bowden, Jamaica, Pt. 2, Gastropods and Discussion of Results. Published by the Carnegie Institu- tion of Washington, Publication No. 385, pp. 1-564. 14 BULLETIN 85 340 EXPLANATION OF PLATE 1 (22) Figure Page 1. Nuclearia gabbiana (Guppy) —.. J ACE) NOS DS oe la iil Middle Miocene, Santo Domingo. 2. Cyprea spurea Linneus See Se ae ee eA) Middle Miocene, Santo Demiance,. 324 Cy prea henekeniiSowerby. -e vee See ee Middle Miocene, Santo Domingo. A’ ‘Gy prea’ spurcoides) ‘Gali w 220 Se ree eee Middle Miocene, Santo Donnie, 5o Cypreay noueli: Maury 22s ee ee eee eee 9 Holotype No. 36984, Invertebrate Paleontology Collection, Cornell University, Ithaca, New York. Length 60 mm. Middle Miocene, Cercado de Mao, Bluff 1, Santo Domingo. G7, (Qyriesey NOS, baer, il, Sig se 3 Holotype No. 483461, U. S. National Museum, Washington, D. C. Length 17 mm. Middle Miocene, U. S. National Mu- seum Loc. No. 8737, Santo Domingo. 8-9. Cyprza henekeni var. potreronis Ingram, n. var. —.. 5 Holotype. Housed in U.S. National Museum, Washington, D. C. Length 61 mm. Middle Miocene, Gurabo formation, Santo Domingo. Osis COQ) nus ey WapRe I, Thy Bids ae 4 Holotype. Housed in U. S. National WMinsewa,, TWashinetont D. C. Length 74.90 mm. Miocene (?), North shore of Nancy’s Cay, Panama. PLATE 22, VOL. 24 BuLu. AMER. PALEONT. No. 85, Pu. 1 AMERICAN — PALEONTOLOGY * VOL. XXIV IrHaca, N. Y. ; rae oe Uae A: BULLETINS OF AMERICAN PALEONTOLOGY Volume 24 Number 86 Reprint of Conrad’s Jackson Eocene Fossils as Described and Illus- tarted inthe Philadelphia Academy of Natural Sciences, Proceedings for 1855, pp. 257-63 and Wailes’ Report on the Agriculture and Ge- ology of Mississippl, 1854, pis. X1¥-XVII. (Note: The editors have endeavored to follow the original copy of these articles irrespective of gross orthographic and typographic errors.) July 75, 1939 PALEONTOLOGICAL RESEARCH INSTITUTION Ithaca, New York Wig Saas [ Proe. Phila. Acad. Nat. Sci., 1855, p. 257 | Observations on the Hocene deposit of Jackson, Mississippi, with descrip- lions of thirty-four new species of shells and corals. By T. A: Conrad: The following table will show the order of succession of the Eocene groups; but is not pretended to be more than an exposition of my limited knowledge of them, though they are doubtless presented in the true order of superposition. Further research may develope intercalated groups. Num- ber 6, is probably synchronous with the Orbitulite limestone of St. Siephens, Alabama, as its two most prominent fossils are very abundant in this stratum at Vicksburg. Number 5, is the lowest bed exposed in the bank of the Mississippi river, at Vicksburg. Col. Wailes found a large Ostrea on the top of the Jackson group, which is probably the shell re- ferred to in No. 5. It would be convenient to designate these subdivisions thus:- Claiborne group; Jackson group; Vicksburg group; St. Stephens group, GROUPS OF CHARACTERISTIC FOSSILS 8. Crassatella Mississippiensis, Arca Mississippi- __ ensis, Meretrix sobrina, M. imitabilis. st Newer Turbinella Wilsoni. Eocene, fs Corbula alta, Natica. Vicksburg. 6. Pecten Poulsoni, Orbitulites Mantelli. 5. Ostrea Georgiana? A, Umbrella planulata, Cardium Nicolleti, Conus Older tortilis, Cypraeca fenestralis, Galeodia Peter- Eocene, soni, Rostellaria extenta, &c. Jackson. 3. Crassatella alta, Pectunculus stamineus, Me- Older retrix aequorea, Gratelupia Hydii, Leda coelata. Eocene, Crepidula lirata &c. Claiborne. De Ostrea sellewformis. il, Cardita densata. Cyclas— Alabama river. Nos. 1 to 3 will represent the Claiborne group; 4, Jackson group; 6, St. Stephens group; 7 and 8, the Vicksburg group. When a group of corresponding fossils is to be found elsewhere, its rel- ative position can be stated by referring to the typical subdivision which contains many identical species. Since my discovery of the Eocene formation of Claiborne, Alabama, in 1832, by means of fossil shells collected by Judge Tait, numerous localities have been found in the southern States, and characteristic fossils have al- so been obtained by Maj. Emory, in Western Texas, and even in California, by Mr. Blake. Localities widely separated contain some species in com- mon, but I did not anticipate that groups would vary to the extent they do in the three localities of Claiborne, Alabama, Jackson and Vicksburg, Mississippi. Col. B. L. C. Wailes, of Mississippi, has lately discovered a new group of Eocene fossils at Jackson, in which none of the Vicksburg species occur; and of forty species, five only are identical with Claiborne fossils. One of the Jackson shells, Cardium Nicolletti, Conrad, was first discovered in the bank of Red river, Washita; and, therefore, this latter locality will probably prove to belong to the same division of the Eocene as that of Jackson. The Mississippi deposit described by, Col. Wailes, is a group of shells chiefly, of more than ordinary beauty and preservation, imbedded in sand of a gray color, consisting of fine angular grains of quartz and minute fragments of shells. One of the species, Cypraea Smreeiralis, is closely —...__" =" 4 BULLETIN 86 344 [ Proc. Phila. Acad. Nat. Sci., 1855, p. 258 ] mle Be Sh OS ey trae _ related to C. elegans, of Deshayes; two remarkable species which have no analogue or kindred shell in the later Tertiary tormations. The state of preservation and the forms of these fossils are closely analogous to those of the Paris basin; and I find no recent nor any Miocene species among them. I believe the group to be newer than the Claiborne deposit, and clearly older than that of Vicksburg. I think it will be found that No. 6, of the above table, represents that extensive limestone, which, in Alabama, contains the Basilosaurus re- mains; the Laganum Regersi, Morton, near Claiborne, and near Brandon, Mississippi, where it has been discovered by Col. Wailes, occupying a higher position than the Jackson group. The limestone of Jacksonboro’. Georgia, described by Lyell, is probably referrible to the same division, and contains the Laganum Rogersi, (Scutella Jonesi, Forbes). The following species of organic remains were collected by Col. B. L. C. Wailes, and are figured in his work on the Geology of Mississippi. Those illustrations are referred to in the descriptions. ‘ CORBULA. 1. C. densata, Geol. Miss., Pl. xiv., fig. 9——Triangular, subequilateral, very thick in substance; surface undulated and having angular concen- trie striae; umbonal slope submarginal and acutely carinated, posterior extremely angular. Related to C. nasuta, Con. but proportionally shorter, thicker, with a more rounded base, &c. The description applies to the larger valve as I have not seen the opposite one. 2. C. bicarinata, Pl. xiv, fig. 3—Elevated, triangular, slightly oblique, thick in substance, profoundly ventricose, with robust reflected concen- tric lines; umbo profoundly prominent and the beak incurved; posterior slope biangulated; space between the angles flattened, direct. Resembles C. oniscus, Con., but is thicker, more elevated, not rostrated, and its slight obliquity is the reverse of that in the former species. I have not seen the smaller valve. LEDA. Schum. L. multilineata, Pl. xiv., fig. 4——Ovato-elliptical, inequilateral ventri- cose, with fine sharp concentric lines, which are somewhat undulated; an- terior side rostrated, with closely-arranged, radiating, minute, tubercu- lated striae; posterior side with unequal fine radiating lines, a few of which are very distinct; a few raditaing lines are continued near the base over the middie of the valves. Allied to N. eewlata, Con., but very distinct. NAVICULA, Blainville. N. aspera, Pl. xiv, fig. 5——Trapezoidal, dise contracted behind the mid- dle, cancellated; concentric lines distant, imbricated; radiating lines largest towards the umbonal slope, subspinous; umbonal slope acutely angulated; posterior slope excavated; series of cardinal teeth uninter- rupted; inner margins crenulated. CARDIUM, Lin. C. (Protocardia) Nicolletti, Pl. xiv., fig. 6—Proc. Ac. Nat. Sci., 1841, p. 33. This shell agrees, except in size, with the specimen originally described from the Washita, and doubtless the beds of that locality will prove to be of synchroncus origin with those of Jackson. A species of Cardium very nearly allied to this, I formerly believed to be the same; but it accom- panies a different group, and presents variations entitling it to be a specific distinction. It is from Pamunkey river, Virg. Compared with C. Nicolletti; umbo less inflated, posterior margin oblique, shell proportionally longer, and the radiating lines 22; in the other 25. The posterior cardinal tooth larger &c. It may be named C. fene, ee 345 JACKSON Fosstts: ConrapD ou [ Proce. Phila. Acad. Nat. Sei., 1855, p. 259 ] CRASSATELLA, Lam. C. flexura, Pl. xiv., fig. 7—Trapezoidal, inequilateral; ventricose medial- ly; slightly contracted anteriorly, and more so posteriorly; umbonal slope angulated and prominent; whole surface with concentric prominent lines, some of which bifurecate anteriorly; inner margin crenulated. Approaches C. protexta, Con., but has the strie over the whole disk, the cardinal teeth more compressed; inner margin with larger erenula- tions, &c. GLOSSUS. G. filosus, Pl. xiv., fig. 8—Orbicular, ventricose, with radiating lines, un- equal, medially flattened, and towards the ends angulated; concentric lines microscopic, series of cardinal teeth uninterrupted, generally large and prominent. Allied to G. stamineus, Con., but very distinct. OSTREA, Lin. O. trigonalis, Pl. xiv., fig. 10—Triangular, flat, surface irregular, with some indistinct radiating lines; muscular impression obliquely suboval, situated nearer the summit than the base; margin somewhat ascending, submargin carinated. A single imperfect upper valve is all that I have seen of this shell, but it is widely different from any other Eocene species known to me. PECTEN, Lin. P. nuperus, Pl. xiv., fig. 11—Suborbicular, ventricose, with about twenty- three angular prominent ribs, crossed by fine, closely-arranged wrinkled lines; ears finely striated obliquely. A single valve with the ears broken is all of this species in the collec- tion. UMBRELLA. U. planulata, Pl. xiv., fig. 1—Suboval, flattened, surface undulated, ris- ing a little towards the apex, which is prominent and acute, and situated much nearer to one side and nearer to one end; lines of growth conspicu- ous; inner side with a very large suboval cicatrix, with radiating inter- rupted lines. This fine species is the only one yet known in North America. Two spe- cimens occur, one of which is marked with some hair-like brown radiat- ing lines, both internally and externally. CAPULUS, Mont. C. Americanus, Pl. xv., fig. 1—Obliquely ovate, longitudinally contract- ed on one side; lines of growth profound; summit very oblique; apex pro- foundly prominent, acute, curving towards the base and projecting far be- yond the basal margin; aperture obtusely oval or suborbicular. TROCHITA, Schum. T. alta. Pl. xv., fig. 3—Conic, elevated, with three or four transverse undulations; radii prominent, rounded, very irregular, interrupted, some- what tuberculated; vertex central, spiral, somewhat prominent. CLAVELLA, Swains. CLAVILITHES, Swains. 1. C. humerosa. Pl. xv., fig. 2.—Fusiform, volutions eight?, rounded; body whorl and penultimate entire, the others with broad rounded ribs; whorls carinated below the suture and with revolving lines, most prom- inent towards the apex; body whorl and penultimate, channeled above and contracted near the summit; body whorl angulated inferiorly; beak long and straight. 29. C. varicosa. Pl. xvi., fig. 7—Fusiform, spire and beak elongated; whorls nine, with distant, rounded, thick ribs and with revolving acute lines, which are obsolete or less prominent on the ventricose portion of the body whorl; papillated apex formed of three volutions; columella near- ly straight, and with microscopic longitudinal lines. C. Mississippiensis. Pl]. xvii., fig. 8 is probably the same species, 6 BULLETIN 86 346 [ Proc. Phila. Acad. Nat. Sei., 1855, p. 260 | MITRA, Humph. Lam. Subgenus LAPPARIA, Conrad Short-fusiform, spinose; apex papillary; beak very short, thick, twisted; plaits as in Mitra. M. (Lapparia) dumosa, Pl. xv., fig. 4.—Short-fusiform, volutions seven, direct, obliquely flattened above, with a series of transversely compressed, distant spines on the two largest whorls; on the contiguous whorl they become nodules; two whorls below the apex papillary, smooth; the next two longitudinally ribbed, and the others longitudinally striated or with prominent lines of growth; whole surface with revolving wrinkled lines; plaits four;* beak profoundly ridged. CONUS, Lin. C. tortilis, Pl. xv, fig. 5—Ovato-turbinate; spire obtusely conical with the apex exserted, acute; whorls obliquely flattened, with revolving im- pressed lines and transverse wrinkles, carinated near the base, dizect be- tween the carina’and suture; lines of growth on body whorl profoundly curved; base with a profound thick fold. Differs from C. saurodens, Con., in having a moze prominent and convex spire, in the large twisted callus at base, &e. ROSTELLARIA, Lam. 1. R. velata, Con., Pl. xv., fig. 7. KR. Lamarclkii, Lea, Cont. fis. 1164: 2. R. staminea, Pl. xvi, fig. 9—Fusiform, spire elongated, subulate above; whorls fifteen; body whorl slightly concave with fine closely-arranged re- volving lines and obsolete longitudinal undulations; three upper whorls with curved longitudinal acute ribs; the remainder covered with a polished calcareous deposit, and excavated at the suture; body whorl angular on a line with the upper margin of the aperture; labrum thin; beak slightly curved. This species occurs at Claiborne in great abundance. j VOLUTALITHES, Swains. 1. V. symmetrica, Pl. xv., fig. 8—Subfusiform; with longitudinal acute ribs terminating above in short spines on the body whorl; volutions exca- vated above, where they are striated but not ribbed, body whorl with raised alternated revolving distinct lines; above the angle they be- come almost microscopic; suture margined below by a series of small points, and somewhat carinated; plaits three, slender. Allied to V. Sayana, Conrad. NATICA, Lam. N. permunda, Pl. xvi., fig. 2—Suborbicular; body whorl somewhat exca- vated near the suture; spire very short; umbilicus very long, profound, with a central broad rounded ridge, and the lower margin subcarinated; columella subrectilinear. APORRHAIS. Subgenus PLATYOPTERA, Conrad. Shell with a profoundly expanded labrum which is entire, or without a rostrum, and with the margin very thin and acute. A. (P.) extenta. Pl. xvi., fig. 3—Shell independent of labrum fusiform, with prominent revolving rounded lines and intermediate fine lines, from one to three,:and longitudinal microscopic lines; volutions rounded, covered towards the apex with a polished calcareous deposit; labrum within with impressed radiating lines, becoming well marked furrows towards the base. 347 JACKSON Fossits: CONRAD if [ Proc. Phila. Acad. Nat. Sei., 1855, p. 261 ] MITRA. Subgenus FustmMitra, Conrad. Elongate-fusiform, smooth and polished with impressed revolving lines; aperture narrow; plaits two prominent, and two obsolete. or much smaller than the others; beak elongated. To this subgenus belongs M. conquisita Con. M. Mississippiensis, and Con. of the Vicksburg deposit. M. (Fusimitra) Mellingtoni. Pl. xvi., fig. 5——Profoundly elongated, fusi- form; volutions ten, convex, six of which towards the apex have revolving impressed lines, with ithe interstices transversely striated; in the contigu- ous whorl they are distant and obsolete, except near the summit, where there are two distinct impressed lines; on the penultimate whorl one dis- tinct impressed line, and the summit of the body whorl obtusely carinated; spire longer than the aperture, which is narrow; plaits four, the two su- perior ones very prominent, robust. Allied to M. conquisita, but much larger, proportionally longer, and with the striae less deeply impressed. It may prove, however, to be the same when many specimens from ‘the two localities can be compared. If it should be identical with the former it is the only specimen common to the Vicksburg and Jackson deposits out of the 40 species of the latter and 100 of the former deposit. CARICELLA, Con. 1. C. subangulata, Pl. xv. fig. 8.—Turbinate; labrum expanded; shoul- der subangulated; body whorl flattened above; spire short, conical, consist- ing of 4% volutions, with microscopic revolving lines near the apex; columella with four remote plaits the two inferior ones most oblique. 2. C. polita, Pl. xvi, fig. 4—Fusiform; smooth and polished, with re- volving lines inferiorly, and on two volutions of the spire; the whorl above is papillary and smooth; columella with closely-arranged microscopic longitudinal lines; plaits four, slender, prominent, remote; beak slightly curved. Allied to C. but proportionally shorter and very distinct. SCALARIA, Lam. S. nassula, Con., Pl. xvi., fig. 6—Foss. Shells of Tert. Form. This shell, though much larger than the Claiborne specimens, specifically agrees with them. Probably Lea’s S. planulata is the same species. ARCHITECTONICA, Bolton. SOLARIUM, Lam. 1. A. acuta, Pl. xvii, fig. 1—Much depressed, very thin and acutely car- inated on the margin; convex above, lower half of the whorls somewhat excavated; revolving striae linear, crenulated, with a minute intermediate crenulated line, and a still finer line or two in some of the interstices; base convex, flattened and somewhat excavated towards the periphery, re- volving striae linear, alternated with a medial smaller line and two mi- nute ones, nearly smooth, except four from the umbilical margin which rapidly increase in size towards the inner margin; The marginal line profoundly crenulated; a carinated beaded line on the middle of each whorl within the umbilicus, which is profoundly scalariform. 2. 7 Nae Leal OO ON Aun KRW D H BULLETIN 86 [ Geol.Miss.: Wailes } Jebeville, SQV |DOVINO|==SisU8iLILS, UNIVALVES. . Architectonica acuta. . Architectonica bellastriata. a. b. Cyprea pinguis. . Gastridium vetustum. Cyprea fenestratis. a. b. Phorus reclusus. . Lurritella alveata. . Clavelithes Mississippiensis. . Morio Petersoni. (Galeodia, of Link.) . Strepsidura dumosa. PL. 26, Vol 24 BULL. AMER. PALEONT No 86, PL.4 Page 289 UNIVALVES Plate XVI! JACKSON TERTIARY SHELLS aya Vln Pyne a 1s BULLETIN 86 308 [ Geol. Miss.: Wailes ] PLATE XVII [XVI]—SHELLS. Oo CON Aun f WD H UNIVALVES. Volutalithes dumosa. Natica permunda. Rostellaria extenta. Caricella polita. Mitra Millingtoni. Scalaria nassuta. . Clavelithes varicosa. . Teredo Mississippiensis. . Rostellaria (young). PL. 25, Vol. 24. Pag@ 289 BULL. AMER. PALEONT. UNIVALVES No. 86, PL. 3 Plate XVI JACKSON TERTIARY SHELLS aS gene JACKSON FossiIts: CoNRAD [ Geol. Miss.: Wailes, p. 289 | Fossil Testacea of the Tertiary Green-sand Marl-bed of Jackson, Determined and named by T. A. Conran, Esq. BIVALVES. Astarte. Lamark. Astarte parilis. Con. Cardita. Brug. Cardita planicosta. Lam. Cardita tetrica. Con. Cardium. Lin. Cardium Nieolleti. Con. Corbula. Brug. Corbula densata. Con. Corbula biearinata. Con. Lam. Con. Glossus. Poli. Glossus filosus. Con. Leda. Leda multilineata. Meretrix. Lam. Meretrix profunda. Con. Blain. Con. Lin. Con. Pecten. Lin. Peeten nuperum. Con. MULTIVALVE. Teredo. Lin. Teredo Mississippiensis. Con. UNIVALVES. Architectonica. Bolton. Architectonica bellastriata. Con. Crassatella. Crassatella flexura. Schum. Con. Navicula. Navicula aspersa. Ostrea. Ostrea_ trigonalis. Arehitectonica acuta. Con. Capulus. J/ont. Capulus Americanus. Con. 19 Cyprea. Lin. Cyprea fenestratis. Con. Cyprea penguis. Con. Conus. Lin. Conus tortilus. Con. Caricella. Con. Caricella polita. Con. Caricella subangulata. Con. Clavalithes. Szain. Clavelithes humerosus. Con. Clavelithes varicosus. Con. Clavelithes Mississippiensis. Con. Gastridium. Gastridium vetustum. Natica. Adan. Natica permunda. Con. Mitra. Hump. Mitra Millingtoni. Con. Mitra dumosa. Morio. Morio Petersoni. Con. Phorus. J/ ont. Phorus reelusus. Con. Rostellaria. Rostellaria vellata. Sow. Con. Con. Lam. Con. Rostellaria extenta. Con. Sealaria. Sealaria nassuta. Strepsidura. Swain. Strepsidura dumosa. Con. Trochita. Schum. Lam. Con. Troechita alta. Con. Umbrella. Lam. Umbrella planulata. Con. Volutalithes. Szwain. Volutalithes dumosa. Con. Volutalithes symmetrica. Con. 19 Miss. AMERICAN PALEONTOLOGY =) 7 3 * VOL XXIV Irnaca, N. Y. Pi era ele hoes » ra ct) peop Yael Peers Pe BULLETINS OF AMERICAN PALEONTOLOGY Volume 24 a SI Number 87 A Group of Pennsylvanian Crinoids from the Vicinity of Bartlesville. Oklahoma By HARRELL L. STRIMPLE July 28, 1939 PALEONTOLOGICAL RESEARCH INSTITUTION Ithaea, New York ASR ee Roaiee ants A GROUP OF PENNSYLVANIAN CRINOIDS FROM THE VICINITY OF BARTLESVILLE, OKLAHOMA By HarreELu L. StTRIMPLE Considerable careful comparison and study together with con- sistent collecting will be necessary before the full scope and var- ious manifestations of Pennsylvanian crinoids will be known. A few are presented herein as follows :— Of the Flexibilia— Amplhicrinus oklahomaénsis, n. sp., is apparently the first com- plete specimen of this genus to be figured from North America. A distinct form from Wayside, Kansas is described as Amphi- crinus poundi, n. sp. With Cibolocrinus robustus, n. sp., we have our first definite knowledge of the flexible arms of this and allied genera. Of the Inadunata— Delocrinus nodosarius, n. sp., is a robust representative of this genus having strong nodes developed, Delocrinus tumidus, n. sp., a small representative having tumid plates and depressions at the upper apex of BB and the lower apex of RR, as well as a vertical facet below the ligamental furrow. Moundocrinus osagensis, n. g., n. sp., represents a new form with prominently protruded posterior side and articular facets no wider than normal thickness of RR. Pentadelocrinus typus, n. g., n. sp., is a new form similar to Delocrinus save in the possession of a pentagonal stem and a different type anal X. Graphiocrinus stantonensis, n. sp., is dis- tinct mainly in the full bowl-shape. Fuerisocrinus waysidensis, n, g., n. sp., is a form with an anal plate within the cup yet very AUG . 4 1638 4 BULLETIN 87 364 close to Erisocrinus typus. Ethelocrinus convexus, n. sp., is quite distinct from all other known species in the possession of IBD visible from side view, having a shallow convex base. Hyx’reiono- crinus deweyensis, n. sp., is distinct in being the only known spe- cies having a cup similar to the genotype, H. woodianus. Decado- crinus regularis, n. sp., is a small, distinct, little known form in the locale under consideration. Melbacrinus americanus, n. g., 0. sp., 18 a distinct small turbinate cup form with anal arrangement as in Hydriocrinus but differing mainly in the branching of the arms and possession of a round stem. Agassizocrinus mcguirei, n. sp., 18 a distinct species based on the IBB cone, being the second species observed above the Morrow subseries. Lecythiocrinus ollicueformis and Lecythiocrinus adamsi are included merely to figure the well nigh perfect calices found in this locale and give a more comprehensive view of these unusual forms. Lecythiocrinus urneformis, n. Sp., is a distinct new form. It 1s with considerable pleasure that I acknowledge the excell- ent collecting and assistance of my wife, Mrs. Melba Strimple. Dr. R. S. Bassler of the U. S. National Museum, and Dr. Edwin Kirk of the U. S. Geological Survey have been kind enough to favor me with much technical information and advice. All specimens figured herein were collected by Mrs. Melba Strimple, or the author, save that described as Amphicrinus poundi, n. sp., which was collected by our friends and occasional fellow collectors, Mr. and Mrs. R. J. Pound of Bartlesville, Oklahoma. Family ICHTHYOCRINIDZ Angelin (Emend. Wachsmuth and Springer) Genus AMPHICRINUS Springer Amphicrinus oklahomaénsis, n. sp. Plate 1, figs. 1, 2 Calyx shallow expanding rapidly, some 60 mm. from axial canal to tips of arms. Proximal columnals in place but usual struc- ture is apparent with BB and part of RR in columnar scar, post. B extending out of basal circlet, RR large heptagonal save those two which are posterior being hexagonal. IBr! large, hexagonal or pentagonal, wider than high. IBr? hexagonal or pentagonal, axillary, wider than RR. IIBr’ hexagonal or heptagonal, slight- ly wider than high. IIBr? hexagonal or pentagonal. IIBr? penta- ‘ K TAT. ~ 365 PENNSYLVANIAN CRINOIDS: STRIMPLE or gonal to heptagonal, axillary. IIIBr are hexagonal to quadrangu- lar branching on fifth or sixth plate, following inner ray com- prised of some 12 plates tapering to tip with no further branching, outer ray branching again on IVBr® or IVBr°, inner ray follow- ing tapering to tip, outer ray branching again on fifth or sixth brachial following, further branching unobserved. It is to be noted that the outer rays expand strongly toward ends. iBr! hexagonal, that of the anterior slightly larger than the others, all followed by a double row of plates, some thirteen to the inter- radii save the anterior which has nineteen observed. ilIBr! normally pentagonal followed by axillary plate and then a double row of small irregular plates, some seven in all. Anal X penta- gonal, axillary, not quite as large as normal iBr' following supple- mentary plates two, hexagonal, followed by two single plates the first pentagonal, the second hexagonal and supporting obliquely to the left a small pentagonal plate, solidly above a large hexag- onal plate, five plates of similar structure and placement are ob- served above. Plates smooth, rather thin and _ undecorated. Proximal columnals very thin, This species is easily distinguished from A. scoticus Springer in that it has a double row of anal plates whereas A. scoticus has a single row. Of A. carbonarius Springer we have only the arms of the poorly preserved holotype and those cups referred to the species by Laudon (1937). The arms of the holotype taper even- ly to the tips whereas in the form at hand there is a noticable swelling distally. Of the cups figured by Laudon, anal X is fol- lowed by two single plates the third being axillary, whereas in the form at hand the anal X is axillary with the following plates following a distinctly different pattern. Occurrence and horizon.—Stanton limestone member, Ochelata group, Pennsylvanian. The mound just west of the city limits of Bartlesville, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Amphicrinus poundi, n. sp. Plate 2, figs. 18, 19 Cup low rapidly expanding, 5 mm. high by 10.4 mm. wide at IBr'. Columnar depression is occupied by the diminutive IBB and rather large BB, with the BB developed as a rim externally 6 BULLETIN 87 366 so that the RR never enter the basal circlet. The sutures of IBB are rather indistinct but apparently five tiny pentagonal plates are present. BB five large pentagonal plates extending very slightly out of basal depression save for post. B which is elongat- ed and truncated for the reception of anal X. RR five large hep- tagonal plates save for those two posterior which are hexagonal. IBr! quadrangular wider than high. iBr! hexagonal, rather large, followed by double row of very small plates. Anal X axillary, rather elongated, heptagonal being comparable in size to iBr!?, followed by two very small plates. This species differs from A. scoticus and those forms referred to A. carbonarius by Laudon in that the anal X of A. poundi is axillary. In A. oklahomaénsis the anal X is axillary but is penta- gonal and smaller than iBr', wherein A. powndi is elongated heptagonal. Also in the present species the RR do not enter the basal depression. We are indebted to Mr. and Mrs. R. I. Pound of Bartlesville, Oklahoma for the figured specimen of this species which takes their name. Occurrence and horigon.—Stanton limestone member, Missouri series, Pennsylvanian, near Wayside, Kansas. Type.—Springer Collection of the U. S. National Museum. Order FLEXIBILIA Zittel Suborder SAGENOCRINOIDEA Wachsmuth and Springer Family LECANOCRINIDZ Springer Genus CIBOLOCRINUS Weller, (Emend. Moore and Plummer) The genus was established to include Permian forms from Texas with 3 IBB and one anal plate within the cup, with Cibolo- crinus typus Weller as the genotype. Moore and Plummer (1938) restricted the genus to those forms having low cup-shaped calices with the IBB not visible from side view, the RR having narrow articular facets that are not appreciably wider than the normal thickness of the RR, and with the stem impression round and slightly depressed. The range of the genus was taken into the Morrow subseries of the Pennsylvanian. From the Ochelata 367 PENNSYLVANIAN CRINOIDS: STRIMPLE ~] group (upper middle Pennsylvanian) now comes a species herein described as Cibolocrinus robustus, n. sp. Cibolocrinus robustus, n. sp. Plate 1, figs. 3, 4 Cup loosely bowl-shaped; base somewhat flattened, diam. of holotype 18 mm. ; height 9 mm.; plates very gently bulbous, tend- ency for posterior side to protrude. “-IBB disk has a diameter of some 7 mm. of which the sharply depressed round stem impres- sion occupies some 5.5 mm., same being 1 mm. deep. IBB 3, un- equal elements, pentagonal, smallest r. post. in position. BB 5 equal hexagonal elements save that of the posterior which is broadly truncated for the reception of anal X, lower margins curved under slightly to form portion of basal area, length and wicth about equal. RR 5 equal pentagonal elements, those of the posterior losing a portion of their width to the anal X, about twice as wide as high, articular facets developed inward very slightly but decidedly so, at the interradial sutures the laterally developed muscle scars attain their maximum width making a shallow basin with the confluent scars of adjoining plates, to the fore is a rather strong cross ridge adjoined by sharp ligamental furrow. The cross ridge does not quite reach the interradial sutures leaving a small but sharp notch. Anal X rather large, heptagonal in outline, resting broadly on post. B, lower half within cup, upper half curving slightly inward and has a tendency to become very thin. | Of the arms only a portion is known. IBr’ low covering entire width of RR, lower edges fitting snugly into interradial notches. IBr? low, axillary. Three secundibrachs observed with no further branching. Surface of entire specimen covered by close spaced small to minute sharp granules. Stem known only from a few thin, smashed proximal column- als. Stem impression in cup is not crenulated. Tegmen unknown. The species is quite distinct from other known forms. Three good specimens have been observed, two with portions of the arms attached. Occurrence and horizon.—Shales associated with the Stanton limestone member, Ochelata group, Pennsylvanian, at the mound just west of the city limits of Bartlesville, Oklahoma. 8 BULLETIN 87 368 Type.—Springer Collection of the U. S. National Museum. Order INADUNATA Wachsmuth and Springer Suborder FISTULATA Wachsmuth and Springer Family POTERIOCRINITID Bassler Genus DELOCRINUS Miller and Gurley This genus is well known and understood so that further dis- cussion is not attempted. Two new species are described herein one as Delocrinus nodosarius, n. sp., the other as Delocrinus tumidus, n. Sp. Delocrinus nodosarius, n. sp. Plate 1, figs. 18, 14, 17 Calyx saucer-shaped, deeply invaginated base, measuring 21 mm. diam. by 9 mm. high; IBB 5, equal pentagonal elements limit- ed to basal depression; BB 5, entering strongly into basal cavity, regular hexagonal elements save for post. B which is truncated for reception of anal X; RR 5, regular pentagonal elements, facets developed inward as horizontal shelves, shallow ligamental groove to the fore backed by strong crenulated cross ridge, shal- low muscle scars developed laterally ; anal X hexagonal slightly higher than wide, half in half out of cup, upper portion curved strongly inward, articular facet possessing shallow rounded groove encircled by sharp ridge; IBr+ axillary, small spine developed at apex but not affecting balance of plate to any great extent. All observed plates covered by thin, small, sharp and closely spaced protuberances with occasional large nodes appearing particularly in distal portions of cup. Tegmen unknown. ‘Relationship.—The species is very close to D. hemisphericus from which it differs in having strongly nodular surface and dif- ferent type of spine development of the IBr'. Occurrence and horizon.—Figured specimens from those shales associated with the Stanton limestone member, Ochelata group, Pennsylvanian, the mound just west of the city limits of Bartles- ville, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Delocrinus tumidus, n. sp. Plate 2, figs. 1-8 Calyx saucer-shaped with invaginated base, measuring from 7mm. to 11 mm, diameter by 3 mm. to 4 mm. height in observed specimens ; IBB 5, pentagonal regular elements limited to funnel- 369 PENNSYLVANIAN CRINOIDS: STRIMPLE 9 shaped basal concavity; BB 5, hexagonal regular elements save for post. B which is truncated for the reception of anal X, tumid plates; RR 5, equal pentagonal elements, very tumid in distal por- tion where more or less vertical facets are developed, articular facets developed strongly inward as horizonal shelves, ligamental furrows backed by sharp cross ridge, shallow muscle scars de- veloped laterally; anal X hexagonal, slightly longer than wide, normally half in half out of cup, followed by single azygous plate, articular facet with shallow rounded depression encircled by sharp ndge; IBr* axillary, height and width about equal, being strongly tumid to slight projections as spines in region just below apex. Plates of cup devoid of ornamentation, sutures slightly depressed, small area about lower apex of RR and upper apex of BB sharply depressed. Growth stages.——Immature calices do not differ greatly from those more fully developed save in the greater tumidity of BB and RR. Relationship.—This species is no doubt closely related to D. hemisphericus from which it is distinguishable in the smaller, shal- lower cup, particularly in the tumidity of the RR and BB with the sharply depressed areas at the upper apex of BB and lower apex of RR, and the vertical facets just below ligamental furrows of RR. Occurrence and horizon.—Figured specimens from the Stan- ton limestone member, Ochelata group, Pennsylvanian, the mound just west of the city limits of Bartlesville, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Genus MOUNDOCRINUS, n. g. Genotype.—Moundocrinus osagensis, 1. sp. The unusual protrusion of posterior side, large nonaxillary anal X, articular facets no wider than the normal thickness of the RR, more or less flattened base, five IBB, BB, and RR, and round stem distinguishes this form from any known genus. Certain manifestations of Cibolocrinus are similar but in that genus there are only three IBB, and the columnar scar is slightly but sharply depressed, usually comparatively large. The form at hand has a small columnar scar with the sides sloping inward and is sharply crenulated which is more in keeping with the Inadunata than the Flexibilia. 10 BULLETIN 87 370 Among the Inadunata Euerisocrinus is quite similar in some respects, but is close to Erisocrinus, having wide horizontal artic- ular facets and lacking the protrusion of the posterior side found in the form at hand. It is doubtful that they are even remotely related. Occurrence and horizon.—Pennsylvanian of North America. Moundocrinus osagensis, n, sp. Plate 1, figs. 5, 6, 10 Calyx irregular outline, anterior side curving evenly upward but posterior side sloping upward at a forty-five degree angle, diam. 21.5 mm., height 9.0 mm., basal area flattened. IBB disk measures 8.0 mm. across, of which the round columnar scar occupies 3.5 mm., depression sloping gently inward and is pierced by minute round axial canal, the circumference being sharply crenulated. IBB 5, small equal pentagonal elements ; EB 5, large equal hexagonal elements, save that of the posterior which is truncated for the reception of the anal X, lower ex- tremities curved under to form part of basal area; RR 5, large equal pentagonal elements and since the posterior side is protruded those of the posterior do not lose any of their width to the large anal X, articular facets no wider than the normal thickness of the plates, outwardly a strong ligamental furrow adjoined by sharp cross ridge, muscle area slopes inward evenly divided by a narrow groove, that area adjoining the cross ridge is backed by small leaf- shaped, low protuberances which join at the groove and in turn possess shallow muscle depressions, the main muscle area is backed at the inward and lateral extremities by a low ridge, the notch developed at the outer extremity of the interradial suture expands inward to form shallow confluent depressions ; anal X large hexag- onal, resting broadly on post. B, extending only slightly above up- per extremity of RR. Plates of cup smooth, only aio depression of sutures, rather thick plates. . Arms and tegmen unknown. Stem unknown save for column- ar scar which is small, round and crenulated. Occurrence and horizon.—Stanton limestone member, Ochelata group, Pennsylvanian, the mound just west of Bartlesville, Okla- homa. | Type.—Springer Collection of the U. S. Museum. Sift PENNSYLVANIAN CRINOIDS: STRIMPLE ay Genus PENTADELOCRINUS, n. g. Genotype.—Pentadelocrinus typus, n. sp. The form here under consideration is comparable to Delocrinus in having a funnel-shaped basal concavity, wide articular facets, a single anal plate within the calyx, and the same general appear- ance, differing mainly in having a pentagonal stem and the anal X being axillary and similar to those found in Cibolocrinus and asso- ciated genera. Parasaplocrinu:s Moore and Plummer is quite close but.has only 3 IBB and a round stem. Thete is a strong possibility that the present form will eventually find its way to the Lecano- crinidz but for the present is assigned to the Inadunata. Occurrence and horizon.—Pennsylvanian of North America, -Pentadelocrinus typus, n. sp. Plates ies eon Cup low saucer-shaped, measuring 20 mm. diam. by 6.5 mm. high, strongly invaginated base. IBB 5, small equal pentagonal elements entirely within basal cavity. BB 5, large equal hexagonal elements save that the posterior which is truncated for the recep- tion of anal X, proximal portion curved under to participate in basal concavity. BB 5, large equal pentagonal elements, those two of the posterior losing some of their width to the rather large anal X, almost twice as wide as high, left lateral side of r. post. R 0.5 mm. longer than normal, right lateral side of 1. post. R 0.5 mm. shorter than normal. Articular facets developed strongly inward as horizontal shelves, slanting slightly inward, sharp ligamental furrow, larger than usual, to the fore adjoined by cross ridge, inner edge strongly notched by ambulacral furrow ending just back of the cross ridge with a minute circular opening, muscle scars very shallow, developing broadly laterally until stopped by a low ridge just short of interradial sutures, which are in turn depressed and widen slightly inward. Anal X rather large, heptagonal, low- er extremity 4.0 mm. wide; 6.5 mm. greatest width; 6.1 high ; right lateral side measuring 4.2 mm. left lateral side 3.1 mm. ; right upper edge 3.6 mm.; left upper edge 2.2 mm. These measure- ments are given to show the unusual shape and ensuing effect on adjoining plates. Upper facets are developed inward and are strongly crenulated to the fore. Several plates which have fallen into the cup cavity indicate through measurements and compari- son, that they are azygous. These plates are low and broad, strongly rounded, and inwardly notched by ambulacral-like 12 BULLETIN 87 372 grooves, as is also the upper facets of anal X. Stem unknown save for columnar scar, same being strongly pentagonal and pierced by minute round axial canal. Arm and tegmen unknown. Occurrence and horizon.—Stanton limestone member, Ochelata group, Pennsylvanian, from road cut 3 miles due west of Ramona, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Genus GRAPHIOCRINUS de Koninck Graphiccrinus stantonensis, n. sp. Plate 2, figs. 11, 12 Cup full, bowl-shaped, basal area mildly convex, diam. 9 mm. by 4mm, high. IBB disk occupied in the main by sharply depressed columnar scar, which is round heavily crenulated and pierced by minute axial canal, IBB 5 small pentagonal elements. BB 5 rather large hexagonal plates, post. B differing in being truncated for reception of anal X. RR 5 pentagonal elements. Anal X ‘hexagonal, protruding well out of cup, followed by single unob- served azygous plate. Articular facets of RR developed inward as horizontal shelves, strong ligamental furrow to the fore backed by cross ridge, pronounced ambulacral groove, shallow muscle scars developed laterally and divided by shallow, small perpen- dicular groove. A specimen with the IBrt attached has been observed but is not figured herein because it is not well enough preserved. [Br axillary, slightly elongated, mildly constricted laterally in median portion, quite similar to that found in G. carbonarius. Plates of cup very mildly tumid, fine granular appearance in some instances; balance of arms, tegmen and stem unknown. Relationship.—The only stratigraphically associated species known is G. carbonarius (Meek and Worthen) which is quite distinct in having the anal X axillary, IBB restricted to depressed area, and strong protuberances at the interradial sutures of the articular facets. The plates of the cup are also more tumid in G. carbonarius. Occurrence and horizon.—Stanton limestone member, Missouri series, Pennsylvanian, near Wayside, Kansas. Type.—sSpringer Collection of the U. S. National Museum. 373 PENNSYLVANIAN CRINOIDS: STRIMPLE 13 Genus EUERISOCRINUS, n. g. Gcnotype.—LHuerisocrinus waysidensis, n. sp. This form is close to Erisocrinus specifically E. typus differing only in the possession of a single anal piate within the cup. Un- fortunately only a single complete cup has been observed to date, however, portions of cups and fragments bear out the conclusion that this form is distinct. The specimen at hand does not have the anal X resting on the post. B, but same has migrated a short distance distally, however, the genus is intended to include those forms having the anal X well established within the cup and the cup being high, slow steady expansion, a flattened base, and round stem. Occurrence and horizon.—Pennsylvanian of North America. Euerisocrinus waysidensis, n. sp. Plate 2, figs. 14-16 Cup high, slow expanding, height 6 mm., diam. 8 mm., flat- tened base, IBB not visible from side view. IBB small pentagonal plates occupying a sharp depression in the center of the flattened basal area, BB 5 large hexagonal plates, lower extremities curved under to participate in flattened base. RR 5 pentagonal plates save those two of the posterior which are hexagonal by virtue of closing behind the anal X, articular facets developed inward as horizontal shelves, notched to the fore by ligamental groove which is adjoined by sharp cross ridge, muscle scars developed laterally, ambulacral grcove prominent. Anal X pentagonal, rather large, resting well within the cup with distal portion only protruding. Relationship.—This form is close to, and no doubt the prede- cessor of, E. typus from which it differs only in having a single anal plate well within the cup and possibly being a little higher than the normal representative of that species. Occurrence and horizon.—Stanton limestone member, Missouri series, Pennsylvanian, near Wayside, Kansas. Type.—Springer Collection of the U. S. National Museum. Genus ETHELOCRINUS Kirk Ethelocrinus convexus, n. sp. Plate 1, figs. 11, 12, 15, 16 Cup broad, bowl-shaped, IBB disk shallow, convex, saucer- shaped, visible from side view, young figured specimen 12 mm. 14 BULLETIN 87 374 wide by 9.2 high, mature specimen approximately 29 mm. wide by 12 mm. high. Median portion of IBB disk is slightly de- pressed and occupie’! by the proximal columnal which is smaller _in diameter. IBB 5 large pentagonal elements. BB 5 large hex- agonal elements, ge.itly tumid, that of the posterior being trun- cated for the reception of anal X and together with the r. post. R carries the radianal. RR 5 regular pentagonal elements, slightly tumid, that of the r. post. encroached on by the radianal, articular facets developed inward as horizontal shelves, very pronounced ligamental furrow to the fore adjoined by sharp cross ridge, shal- low muscle scars developed laterally. Anal X elongated, hep- tagonal, approximately one-fourth extending out of cup and fol- lowed by two unobserved azygous plates, measuring in smaller specimen 4 mm. high by 1.5 mm. wide, larzve specimen 9.6 mm, high by 6 mm. wide. Radianal placed oblicuely resting on post. -B and supported by r. post. B, quadrangular loaf-shaped. No ornamentation, plates thick. Of the arms our knowledge is limited to IBr? of the 1. post. and r. ant., 10 cuneiform arms indicated. IJIPr* axillary, height and width about equal, mildly constricted laterally in median portion. Of the stem we have only the proximal columnal, same being round, small, heavily crenulated, and pierced by minute round axial canal. Tegmen unknown. Relationship.—This species is close to F.. plattsburgensis Strim- ple which species however, has a concave basal area. Occurrence and horizon.—Stanton limestone member, Ochelata group, Pennsylvanian, the mound just west of Bartlesville, Okla- homa. Type.—Springer Collection of the U. S. National Museum. Genus HYDREIONOCRINUS de Koninck (Emend. Moore and Plummer) Whereas Hydreionocrinus and allied genera are quite prolific in these formations, considerable more research will be necessary for their proper handling. Moore and Plummer (1938) restricted Hydreionocrinus to those forms having the IBB visible from side view with the notation that no forms from North America con- 375 PENNSYLVANIAN CRINOIDS: STRIMPLE 15 forn-ed in that respect, taking into consideration also the arm structure. Although the arms are unknown there is at hand a cup with the IBB visible from the side, same being presented herein as Hydreionocrinus deweyensis, n. sp. Hydreionocrinus deweyensis, n. sp. Plate 2, figs. 13, 17 Cup low, widely expanded, with the IBB visible from side view, measuring 9.5 mm. diam. by 3.9 mm. high. IBB 5 small pentag- onal elements, distal portions curved under to form flattened area occupied by small columnar scar, same area being round, heavily crenulated, and pierced by minute round axial canal. BB 5 rather large hexagonal plates save for the post. B which is truncated for the reception of anal X and together with r. post. B assists in the support of the radianal. RR 5 equal pentagonal elements, slightly wider than high, those adjoining the anal series losing some of of their width, facets developed inward as horizon- tal shelves, notched to the fore by a short ligamental furrow which is adjoined by sharp cross ridge, shallow muscle scars developed laterally and ambulacral groove pronounced. Anal series com- posed of three plates within the cup, anal X hexagonal, slightly elongated, resting solidly on post. B, distal portion extending out of cup, radianal rather large, slightly elongated, pentagonal, placed obliquely on post. B, supported by r. post. B, assisting in the sup- port of anal X, and supporting the small right tube plate, which is missing. Arms, tegmen and stem unknown. Cup covered by irregular, wide spaced, minute spinelike nodes, entire surface rough. Relationship—The only known form comparatle with this species is the genotype, H. woodianus de Koninck from the Scot- tish Carboniferous, which although larger has the same general appearing calyx. Occurrence and horizon.—Dewey limestone, Pennsylvanian, Dewey Portland Cement Quarry, Dewey, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Genus DECADOCRINUS Wachsmuth and Springer Of this genus a single representative has been observed in the specimens at hand, same proving distinct and is described herein as Decadocrinus regularis, n. sp. 16 BULLETIN 87 376 Decadocrinus regularis, n. sp. Plate 2, figs. 20,°21 A small species, cup evenly expanded, very small basal area rather flattened, diam. of cup 4.3 mm., height 2 mm. IBB 5 small pentagonal plates. BB 5 large hexagonal elements save that of the posterior which is truncated for the reception of anal X and together with r. post. B supports the radianal, lower ex- tremities slightly entering basal area. RR five, not so large as BB, pentagonal equal elements save where encroached on by anal series, facets developed inward as shelves but unobserved beyond the ligamental furrow. Anal series of three plates, anal X hexagonal, resting solidly on post. B, small, extending slight- ly out of cup, radianal elongo-pentagonal, resting obliquely on post. B, lower extremity supported by r. post. B, carrying above the small hexagonal right tube plate. Arms ten to the IIBr*, cuneiform, branching on IBr?. IBr'’ elongated, median portion slightly constricted laterally. Brachi- als following slightly elongated, strongly rounded, median por- tions mildly constricted. Left and right anterior [Br’, shorter than other three. Stem unknown save for the columnar scar which is small, round, crenulated and pierced by minute round axial canal. Tegmen unknown. No ornamentation observed. Sutures of the cup very slight- ly depressed. Relationship—This species is readily distinguished from other known species of the genus and is considerably higher stratigraph- ically. It is comparable in size to those known. Occurrence and horizon.—Stanton limestone formation, Oche- lata group, Pennsylvanian, the mound just west of Bartlesville, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Genus MELBACRINUS, n. g. Genotype.—Melbacrinus americanus, n. sp. There has appeared a very distinct form with a turbinate shaped cup and the anal series of three plates not extending above the upper extremity of the RR. Hydriocrinus Trautschold is im- mediately thought of but in that genus the stem is pentagonal and _ the branching of the arms different. It is with pleasure that I dedicate this genus to my wife, Mrs. Melba Strimple. 377 PENNSYLVANIAN CRINOIDS: STRIMPLE iV; Occurrence and horizon.—Pennsylvanian cf North America. Melbacrinus americanus, n. sp, Plate 3, figs. 1-4 Cup high, turbinate-shaped, largest observed specimen approx. 5 mm. high by 4.5 mm. wide. Columnar scar round, circum- ference deeply crenulated, and pierced by minute round axial canal. IBB 5 pentagonal elements rising sharply above columnar area, higher than wide. BB 5 equal hexagonal elements save that of the posterior which supports the anal X above and the radianal to the right along with the post. B. IRR 5 equal pentagonal ele- ments, slightly wider than high . Radianal pentagonal, resting to the right on r. post. B to the left on post. B, assisting to the right in the support of r. post. R supporting above the right tube plate, and assisting to the left in support of anal X ; anal X penta- gonal ; right tube plate quadrangular. The anal series do nct ap- preciably extend above the upper margin of RR; ana! X and rt forming a horizontal line. The articular facets of RR are not appreciably produced in- ward, shallow muscle scars developed laterally, ligamental notch to the fore very pronounced. The upper facets of the anal X and rt plate are similarly developed. Of the arms we know the following: In the immature speci- men the r. post. and |. post. IBr’ are axillary, very elongate and narrow, |. post. IBr? is 4.4 mm. in length, r. post. [Br* 3.5 mm. Ant. IBr! is elongated, 3.6. mm, in length, and followed by a sin- gle series of brachials. L. ant. [Br* and r. ant. 1Br* are not axillary, measuring 1.7 mm. and 1.8 mm. respectively, the IBr? following being axillary and measure 2.1 mm. and 1.9 mm. in length respectively. A light raised ray extends the length of the brachials and is retained in the single IBr' preserved with the more mature specimen, same being 1. post. Relationship.—The closest known form is Hydriocrinus pus- illus Trautschold which, however, has a pentagonal stem and the arms branch on the primibrachs of all rays. Occurrence and horizon.—Stanton limestone member, Oche- lata group, Pennsylvanian, the mound just west of the city limits of Bartlesville, Oklahoma. Type.—Springer Collection of the U. S. National Museum. Genus AGASSIZOCRINUS Owen and Shumard Whereas the practice of using fragmentary remains is to be 18 BULLETIN 87 378 discouraged, a distinct fused IBB cone of this genus is consid- ered worthy of specific segregation. Agassizocrinus mcguiret, n. sp., is proposed, named for the so ardent collector, Mr. Paul MoGuire of Fairfax, Oklahoma, who called attention to the loca- tion where this form occurs. Agassizocrinus mcguirei, n. sp. Plate 2, figs. 9, 10 Fused narrow IBB cone measuring 8 mm. diam. by 8.2 mm. high, no evidence of sutures or stem attachment, expanding very slowly until approximating upper extremity where there is a sudden flare. This is quite different from any other known spe- cies. The only other known Pennsylvanian representative above the Morrow subseries is 4. farri Strimple, in which the cone expands evenly. Occurrence and horizon.—A friable limestone associated with the Nellie Bly formation, just north of the city limits of Ramona, Oklahoma. Type.—Springer Collection of the U. Ss. National Museum. Family CYATHOCRINIDA Roemer (Emend. Wachsmuth and Springer) Genus LECYTHIOCRINUS White As there are at hand well nigh perfect specimens of the only two previously known species L. ollicueformis White and L. adamsi Worthen, same are figured herein to give a more ade- quate perspective of these unusual forms. From the Stanton lirnestone near Wayside, Kansas we have a new form, described herein as Lecythiocrinus urneformis, n. sp. Lecythiccrinus cllicueformis White Plate 3, fics. 5-7 Lecythiocrinus ollicueformis White, 1880, Proc. U. S. Nat. Mus., vol. II, p. 257; White, 1880, Geol. Survey ‘of the Territories, p. 124, pl. 35, figs. Qa, b. Lecythiocrinus ollicueformis Wachsmuth and Springer, 1886, Rev. Paleo.; Part III, Acad. Nat. Sei. Phila., Proe., vol. 38, p. 152. The species is now well established and known but the holotype was poorly preserved. Information derived from the well pre- served specimens herein figured is thought interesting. White noted only three IBB but anticipated five. There are usually three, occasionally five IBB in observed specimens. The 379 PENNSYLVANIAN CRINOIDS: STRIMPLE 19 peculiar rather large interradial aperture occuring entirely with- in the cup was no doubt obliterated in the holotype. Whether the small tubes observed emerging from the aperture perform some function of the animal or are foreign is necessarily a mat- ter of conjecture. If the aperture was an anal opening, as seems quite probable, such as found in Edapocrinus rugosus Wright for example, one would expect either a smaller opening, cover- ing plates, or a leathery peristome. There is no evidence of cov- ering plates. With a leathery covering one would expect some attempt to place the anus at some distance from the calyx, which function could be performed by the tubes. - Similar tubes hav been observed in L. adamsi: Occurrence and horizon.—Holotype— Upper Coal Measure strata, “30 miles W. of Humbolt, Kansas.” The type specimen is in the U. S. National Museum. Figured specimen.—Dewey limestone, Pennsylvanian, Dewey Portland Cement Quarry, Dewey, Oklahoma. Specimen in the Springer Collection of the U. S. National Museum. Lecythiocrinus adamsi Worthen Pl. 3, figs. 8-10 Lecythiocrinus Adamsi Worilen, 1882, Ill. State Mus. Nat. Hist., Bull. I, p- 37; 1883, Geol. Rep. Lll., vol. VII, p. 317, pl. 30, fig. 8. } Lecythiocrinus Adamsi Wachsmuth and Springer, 1886, Rev. Paleo.: Part Til Acad: Nat. Sci. Phila., Proe., vol.-38, p. 152. The specimen on which this species was based merely indi- cated the interradial aperture shown herein. In other respects the species was well described and illustrated, however, 5 IBB are shown by Worthen, and only three have been noted in speci- mens at hand. . Occurrence and horizon.—Holotype—about the horizon of Coal No. 8 of the Lower Coal Measures, Sec. 13-11N-6E, Peoria County. Figured specimen.—Dewey limestone, Pennsylvanian, Dewey Portland Cement Quarry, Dewey, Oklahoma. Specimen is in the Springer Collection of the U. S. National Museum. Lecythiocrinus urneformis, n. sp. Plate 3, figs. 11-13 Calyx high, urn-shaped, with full base and constricted distal portion, diam. at upper extremity 5.9 mm., greatest width 7.9 mm., height 8.9 mm. IBB 5 small pentagonal plates, gently con- vex basal disk, columnar scar small, round, with heavy crenula- 20 BULLETIN 87 380 tions and pierced by minute round axial canal. BB 5 large hex- agonal plates. RR 5 comparatively small pentagonal plates, facets not filling distal face of RR and slightly protruded. Fully within the cup and interradial in position is a small oval-shaped opening. The basal is mildly protruded just below the opening, and on all other BB at similar location. There is no evidence of cover- ing plates. Relationship.—This species is very close to L. ollicueformis differing only in having comparatively greater height and the quite distinct urn-shape. Occurrence and horigon.—Stanton limestone member, Miss- ouri series, Pennsylvanian, near Wayside, Kansas. Type.—Springer Collection of the U. S. National Museum. REFERENCES Bather, F. A. 1900. The Echinoderma, in E. Ray Lankester, A Treatise on Zool- ogy; Part III, pp. 1-333. 1911. Notes on Hydreionocrinus: Trans. Edinburgh Geol. Soce., vol. X, part 1, pp. 61-79, 8 pls. Kirk, Edwin 1937. Hupachycrimus and related Carboniferous crinoid genera: Journ. Paleo., vol. 11, pp. 598-607, pl. 84. Laudon, L. R. 1937. New occurrence of the Upper Carboniferous crinoid genera Amphicrinus and Synerocrinus: Journ. Paleo., vol. 11, PP .706- 708, figs. 1 and 2. Meek, F. B. and Worthen, A. H. 1865. American Journ. Sci., ser. 2, vol. 39, pp. 174-350. Miller, S. A, and Gurley, Wm. F. E. 1890. Some new genera and species of Echinodermata: Journ. Cin- cinnati Soe. Nat. Hist., vol. XIII, 25 pp., 4 pls. Moore, R. C. and Plummer, F. B. 1938. Upper Carboniferous crinoids from the Morrow subseries of Arkansas, Oklahoma, and Texas: Denison Univ. Bull., Journ. Sei. Lab., vol. 32, pp. 209-313, 5 pls. Springer, F. 1920. Crinoidea Flexibilia: Smithsonian Inst., Publ. 2501. 1926. Unusual Forms of Fossil Crinoids: Oy S. National SUES, Proe., vol. 67, Art. 9, pp. 1-137, 26 pls. Strimple, Harrell L. 1938. A Group of Crinoids from the Pennsylvanian of Northeavtenn Oklahoma, 12 pp., 2 pls. Trautschold, H. 1867. Hinige Crinoideen des jungeren Bergkalks Moskau: Bull., Soe. Imp. Nat. Moscou, vol. 40, No. 3, 49 pp., 3 pls. 381 PENNSYLVANIAN CRINOIDS: STRIMPLE 21 Wachsmuth, Chas, and Springer, F. 1879, 1881, 1886. Rev. Paleo.: Parts I-III, Proce. Acad. Nat. Sci., Phila. Wanner, J. 1916. Die permischen Echinodermen von Timor: Paleontologie von Timor, Lief 6, 329 pp., 19 pls. 1924. Die permischen Krinoiden von Timor: Jaarboek van het Misnwezen in Nederlandsch Oost-indie, Verhandlelingen 1921, 3rd Gedeelte, 347 pp., 22 pls. Weller, Stuart 1909. Description of a Permian crinoid fauna from Texas: Journ. Geol., pp. 623-635, 1 pl. White, C. A. 1880. Proce. U. S. National Museum, vol. ii, p. 257. 1880. Geol. Survey of the Territories, p. 124, pl. 35, figs. 2a, b. Wright, James 1935. Geol. Mag., vol. LX XII, No. 851, pp. 195-197, 7 pls. 22 Figure i, Dy 3, 4. 5, 6, 10. Up telnasls 1 12555 16: 138, 14, 17. BULLETIN 87 EXPLANATION OF PLATE 1 (27)* Amphicrinus oklahomaénsis, n. sp. 9 -—~—----——--—---— Fig. 1, view from below; fig. 2, view from above; .pos- terior lower extremity, Stanton limestone, mound west of Bartlesville, Oklahoma. Cibolocrinus robustus, n. sp. 2 atlas kee Fig. 8, posterior to the left; fig. 4, anterior view. Moundocrinus osagemnsis, n. g., n. sp. . — ——---. = Fig. 5, posterior view; fig. 6, view from below; fig. 10, view from above, Stanton limestone, mound west of Bartlesville, Oklahoma. Pentadelocrinus typus, n. g., n. sp. - Ae et Fig. 7, posterior view; fig. 8, view from below; fig. 9, view from above. Stanton limestone, mound west of Bartlesville, Oklahoma. Ethelocrinus convexus, n. sp. - iS Figs. 11 and 15, mature form; fig. 11, view from below; fig. 15, posterior view; figs. 12 and 16, young form; fig, 12, posterior view; fig. 16, anterior view. Stanton limestone, mound west of Bartlesville, Okla- homa. Delocrinus, nodosarius; ne ‘sp. 22 Fig. 18, view from below; fig. 14, view of another specimen from above showing [Br1; fig. 17, same as fig. 13, posterior view. Stanton limestone, mound west of Bartlesville, Oklahoma. *(All Figures Natural Size) 382 10 fe 13 Pu. 27, VOL. 24 BULL. AMER. PALEONT. No. 87, Pu. 1 24 BULLETIN 87 384 EXPLANATION OF PLATE 2 (28) Figure Page 1-8) Delocrinus tumidusyin) spo ee 8 Figs. 1-3, mature specimens; fig, 1, view from above; fig. 2, view from below; fig. 3, posterior view, natural size; figs. 4 and 5, specimen with nonton of arms attached; fig, 4, poste- rior to the left; fig. 5, anterior view, natural size; figs. 6-8, young specimen; fig. 6, view from above; fig. 7, view from below; fig. 8, posterior view, X 2. Stanton limestone, mound west of Bartlesville, Oklahoma. 9,10. Agassizocrinus meguirei, n. sp. — 18 Fig, 9, IBB cone from side; fig. 10, from below. Unidentified limestone associated with Nellie Bly formation, near Ra- mona, Oklahoma. Natural size, 11,12. Graphiocrinus stanionensis, n. sp. 12 Fig. 11, posterior view; fig. 12, view from below. Stanton limestone, near Wayside, Kansas. X 2. 13,17. Hydreionocrinus deweyensis, n. sj 9 15 Fig, 13, posterior view; fig. 17, view from below, Dewey lime- stone, Dewey, Oklahoma, X 2. 14-16. Euerisocrinus waysidensis, NSS Pigg eh ee ee ee 13 Fig. 14, posterior view; fig. 15, view from above; fig. 16, view from below. Stanton limestone, near Wayside, Kansas. X 2. 18,19. Amphicrinus poundi, n. sp. ____.-- 5 Fig. 18, view from above; fig. 19, view from below. Stanton limestone, near Wayside, Kansas. X 2. 20,21. Decadocrinus reguiaris, n. sp. 16 Fig. 20, posterior to the left; fig. Zile anterior view, Stanton limestone, mound west of Bartlesville, Oklahoma. X 2. PL. 28, Vou. 24 BULL. AMER. PALEONT. 26 BULLETIN 87 386 EXPLANATION OF PLATE 3 (29)* Figure Page 1-4. Melbacrinus americanus, n. g., n. sp, —-------- peas 17 Figs. 1 and 2, mature specimen; fig. 1, posterior view; fig. 2, anterior view; figs. 3 and 4, immature specimen; fig. 38, anterior view; fig. 4, posterior to the left. Stanton lime- stone, mound west of Bantlesville, Oklahoma. ; 5-7. Lecythiccrinus ollicueformis White 18 Fig. 5, view showing interradial aperture; fig. 6, side view; fig. 7, view from below. Dewey limestone, Dewey, Oklahoma. 8-10. Lecythiocrinus adamsit Worthen —_-.... 19 Fig. 8, view showing interradial aperture; fig. 9, side view; fig. 10, view from below. Dewey limestone, Dewey, Oklahoma. 11-13. Leeythiocrinus urneformis, n, sp. 19 Fig. 11, view showing interradial aperture; fig, 12, view of anterior; fig. 13, view from below. Stanton limestone, Dewey, Oklahoma. *(All Figures X 2) PL. 29, VOL. 24 BULL. AMER. PALEONT. No. 87, Pu. 3 INDEX TO VOLUME XXIV Note:- Light face figures refer to the volume paging and not to the paging of the separate } plate numbers. A ACTOS pInIeTs sae ee olssoni US a ay aS aly Nice ONC aa seen See Acteon Andersoni —.. Actinoptera poudt a eee ApPOMUS ACI ween textilis SEROUS), Ss eS ape pe yulletins. 248, 236 Agassizocrinus Ee RRA. >. 377 meguirel 28 364,378 ANGARUO ER HIS) ee eae nl 49 Aldrichia te Bu So gato 3 ATU ONGrMiS) eee 364 eu DOMAINS = = se ee 365 oklahomaénsis 27 +«9363, 364, 366 [OWN 2 ee PAS) BB Boi okela) ScouGusm a=. fen 365 Amphidesma mississippiensis 350 Amphistrophia 198, 199, 200 continens senilis 209 Podolica = L. —_ 205 Anastomopora quebec- ensis - pal CO), ile SLE Anomia reticularis 239 Anomites lineatus —_ 245 striatus __ bess 248 Anoplotheca cf, -silvetti 13 ile}, Bes} Antispirifer harraldi - 255 IAT OID NIN ee er ae Sor extenta s oot eee PT eS Arca mississippiensis | pee 342,350 Architectonica Be 351 EEYCUDEEDICS, se Aes Beeler Beer 26 347 bellastriata 26347 Arnold, Chester NOY on fossil plants of (Gaspe == = Tt NStarten parilis~== =.= = 92d) GO2.a09 Auhiyrist neta) =. 25.2 69 INGAG OG Set sae eee ee ee 139 aesquamata =e 241 desquamata magna 2 241 harrisi 13, 16, 17° «113, 240 Harris enasUitcan ese ee fp able Bee reticularis 67, 241 unisuleata —_ 271 Australospirifer eee 249, 259 ef. antarcticus, var. 1 18, 19_ 113, 262 ef. antarcticus, var. 2 13,19 113,263 Australostrophia Ave 14 1939 129, 174, 183 Heavy face figures refer to Avicula argentea Aviculopecten, sp. A _18, Sp: (bi eeoesuee Axinea _._ Baton River beds, New Zealand a Beachia amplexa Bellerophon leda __ Bisoarca lima Mississippiensis protracta Botryllopora socialis Brachyprion newsomensis ee ene plantas schuchertana shaleri subinter striatus ensis Brachyspirifer palmere Buccinum Bulbifusus 16, Mississippiensi Bulla crassiplica a. Busycon SS Ce Cachira series Calceola beds __ Callograptus salteri strictus Callonema ef, bellatulum Camarotcechia semiplicata sp. Cancellaria funerata Mississippiensis Cape Bon Ami beds Cape Rosier beds Cape Rosier lighthouse section Capulus Americanus Cardita bilineata densata planicosta subquadrata subrotunda tetrica vigintinaria Cardium diversum eversuni. =e Nicolletti s Vicksburgensis —-—-—-- Caricella demissa 18 13 113,275 the volume 350 19 118, 278 19) 1135278 351 65, 69, 74, 89 128, 131, 133 133 133 147 134 ‘serent- 134 248, 264 113, 264 S 350 5 350 351 110, 284 182 44,50, 55 46, 50, 52 46 a og “gan 350 343,350 359 350 350 359 350 350 350 ~ 343, 844, 359 350 350 388 INDEX, VOL. 24 388 MO Olas ce rs NE Psy BVM Bae) (Callens, - a 3 subangulata 24 347,359 Colombian Bryozoa — 276, 279 Glaciaigiamig: Lj 348 EHR) OCIS eee a iI) Bre) Iiniveas ee Sen 350 OSitracodes j= === asses 277, 280 Cassiduila | 22 351 Conrad, T. A., on Jackson Gassis) czlatunae =e 350 Eocene Fossils 343 Mississippiensis __. 350 Conus [sauridens| — 346 Caster, K. E., on Colombian COUCHES ee Oat 343,346, 359 Devonian fauna __.__._- 107 Corbis staminea 350 Catopygus Conradi 350 Conbullas salva 343, 350 Centronella silvetti 244 ICE NOINED, ee Pas Byala! Cerithum Claibornensis — 850 déensata. ee 23 344 TYVEUS SU Es Bay aoe ce ee 350 CENA MENERY a 350 SU bile Euniryt ee ae ee et ee 350 Tae SENN 350 S@litteymlpyry 350 THES MGs see ee eee 344 Chama Mississippiensis _ 350 ONUSCIUS yet reeoeeiene 344 Chemungia- — 198, 200, 208, 206 Corynoides gracilis. aaa 47 Chenopus liratus —..--..-_- 350 Coscinium striatum — 74, 90 GON CeCe see aa ees Pash (Ore CIAO CANNES Lo 28 PASO ie, eee ot Heenan sd La popxe).) (Ghee outlet 116 AT CULE US ae es ec ne 221 Crassatella saliva === 343 ROU DET TS ego eres oe SIE ee 69 flexura : i 23 345 cia billinigsi] = 16} dale. ABS By Mississippiensis | cm iain 343, 350 che compilanaitay = a= 69 FOMAOLLED, cree neste AN er 345 OOS HOO aT pil Creeyoptahwilless Mase ye, 343 cCononatus) == eee 231,234 Cromyocrinus — —__- nee 25 CWE AE DIS cy eee 236 Cryptograptus trieornis _ 48 faliclandicuis\s sa 230, 285,282 Cryptonella, sp. . 1G) WS, AIG HUESTEC ETSI S ee ree 2385 Crystolites expansus ie 66 Ree yee le ones eter 235 Cyathophyllum bolivianum 277, 280 herbertsmithi = 236 Cyclas —__- oe 343 hudsonicus gaspensis 64, 69 Cymostrophia _ 128, 132, 135, 148 UCT Oa CUIS eens eee 231 dickeyi OMOMOINUS oe 236 9,10,14 112, 145, 155, 156, 159 TEVeUSA oe lire ean aNs 237 schucherti 7,8, 9,12, 16 APL Csi SUES Se es Se 233 112, TAA VAS 55, 09 SOO MHS nena 235 waringi 8,9 _ 112, 153, 154, 159 SUMIAtISSimMus) = ee 235 Cyphaspis, sp. —_ 20 1138, 281 Gir, Sulloet LA A 2381h 2383 Oyiprceabarcsc hin assess Pil B28) subhemispherica _.... 226, 229, 236 campbelliana i 329, 333 VW aliniovleang) a 231 CMvUlOme, = Dl Be venezuelensis 237 Ginlenes) uaa seas 324, 329, 333 AUDI TTS 1S yg eel 231 cinerea morinils) see) ee Bil ae Chonostrophi aes 237 dominicensis =. 333 @ompllanalt eae hier eee tine 238 elle ganic ae eemene 344, 348 Caw: S Ong ase ewer een 64, 69, 238 fenestralis 26 348, 348, 359 knodi) 22 ects al ace 15) 113-237 AR NAS Le i 22 329 Cibolocrinus robustus at BEB, BOY enorme 2 PL RL Claiborne group 343 henekeni )s.s)) Stee. PL BA). BSL @ilevelllleigiie ec ueat sce ree 4 henekeni poucreute 22 331 Glavilithes! cee 4, 845 isabella 222 secre 324, 335 humerosus 22222555 24 345, 359 isabella mexicana —___ 355 Kon gcevilisi es ee 5 Vinite a) (oe ee te 350 Mississippiensis 26 345,359 merriami an 22eooU pachyleurus; =.= 350 PL OVUL elise ccm dn eae 22 329, 335 AFISIe NS1S) ee 5 DPatrep atric 335 MATIUCOS See ee 25 345,359 OUT GU sive ee ale 26 348 Vieksburgensis 350 raymondrobertsi 324, 329, 336 E€limacograptus bicornis _ 48 raymondrobertsi bow- modestus 48 denensis 0 i a aie 336 Pans aie eo es 2 AS) I SOMON lee es ee eee 330 3°) i sphwenoides) =. 348, 350 spurea 22 329,336 spurcoides 22 329,337 Cypredia 2a 348 €ypricandinia —.--_—- ae 276 ef. subindenta eis PAD = alls vr /y) Cyrtina hamiltonensis 69, 74 Cytherea astartiformis —~ 350 CRE) ea ae eee 350 THOTT Op ee ee eee 350 nicer) a2ye Biase ee ee 350 HE Mage eee ee 350 Mississippiensis 350 WEG REVS bee ee 350 DARE, ae ee 2 350 Semepunictata =... 350 SOpiiitanee = 359 subimpressa sete 350 D Dalmanella penouili _- 69 Dalmanites cf. patacamaya- ensis eee 20) 38-281 Decadocrinus ‘multinodosus serratobrachiatus 23 memulanrisy = 28 364, 31D, 018 spinobrachiatus 2 14, 23 Delocrinus hemisphericus 368 nodosarius —.-- ————«<— TT: 8G, BGS tumidus. —— cee PA BKB RSS: Delthyris fim briatus 245 granulosa — 267 Dendrograptus fructicosus 47 Dentalium Mississippiense 350 Derbyiana, sp. - 7 -alalss, Par Deronaster eucharis gold- TET ay 20-2 ps A 66 Diaphorostoma perceense Tal Dicellograptus divaricatus 48 moffatensis d 48 SextanseexXilis) |. - 48 Dicranograptus furcatus 48 gramosus pars 48 ramosus 48 Dictyonema approximatum 47 perbextuml, 2. 47 Dictyostrophia "128. 132, 135, 140 cooperi 8,10, 12 112, 144, 155, 163 Didvmograptus gagitticaulis 48 Diplodonta eee s 351 Diplograptus acutus. 48 euglyphus oe 48 Discoidea Haldermani 350 Douvillina 128 Glin ae 170 Wornvallimeltay os 128 Douvillinini 128, 170 E Eatonia 53 peculiaris 69, 72 Edapocrinus rugosus 379 VoL. 24 389 Bly thepe eet 245 columbisna = — 19 113,246 Endopachys alticostatum 349 expansum, = 349 TANS ae 349 Hodevonsria. 221 arcuatas= 225 hudsonicug 230, 236 hudsonicus gaspensis —— 74 imperlalis USPS) 1135 2235 237 imperialis parva 13,15 113, 226 imperialis transversa 13,17 118,228 reedi 2 ee 13, 15 229,236 thomasi 233 Eospiriferina lachrymosa’ 254 Erisocrinus typus: —______ 364 Escuminac beds —.___ {ia Ethelocrinus convexus 27 364, 373 Euerisocrinus waysidensis 28 363, 373 hulimella:=— = 6 GRASSI as Gee eee es 7 MacAndrel.==—— == if Scully ! a ene e 7 Euphemus ? quebecensis — 70 F HAVOSItC Ss). =. ee Ss 69 Fenestella venezuelensis 113; 279 Fenestrellina erectipora — 93 gaspiensis _..._. 5 65, 69, 74, 90, 92 oceidentalis) 2-2 16 74, 90. 92 SLT OS ea een ee ee 93 Fimbria aes 351 Fimbriata uncispinei ieee 254, 262 Fissurella Mississippiensis 350 Flabellum Wailesii 349 Floresta series __ Be 110 Four Mile Brook beds 74, 79, 89 Four Mile Brook section 73 Fritz, M. A., on Gaspé De- vonian Bryozoa _------ 89 Fulgoraria Mississippi- ensis Se aa 350 Fulegur nodulatum > 350 Fusimitra = 347 Fusus Mississippiensis ee 350 now = eee 5 papillatus a 348 Spinigeri) == 350 Vicksburgensis ~~... 350 G Galeodia funiculosa —_. 348 Galeodia [Morio} petersoni Gaspé limestone Gaspé sandstone 25 348, 359 55 3 44,53, 58, 63, 65, 78 Gastridium vetustum 26 348, 359 Giron series 109 399 Glossus* flosuss = 23 345 Ge ITIAETNS, | pe 345 Goldring, W., on Mackenzie River Basin crinoids _- ial Goniophora cf. carinata _. 74 perangulata var. —-... 55 Cia LUSOS ay ee ee 74 bechiyist ese ae eee 70 Grammysia canadensis 70 Grande Gréve limestones 3 44,50, 55, 78, 118, 209 Graphiocrinus carbonarius 372 SHEVONEOPGINS TS es 28) -s63 ane Gratelupia Hydii + 7 343 Griffen Cove River beds 44,49, 51 Griffen Cove River section 51 Gyvichnites gaspensis 69 H Hartt and Rathbun, Para Devonian __ ewe: 284 Hederella blainvillii 69, 90 Istallnoyolayhwyod 74 Heppel formation _ 78 Hexacriniws) hum ei. 13 Holopea gaspesia ____ 65, 71 AVM oN 71 Homalonotus, sp, Hydreionocrinus dewey- 20 113,281 EIST Spee saree Nile Se Py) Breve, Bs il) woodianus bar 364 Hydriocrinus pusillus” aa 377 Heolithuismetanachic: =e 71 IBAVSUIENC AUIS 69 I Icthyoecrinus __. Dall Infundibulum trochiformis 350 Ingram, W. M., Cypreas from Florida and Costa. Rica 321 Cypreide of Dominican Republic eee ee iad ee 329 Ischadites cf. squamifer 65, 69 VACKSIOM ROU) 2a 343 K Kindle, E. M., on Devonian faunas of Gaspé _._ 40 L Laganum Rogersi —...... 344 PGayp Parle: ele aL nee 346 Lecythiocrinus adamsi _ 29 364,379 ollicusformis 29 364, 378 TINGE OSNNIS ees 29 364, 379 dbrayoley aera Ut A eee era 351 brevirostris 70 colatauieers seme 843 multilineata: lolli 23 344 Leptena boyaca Ue 1K) ata, als) GOSH, ee i 135 ONIN, se 185, 1386 INDEX, VOL 2} 390 TN@snidoniolelicy 120, 124 rhomboidalis ventricosta 125 SHES O MEIN See ates ee 139, 148 ATDEOIOOMUIS see ee 47 Lepvoccelia flabellites 64, 69, 71, 244, 282 Leptodesma cf. rogersi _— 74 epuostrophiay == 129, 174, 195 aSSellan or Sen eee 182 eC katie eee er ees 176 Jovlianbaygullit oe 64, 69, 175 Ca sypiiratt ieee oe 181 GAMER aba es. Naeem 129, 181 CATH OOO NEY 124 exsplamiatae eae 129, 180, 197, 283 ULM arco ete ce, San eioeieAee 129, 174. 180 TONEKeOUNGOE) 185, 195 MAMI OUNCE, Ge 197 MNS vl eNe), 184, 194 onisikeanian =o 2S 124 POSTPONING) A 129, 175, 178 (pel GbreNey sn 181 HOCIUOMENESIS 186 TOC MFONAGISINE NE) Les 181 TALL he Sacer 185 ISSO COS EOI O aulisaN 128, 173, 174 Licnograptus elegans _____ 47 Ibghoney Swevomain@g 2 350 Limoptera macroptera —_ 65, 70 Lingulapholis floreste __ G UNG ILMAOCMAINUWIG! — a eee ZY kindlei bee An pee lee et my, NAL 217, 28) Lithophaga Carolinaensis 350 @latibiomnien'sisies = 350 Lucina perlevis ___ ee 350 Lingulapholis terminalis _ SL, Littorina rudis greenlan- dica eR Sea weds eb es 61 Loripes eburnea 350 turgida ieee Wade 350 Lunulicardium 2. convex- BLO Myce Uae Nee ee new vl Pe 70 M IWIN) sewn EY 350 Mississippiensis : 350 Wiageubiing), t 5 IMicl Ganiici me emen ene ae 128, 131, 165 Megalanteris suessi ___ 273 Gig AWGN 64 IMLS pere MeN, | Ee 273 australis ____ 15, 19 “TB, 273, 274 ChiIOlnOATAS * so 274 neozelanica 274 OV ANS eer EE eee 274 al bale bye eet ene 274 Megastrophia — 128, 130, 135, 137, 142 CON CAV ae eee ean 145 hemispherica 137, 145 la@plabash ek MS I) 112, 142, 145 pygmea 9,11, 12,14 112, 145° 391 INDEX, Megistocrinus depressus _ 20 Melbacrinus americanus 20 304, 37% Melocrinus bainbridgensis 18-20 borealis Ss SSeS 11,15 breviradiatus, 22-— —__ 14 Canadensis, —— <== =e ills az Clan ele eee ee = aa 14,19 Bienos)lyiphicus) 2 == 12 LAD AC i ee te 12519 Leite ii teen Ses ee 12 Sqoranavol i ee ee 12 MAC KenZilei see 12), TUS) Cyoloblispeeieeqeyss 2 al alloe alts} SUC OSUG UN aa it abeally ENTS US yee ee eee iil whittakeri 12, 15-19 Melongena crassicornuta 3850 Mendez-Alzola, on Devonian of Uruguay — 282 Mere bisbx ee 2s ee 351 wKUOtea = 343 IMMIGAD Se — === 343 EO a 359 TSI) oD) E11) 6 (2 le aa eet Beene Ce ee 343 Memnista dlisevailsy = 269 Meristella 269 alnvziaayollaitien ee Briel hoskinsi 270 Tl ceitceay sa ae i Seen aes paral MUSE ee ee 271 DIE WMICC S| eee a a Pat ERIS OWIS Kevin eee es eee 270 septata aie eee eee 270 wieder 2. es ee I aig Ass Michelinocerds — —2.2.-—— yal Mitra cellulifera — 350 conquesta feonquisita | 347, 350 Cong uUUsita = 347 dumosa 24 346, 359 Georgiana : 350 mellingtoni 'millingtoni | 26___347, 359 Mississippiensis - 347, 350 staminea : 350 teretretormis ===. — 350 WOiMGOIY 2 4 Modiella modiola —-...-- 70 pygmea = 2 64, 70 Modiola Mississippiensis 350 Modiomorphia inornata —- 70 mytiloides ee Ee 70 it, Rilolieyed, 2 = 74 Morio 2 851 Moundocrinus ‘osagensis 27 363, 369, 370 Murchisonia egregia —_--- 71 Murex Mississippiensis —- 350 Muscle diaphragms ------ 189, 196 Mytilarca cf. nivide eet 70 Narica Mississippiensis —--. 350 VOL. 24 Natica Mississippiensis —_ pPermiund a) Vicksburcensis: === Navicula ASD Chae eee Nemagraptus gracilis FreKohis. Graig) l- 2. gracilis linearis Neospirifer Nervostrophia Normanskill age — _ ~ North Fork River section. 26 Nuclearia gabbiana 22 Nucleolites Lyelli Wikoeeoy svi eee ee Nucleospira ventricosa Nucula Claibornensis — corbuliformis improcera parilis _ (eee oS Gis randalli- serica =e V icksburgensis a Nuculites oblongatus —— : ef. oblongata - triquetrus (0) Olsson, A. A, and Caster, K. E., on Colombian stratigraphy —.-—- a Omisciaehninpula ses Orbiculoidea montis ~ Orbitulites Mantelli Orthis) concinna = Sli ani ee Orthotetide aS Orthothetes becraftensis. os chemungensis Sulivani =. — pte Osteodes irroratus Ostrea Georgiana selleformis trigonalis Vicksburgensis —- Ip Palzeoneilo cf. constricta MVAKIN), cee ee muta planaee = a Paleopinna flabellum Palmer, K. V. W., on nomen- clature of Eocene Mol- lusea ee Panopea oblongata nee Papillina : Mississippiensis Paracyclas lirata cf. tenuis =e Paradental plates Parasaplocrinus 391 350 346, 359 129,174, 179 47 53 329, 337 350 4 26 348 74 70 171, 176 371 392 INDEX, DAE. Sioiebi@re “Ws 248 SD emer Se ae ne ee 18 267 Pecten elixatus pees Tt 350 MUON we: | 8} aD To wUlls@ri” Se 343 Pectunculus arctatus ____ 35U Mississippiansis 35) stamineus ate 343 Pentadelocrinus typus 2 A BSS Ul JPXSFONERUERONONE) oe 271 gemmisuleata _____.___ UG} iss ay Phacops ciasbutowrana 66 TAT Aileen ee ee alert Git, Saleen 2 20S walsee2 ul Rholas tang uetra ean 350 TEAROUNGOOS 9 ee TA. 116 an efev ele renltzh, Numa Denia emacs e 118 areolata 116 bellula 118 flomesitees He aes etl Ge sat, aS Osetia Seek 118 Fholidostrophia 198, 202 IPinorewiss Ioysremiibls a 359 THVT 2 ee 26 348.359 Phricodothyriins 245 Phricodothyris lucerna _- 245 Phthonia cylindrica _.._. z 70 Pika, EWAN, 350 IPDIBIANV CCIE Li 66 (REVS oS MKey ie i7all RlAtyOpteray sess sels 345 Pleurotoma abundans 350 @@@mlegirig Lo 350 COMBOSUE), 359 GHEE eee 350 Gece panes ae 350 GOORONCIS 2 350 Mississippiensis 350 | porcellana 350 Tobe dens: ye ee eee 350 GevmiG ull aes oe ea ae 350 bean eileen screint a Rceam. 0 ia 350 Senvyaitasmpeeees z 350 Pleurotomaria sulcomargin- atamleclencaimesssananen 70 Point bevis fauna 2 47 Polypora orientalis 5,6 65, 69, 74, 90, 92 LXATMMAOG IMME 1 25 TOO WIS. ee py IAL AS IPSOCOSS, Tomes ale IPreo@lmoweilley, 2290 Spinuiliicostaye samen 163 L183. BB Productus lamellosus 243 TSU 20'S ey ae reer 119 subaculeatus __ ay 22) Proetus cf, protuberans a 55 Protoleptostrophia fa 129, 174 Protomegastrophia 128, 130,135, 136 Psammobia papyria — 350 Mississippiensis 350 Piteriniea) ee meee 7, 113, 279 WOK. BA 392 Rteny cobs se ereeen ee 71 Ptychoglyptus 140 Pusiuiaria nucleus 337 R Nafinisquina miinsteri 140 ringerikensis 140 ? sehmidti 140 Rafinesquinide 119 Rensselewria —_ 2a 274 ovoides gaspensis. Sen eeoe 64, 70 Reticularia 121 reticulata 245 Reticulariine é, 244 Rhipidomella musculosa _ 213 vanuxemi var. _ apres 74 Rhopalithes — . 5 Rhynchospira ef. globosa | 54 Rhytistrophia 2 Oras Age: 186 beckil tennesseensis 194 caribbeana é 194, 283 caribbeana colombia VT, WP ef 2 ay, UBT, BSS} Ringicula Mississippiensis 350 Rostellaria extenta 343, 359 Lamarckii 346 staminea ee FO BES velata ee 24 346, 359 Ss SASGICENVE) WHIEOSE, soot! 41 Sealaria nassula {nassuta | 26 347, 359 trigintanaria 350 Schellwienella 211 agassizi 169, 2138 baini sine 216 goldringe 2 ey a4b 1Bs BIG goldringe juvens Zales salals}. 21%) sulivani ae 169, 212 Schizodus appressus __ t 70 Soldwelacmrewle, 211 SEEN Aliya Ate aihele ee er le 216, 282 baini 216 becraftensis 54, 64 sancticrucis 212 woolworthana gaspensis 219 Scutella Jonesi 344 Seyphocrinus 51,54 Schaleria 128, 181. 133 Shirley, on Devonian of New Zealand 282 Sigaretus Mississippiensis 350 “olarium triliratum 350 Sonneau Brook beds 44,78, 89 Sphenotus truncatus 70 Spinocyrtia 262, 267 Spinocyrtia cf. valenteana 19 113; 268 Spirifer antarcticus 248, 259.262 aristookensis 258 - audaculus var. 74 audaculus zulianus 266 INDEX, Vora 24 393 buarqueanus ~ 269 cameraius 250 carinatus 264 cnuquisaca 260 contrarius 260 crenistriata Paw cultvijugatus 207 cyclopterus 2a4 disparilis 243 divaricatus 248, 249 duodenarius 267 gaspensis 254 hartti Zoo hawkinsi 260 hercyniz 2o4 iheringi 260, 282 eG Clete eee ee 264 kayserianus . 2425 249,257, Zoy kingi 14, 1G: 8) 3), 251 lauro-sodreanus —_._. | 260 marcyi Sei 260, 262 medialis ___ uf ee A 66 meridioamericanus 247 murchisoni 254, 258 @relieaNiay Se es ee 259 parana 260 pedroanus 260, 265 pennatus posterus nhs 74, 76 planoconvexus) ===> = 242 Carb uighy ee en et 243 [primes ys 253 raricostus 254 SGPT AU US pen. eee ons 250 Vihent@ana, 2 268 \WEinws Riis 198, 203, 206 SD = 16 170 stephani Ya Se 1538, 283 Stropheodontide rea 126 Stropheodontine — 127 Stropheodontini 128, 130 Strophomena bainj _____ 215 demissaten- Se ae 165 geniculatum See 202 Jol ai 2h Speen Ae as Oe 169 inequistriata ___. 128, 171 ed age ese. 2 ee Se 133 ornatella : 133 [DGUCING ae ee 200 SGuylevbay es lea Senco 198,200 Strophonella 2225s 198, 203, 207 ampla 198, 200, 209 continens ____ _. 198, 200, 205 MOWeT ee ee SEG 118.209 lonydopencle, Leena 206 hybrida pondero: osa 206 Jjamesonie = : 199, 204 leavenworthana — 198, 199, 204 meridionalis —_ 10,14 112,207 Silica bela ee ae eee ee 134 williamsi —_ 198, 200, 204 Strophonelline 127, 128, 198, 199 Strophonelloides 198, 200, 208, 206 Strophoprion 199, 200 St. Stephens limestone Suleatiostr ophia 343 129, 174, 181, 206 Suleoretepora cf. incisurata 74, 90 Synaptocrinus nuntius Zi (2?) rotundatus 1 14, 20 T Teniopora exigua —-..... 74,90 pennifonmis 74, 90 Taonurus 50 Tellina lintea 350 pectorosa 350 perovata 350 serica : 350 Vicksburgensis = : 350 Tentaculites cf. bellulus 75 Carter 64, 69 394 Terebra diversum weNnwOUley) Terebratula archiaci amibricava mee E Tetragraptus similis MAEMO KATANOUS © sam Oca) en ah Tribrachiocrinus Triton abbreviatus crassidens Mississippiensis subalveatum Trochilea Trochita alta : Propidocarilsy belli 2 —= Tropidocyclas brevilineatus rotalinea See ak, Turbinella Weill Oni ye eee eS Turbinellus protracta _-_- Wilsoni Turbinolia caulifera ___.__ eyanthus lumuilitikorms) ee Turbinolopsis INDEX, 351 345, 359 Vor. 24 Toermicilla ailkyegie, 2... obruta lineata eee ees Mississippiensis — U Umbrella planulata 23 Uncinulus globulus Vi Venus mercenaria Wielkslomias syeotiy 2 Wiheniing, S15. 18}, pustulosa ey sin Volutalithes dumosa . Sayana SAAMI), | WwW Wachsmuthicrinus —._.. : Whitfieldella cf, minuta Y York River beds Zeacrinus 3.13, 345, 359 25 349 349 349 35) 54 44, 62,78 25 Cet! 7 : f U , rt ee ANS Gaal »> yo» 33 > >» D2 »») yy» 297 BS DS D » » yD» >» PD DD® SS DLDZ_P IP ZS DIP Inds DD 2D») 2» Sy > 22) > Sp >> >D) » >» > »») Jy») BW) HD » >> >> y» >» > D 2» »)»y yy» > D) > IP Dy y SSSR > 2 yy» 2» _¥ DS» Iam py ) » FD D5 IW YI D> >>>’ -¥ D yy ) D> » YD »> DY ) I) DDS Iw yy > » I> IBS DI bd) DP)» » D> DY » PI) D) yy; y Dy») dS DY DIIWIID) “> Dy yyy D2 eS DY >)D II Ip 2)» yy yy >» »» d PY » »> TDPIP ID IND Sy PD» DI DDID Wy yy» wy» DD DD» yy» yyw» > — »D> » >» > 23 Seas D> ) yD» Dy) D> y D»s Dy) )) yD » >> yy » )) D» > D) D) vw ») yy Dy) > yy »» > D> yy »> > 3 UR ee y yp) i ) Dy yy» yD DS »») Dy »> » Dy DS »y » ps D : >» » yD» yy ) »» 3 >») D ) »D)) yD Dy) >») ») Dy * 3 >> ee » 22 a Py 3p I> Iw 33> yD 23 ey» DI By SS By DDT ) ») > ) » Sy ) Dy > > »»yY D5 >») » » DVD DI) » IID Ea >> >s 2 » ID —Way>d > J DDD —DPYPS » » ry DDD I 3 » rw DDD » >wy DY >» IBY DIDI 2)» WD yy> a » wy >>)» 2 >} IBD Ia p » J fe ») DD») p> »D» D DD dy > Ip pp ) yD DD _y ») » a > > pv ] _ ILD VT DD 2 DE aes 2 >. >> > >- 2) WD yy > >> D> Die eee >) 3 $3 sii eS >» > 2 23>») Wy» . > > »> » D> > >> »»>>) >» > >. Ps D> DIB Dy») » SS > . ID? >) dB DP > IP > 2» >» So >>D> a> yD > » »>d »>) _»> > >>» >» >>: JID YD) ED SDD >> > DD) > Dy EP res >> 35 BP ee D>) : ) >> >> > » y D> »») $ >) . 353 =, 222 a gD» »> | ») > >. ? yp 2 >) 323» DD 92. 2 >> > » yy >> 22>» >> 2 D» DD» > Pa BPW 2> > » ») > yD) > >» ee 2 >>> J ») ») » Fi p> iB? wD >> 2 > 33 b> >> 4) PD» »» Dr 333 > SD DDD» » >. 2» >» > 3 Sees 223 ee : > >>> ) - > > >>» Pe 3 2s Pp». »» >» » Dp) D) » » »»» Dp». Be > - » >> > > ) Dp» >: ) » 2 » ) D»»» yy y b>» » y 2» a) >) > BDP> > . 2 5 4 >)». > ») yy > » DP ») > » >)» ») Dp ») >») > >» Dp) 2 ) > »)) 2 | D yp» 2? Sp 3 D »» > » 2 ) D2 Ss > > > ID» > > » 2» »-s > D > > > > > > > »)) >> DP J P 2 > » » y >» yp » » DD 3 Sp > PD! 2» 52> Dt» >) >» D> » se )>_D yy > > i | | 3