So ST Rr ete se ar if ur tae er ed er re er HF OF OPO b- 3 <- ioe -5~8 Seiten eeeaters ses pnnigesiites mesenger apa ; ere otra anne eee paceraces . erst peteescscectcen raze = ae Parasites psticeepeesesen paatiessztores poxerey st artsy tare re seagencsstesee i323 i= oes : 3is pocteecese 2 ie sesserecisetacesite spesasssagreace, 33 2 peenets ae og te te booms Sete Bete a Cpa ee Pests estate eee ate Oe cab peirarts ADitieietat stot ese Tires ites et erat Teistrists Sates Sixissceseesrer isis: ae Seisssieretes Srstrisseres Srsstatesssastssstssseeee roses spestessssecesstass Rise -ssttact ae ag oe neat SSS ITF Sr sreeereessresastesstasen 33 sastseesstesesessssetass opt ch serrate eect erecetetetottesesarseecseasSosess STIs pererttsrtisteree stress $SeSeesSe Se Sess i : ‘ es é - — E 7 7 : f 7 we 1 - 7 — BULLETIN OF ihe BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY VOle od 1965—1966 TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) LONDON: 1967 DATES OF PUBLICATION OF THE PARTS No. i. oth July 5 : . : . 1965 No. 2. 7thSeptember . : 5 : . 1965 No. 3. 8th December . 0 : c - 1965 No. 4. 8th December . : : é - 1965 No. 5. 17th December . : : 0 . 1965 No. 6. 7th January : ; : ¢ . 1966 No. 7. 11th January : : . : . 1966 No. 8. 21st January : c 0 0 . 1966 No. 9. 28th January ‘ 6 : 0 . 1966 No. to. 19th April . ; : “ a . 1966 PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED BARTHOLOMEW PRESS, DORKING No. I No. 2 No. 3 No. 4 No. 5. No. 6. INO: 7: No. 8. No. 9 No. Io. CON TENG s GEOLOGY VOLUME XI The head of Dzupterus valenciennest Sedgwick & Murchison. IS. le \Waeouns Some Lower Cretaceous Terebratelloidea. E. F. OWEN Middle Jurassic Ostracoda from the Grey Limestone Series, Yorkshire. R. H. BATE Human skeletal material from Ceylon, with an analysis of the Island’s Prehistoric and Contemporary Populations. K. A. R. KENNEDY Some new British Albian Ostracoda. P. KAYE Tertiary Red Algae from Borneo. J. HARLAN JOHNSON British Wealden Sharks. C. PATTERSON On certain Triassic and Liassic representatives of the family Pholidophoridae s. str. O. NYBELIN Some British Jurassic and Cretaceous Ostracoda. FF. W. ANDERSON & D. BARKER Trilobites of the Henllan Ash, Arenig Series, Merioneth. H. B. WHITTINGTON Index to Volume XI PAGE 433 489 507 EEE eA) OF DIPTERUS VALENCIENNESI Sedgwick & Murchison ERROL IVOR WHITE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 1 LONDON : 1965 THE HEAD OF DIPTERUS VALENCIENNESI Sedgwick & Murchison CSS man ee ge i) JUL ry xy CL; ~S Tay |- (re Soc \7 16 Fics. 14-17. Dipterus valenciennest S. & M. Fig. 14, Skull-roof in which the sensory canals have been wholly dissected out on the left side, partly on the right. Thurso, Caithness. 33165. 1:3. mill, main lateral line. For other lettering see Fig. 15. Fig. 15, Skull-roof in which the sensory canals of the snout have been naturally weathered out and some of the remainder dissected out. ‘“‘ Caithness”. P.6087. x2. ie, infraorbital canal. voc’, rostralcommissure. soc’, anterior aperture of the supraorbital canal. Fig. 16, Median part of upper lip showing both external apertures of the supra- orbital canal (soc’), with part of rostral commissure (voc’) and supraorbital canal itself and branches leading to external pores (ep) exposed by a natural longitudinal fracture. Thurso, Caithness. 33166. 2. Fig. 17, Skull-roof with part of canals on snout naturally weathered out. Foreshore, Clardon Haven, Caithness. P.34546. (Colld. D. L. Dineley, 1957). 1-3. ifc, infraorbital canal. soc, supraorbital canal. LAE HEAD OF DIPTERUS VALENCIENNESTI 15 The main canal runs through Z, in which it gives off the occipital commissure, first to the hinder end of I, and then through A. From Z the main canal normally passes through Y, and Y,, to X, in which it turns down to pass through circumorbital 4 to form the infraorbital canal, at the same time sending a short branch to K to anastomose with the supraorbital canal. The infraorbital canal runs through the circumorbital plates 5, 6, 7, I,, I, up on to the snout, crossing the antero-lateral corners, as may be seen in most hard-snouted specimens. It may finish in a pore under the snout like the supraorbital canal, but if so the pore is often very small, and not always to be distinguished easily from the others present in this region; for the whole canal under pressure tends very readily to dissolve in the system of pores and spaces which permeate the bone there. To give a general impression of the canal-system, the details from Text-figs. 14, 15 etc. have been superimposed on the left side of the outline of a well known, more complete skull of “ platycephalus”’ type (P.7834, Text-fig. 18) to which plates Z—A-Z have been added and, in fig. A (P.6507), the upper lip of another specimen. The skull-roof has been figured by three other authors (Jarvik 1950, text-fig. 6; Gross 1956, text-fig. 133B; Orvig 1961, text-fig. 76), each giving a somewhat different interpretation of the plates, and all different from that given here. The right side shows the actual distribution of the pores in this specimen. From this general pattern there are a number of significant deviations, as there are from the standard patterns of the head-plates, and since they are interdependent, sensory canals and head-plates will be considered together, in the light of Parrington’s axioms (p. Ir swpra) and of his theory of “ Patterns of dermal bones ”’ (1956). It is in the lateral line series, Z—X and K—P that most difficulties arise, but there are four plates and areas that have sufficiently constant features to make them readily recognizable, Z and Y, at the back of the skull-roof and N and P near the front. Z, formerly H, is the hindmost plate and always roughly triangular. It receives the main-line canal from the supracleithrum and passes it on to Y, after the canal has been joined by the occipital commissure. Only once in the series available is it in any way abnormal. Y, always occupies the same position, just overlapping the front of the operculum, and so having a characteristic projecting angulation on the outer side. On the mesial side it lies alongside the hinder part of J and a small part of I,. Very rarely is it replaced by, or fused with, X (Text-fig. 25; Westoll 1949, text-fig. 2B). N is the plate, often fused with the snout, where the supraorbital canal turns outwards at the start of the sigmoid curve in front, and P (Westoll’s “Q”’) is the plate or area in which the canal turns forward again. A survey of the available material will make clear the advantage of a purely alphabetical system of bone nomenclature. The first specimen to be considered must obviously be Forster-Cooper’s standard specimen (P.22187, Text-figs. 11-13). Fig. 12 shows the actual specimen as it now is, partly dissected. It also shows the original lettering except that G and M are omitted, G because it was only a scale and M because its supposed presence was due to a crack in L. 16 APISWID, JSMIBINID) (OY PIS IBIS WV ZUILIB IN EM Ia IN NID Sth The sensory canals are perfectly orthodox, as may be seen in the revised diagram (Text-fig. 13), but the series of lateral line bones show two interesting deviations, apart from the absence of M. A single bone occupies the space not only of M but of both the ‘‘ L”’ bones and there is clearly only a single centre of growth, which from its central position is judged to be that of the middle bone of the three, L,. There is Fic. 18. Dipterus valenciennesi S. & M. Skull-roof of large head with plates Z—A-Z added, showing external pores of sensory canals on right side, and restoration of the canals on the left side, based on Figs. 14, 15. (cf. Jarvik 1950, text-fig. 6; Gross 1956, text-fig. 13B; @rvig 1961, text-fig. 7B). Coast near Thurso, Caithness. P.7834. x1-5. A, underside of snout of similar specimen. Toldale Quarry, between Thurso and Reay, Caithness. P.6507. 1-5. fc, infraorbital canal. mill, main lateral line. soc, supraorbital canal and aperture. AWISIIS, WED LNID) (OND IU ID IID IR (OLS WAIL IF IN (CIID IN INIT ye 17 also only a single Y bone, but this is judged to be compound, Y, + Yo, by reason of its shape and the distribution of the pores which seem to indicate two centres. In 33165 (Text-fig. 14) the two sides make a most interesting contrast. On the wholly dissected left side a large single bone in the M—L, area does seem to have had two centres and is accordingly labelled as a compound bone, and the large odd-shaped bone between L, and Y,, is almost certainly a compound K + X. On the partly dissected right side the L, to M region is similar to the left, but the K and X area is very different, for K is not only small but is excluded from contact with C by an extension of L, and has lost to X the beginning of the supraorbital canal where it passes out of J. This is most unusual, but the result can be simply explained a la Parrington, by delay in the development of K’s rudiment, which has allowed the capture of the beginning of the supraorbital canal by X and the invasion of this territory on the other side by 1. P.6087 (Text-fig. 15) shows the suppression of M on both sides, in favour of N on the right, and probably of L, on the left. On both sides the important K—X region is occupied by a single, not very large bone, with a single growth centre but containing both the turn-down of the main canal at the beginning of the infraorbital canal, typical of X, and the continuation of the supraorbital canal from J, typicalof K. Itis not clear which has been suppressed, but it is probably K, for L, on both sides is much larger than usual and extends further backwards into the K area to meet J, as it also does on the right side in the previous figure. In a short-headed specimen (P.34546, Text-fig. 17) two bones only occupy the M-L position, each with a single centre and those of the left side are only superficially separated. It is probable from their forward position that the anterior bone on each side represents M, but that the hinder bone on the left side, judging by the position of the growth-centre, seems to be Ly, whereas on the right L, is the more probable survivor, particularly as M is unusually large, as if it had partly replaced L,. The fine narrow head from near Thurso (P.7834, Text-fig. 18, p. 13 supra) is undissected, the sensory canals on the left side being diagrammatic. It is one of the longest heads examined, 6-9 cm. from the hinder margin of B to the tip of the snout, being only two millimetres or so shorter than the Toldale head (P.6507, Forster- Cooper 1937, pl. 7, fig. 14) and a specimen from Edinburgh (1859.33.622), but 8 mm. less than the record head in that Museum (1901.153.2) which measures 7-7cm. The head in Text-fig. 18 shows quite markedly contradictory tendencies, a tendency to the decrease in size of plates in the middle lateral regions (K, L,, L,) especially on the left side ; but in front there is a very obvious increase in size through the suppression of M, to the advantage of N only on the left side, but also for the benefit of L, on the right. A marked difference in development of the sensory canal plates of the two sides is illustrated by a Clardon Haven specimen (P.46690, Text-fig. 19). Of considerable interest in showing the relationship between the distribution of pores and the actual course of the canals is the specimen (P.22189) on which Westoll (Graham-Smith & Westoll 1937 : 244, text-fig. 2a ; Westoll 1943 : 89, text-fig. 7c ; 1949 : 126, text-fig. 1) based his modified scheme of lettering. In the original somewhat diagrammatic and restored drawings X and K are shown as a compound plate. The supraorbital canal entered this plate from J on its pos- 2 18 DHE HEAD OR D7 PERU S \VcAEE NiGIEENUNG ES terior-inner side, and almost opposite this the main canal is supposed to have come in from the anterior Y plate and then turn down into circumorbital “‘ 4’ to form the infraorbital canal. In fact dissection (Text-fig. 20) shows that the main canal never reaches the ““ X and K”’ bone but turns down in Y, (Y, of Westoll), and that the connection between the main and supraorbital canals is not only drawn out but apparently double. Examination of the structure of this supposedly compound plate shows clearly enough that it had a single growth-centre; X has failed to Fic. 19. Dipterus valenciennesi S. & M. Skull-roof showing asymmetrical development of lateral line plates. Clardon Haven, Caithness. P. 46690 (Colld. R. V. Collier, 1964). X 2:25. THE HEAD OF DIPTERUS VALENCIENNESI 19 develop and its space has been largely taken by K, but the lateral line has been largely captured by Y,. The L plate is also shown as a compound L, + L,, but this has a single centre and seems to be composed of L, only which has taken over the territory of a missing L,. The pores in the very small head illustrated in Text-fig. 41 (P.22194) seem to indicate a similar situation in regard to the XK area, and here also the area of M has apparently been taken over by Ly. The capture of the infraorbital curve by Y,, is even more clearly demonstrated in Text-fig. 21 in a very abnormal short-headed fossil from Achanarras (P.17642, cf. Fic. 20. Dipterus valenciennesi S. & M. Left side of skull-roof and cheek of original of Westoll’s standard specimen (see Graham-Smith & Westoll 1937: 244, text-fig. 2a; Westoll 1949 : 126, text-figs. 1A, C) with sensory canals partly dissected out and plates partly re-lettered. Achanarras, Caithness. P.22189. X3°5. 20 APISEID, IRMBINID) (OMI IMI IZ IOS WY ZUILIZIN (E1013 IIE SIL Forster-Cooper 1937, pl. 5, fig. 9). Plate X apparently did not develop, to judge from single-centred K, but there seems to have been some opposition by K to the annexation of the main canal by Y,, judging by its subdivision. This fragmentation of canals owing to disturbance by fusion or elimination of plates is probably not unusual (Text-figs. 22-24). In 42403 (Text-fig. 22), a much resorbed specimen from near Wick, K and X do seem to have fused to show what appear to be double centres of growth, but nevertheless the start of the infraorbital canal has gone to Y. In Text-fig. 21 another most interesting abnormality is illustrated. The supra- occipital commissure shows much disturbance owing to the non-appearance of Yog, but instead of its space being occupied by Y,, that bone is of the average size, and Z has grown forwards taking with it the outer end of the commissure, which crosses I, much further forward than usual. In contrast, the occipital cross-commissure in two other fishes from Achanarras (P.17641, P.17643, Text-figs. 23, 24), instead of its usual path Z—I,—-A-I,-Z, misses I, altogether, although these plates seem normally developed. In P.17641 (Text-fig. 23) K and X seem undoubtedly to have fused on the left side, but on the right K has been eliminated to the advantage of L,, which is rather unusual (but see P.6087, Text-fig. 15). Generally K is a better “ stayer”’ than X, and it is worth noting that its division from J is often only superficial, the basal layers being completely fused (cf. Text-figs. 15, 22 etc.). On both sides the start of the infra- orbital canal has been largely captured by Y, and only a small subdivision of it is retained by X. There is some doubt whether the missing plate on the right side is K or X, but the arrangement is partly paralleled in Text-fig. 15 where there was similar expansion of L, into the K area and disarrangement of the canals. L, and L, in P.17643 (Text- fig. 24) show a very obvious case of imperfect fusion. Of the two “ brachypygopterus”’ types from Banniskirk illustrated (42480, R.S.M. 1859.33.624, Text-figs. 5, 6), the first shows a fairly obvious X and K fusion and the L,-L,, M and possibly N area is occupied by a single plate. The second shows a large N and on the left side only one small plate (labelled L,) between N and K, also a large Y,; andno Y,. The naturally exposed sensory canals are standard as far as they may be traced. Attempts to Identify the Head-plates The difficulty of tracing the canals by the pores led Stensi6 (¢n Holmgren & Stensi6 1936 : 366, text-fig. 280A—C) to misplace the connection between the supraorbital canal and the main canal in the three rather difficult Edinburgh skulls on which he based his nomenclature (Text-figs. 25-27). Stensio’s choice of a key-specimen (R.S.M. 1878.5.164) on which to base his identifications was not altogether fortunate, for it suffers very much from multiplica- tion of plates in the central area and in the front lateral line series. His figure (Text- fig. 28), was formalized and partly restored, cf. Text-figs. 25, 29. Neither of Stensié’s supraorbital bones (SO,, SO,) belongs to the circumorbital series, but his “‘ dermosphenotic + postorbital”’ (Dspf + Po) does, and the orbit is farther out and certainly smaller. This bone is not compound but is the circum- orbital 3 of Forster-Cooper’s notation. In the original specimen it is preserved only THE HEAD OF DIPTERUS VALENCIENNESI 4 2| Figs. 21-24. Dipterus valenciennesiS.& M. Fig. 21, Left posterior part of skull-roof and cheek with sensory canals dissected out (cf. Forster-Cooper 1937, pl. 5, fig.9). _Achanarras, Caithness. P.17642. %*x<4. Cl, cleithrum. Op, operculum. Ovb, orbit. Sc, scale (Forster-Cooper’s “G’’). SCJ, supracleithrum. s.com, occipital cross-commissure. Fig. 22, Posterior half of much resorbed skull-roof with sensory canals partly dissected out to show probable fusion of plates K and X on each side. Killimster, near Wick, Caithness. 42403. 4. Fig. 23, Skull-roof with sensory canals dissected out (cf. Forster-Cooper 1937, pl. 6, fig. 11 ; Westoll 1949, text-fig. 2E). Achanarras, Caithness. P.1764I. X2°5. s.com, occipital cross-commissure. Fig. 24, Right posterior part of skull-roof with sensory canals dissected out (cf. Forster-Cooper 1937, pl. 4, fig. 8). Achanarras, Caithness. P.17643. x2. ifc, infraorbital canal. LLSc, lateral line scale, Sc, body scale, SC/, supracleithrum, s,com, occipital cross-commissure, 21 22 WDISUID, ISIIB/ANID) (QUIN IVI MITICIU SW ANIL IB INC IE IZ IN INTE, SIE Fics. 25-27. Dipterus valenciennesi S. & M. Camera lucida drawings of originals of skull- roofs of “ platycephalus ”’ type, sketched by Stensid (in Holmgren & Stensi6 1936 : 366, text-figs. 280A-C). Fig. 25, Large head (6-1 cm. long) with median mosaic and unusually numerous plates (6) between K and P. Locality unknown. R.S.M. 1878.5.164. X1-3. Fig. 26, Medium sized head (4-6 cm. long) with anterior mosaic and only three plates between K and P on right side. Firth, Orkney. R.S.M. 1898.163.6. X1-3. Fig. 27, Large head (6-3 cm. long) showing tendency to subdivision of plates between K and P, Locality unknown. R,S,M. 1859.33.32. 1-3. DHE HEAD OF DIPTERUS VALENCIENNEST 23 on the left side and carries an unusually large branch of the supraorbital canal, for as a rule the pores are few, often absent. This skull-roof, in spite of its oddities, is reasonably reduced to order under Parringtonian influence with Forster-Cooper’s amended lettering (Text-fig. 25). The only point of real importance is that Yj, normal on the left side, has most exceptionally not developed at all on the right side, where its place has been covered very largely by a much expanded X, and only to a small degree by the advance of Y,. Both of Stensié’s other two specimens show peculiar features. The small broad head (R.S.M. 1898.163.6, Text-fig. 26) shows division of J on the left, and no M and a possible fusion of K and X on the right; the larger head (R.S.M. 1859.33.32, Text-fig. 27) shows fusion of K and X on the left side, fusion of L, and L, and im- perfect subdivision of N on the right. Lehman (1959) in his work on the Upper Devonian Dipnoans from Greenland followed Stensi6 in endeavouring to name the plates of the skull-roof, although by no means agreeing on the terms to be used, nor for that matter on the homologies when he did. Comparison of Text-figs. 28-30 well illustrates the confusion that can be caused by efforts to identify these head-plates with the bones of other fishes. Stensi6’s supposedly compound “‘ dermosphenotic and postorbital’”’ (Dsph + Po, Text-fig. 28) is not the homologue even in part of Lehman’s “ Dermosphénotique ”’ (Dsph, Text-fig. 30), as a glance at Text-fig. 29 will show, for Stensi0’s plate is a simple circumorbital, no. 3 of Forster-Cooper. Lehman’s “ Dermosphénotique ”’ is a single- centred plate (Lehman 1959, pls. 1-14, 16) and represents the survivor of K or X, probably K, which has taken over the other’s territory and acquired all the canals (cf. Text-figs. 19, 20, 24). It is therefore strictly homologous with the most posterior member of Stensid’s “ laterale Frontalserie” (Fy. l, Text-fig. 28) of which Lehman’s “ Nasaux ”’ are the front members. Again, Stensi6’s supposedly compound “ Supratemporo-Intertemporale ”’ (It + St, Text-fig. 28) is also a simple bone, the X of Text-figs. 25, 29, which has invaded Y,’s territory, and Y,, the equivalent of Lehman’s “ Dermoptérotique antérieur ”’ (Dpt. 1) has most exceptionally not developed at all. But Lehman’s “ Dermo- ptérotique postérieur”’ (Dt. 2) is the Y, of Text-fig. 29, and thus the homologue of Stensi6o’s “laterale freie Extraskapularplatte”’ (Ext. 1,, Text-fig. 28). This last plate, which is laterally imperfect in the original (Text-fig. 25) does not show any part of the supratemporal cross-commissure as indicated by Stensid (Text-fig. 28) ; it would have been present on the plate behind (Z), which is missing. Stensi6’s “ hintere laterale Parietalplatte + einer lateralen Extraskapularplatte ”’ (Pa. 1, + Ext. l,, Text-fig. 28) which is identical with Lehman’s “ pariétolatero- extrascapulaire ’’ (PalExSc, Text-fig. 30; Lehman, 1959: 18, pls. 1-21, text-figs. 2, 4, 5, 7-13, 21-23, 26), presents a very different problem, and indeed provides the key to the homologies of the plates of the earlier Dipnoi. These paired plates are the I, of Text-fig. 29 and show in the available material of Dipterus and Lehman’s illustrations a single growth centre; nevertheless the suggestion that there is an extrascapular element is supported by Westoll (1947 : 134, text-fig. 37) who states that “‘ it is likely that at least bone I also arose from more than one rudiment since a few specimens show ossicles enclosing the appropriate section of the canal”’. And 24 THE HEAD OF DIPTERUS VALENCIENNES Fic. 28. Dipteyus valenciennest S. & M. Right side of Stensid’s diagram, text-fig. 280A (see Fig. 25), with original lettering. Dsph + Po, Dermosphenotiko-Postorbitale; Ext. 1,, laterale freie Extraskapular- platte ; Fyr.c, zentrale Frontalserie ; Fyv./, laterale Frontalserie; Fr.m,, Fr.m,, Platten der medialen Frontalserie ; Jt, Intertemporale ; /¢ + St, Supratemporo-Intertemporale ; Pa.c, zentrales Parietale; Pa.l,, vordere Platte der lateralen Parietalserie; Pa.l/, + Ext.l,, laterale Parietalplatte + einer lateralen Extraskapularplatte; Pér,, hintere Postrostralia ; SO,, SO,, Platten der Supraorbitalserie (S.O. (=3) is omitted in Fig. 30) ; ap, vordere “ Pitlinie’’ ; zfc, Infraorbitalkanal ; mp, mittlere “‘ Pitlinie”’ ; pp, hintere “ Pitlinie ” ; s.com, Supratemporalkommissur ; soc, Supraorbitalkanal., A 1RHID, ISB ANID) (OQ) IDL IPINIIIGNOL SS WP IEIFIN( GIL IN IND St 25 29 : Ke 30 et, AVN a S28 -C o, & ns Fic. 29. Dipterus valenciennesi S. & M. Right side of same diagram (Fig. 28) partly redrawn from original specimen (cf. Fig. 25), with new interpretation of sensory canals and plates re-lettered according to modified FCAN. Other lettering as in Fig. 28. Fic. 30. Soederberghia groenlandica Lehman. Right side of diagram of skull-roof by Lehman (1959, text-fig. 2) showing original lettering (with broken indication lines) and also modified FCAN. Dpt,, dermoptérotique antérieur ; Dpt,, dermoptérotique postérieur; DSph, dermos- phénotique; vb, composant antérieur distinct du frontal latéral; fyi, frontal latéral; Fym, frontal médian ; Na,—Nag, nasaux; Orb, orbite; Pac, pariétal central ; Pal, pariétal latéral; PalExSc, pari¢to-latéroextrascapulaire ; Pz, plaque pinéale; Po, postorbitaux ; Pty, postrostral; So, supraorbital. Sensory canals as in Fig. 28, 26 ADISUIS, Teli ANID) ON IDMIDIDITICUES) |Y AUILIB IS CM IE INGN JES IL indeed the Belgian specimen of Rhinodipterus secans (White 1962 : 3, pl. 1, text-fig. 1) shows a similar “‘ extra’”’ plate enclosing the canal. Subdivision or proliferation of lateral line and general plates is not necessarily significant except that perhaps it points to the ancestral mosaic, and is of common occurrence in Dipterus, as the front of Stensié’s specimen shows. But in this instance it throws light on the presence of part of the commissure on I, which is anomalous, for it is clearly out of line. This points to the former presence of a plate on each side between A and Z, now lost, and that the section of the cross-commissure belonging to it has been captured by I,, that is, usually, for as noted above (p. 18, Text-figs. 23, 24) the commissure does sometimes go straight across through Z—A-Z, as it must have done when there were five plates in the “‘ extrascapular ”’ series, even as Save-Sdderbergh (1932 : 98, text- figs. 18, 20) postulated for his ““ common ancestor ’’ of the ichthyostegids and crossop- terygians ; but it was not the bone I, (Save-Séderbergh’s and Stensid’s Pa.l, + Ext.l,) that belonged to that row and originally bore the commissure, but a plate behind it, represented rarely by the little plate I,, noted by Westoll, and seen also in Rhinodipterus. The above correlations may be summarized as follows : Stensi6 1936 Forster-Cooper 1937 emend. Lehman 1959 (Text-fig. 28) (Text-fig. 29) (Text-fig. 30) Dsph + Po (CO) 3 SO (omitted) (Posterior) Fr. 1 K DSph Tt + St xX absent absent absent (Y,) Dpt, Ext. l, ve Dpt, Pal, + Ext. l, le PalExSc To trace back the homologues of the skull-plates of Dipterus in the earlier genera of dipnoans, the two species of Dipnorhynchus from the Lower Devonian (Siegenian)?, presents considerable difficulties. The roof-patterns of D. stissmilchi and D. lehmanni are reasonably well known, thanks to E. S. Hills (1941, pl. 9, text-fig. 5; 1943, text-fig. 1B), Westoll (1949 : 140-143, text-fig. 5) and Lehmann & Westoll (1952, pl. 24, fig. A, text-figs. 44, 5A). The most obvious points about the pattern of these skull-roofs (Text-figs. 31, 34) is the presence of a pineal foramen and the separateness of the supraorbital and infra- orbital sensory canals. The back row of plates, presumably the extrascapulars, were finally shown in D. lehmanmi (Text-fig. 31) to consist of two pairs of plates identified as H(Z), the smaller outside pair, and I, the large median pair, instead of the Z—A-Z arrangement of Dipterus. The occipital cross-commissure, apparently not seen in either specimen is assumed to run across H(Z)-I-I-(Z)H. There is no A plate but that immediately in front, a smallish median element, is considered to be B, and in front of that again a pair of larger plates completely separate in the German species, but partly attached to one another in the Australian. Westoll (1949 : 141) has resisted the temptation to label these as the “C ”’ pair, as at first sight would seem obvious, on the grounds of their posterior relationship to the orbits and then to bones of the X-K area. This pair is labelled B, and along side the B 1G, M, Philip & A. E. H, Pedder (1964 : 1323). THE HEAD OF DIPTERUS VALENCIENNESI 27 plates and bearing the supraobrital canal are L, K and three J plates on each side, separate in D. stissmilcht, or partly fused in D. lehmanni, the most posterior of the J plates being very large (Lehmann 1956, text-figs. I, 2). It is unexpected that B should be in three parts, two paired, and at least two other arrangements seem possible. Taking first the obvious one rejected by Westoll (Text-fig. 32) one might consider that in the later Dipterus the orbits had moved backwards from their position in Dipnorhynchus, bringing them opposite the anterior ends of the C plates (Text-fig. 35), and this would account for the difference in the relative position of K, X etc. in the two genera; and their movements outside of J might have been a factor Fias. 31-33. Dipnorhynchus lehmanni Westoll. Fig. 31, Reconstruction of skull-roof after Lehmann & Westoll (1952: 411, text-fig. 5A) with the original lettering (anterior Y-plates should be Yo, posterior, Y,). Fig. 32, The same, but re-lettered according to first alternative, Fig. 33, the same, but re-lettered according to second alternative, 28 APIs, ISEB INID) CO) POU IPIMID Ice UPS) Wh BLIL IZING ID IN IMI, SIE Fic. 34. Dipnorhynchus stissmilchi (Etheridge). Outline reconstruction of skull-roof, after Hills (1933 : 637, text-fig. 2) with addition of the sensory canals from Hills (1941 : 646, text-fig. 1) and the back of the skull-roof restored in outline, after Westoll (1949: 142). The lettering according to second alternative (cf. Fig. 33). Fic. 35. Dipterus valenciennesi S. & M. Reconstruction of skull-roof, after Graham- Smith & Westoll (1937, text-fig. 2a) and Westoll (1949, text-fig. 1A), the lettering modi- fied after Fig. 18. Fic. 36. Scaumenacia curta (Whiteaves). Reconstruction of skull-roof, after Westoll (1949, text-fig. 6D) with the lettering modified. Fic. 37. text-fig. B) with the lettering modified, and Watson & Gill (1923, text-fig. 21). Fia. 38. Fic. 39. Fic. 40. THE HEAD OF DIPTERUS VALENCIENNESI ait, ots, Pee Jagaosrere, * Ctenodus cristatus Agassiz. Reconstruction of skull-roof, after Westoll (1949, Ceratodus sturi Teller. Reconstruction of skull-roof, after Westoll (1949, text-fig. 9A), with lettering modified. Ceratodus forymosus Wade. Reconstruction of skull-roof. als, approximate anterior limit of scale covering. Neoceratodus forsteri (Krefft). Skull-roof, after Holmgren & Stensié (1936, text- fig. 288B), re-lettered. 29 30 TAE HEAD OF DEPT E RUS Vea NG ile IN NEE Sa contributory to the anastomosing of the infraorbital and supraorbital canals. The paired “ B,”’ plates would then be C plates and instead of a proliferation of J plates we would have a more normal series of K, J and I. B would be small and A un- accounted for, unless it was the result of fusion of the very large paired plates labelled I by Westoll, for it seems unlikely in any event that the supposed I plates would later move forwards from the “‘ extrascapular ’’ row to embed themselves in the skull-roof between B and Y as in Dipterus. On the other hand, if we accept the possibility that the two anterior B, plates of Westoll’s interpretation did fuse to form the B of Dipterus, as seems indicated in Dipnorhynchus stissmilchi (Text-fig. 34), then an even more interesting situation could have held (Text-fig. 33). B, of Westoll, would be A and J, would be I, (missing on the right side) with J, as I,, the plate that is still occasionally preserved as a small entity carrying part of the occipital cross-commissure in Rhinodipterus (White 1962, pl. 1, text-figs. 1, 2) and Dipterws (Westoll 1949, text-fig. 37). On the outside of it is Z(H). At this level of development, which Westoll (1942 : 142) claims is a fair approxima- tion to the hypothetical ancestral condition, this row did not carry the cross- commissure, which was supposedly borne by the row behind, H—-I-I—-H of Westoll (Text-fig. 31), but here supposed to be a row of plates (f-a-«-f, Text-figs. 33, 34) that has now disappeared in other known fishes. This is in keeping with the important point shown by Westoll (1938, text-fig. 2) in his brief paper on the ancestry of the Tetrapods, that in vertebrates there is a movement backwards of the bones of the skull roof with additions at the anterior end by orderly development from the variables, and a loss of those behind with a shortening of the occipital region, features shown in some degree by Dipterus relatively to Dipnorhynchus. Between Dipnorhynchus and Dipterus there must have been a stage when the hindmost row f-x—-a—£ was lost and the cross-commissure had been taken over by the full “ extrascapular’’ row Z-—I,—A-I,-Z, before the two I, plates lost their territory to their neighbours and their remnants fused with I, in front taking with them a segment of the commissure—as noted above, only rarely does this remnant of I, ever show itself as a minute independent plate. From Dipterus one can move forward in time through a perfect morphological series based on Westoll’s (1949, text-figs. 6D, 8B, 9A) outline restorations of the later dipnoans Scaumenacia, Ctenodus and Ceratodus, which show progressive diminution and final loss of the “‘ extrascapular’’ Z—-A-Z row, with the cross- commissure being finally captured by Y,—-B—Y, (Text-figs. 35-39). At the same time the series shows one other powerful trend coupled with the relative movement back of the head plates, and that is on simplification of the pattern, by invasion and loss rather than by fusion as Westoll indicates by his lettering. This trend is after all a continuation of the process seen in passing from Dipnorhynchus to Dipterus. The apparently unstable many plated mosaic inherited from the ancestral form persists in the CDE area in Dipnorhynchus ; in Dipterus the C area generally, and that of D and E quite often, has settled down in a more constant pattern of a few large plates, which is characteristic of the later terms in the series. It is interesting to note that in C. formosus Wade (1935: 1, pl. 1; see also Text-fig. 39) the dermal skull pattern can be reasonably interpreted in a form comparable with that of C. stuvi Teller from REE EAD On DEPT EROS VAEENGIEN NEST 31 slightly earlier beds in the Austrian Alps (Text-fig. 38), the most striking difference being that F has increased in size at the expense of D, which in the living Neoceratodus has disappeared altogether (Text-fig. 40). The various trends in this morphological series may be summarized as follows : Posterior Centre Lateral line row of plates plates, Age plates front to rear fp—X, a-P Dipnorhynchus lehmanni late appa Ant. mosaic 4, 4 Lower BAa Devonian D. stissmilcha é : do. appa Ant. mosaic 37+ BAa Dipterus 5 : ; Upper ZAZ Ant. mosaic Ai M.O.R.S. CBA or FEDCBA Scaumenacia. : : Early ZI,AI,Z ECBA Al 3 U.O.R.S. Clenodus : i 9 U.Carbo- Y>.ZI,BI,ZY,; KFEDCB 4, 3 niferous Cervatodus sturi A M. Trias Y,BY, FDB iy 3} C. formosus . ‘ F M. Trias Y,BY, FDB it, 3 Neoceratodus. F : Living FB ity The Endocranium The undersurface of the skull of Dipterus has been tolerably well known since the days of Hugh Miller (1849 : 62, text-fig. 20). Pander (1858 : 10, pl. 3, figs. 11, 13, 14) was the first to deal seriously with this part; Traquair (1878:5, pl. 3, fig. 1) established that the suspension was autostylic ; then Woodward (1891 : 234, text- fig. 36(i)) gave a restoration of the upper and lower jaws, and Goodrich (1909 : 242, text-fig. 210A) one of the whole palate. Finally Watson & Day (1916: 29, text-fig. 6) published a general description of the palatal aspect of the head with an unlettered restoration which has been freely copied by later authors, either as originally published (Holmgren & Stensid 1936 : 365, text-fig. 279) or somewhat modified (Graham-Smith & Westoll 1937: 251, text-fig. 8D). Save-Sdderbergh (1952 : 22) noted further details “. . . similar canals [to those of Chirodipterus] for the efferent pseudobranchial arteries, grooves for the lateral dorsal aortae and orbital arteries, canals for occipital arteries, and grooves for the internal jugular veins ’’, but he gave no figures. Two years later Jarvik (1954 : 69, text-fig. 36B) figured a fine specimen in which he labelled the foramen for the occipital artery, a large paired fossa on the lower side of the otic region, and the groove for the lateral dorsal aorta. Kesteven (1951 : 108) has given a clear warning against the too positive identifica- tion of foramina and other features in fossil skulls. The force of this warning is proportional to the distance in the relationship between the fossil and a living species. In the case of the Dipnoi, it can be shown that the endocranium of Dipterus is sufficiently close to the only other described fossil Dipnoan skull, that of the Upper 32 AslId TELIDINID) CO) IDI ISI IDI OS WY AAIE IB IN (GM IB IN ISIE SL Devonian Chirodipterus (Save-Séderbergh 1952)! and to that of the living Neoceratodus (Giinther 1871; Huxley 1876; Kesteven 1931; Holmgren & Stensi6 1936 : 372) that comparisons may be reasonably made. The most important specimen of Dipterus is an incomplete but uncrushed skull from ‘‘ Caithness’’, P.17410, with a curiously asymmetrical pattern to its roof- plates (Text-figs. 42-47). This has lost the snout and most of the right side of the skull roof and endocranium, but the left side is reasonably well preserved, although the perichondral bone has flaked off in places and intractable matrix obscured others. The articular head of the left palatoquadrate is displaced upwards. This Fic. 41. Dipterus valenciennest S. & M. Head of small specimen showing part of lower dentition of each side. Achanarras, Caithness. P.22194. 5. AG, anterior gular; By, branchiostegal ray; Cl, cleithrum; LafL, lateral line; Op, operculum; PG, posterior gular ; SOp, suboperculum. 1 Save-Sdderbergh pointed out (p. 8) that the snout of Chivodipterus if broken off would be identical with the fossils described as Ganorhynchus. The locality and formation of the unique holotype (44627) of the type-species, G. woodwardi Traquair (1873 : 555, pl. 14) are unrecorded, but Mr. H. A. Toombs has established that the matrix is so similar to that of Middle Devonian (Couvinian) fossils from the neighbourhood of Gerolstein in the Eifel as to leave little doubt as to its source. THE HEAD OF DIPTERUS VALENCIENNESI 33 skull was most probably developed by Save-Séderbergh and formed the basis of his brief remarks on the endocranium of Dipterus (1952 : 22) in his classical description of the skull of Chivodipterus. The bony structure of the endocranium is exactly as that described in Chirodipterus (Save-Sdderbergh 1952 : 6) consisting of inner and outer shells of thin laminar peri- chondral bone which also lined the vascular and nerve canals connecting the two capsules. The space between is largely filled by cancellar bone. The latter is frequently exposed in this specimen owing to the loss of the perichondral layer, and it is not always a simple matter to determine whether the supposed foramina are genuine or not. As often observed in this genus and noted in Ciivodipterus the neurocranium is undivided and forms a single unit with the palatoquadrates which Fic. 42. Dipterus valenciennesi S. & M. Imperfect and irregular skull-roof. Same specimen (P.17410) as in Figs. 43-47. X2°5. 34 ADISIIS, IGUBINID) ONO IDIIPIPIZI ROS) WAIL IB IN| CI IE IN INTE Sl S Mip=e = = Yip! DAH ») a. EY, BYP yyy, Fics. 43-47. Dipterus valenciennesi S. & M. Imperfect skull. Locality in Caithness unknown. P.17410. Fig. 43, Left side view. Palatoquadrate removed. %X2°5. Fig. 44, Planum antorbitale. <4. Fig. 45, Outline of left orbitotemporal space. x2:5. Fig. 46, Three-quarters ventral view. %2°5. Fig. 47, Back of skull. x2.5. a.occ, foramen for occipital artery ; a.oy, foramen for orbital artery; a.or.p, posterior opening for orbital artery ; aps, posterior opening of efferent pseudobranchial artery ; THE HEAD OF DIPTERUS VALENCIENNESI 35 form on each side a high transverse lamina, in this specimen with a forward slope of about 45°. A major difference between the skulls of Dipterus and Chirodipterus is that in the former the neurocranium is in contact with the cranial roof throughout its length, so there is no median fossa or crest over the otic and occipital regions, the fossa for the temporal muscle (¢f) on each side being bounded by the neurocranial wall medially and by the thin dorso-lateral crest (dlc) on the outer side (P.17410, Text-fig. 47). This crest is probably continued backwards under the “ tabular horn ”’ which is a projection of plate I, itself and seems to comprise all three layers (2246761, P13; figs! 1, 2). The lateral crest (/c), in continuation of the palatoquadrate, forms the outer wall of the masseter fossa (m/f) on the inside and the roof of the branchial cavity on the outer. Of the ethmoid region in front of the planum antorbitale there remains only a short length and cross-section of the perichondral lining of the right olfactory canal (mJ, Text-fig. 46), through the post-nasal wall, and it is clear that, as in Chirodipterus, ossification was in this part only perichondral. The anterior ramifications of the profundus in the roof of the nasal capsules along with other canals and vessels, are well displayed in R.S.M. 1859.33.612, and to a lesser extent in P. 466093. The ethmoidal part of the orbital space, preserved only on the left side, is very different from that of Chivodipterus or Neoceratodus (Text-figs. 43-45), for it is much longer, occupying as much of the space as the orbito-temporal region. The posterior face of the planum antorbitale is completely ossified and runs diagonally from the outer margin backwards and inwards to the neurocranial wall. The outermost part is formed by a very large buttress (b,) which joined the entopterygoid to support the dental plate. Narrow at the base it had a wide triangular section dorsally and a small space-filling talon (¢a) outside it. Separated from it by a deep cavity containing the main profundus foramen (prof) was an even larger multiple buttress (b,) forming the neurocranial wall and pierced by two or three small foramina for branches of the profundus. The lower half of this region, which is probably a little crushed down- wards, is obscured by débris (de), but a diagonal fracture in P.755, apparently through the second buttress immediately above the inner posterior angle of the left tooth-plate shows the buttress passing below into a horizontal lateral shelf above the entopterygoid. These buttresses provide a necessary support, from the skull-roof, to massive art.s, articular surface of palatoquadrate ; b,, b,, buttresses ; ch, notochord ; c.i, groove for internal carotid; db, débris; div.end, diverticulum of endolymphatic duct; dic, dorso-lateral crest ; dpt, tooth-plate; Enpt, entopterygoid; forv.m, foramen magnum ; hm, exit of truncus hyomandibularis VII; l.c, lateral crest; /cv, foramen for lateral cerebral vein; mf, masseter fossa; Pq, palatoquadrate (cut away in Fig. 43); prof, main profundus foramen; prof + vju, foramen for profundus nerve and jugular vein ; Psph, parasphenoid ; s.a, groove for lateral dorsal aorta; ¢a, talon of buttress; #f, temporal fossa ; vca, foramen for anterior cerebral vein ; Ve, fused vertebra ; vju, groove for internal jugular vein ; v.pit, groove and foramen for pituitary vein. nI, olfactory nerve; II; III; IV; Vg, 3, opening for mandibular and maxillary branches of trigeminal nerve; VJIIp, posterior branch of acoustic nerve; 1X; X ; Xbr, branches of vagus. 3§ 36 ADIS(ID, WMIDANID) (Ol) ID) MIP ID ID Ire (OWS) VAL IL IE IN| (G IID INN IE SIL dental plates, which with their strongly denticulated surfaces, were clearly capable of dealing with very rough food. Mollusca are not very evident in these beds, and such denticulated plates usually with no obvious facets of wear, speak rather for gripping and crushing struggling animal prey than for cutting inert vegetable matter, so arthropods, worms and perhaps other fishes suggest themselves. That Chivodi- pterus had feebly sculptured plates and rudimentary buttresses—Save-Sdderbergh (1952: 11, pl. 3, fig. 2) speaks only of “a vertically elongated prominent area ’’— suggests that it fed on softer unresisting matter such as carrion. On the other hand the earlier Australian marine Dipnorhynchus had a large free buttress at the outer margin of the entopterygoid in the middle of the orbito-temporal space (P.33699) to support its large massive dental plates for holding and crushing shelly prey, such as the numerous brachiopods with which its remains are associated. Only the outer face of the left wall of the orbito-temporal region of the neuro- cranium of Dipterus is reasonably preserved (P.17410, Text-fig. 43). The brain-case is short and narrow, and the wall, as preserved, curves gently inwards towards the skull roof. There is a wide subocular shelf, a continuation of that already noted, and this passes imperceptibly into the processes of the palatoquadrate. In front high up and just behind the second buttress is a large foramen for the anterior cerebral vein (v.ca). Below and behind this is a very large dumb-bell shaped aperture for the optic nerve (IZ). Whether this is its true shape is not quite certain, but if so, it is unlike what is found in either Chivodipterus or Neoceratodus. As in Chirodipterus, above and behind in a slanting line are three foramina, the topmost for the fourth nerve (JV), the middle and largest for the oculomotor (J//) ; and on a level with the foramen for the optic nerve, a groove and foramen for the pituitary vein (v.fit). Immediately behind and below this vein is the large foramen for the profundus nerve and the internal jugular vein (prof+-v.jw). This foramen lies at the junction of the endocranium and the fused processes of the palatoquadrate and faces forwards. High above this foramen is a large opening corresponding to that identified in Chivodipterus for the mandibular and maxillary branches of the trigeminal nerve (V2,3), but it is relatively farther back, and below it, instead of at the side, is the smaller foramen for the orbital artery (a.07). Behind the palatoquadrate on the left side the ventro-lateral surface of the otic and occipital regions is reasonably well preserved (Text-figs. 43, 46, 47), although the upper part formed by the lateral crest (/c) is possibly a little distorted by crushing. Below it the wall of the endocranium is almost straight and vertical except for the slight swelling over the saccular division of the labyrinth cavity. The back of the skull slopes somewhat irregularly at an average angle of 60° to the base of the occipital region as preserved, but although two vertebral elements (Ve) are seen to be incorporated in this part it is clearly imperfect. A number of features are readily identifiable with those recognised in Chirodipterus. Immediately above the labyrinth swelling lies the groove for the internal jugular vein (vjw) which leads to the foramen piercing the palatoquadrate, and just in front of this foramen is another, almost as large, probably for the exit of the truncus hyomandibularis VII(im), and below it the posterior opening of the canal for the orbital artery (a.orv.p) to which a groove leads from that of the dorsal aorta (s.a). THE HEAD OF DIPTERUS VALENCIENNESI 37 Where these two grooves meet a third groove, for the internal carotid artery, passes slightly medially into a notch (cz) and then under the parasphenoid. Above it is the posterior opening of the efferent pseudobranchial artery (aps). All these features are little more than a paraphrase of what has been found in Chivodipterus, but not seen in that genus is a very large opening, possibly enlarged by fracture, just behind the junction of these same three grooves, probably for a posterior branch of the acoustic nerve (VIJIp). Above and behind this is the foramen for the glosso- pharyngeal nerve (/X). The short occipital region is continuous with the otic but the sides are flatter and more nearly vertical. Owing to the slope of the back of the head, about 60° but exaggerated by breakage, the dorsal flange formed by the lateral crest is present only in the anterior part. At least two vertebrae are fused on to the basioccipital region, and in other specimens (DMSW.P.149, Pl. 1, fig. 2; also P.755 and R.S.M. 1859.33.617) it seems evident there are three. The main features of the otic region, the internal jugular groove (vju) above and that of the lateral dorsal aorta (s.a) below, are continued on to the occipital region. The vagus foramen (X) is in the jugular groove in much the same position as in Chirodipterus (Save-Séderbergh 1952: 19, text-fig. 8). Immediately above and at right angles to it is a large foramen running into the back of the skull, presumably for the lateral cerebral vein (/cv, Text-figs. 43, 47), and above this are two smaller openings for branches of the vagus (Xbr). However, the most conspicuous feature in the back of the skull is a pear-shaped opening in the centre near the skull-roof, presumably a cross-section of a median diverticulum of the endolymphatic duct, (div. end) as in Eusthenopteron (Stensid 1963 : 97, 100, text-figs. 50A, 51A). The groove of the lateral dorsal aorta (s.a) widens to the rear and joins that of the right side behind the parasphenoid (P.34549, Text-fig. 49; cf. Jarvik 1954, text- fig. 36B). Just before it does so, it is pierced at its upper margin by a foramen for the occipital artery (a.occ), which is nearly horizontal owing to the swelling out of the bone. It penetrates the bone for almost 3 mm. by a somewhat sinuous course, and re-emerges above and slightly behind by a similar foramen above, then continues upwards by a well-marked groove (P.34544 ; also DMSW.P.140) which apparently bifurcates before the vessel re-enters the skull through two distinct foramina (Text- fig. 48). The right side of the occipital region of P.17410 is too imperfect to show these features, but the vagus foramen, much enlarged, is present and a little below and behind it, as in Chivodipterus, two or three foramina for spino-occipital nerves. Of the dermal bones of the palate in this specimen, only the right half of the parasphenoid and part of the right entopterygoid is preserved. The former is sunk into the endocranium so that the surfaces are level, and they give the impression of being joined by a suture; and the entopterygoid, to which the parasphenoid is truly sutured, has the same appearance in regard to the palatoquadrate. The features of the palate are better seen in other specimens, in particular P.34540, a specimen collected by Dr. D. L. Dineley in Clardon Haven, near Thurso (Text-fig. 49, Pl. 1, fig. 1). It is a crushed head with very well preserved, but flattened palate. 38 THE HEAD OF DIPTERUS VALENCIENNESI The anterior angle of the parasphenoid in this specimen shows a very well marked hypophysial foramen (/y. f) which runs inwards (upwards) and somewhat backwards. There is a broad, flat or slightly concave selvage along the antero-lateral sides which is widest in front medially and decreases laterally to vanish at the greatest breadth. On the medial side of this selvage the parasphenoid shows a complementary depres- sion or groove, narrowest medially in front and widest behind at the lateral angle. Between these grooves the parasphenoid is gently convex. Near the front of this Fics. 48, 49. Dipterus valenciennesi S. & M. Fig. 48, Basioccipital region to show course, partly excavated, of occipital artery (a.occ). Clardon Haven, near Thurso, Caithness. P.34544. 2:25. Fig. 49, Palatal view. See also Pl. 1, fig. 1. Clardon Haven, near Thurso, Caithness. P.34549. 2:25. Both specimens colld. D. L. Dineley, 1957. fd, fold ; hy.f, hypophysial foramen; s.d, secondary denticles, Other lettering as in Figs. 43-47. THE HEAD OF DIPTERUS VALENCIENNESI 39 median area and some distance behind the hypophysial foramen there is a very conspicuous crescentic, transverse slit or infolding of the surface of the parasphenoid directed forwards (fd) with a corresponding broad and shallow depression behind it in which are four or five well marked denticles on low ridges. The depth of the pocket cannot be seen, but probably it does not extend far in. It is seen in a few other specimens, in varying degrees of development ; well developed although partly masked by transverse cracks in P.755, P.34556 and M.M. no. L.11577 ; irregular in R.S.M. 1859.33.614 ; replaced by a number of smaller shallow tucks in M.M. no. L.10861, and absent in R.S.M. 1859.33.33 (Traquair 1878, pl. 3, fig. 1) and other specimens. This feature is therefore no more than an individual defect, possibly based on an atrophied organ, in the laminar surface layer of the parasphenoid, which is frequently wrinkled or irregular when worn (P.34549 and DMSW.P.149, Pl. 1, figs. I, 2) and possibly was denticulated over much of the palatal surface (M.M. no. err 577): The hypophysial foramen varies quite obviously in its development, being con- spicuous in some specimens (PI. 1, fig. 1) but in others reduced to a very small size and accompanied by other diminutive foramina (P.34556). It was figured without comment by both Pander (1858, pl. 3, figs. rr, 13) and Giinther (1871, pl. 34, fig. 4) and has been noted by Westoll (1949, text-fig. 1D). It is also present in one of the specimens of Rhinodipterus secans, but Gross (1956 : 28, text-fig. 21c) dismissed it as “vielleicht sekundaér entstanden’”’. No such foramen has been recorded from Upper Devonian dipnoans. The dental plates are continuous with the entopterygoids and appear as mere outgrowths of the supporting bones. Fine new specimens of the mandible of the earlier Dipnorhynchus from New South Wales (P.33714, P.46773) clearly show that the rather formless swellings that do duty as dental plates are in fact just thickenings of the vascular and cosmine layers of the bone of the jaw, and it is evident the more definitely patterned “ plates’ of Dipterus and later dipnoans are of like origin. In well preserved specimens they are clearly differentiated by their thick surface layer, which has much the same appearance as the cosmine of the external head-bones but with rather fewer openings of the fine tubuli, well marked in the low flat longitudinal medial areas along which the two plates meet (P.34543, Pl. 2, figs. 1,2; R.S.M. 1859. 33-33, Traquair 1878, pl. 3, fig. 1). In P.755 and 33166 (PI. 2, figs. 3, 4) there is irregularity along the symphysial margin due to resorption and in P.34556 (Pl. 2, fig. 5) asin P.4669r and R.S.M. 1859.33.612, much of the thick cosmine layer has disappeared showing underlying trabecular layer. In extreme cases not only has the whole of the surface between the dental plates gone, as in the relatively small original of Pl. x, fig. 2 (DMSW, P.149), but secondary tubercles (s.d.) may be developed on the rough trabecular bone, and an unusual number of denticles appear between the regular rows on the biting surfaces (P.34549, Pl. 1, fig. x, Text-fig. 49). In this specimen small irregular areas of the right dental plate (left of figure) have disappeared, but this degree of resorption is not seen in any other specimen and must be due to decay in an old specimen. Resorption of the plate is not correlated with size for the completely resorbed originals of Pl. 1, figs. r and 2 are respectively larger and smaller than Traquair’s specimen which is almost perfect. The vomerine dentition is not well displayed in any of the available specimens, 40 ADISIID, IEUDINID) (ONT IDIMIL ID IRI (OS WY AUIL ID ING MIDI IN 1B, SIE although some show parts of the supporting bone, but one seems to show remains of the dentition that can be reasonably interpreted (33166, Pl. 2, fig. 4). Watson & Day (1916 : 33, text-fig. 6) and others following them showed a pair of elongated plates with three transverse rows of small denticles and a narrow hinder end inserted medially between the front of the entopterygoid tooth-plates. None of the available specimens shows evidence to support this arrangement, which may be based on the uneven resorption of the surface of the entopterygoid tooth-plates, as this tends to start along the mid-line and in front (P.34543, P.755, 33166, Pl. 2, figs. 2-4) and the narrow extension of these plates is usually broken away. The last of these specimens shows a very different arrangement : on each vomer there are two or perhaps three Fic. 50. Dipterus valenciennesi S. & M. Restoration of Skull, palatal view. REE EAD OR DEPT RUS VWAEE NCE NN ESL 41 slightly diverging longitudinal ridges which most probably bore denticles, as Watson noted, although there is little evidence of these now. The vomers are well separated from the hard “ lip”’ and must have bitten against a cartilage pad in the front of the much shorter lower jaw—Giinther’s (1871 : 525, pl. 35, figs. 1-3, /.a.) “ lower labial cartilage ”’ in Neoceratodus—and their function could have been merely to hold before gulping. Gtinther (1871 : 518) refers to the “‘ double kind of action ” that the main upper and lower tooth-plates have to perform, crushing or grinding food on the “flat surfaces ’’ and cutting food with the “ sharp lateral ridges’ which “ fit into the notches of the opposite tooth like the shells of a Cardium’’. So very nicely do they fit that there can have been no grinding action involving any degree of lateral displacement, so that the movement must be a simple up and down champing, not quite vertical, for the lower plate is set at a small angle to the horizontal, sloping inwards. And the tooth-plates in Dipterus, with their highly denticulated ridges, must have functioned in much the same restricted manner, and such few facets of wear as the denticles show bear this out (P.34549). Generally speaking the wear is due to friction with the food and just blunts the tubercles which fit into pits of their own making between the radiating rows of denticles on the opposing plate (e.g., 33166, Pl. 2, fig. 4), and in the lower jaw (P.46692).1_ In old worn dentitions, action was much more irregular and new denticles were sometimes developed between the ridges (2234540, Pl. x, fig. 1). A restoration of the palatal aspect of Dipierus is given in Text-fig. 50. 6 SUMMARY 1. The pattern of the dipnoan skull-roof has developed from an ancestral mosaic principally by loss and invasion of roofing plates ; fusion occurs for the most part as an individual aberration. 2. The dipnoi were already widely separated from the Crossopterygii (Jarvik 1960 : 32) at their first appearance, and this separation took place when the skull- roof was in a mosaic stage. It is therefore not possible to correlate the plates of the dipnoan skull-roof meaningfully, and attempts to do so must result only in geographical approximations for which compound names have no validity. 3. The use of nomenclature based on that of the skulls of higher vertebrates is therefore both misleading and undesirable : a special notation is required and a modified edition of Forster-Cooper’s alphabetical scheme, properly related to the standard sensory canal pattern adequately serves the purpose. 1 Dr. Gwynne Vevers’ observations on the feeding of Neoceratodus at the Zoological Gardens in London (in lit. 25 Nov. 64) are very much to the point—‘‘ We feed the present specimen on strips of raw horse meat or fish approximately 3 x 4 Xx inches. The previous specimens here have all eaten a great deal of lettuce, but the present specimen does not appear to enjoy this at all... There is no lateral movement of the jaws in chewing but a very pronounced up and down nibbling movement along the length of the meat strip or fish. The food is then extruded from the mouth and the operation repeated. This may happen several times before the food is finally swallowed. These are straight observations. I suppose the nibbling movement would be a very suitable way of dealing with fresh water crayfish or a similar object.” In a film recently shown by Mr. I. R. Bishop at the Linnean Society of London (20 Feb. 1965) Lepidosiven is seen feeding on live worms with exactly the same action as that described above in Neoceratodus. Dipterus may very well have eaten in a similar fashion once food was obtained, but its heterocercal tail suggests less sluggish movement than in the living lung-fishes and therefore the possi- bility at times of livelier prey. 42 ADI GIS IGLIDINID) (OY JOIIP IMIR OLS) | AUILIE INE ILID IN IN IBS IE In the early Dipnoi there was an additional row of plates behind the “ extra- scapular ”’ row of Dipterus (Z—A-Z) which bore the occipital cross commissure and was later lost. Parallel with the relative movement of the roof-bones in Crossopterygii and higher vertebrates demonstrated by Westoll (1938), there was a movement backwards of the roofing bones of the Dipnoi; while standard plates were developed from the primitive mosaic in front, presumably by loss and inva- sion, rows of plates were successively lost at the back of the skull-roof, the occipital cross-commissure being in turn captured by the row in front until in Triassic times it was established in Y,BY,. After the Devonian period there was also an overall reduction in the number of the plates, by invasion rather than fusion, a process continued to the present day. The effect of §2 is further to emphasize the difference between the lung-fishes on the one hand and the rhipidistians and coelacanths on the other. It seems, indeed, on the evidence of the skull-roof patterns that the Dipnoi split off from all the other Osteichthyes at the skull-mosaic stage and developed an independent pattern of dermal bones in the skull-roof which has little relevance to that of the “ Teleostomi”’ in spite of Westoll’s ingenious arguments to the contrary and his bold assertion (1949 : 159, text-fig. 10) that “‘ The evidence of Devonian forms shows very clearly that the early Dipnoi were very closely related to the contemporaneous Rhipidistia ’’; my reading of the evidence favours very much Jarvik’s (1960 : 31-35, text-figs. 28, 30) interpretation of their relationships as an isolated offshoot from “‘ Preteleostomi’’, and there is a lively expectation that the resemblances that they appear to show towards the contemporary Rhipidistia will prove of decreasing systematic significance. This supposed isolated position is supported by the development of the tooth- plates. In regard to the classification of the Osteichthyes, we are forced back to the ideas of the early years of this century, as expressed by Goodrich (1909) when the first major cleft in the group was placed between the Dipnoi (without, of course, the Arthrodires) and the Teleostomi, equally rated as sub-classes. Lehman (1959 : 8) rates the Dipnoi as a Class, but whatever their evaluation, they are an isolated group, and the somewhat heated argument as to the significance of the term “‘ Crossopterygii ’’ and the appropriate label to cover all the Dipnoi Rhipidistia and Coelacanthini (Romer 1955; Trewavas, White, Marshall & Tucker 1955) is so much beating of the air—the Dipnoi stand apart, and the term Crossopterygii covers only the rhipidistians and the coelacanths. ACKNOWLEDGMENTS Most of the material studied is in the British Museum (Natural History), and to the collections there substantial and important additions have been made lately by Dr. D. L. Dineley and Mr. John Saxon and his son Alan ; and Mr. Saxon has also been able to send valuable information regarding a number of older localities. As usual, ready and patient help with invaluable material on loan from the RoyalScottish Museum was given by Dr. Charles Waterston, and a number of important fossils THE HEAD OF DIPTERUS VALENCIENNEST 43 from the Manchester Museum were generously lent by Dr. R. M. Eagar. These last were from the D.M.S. Watson collection, and one from Professor Watson’s own col- lection was made available for study as well. Dr. F. R. Parrington also kindly sent me material on loan from Cambridge. In the Department I had my usual full backing by Mr. Harry Toombs, well supported in the development of the fossil heads by Mr. Ian Macadie, who also influenced the offer of a number of valuable donations from his native Caithness to the right quarter. The photographs were dexterously taken by Dr. W. T. Dean and Mr. John Ferguson, while in putting the final touches to this document Mr. R. H. Spires, Mr. F. M. Wonnacott and Mr. R. Baker exercised their respective skills. To all these I offer my sincere thanks. Fic. 51. Dipteyvus valenciennest S. & M. Skull-roof of small but aged specimen. Lecto- type of “ Polyphractus platycephalus’’ Agassiz (1844:5, 29, pl. 27, fig. 1). Orkney. PSB 78d | Xo keb: 44 THE HEAD OF DIPTERUS. VALENCIENNES LE REPERENCES Acassiz, L. 1833-44. Recherches sur les Poissons fossiles, II. xii + 310 + 336 pp., atlas 149 pls. Neuchatel. 1844-45. Monographie des Poissons fossiles du Vieux Grés rouge ou Systeme Dévonien (Old Red Sandstone) des Iles Britanniques et de Russie. xxxvi + 171 pp., atlas 43 pls. Paris. BrripGe, T. W. 1898. On the Morphology of the Skull in the Paraguayan Lepidosiven and in other Dipnoids. Tvans. Zool. Soc. Lond., 14 : 325-376, pls. 28, 29. FORSTER-CooPER, C. 1937. The Middle Devonian Fish Fauna of Achanarras. Tvans. Roy Soc. Edinb., 59 : 223-239, pls. 1-8. GoopricH, E.S. 1909. Cyclostomes and Fishes. Jn LANKESTER, R. A Treatise on Zoology, @). i, 1925. On the Cranial roofing-bones in the Dipnoi. /. Linn. Soc. (Zool.), London, 36: 79-86, 7 figs. 1930. Studies on the structure and development of Vertebrates. xxx + 837 pp. London. GRAHAM-SMITH, W. & WESTOLL, T. S. 1937. On a new long-headed Dipnoan Fish from the Upper Devonian of Scaumenac Bay, P.Q., Canada. Tvans. Roy. Soc. Edinb., 59 : 241-266, OS, uw, 2. GreiL, A. 1913. Jn SEmMoN, R. Zoologische Forschungsreisen in Australien und dem Malayis- chen Archipel. Denkschy med.- naturw. Ges. Jena, 4, 2. Gross, W. 1956. Uber Crossopterygier und Dipnoer aus dem baltischen Oberdevon im Zusammenhang einer vergleichenden Untersuchung des Porenkanalsystems palaozoischer Agnathen und Fische. K. svenska VetenskAkad. Handl., Stockholm (4) 5, 6: 1-140, pls. I-16. GUNTHER, A. 1871. Description of Cevatodus, a genus of Ganoid Fishes, recently discovered in Rivers of Queensland, Australia. Philos. Tyans., London, 161 : 511-571, pls. 30-42. Hitts, E.S. 1933. Ona Primitive Dipnoan from the Middle Devonian Rocks of New South Wales. Ann. Mag. Nat. Hist., London (10) 11 : 634-644, pls. 11, 12. 1941. The Cranial Roof of Dipnorhynchus stissmilchi (Eth. fil.). Rec. Aust. Mus., Sydney, 21: 45-55, pl. 9. 1943. The Ancestry of the Choanichthyes. Aust. J. Sct., Sydney, 6 : 21-23, fig. HormGrEN, N. & StENSIO, E. 1936. Kranium und Visceralskelett der Akranier Cyclostomen und Fische. Handb. vergl. Anat., Berlin, 4 : 233-500, 171 figs. Huxtey, T. H. 1876. Contributions to Morphology. Ichthyopsida, No. 1. On Cevatodus forsteri, with Observations on the Classification of Fishes. Pvoc. Zool. Soc. Lond., 1876: 24-59, II figs. Jarvik, E. 1942. On the Structure of the Snout of Crossopterygians and Lower Gnathostomes in General. Zool. Bidy. Uppsala, 21 : 235-675, pls. 1-17. 1948. On the Morphology and Taxonomy of the Middle Devonian Osteolepid Fishes of Scotland. K. svenska VetenskAkad. Handl., Stockholm (3) 25, 1 : 1-301, pls. 1-37. 1950. Middle Devonian Vertebrates from Canning Land and Wegeners Halv6 (East Greenland), II. Crossopterygii. Medd. Gronland, Kobenhavn, 96, 4 : 1-132, pls. 1-24. 1954. On the visceral skeleton in Eusthenopteron with a discussion of the parasphenoid and palatoquadrate in fishes. K. svenska VetenskAkad. Handl., Stockholm (4) 5, 1 : 1-104, 47 figs. 1960. Théories de l’Evolution des Vertébyés. 104 pp., 30 figs. Paris. KESTEVEN, H. L. 1951. The origin of the Tetrapods. Pyvoc. Roy. Soc. Victorvia, Melbourne, 59 : 93-138, 8 figs. LEHMAN, J.-P. 1959. Les Dipneustes du Dévonien supérieur du Groenland. Medd. Gronland, Kobenhayn, 160, 4 : 1-58, pls. 1-21. LEHMANN, W.M. 1956. Dipnorhynchus lehmanni Westoll, ein primitiver Lungenfisch aus dem rheinischen Unterdevon. Paldont. Z., Stuttgart, 30 : 21-25, pl. 1. LEHMANN, W. M. & WesTOLL, T.S. 1952. A primitive dipnoan fish from the Lower Devonian of Germany. Pyvoc. Roy. Soc. Lond. (B) 140 : 403-421, pl. 24. MarsuHaLt, A. J. 1962. In Parker & Haswell “ A Text-book of Zoology ’’. Seventh Edition. xxiii + 952. London. APTaLID: VENI INID) Ol ID NIP IL ISZIZ IS Y Zol ITE IN[ OID IN IND SIE 45 Mitter, H. 1849. Footprints of the Creator : or the Asterolepis of Stvomness. xvi + 313 pp. London. Orvic, T. 1961. New Finds of Acanthodians, Arthrodires, Crossopterygians, Ganoids and Dipnoans in the Upper Middle Devonian Calcareous Flags (Oberer Plattenkalk) of the Bergisch Gladbach—Paffrath Trough, 2. Paldéont. Z., Berlin, 35 : 10-27, pls. 1-6. PANDER, C.H. 1858. Uber die Ctenodipterinen des Devonischen Systems. viii + 64 pp., 9 pls. St. Petersburg. ParRRINGTON, F. R. 1949. A theory of the relations of lateral lines to dermal bones. Proc. Zool. Soc. Lond., 119 : 65-78, 2 figs. 1950. The skull of Dipterus. Ann. Mag. Nat. Hist., London (12) 3 : 534-547, 5 figs. 1956. The patterns of dermal bones in primitive vertebrates. Pyvoc. Zool. Soc. Lond., 127 : 389-411, 9 figs. Purp, G. M. & PEppER, A. E.H. 1964. A re-assessment of the age of the Middle Devonian of south-eastern Australia. Nature, Lond., 202 : 1323-1324. Romer, A.S. 1936. The Dipnoan Cranial Roof. Amer. J. Sci., New Haven, (5) 32 : 241-256, 4 figs. — 1955. Herpetichthyes, Amphibioidei, Choanichthyes or Sarcopterygii? Nature, Lond., 176 : 126. 1962. Vertebrate Evolution. Copeza, 1962 : 223-227. SAVE-SODERBERGH, G. 1932. Preliminary Note on Devonian Stegocephalians from East Greenland. Medd. Gronland, Kobenhavn, 94, 7 : 1-107, pls. I-22. 1952. On the Skull of Chivodipterus wildungensis Gross, an Upper Devonian Dipnoan from Wildungen. K. svenska VetenskAkad. Handl., Stockholm (4) 3, 4 : 1-28, pls. 1-7. SEpGwick, A. & Murcuison, R. I. 1835. On the Structure and Relations of the Deposits contained between the Primary Rocks and the Oolitic Series in the North of Scotland. Tvans. Geol. Soc. Lond. (2) 3 : 125—160, pls. 13-17. StEensio, E. A. 1963. The Brain and the cranial Nerves in fossil, lower craniate Vertebrates. Skr. norsk. VidenskAkad. Oslo, I. Mat. Nat. Kl. (N.S.) 13: 1-120, figs. 1-54. TELLER, F. 1891. Ueber den Schadel eines fossilen Dipnoérs Cevatodus Sturii nov. spec., aus den Schichten der oberen Trias der Nordalpen. Abh. geol. Reichsanst. Wien, 15, 3 : 1-38, pls. 1-4. Traguair, R. H. 1873. On a new genus of Fossil Fish of the order Dipnoi. Geol. Mag., London, 10 : 552-554, pl. 1. —— 1878. On the Genera Dipterus, Sedgw. & Murch., Palaedaphus Van Beneden and De Koninck, Holodus, Pander, and Cheivodus, M’Coy. Ann. Mag. Nat. Hist., London (5) Qe 070 Pls 3p TREWAVAS, E., WuiteE, E. I., Marsuatrt, N. B. & TucKER, D. W. 1955. [Discussion of A. S. Romer 1955]. Nature, Lond., 176: 126. Wave, R. T. 1935. The Triassic Fishes of Brookvale, New South Wales. xiv + 89 pp., to pls. British Museum (Nat. Hist.), London. Watson, D. M. S. & Day, H. 1916. Notes on some Palaeozoic Fishes. Mem. Manchr. Lit. Phil. Soc., 60, 2 : 1-52, pls. 1-3. Watson, D. M.S. & Girt, E.L. 1923. The structure of certain Palaeozoic Dipnoi. J. Linn. Soc. (Zool.) London, 35 : 163-216, 34 figs. WESTOLL, T.S. 1938. Ancestry of the Tetrapods. Natuve Lond., 141 : 127-128, 2 figs. 1943. The Origin of the Tetrapods. Biol. Rev., Cambridge, 18 : 78-98, 9 figs. 1944. The Haplolepidae, a new Family of late Carboniferous Bony Fishes. Bull. Amer. Mus. Nat. Hist., New York, 83, I : 1-122, pls. I-10. 1949. On the Evolution of the Dipnoi. In Genetics, Paleontology, and Evolution (Ed. Jepsen, G. L., Simpson, G. G. & Mayr, E.) : 121-184, 11 figs. Princeton. Waite, E.I. 1962. A Dipnoan from the Assise de Mazy of Hingeon. Bull. Inst. voy. Sct. nat. Belgique, Bruxelles, 38, 50: 1-8, pls. 1, 2. Woopwarpb,A.S. 1891. Catalogue of the Fossil Fishes in the British Museum (Natural History), 2. xliv + 567 pp., 16 pls. British Museum (Nat. Hist.), London. PAA Dipterus valenciennesi S. & M. Fic. 1. Undersurface of skull of old specimen showing decay (or irregular resorption) of right tooth-plate (left of figure) and complete resorption of surface of entopterygoid between tooth- plates with formation of secondary denticles. Colld., D. L. Dineley 1957. Clardon Haven, Caithness. P.34549. 2. See also Text-fig. 49. Fic. 2. Undersurface of skull showing vertebral elements fused to basioccipital region. Loc. unknown, Caithness. DMSW.P.149. x2. Bull. B.M. (N.H.) Geol. 11, 1 PLATE 1 PAL ASE 2) Dipterus valenciennesi S. & M. Fic. 1. Imperfect entopterygoid tooth-plates showing cosmine surface and hypophysial foramen. Colld., D. L. Dineley, 1957. Clardon Haven, Caithness. P.34543. x3. Fic. 2. Same specimen powdered to show initial resorption along median suture. X1I°5. Fic 3. Another specimen showing increased resorption along median suture. Orkney. 755 ON bee Fic. 4. Specimen showing advanced and irregular resorption along median suture and worm vomerine dentition in front. Near Thurso, Caithness. 33166. X1I°5. Fic. 5. Palatal view of skull. The median area of the entopterygoid tooth-plates has lost the cosmine surface completely, and only the base of the vomerine dentition remains. Colld., D. L.Dineley, 1957. Clardon Haven, Caithness. P.34556. XI°5. Bull. B.M. (N.H.) Geol. rr, 1 PLATE 2 PLATE 3 Dipterus valenciennesi S. & M. Fic. 1. Skull-roof showing pores of sensory canals enlarged by resorption and the post- cranial processes (“tabular horns’’). Colld. J. Saxon, 1964. Weydale Quarry, Caithness. IP ACO, == Be Fic. 2. Hinder end of same specimen. The left post-cranial process and plate I, have lost outer and part of middle layers exposing section of occipital cross-commissure with short branch forwards. X4°5. Fic. 3. End of snout showing pores of supraoccipital canal enlarged by resorption. (See also Text-fig. 19) Colld. R. V. Collier, 1964. Clardon Haven, Caithness. P.46690. x 4:3. BUATE 3 Bull. B.M. (N.H.) Geol. 11, 1 v4 " SE UM LN ATURAL HISTORY) _ _ Vol. sue No. 2 SOME LOWER CREMmACEOUS, TEREBRATELLOIDES BY EELS, FREDERIC OWEN Pp. 47-72 ; 3 Plates; 13 Text-figures BUTE BUN (OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 2 LONDON : 1965 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted im 1949, 1s issued in five series corresponding to the Departments of the Museum, and an Historical series. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. In 1965 a separate supplementary series of longer papers was instituted, numbered serially for each Department. This paper is Vol. 11, No. 2 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1965 DRIUS TE BES) OF CEE BRATS El VEO Sie Wii (CNPAG TU RSAUIE eral Sain One) Issued 7 September, 1965 Jevgian deat SOME LOWER CRE WACEOUS FEREBRATELLOMEA By ELLIS FREDERIC OWEN SNF INICOENS IES) Several Lower Cretaceous species previously described by Walker, Davidson, Meyer and Keeping are revised and referred to Tamarella gen. nov. or Vectella gen. noy. Serial sections of representative species of these forms are shown for the first time. In addition, two new species of Rugitela from the Neocomian of Lincolnshire and Yorkshire are described and related to forms previously recorded from North Germany and France. INTRODUCTION REPRESENTATIVE genera of the Mesozoic brachiopod family Zeilleriidae have been recorded from the Trias to the Lower Cretaceous but few of the early records have been further investigated, and the Jurassic genus Zev/leria has often been quoted from the Neocomian and, more recently, with those of Awlacothyris and Ornithella by Middlemiss (1959) and Casey (1961), from Aptian and Albian beds of the Lower Greensand. Some of these Cretaceous brachiopods, while possibly belonging to the same families as Jurassic species, have characters which merit generic distinction. Until the researches of Muir-Wood (1934, 1936) the internal structures of the British Jurassic Terebratelloidea had been known only from dissections or by chance exposure of brachial loops and cardinalia in damaged specimens. Minor differences in loop development had previously escaped critical investigation and no real advance had been made in their further classification. It is not the purpose of this paper to add to or further discuss the classification of the terebratelloid brachiopods. However, a systematic study has been made of some of the Lower Cretaceous species within this suborder together with comparisons of genera already described by Muir- Wood (1934, 1936) and Cooper (1955). Some of the early Cretaceous terebratelloid species formerly described by Walker (1867, 1868, 1870) from the Lower Greensand of Upware (Cambridgeshire), Brickhill (Buckinghamshire) and Potton (Bedfordshire) have been revised and referred to Tamarella gen. nov. Others, which had been broadly interpreted as Waldheimia, Zeilleria, Aulacothyris and Ornithella are placed in Vectella gen. nov. Keeping (1883 : 24) suggested a possible line of evolution of some of Walker’s species and related them to others described by Morris (1854) and J. de C. Sowerby (1836) from the Lower Greensand of the Isle of Wight. His suggestions were, however, based mainly on a comparison of external morphology, and an examination of the species listed by him has revealed a degree of homoemorphy. Information obtained from transverse serial sections of these forms has now made possible a revision of his ideas regarding their relationships. In a description of a brachiopod fauna from the Mural Limestone (Middle Albian) of Arizona, Cooper (1955 : 10) referred Terebratula tamarindus J. de C. Sowerby to his new genus Psilothyris. He figured a series of etched specimens showing (pl. 3) what 58 50 SOME LOWER CRETACEOUS TEREBRADELLOIDEA he described as precampagiform, terebrataliform and dalliniform stages in the development of the brachial loop of certain specimens. On the same plate, for comparison, he figured the cardinalia without brachial loop of a dissected specimen which he identified as 7. tamavindus from the Lower Greensand of Faringdon, Berkshire. If, as he suggested, the etched loops of his specimens exhibit dallinoid development stages, then Psilothyris would appear to belong with the Dallinidae. Fic. 1. Diagram illustrating the cardinalia of (A) Tamarella, showing the broad hinge- plates with the anterior plication, and (B) a reconstruction of the brachial loop showing the thickening of the descending branches. (C) Vectella, showing the deep hinge-trough and long dorsal septum, and (D) reconstruction of the brachial loop. Although many young specimens of 7. tamarindus have been examined from the Lower Greensand of Faringdon, Berkshire, no dallinoid stages in loop development have been found, but a remarkable specimen of 7. tamarindus from Faringdon (PI. 1, fig. 8c) shows what appears to be a thickening of the descending branches some 6 mm. below the dorsal umbo in a brachial valve, 15 mm. long. It is possible that this thickening may be the remains of an early connecting band or attachment. A similar thickening is seen on the descending branches of the brachial loops of speci- mens of Psilothyris figured by Cooper (1955, pl. 3, figs. 18, 19). The lack of serial sections of the genus Psilothyris make absolute comparison with British Upper Aptian species of Tamarella impossible and the absence of any evidence of early loop attachment in even very young specimens of Tamarella means that any attempt to relate the two genera is purely conjecture. However, the two genera have much in common and it may be necessary at some future date to emend the present study. The varied brachiopod fauna contained in the condensed Neocomian beds represen- ted in the Claxby Ironstone of Lincolnshire and the Speeton Clay of Yorkshire is SOME LOWER CRETACEOUS TEREBRATELLOIDEA 51 comparable to that described by Roemer (1836-1840) from the Hilsconglomerat of the Hanover district, north-west Germany. Species of zeilleriid brachiopods described as Terebratula hippopus and T. longa by Roemer are revised here and assigned to the genus Rugitela Muir-Wood from the Fuller’s Earth Rock of Somerset and said at the time to have a probable range of ? Inferior Oolite to ? Upper Jurassic. Judd (1868) was the first to recognize Roemer’s species 7. hippopus in Lincolnshire where it occurs fairly abundantly in the Claxby Ironstone. Davidson (1874) erected a varietal name for the British form on the grounds of its larger dimensions, but these are directly proportionate to specimens from the type locality at Berklingen and from Saltzgitter, near Hanover. Furthermore, the pattern of variation in outline, width of shell and depth of dorsal sulcus is identical in both forms and there seems no point in upholding Davidson’s varietal name. SYSTEMATIC DESCRIPTIONS Family ZEILLERIIDAE Allan, 1940 Genus VECTELLA nov. Diacnosis. Shell elongate-oval, sulco-carinate to biconvex, rectimarginate to uniplicate or sulcate. Umbo massive, beak suberect. Foramen circular. Beak- ridges sharp, mesothyrid ; interarea short. Test smooth, finely punctate. Delti- dial plates conjunct. Short, subparallel dental lamellae embedded in callus; in mature individuals these lamellae support massive, peg-like, inwardly directed hinge- teeth. Cardinal process poorly developed. Acute septalium forms broad V-shaped hinge-trough supported by short, thick, persistent dorsal median septum. Crural bases triangular, giving rise to zeilleriiform brachial loop developed ventrally. TYPE SPECIES. Waldheimia celtica Morris 1854 : 158. LOCALITY AND HORIZON. In addition to the type-species, which comes from the Upper Aptian, Parahoplites nutfieldensis Zone, Shanklin, Isle of Wight, Vectella is represented at an equivalent horizon at Upware (Cambridgeshire) by V. woodward: (Walker) and V. angusta (Walker) and by V. morrisi (Meyer) from the Bargate Beds of Surrey, from Brickhill (Buckinghamshire) and from Shanklin, Isle of Wight. REMARKS. Vectella is probably a further development of the Jurassic genus Ormithella but differs in its more acute septalium and deeper hinge-trough, more acutely triangular hinge-plates and crural bases, and more extensive inner socket- ridges. It differs from Tamarella nov. and Rugitela in its thicker shell, fused dental lamellae and narrower, anteriorly plane, septalium. Vectella celtica (Morris) (Pl. 1, figs. 4a-—c; Text-fig. 2) 1847 Tevebratula longa Roemer ; Davidson & Morris : 255, pl. 19, figs. 1 & 1a—d. 1854 Waldheimia celtica Morris : 158. 1855 Waldheimia (Terebratula) celtica Morris ; Davidson: 73, pl. 9, figs. 32-35. 1874 Waldheimia celtica Morris ; Davidson : 47, pl. 6, fig. 15. DEscRIPTION. Biconvex, elongate-oval zeilleriid brachiopod, approximately 52 SOME LOWER CRETACEOUS TEREBRATELLOIDEA Fic. 2. A series of sixteen transverse sections through the umbonal part of Vectella celtica (Morris) from the Upper Aptian, Shanklin, Isle of Wight. BM. B.25801. x2. 30 mm. long, Ig mm. wide and 18 mm. thick. It has a short, massive umbo, sub- erect beak and well exposed symphytium. The circular foramen is fairly large with mesothyrid beak-ridges bordering a short interarea. In young forms a marked sulcus is developed posteriorly in the brachial valve and may still be seen in some specimens of mature forms. LEcTOTYPE. Morris was the first to use the specific name Waldheimia celtica for a brachiopod from the Lower Greensand of Great Britain and listed it in the second edition of his catalogue (Morris 1854: 158). He illustrated the species by indicating specimens in Davidson’s Cretaceous Monograph which was not, in fact, published until a year later (Davidson 1855: 73, pl. 9, figs. 32-35). One of these specimens had certainly been taken from a series of four given to Davidson by Morris and now in the Davidson Collection at the British Museum (Natural History), registered number B.6757. The specimens are here regarded as syntypes of V. celtica (Morris) and one, figured by Davidson (1855, pl. 9, fig. 33) is chosen as lectotype and has been re- registered BB. 42915. SOME LOWER CRETACEOUS TEREBRATELLOIDEA 53 MATERIAL. The lectotype and numerous specimens in the Davidson Collection and general collection of the British Museum (Natural History) from the Upper Aptian, Parahoplites nutfieldensis Zone of Shanklin, Isle of Wight, registered as B.6757, B.25800, B.25802-25810. Aspecimen, B.14728, in the Sedgwick Museum, Cambridge shows the brachial loop and cardinalia in an almost perfect state of preservation. REMARKS. This species probably illustrates an evolutionary link with some species of the Jurassic genus Orynithella. Externally, the long tapering oval outline, massive umbo and carinate pedicle valve suggest such species as O. ornithocephala (J. Sowerby) and O. bathonica (Rollier). Internally the thickened valve walls and callus embedded dental lamellae in the pedicle valve and the thick septum, sup- porting short, stout hinge-plates, have much in common with those of O. bathonica as shown in transverse serial sections by Muir-Wood (1934: 544). It can be dis- tinguished from either of the two Jurassic species in its shorter umbo, larger foramen and less acutely tapering valves. It is distinguished from other Lower Cretaceous forms, such as V. woodwardi (Walker), by its more regular oval outline, more in- curved beak, smaller foramen and absence of anterior sulcation of the brachial valve. It differs from V. morrist (Meyer) in its greater dimensions, stronger biconvexity and lack of uniplication of the anterior margin. It bears a superficial resemblance to V. angusta (Walker) from Upware and Brickhill but differs in having a more incurved beak and more robust general outlines. It is nevertheless very near this species. Vectella angusta (Walker) (PI. 1, figs. 5a—c) 1868 Waldheimia mutabilis var. angusta Walker : 400, pl. 19, fig. 5, 54. 1870 Waldheimia mutabilis var. angusta Walker ; Walker : 562. 1874 Waldheimia wanklyni var. angusta Walker ; Davidson : 51, pl. 7, figs. 26-28. DEscriPTION. Elongate-oval to fusiform Vectella about 34 mm. long, 18 mm. wide and 12 mm. thick. Maximum width midway between umbo and anterior margin. Biconvex with acute carination of pedicle valve. Large foramen domina- ting a slightly produced suberect umbo with wide interarea and well exposed sym- phitium. Extensive beak-ridges are well defined. Anterior commissure plane. HoLotyre. British Museum (Natural History) no. 67601, figured by Walker (1868, pl. 19, figs. 5, 5a) as Waldheimia mutabilis var. augusta. From the Lower Greensand of Upware, Cambridgeshire. MATERIAL. In addition to the holotype, eight specimens from Upware, registered no. B.2712 and seventeen specimens from Brickhill, Buckinghamshire, registered no. B.25503, all in the British Museum (Natural History). Remarks. Asin the case of Tamarella elliptica (Walker), this species was originally described as a variety of Waldheinua mutabilis, a name already used by Oppel (1861 : 538) for a Liassic species. On Walker’s instructions the name W. wanklyni was substituted by Davidson (1874 : 51) for W. mutabilis Walker. Walker neither selected nor indicated a type specimen for W. mutabilis [= W. wanklyni| and the species is, therefore, in doubt. The variety augusta, however, is 58§ 54 SOME LOWER CRE DAC E OWS) TERE BRAC Eee ONDA easily recognizable as a form from the Lower Greensand of Upware and Brickhill. It is in no way related to the species Tamarella elliptica and is here promoted to specific rank. Somewhat resembling V. ce/tica (Morris) in general outline, V. augusta differs in less clearly defined features, more acutely convex brachial valve and produced umbo. The form from Brickhill has a shorter umbo and is more regularly elongate-oval in outline than the typical form. It also has a slight depression or sulcus visible on the brachial umbonal region. This depression is more marked in younger specimens. Vectella woodwardi (Walker) (Pl. 1, fig. ta-c; Text-fig. 3) 1867 Waldheimia woodward: Walker ; 455, pl. 19, fig. 3. 1868 Waldheimia woodwardi Walker ; Walker : 404. 1874 Waldheimia woodwardi Walker ; Davidson : 52, pl. 6, figs. 1—-5a. 1883 Waldheimia woodwardi Walker ; MKeeping: 21. DEscRIPTION. Elongate-oval Vectella, about 37 mm. long, 20 mm. wide and 19 mm. thick. The pedicle valve is acutely carinate with steep flanks. It has a short, suberect umbo truncated by a large, circular foramen. Sharp mesothyrid beak- ridges border a short, flat interarea. The finely punctate shell surface is covered by numerous fine growth-lines. A shallow sulcus originating from the umbo inthe brachial valve broadens anteriorly. The internal characters are as described for the genus. Lectotype. The original specimen described and figured by Walker (1867) is one of the two syntypes originally registered with the Walker Collection in the British Museum (Natural History) as 62202. It has been re-registered as BB. 42910 and is here selected as lectotype. MATERIAL. Eighteen specimens in the general collection of the British Museum (Natural History) nos. B.25717, B.25801, 62202, BB.21130-38, BB. 42922, all from the type locality at Upware, and several internal moulds from Potton. RemArKsS. Vectella woodwardi is known from the type locality at Upware (Cam- bridgeshire) and from Potton (Bedfordshire), where it occurs in the Parahoplites nutfieldensis Zone of the Upper Aptian. In many respects it resembles V. celtica (Morris) which occurs at the same horizon at Shanklin, Isle of Wight but differs in shell convexity and beak characters. These differences in morphology may be due merely to change of environment. The broad sulcation of the brachial valve in V. woodwardi and the shorter, more massive umbo and larger foramen distinguish the species from V. morrisi (Meyer) and V. angusta (Walker). Vectella morrisi (Meyer) (Pl. 1, fig. 7; Text-fig. 4) (Pl. 3, figs. 7-9) 1863 Tevebvatula moutoniana Lankester : 414, pl. 19, figs. I-3. (non d’Orbigny 1848). 1864 Waldheimia moutomana Lankester ; Meyer : 251, figs. 12-14. 1868 Waldheimia morrisi Meyer : 269. 1874 Waldheimia morrist Meyer ; Davidson : 47. SOME LOWER CRETACEOUS TEREBRATELLOIDEA 55 0°4 0:3 0°3 Fic. 3. Sixteen transverse serial sections through the umbo of Vectella woodwardi (Walker) from the Upper Aptian, Upware, Cambridgeshire. BM. B.42922. 2. Description. Although somewhat terebratulid in general aspect, Vectella morrist is characteristically oval in outline with a shorter, more incurved umbo than other described species of Vectella. Both valves are comparatively flat but the brachial valve has a marked median ridge extending two-thirds of the shell length and giving rise to fairly steep flanks. The pedicle valve remains gently convex. The anterior commissure is marked by a faint uniplication distinguishing it from 56 SOME LOWER CREDACE OMS Wb Roe BivAw tL iWwOm Dies V. celtica (Morris) which it somewhat resembles. The typical form occurring at Shanklin, Isle of Wight reaches a maximum length of 23 mm., a width of 17 mm. and attains an average thickness of II mm. Lectotype. Meyer (1864 : 249) first described the species as Terebratula moutoni- ana Lancaster and later (1868 : 269) re-described it under the binomen Waldheimia morrisi referring to the figured specimens (1864: figs. 12-14) of his earlier work. Two of the specimens figured by Meyer (figs. 12, 13) were stated to have been collected from the Pebble-bed of Shanklin, while the other (fig. 14) was said to have come from the Pebble-bed of Godalming, Surrey. These specimens, here regarded as syntypes of V. morrisi, are in the Sedgwick Museum, Cambridge and are registered as B.14770-71 and B.16785. The lectotype, here selected, is B.14771. It was originally figured by Meyer (1864, pl. 12, fig. 13) and was collected from the Pebble-bed, Parahoplites nutfieldensis Zone, Shanklin, Isle of Wight. 0:3 0:2 0:5 0: Fic. 4. Fourteen transverse serial sections through the umbo of Vectella morrisi (Meyer) from the Upper Aptian, Shanklin, Isle of Wight. BM. B.21937. x2. MATERIAL. Apart from the type specimens mentioned above, there are forty-two specimens in the Sedgwick Museum, Cambridge, twenty-nine from Shanklin and thirteen from the Bargate Pebble-bed of Godalming, Surrey, registered B.16838—50, B.14761-69, B.14776-89. Also there are fifty-four specimens in the general collec- tion and Davidson Collection of the British Museum (Natural History) registered B.25506-9, B.6740, B.25815-16, BB. 42914, BB.42916-18. Remarks. A smaller form of V. morrisi, though with similar proportions, occurs in the Bargate Pebble-beds of Surrey and beds of equivalent age at Brickhill, Bucking- hamshire. The latter has often been confused with Tamarella juddi (Walker) but differs from this species in its more regular oval outline, less acutely convex valves, steeper flanks, shorter umbo, less extensive interarea, rounded beak-ridges and simpler zeilleriform brachial loop. It is approximately 17 mm. long, 11 mm. wide and 8 mm. thick. SOME LOWER CRETACEOUS TEREBRATELLOIDEA 57 Genus TAMARELLA nov. DiacGnosis. Shell biconvex. Circular to elongate-oval to pentagonal in outline. Folding rectimarginate to incipiently uniplicate to ligate. Umbo massive, suberect ; beak-ridges sharp. Deltidial plates conjunct. Foramen large, mesothyrid. Shell surface often with marked concentric growth-lines. Median septum short, extending less than one third the length of brachial valve. Septalium broad, shallow, anteriorly arched. Hinge-plates fused, overlapping. Long brachial loop unattached to median septum in adult stage, given off ventrally. TYPE SPECIES. Terebratula tamarindus J. de C. Sowerby 1836. RANGE. Upper Aptian. Fic. 5. Twelve transverse serial sections through the umbonal part of Tamayvella tamarindus (J. de C. Sowerby) from the Upper Aptian, Shanklin, Isle of Wight. BM. B.506. x2. Tamarella tamarindus (J. de C. Sowerby) (Pl. 1, figs. 2a—c, 8a—c, 10a-c, Pl. 3, figs. 5a—c, 6a—c; Text-figs. 5, 6) 1836 Tevebratula tamarindus J. de C. Sowerby : 338, pl. 14, fig. 8. 1843 Tevabratula tamarindus Sowerby ; Morris : 137. 1855 Waldheimia tamarindus (J. de. C. Sowerby), Davidson : 74, pl. 9, figs. 20-31. 1868 Tevebratula tamarindus var. magna Walker : 465, pl. 109, fig. ro. 1874 Waldheimia tamarindus var. magna (Walker) Davidson : 49, pl. 6, figs. 16-194. 1955 LPsilothyris tamarinda (Sowerby) Cooper: 14, pl. 3, fig. 25. DESCRIPTION. Jamarella subcircular to elongate-oval in general outline. Acutely biconvex to almost orbicular. Approximately 1m mm. long, 10 mm. wide and 8 mm. thick. The massive umbo is truncated by a large foramen. Well marked meso- thyrid beak-ridges border extensive interarea. Shell surface evenly punctate and ornamented by fine concentric growth-lines. The anterior commissure is rectimar- ginate to incipiently uniplicate. There is sometimes a shallow anterior depression or sulcus bordered by faint carinae noticeable on the pedicle valve. Shell margins show a tendency to gerontic thickening. NeEotyPe. J. deC. Sowerby im Fitton (1836 : 338) described a zeilleriid brachiopod from the Lower Greensand near Hythe, Kent, under the name Terebrvatula tamarindus. No type material was indicated but a specimen, stated to belong to the Sowerby 58 SOME LOWER CRETACEOUS TEREBRATELLOIDEA Collection, was figured (pl. 14, fig. 8) and is now lost. Sowerby’s description refers to a specimen with an orbicular outline and with beak-ridges extending some distance down the sides of the valves. The figured specimen appears flattened in lateral profile and with short beak-ridges which are not clearly defined. Although its geo- > a 0°5 NCE Fic. 6. Seventeen transverse serial sections through the umbo of Tamarella tamarindus (J. de C. Sowerby) from the Upper Aptian, Upware, Cambridgeshire. BM. B.25693. x2. SOMES WOW ER CRETACEOUS TERE BRAT EE LOLrDE A 59 logical age was stated to be Lower Greensand, no precise beds were mentioned within this horizon nor was any associated fauna listed. Its geological age must, therefore, remain broadly defined. Furthermore, the type locality, said to be “near Hythe, Kent ’’, although suggesting a Hythe Beds origin, might alternatively have indicated the Sandgate Beds (Upper Aptian) or even an Albian horizon. The present widely accepted connotation of a brachiopod under the specific name tamarindus is Davidson’s description (1855: 74) under the binomen Waldheimia tamarindus (J. de C. Sowerby). Davidson figured six specimens from various localities of Lower Greensand age, including one specimen (pl. 9, fig. 26) stated to have been collected from between Hythe and Sandgate, Kent. Unfortunately none of the six specimens figured by Davidson is available in the Davidson Collection and they cannot be traced elsewhere. Although Sowerby’s definition of Terebratula tamarindus was vague, the name is so entrenched in literature that it would seem a pity not to preserve it. The preserva- tion of the name, however, depends on the erection of a neotype to replace Sowerby’s original figured specimen. The Lower Greensand includes, among other horizons, the Sandgate Beds (Upper Aptian), which crop out between Hythe and Sandgate, Kent, and from which Sowerby’s specimen is likely to have been collected. It is no longer possible to collect brachiopod specimens from the Sandgate Beds exposed in this area but beds of an equivalent age falling within the Parahoplites nutfieldensis Zone occur at Shanklin, Isle of Wight and occasionally contain specimens which agree with the original description of Terebvatula tamarindus J. de C. Sowerby. They also compare favourably with specimens figured by Davidson (1855) from “ near Sandgate’ and from the Isle of Wight. A specimen probably used by Davidson (1855 : 74) in his description of the species is figured here (Pl. 3, fig. 5) and proposed as neotype. This specimen, originally registered as B.6724, forms part of the Davidson Collection in the British Museum (Natural History) and was collected from the Upper Aptian beds of Shanklin, Isle of Wight. It has been re-registered as BB. 42907. MATERIAL. A further sixty-one specimens of Tamarella tamarindus (J. de C. Sowerby), all from the Isle of Wight, are contain in the general collection and the Davidson Collection at the British Museum (Natural History) and are registered as B.506, B.6724, B.25820, B.25187, BB.42905, BB. 42906, BB. 42919, BB. 42920. Remarks. A larger form of this species approximately 24 mm. long, 21 mm. wide and 15 mm. thick occurs in the Upper Aptian at Upware and Potton. It was originally described and figured as Terebratula tamarindus var. magna by Walker (1868 : 465, pl. 109, fig. 10) but agrees in every detail with the typical form and has proportionately similar dimensions. TJ. tamarindus is distinguished from other described species of Tamarella by its acute biconvexity and almost circular general outline. Tamarella bonneyi (Keeping) (Pl. 1, figs. 3a-c ; Text-fig. 7) 1883 Waldheimia bonneyi Keeping : 129, pl. 7, fig. 4. 60 SOME LOWER CREDNCEOUS LEME BRAT EVO TDi DESCRIPTION. Oval to pentagonal, somewhat cinctiform in outline, this species attains an approximate length of 32 mm. with a maximum width of 24 mm. and thickness of 17 mm. Each broad, flattened valve has a shallow anterior sulcation and is ornamented by numerous distinct or clearly defined concentric growth-lines. The umbo is not so produced as in other species of Tamarella but shows thesame broad, flat interarea bordered by sharp beak-ridges. Although constantly biconvex, the degree of convexity of each valve varies considerably in T. bonneyi. Some forms have an almost flat pedicle valve with an acutely convex brachial valve. Others are almost equally biconvex and, apart from their greater length, resemble 7. tamarindus from Upware. Fic. 7. A series of fourteen transverse serial sections through the umbo of Tamarella bonneyi (Keeping) from the Upper Aptian of Brickhill, Buckinghamshire. BM. B.25493. x2. HoLotyPe. B.26803 from the Lower Greensand, Upper Aptian of Brickhill, Buckinghamshire in the collections of the Sedgwick Museum, Cambridge. MaTERIAL. In addition to the holotype, thirty-eight specimens from Brickhill are in the Sedgwick Museum registered as B.80836—39, B.26509-19, B.25490-513 and thirty-six specimens in the general collections of the British Museum (Natural History) registered as B.25492-93, B.25499. REMARKS. There is a marked similarity between this species and Tamarella vesta sp. nov. and attempts have been made to illustrate gradation from onc species to SOME LOWER CRETACEOUS DEREBRATELLOIDEXA 61 another. A series registered B.25490~-531 in the Sedgwick Museum illustrates this variation. 7. bonneyi, however, appears to be confined to the deposits at Brickhill and 7. vesta is only rarely found at this locality while it is abundant at Upware. Tamarella bonneyi is distinguished from 7. vesta and other species of Tamarella mainly by its larger dimensions, broad cinctiform outline and more marked concentric growth-lines. Tamarella juddi (Walker) (Pl. 1, figs. 11a—c ; Text-fig. 8) 1868 Waldheimia rhomboidea Walker : 400, pl. 18, figs. 3, 4. 1870 Waldheimia juddi Walker : 562. 1874 Waldheinia juddi Walker ; Davidson: 50, pl. 7, figs. 15-18. DESCRIPTION. Tamarella about 19 mm. long, 14 mm. wide and 12 mm. thick. Elongate-oval in outline with a maximum width just less than half the distance from the umbo to the anterior margin. Acutely biconvex in lateral profile with marked marginal growth-lines. Short massive umbo and suberect beak with large foramen and sharp mesothyrid beak-ridges. An extensive interarea exposes conjunct deltidial plates. Anterior margin laterally compressed and incipiently uniplicate. LectotypPe. Although the species was originally described by Walker (1868 : 400) as Waldheinia rhomboidea, it has never been referred to under this name. Walker mistakenly thought the name to be preoccupied by Terebratula rhomboidea Barondi 1855, from the Tertiary of Italy and subsequently changed the name (1870 : 562) to Waldheimia guddi. As a junior synonym of W. rhomboidea, W. juddi has Fic. 8. Thirteen transverse serial sections through the umbo of Tamayella juddi (Walker) from the Upper Aptian of Upware, Cambridgeshire, BM, B.25681. x2. 62 SOME LOWER CRETACEOUS TEREBRATELLOIDEA since become entrenched in literature as the name of a well known fossil brachiopod and is maintained here. Walker’s (1868) original description was accompanied by figures of two specimens from the type locality at Upware which are in the general collections at the British Museum (Natural History) registered as 67602. The larger has been re-registered as BB. 42929 and is here selected as lectotype (PI. 1, figs. 11a—c). MATERIAL AND LOCALITY. In addition to the type material, there are fifty-eight specimens in the British Museum (Natural History) registered as B.25681 and B.25682. Among seven specimens in the Sedgwick Museum from the Upper Aptian of Faringdon, registered as B.18400-06, one specimen, B.18400, shows a well exposed brachial loop and cardinalia which appear to be very similar to Cooper’s figure of Psilothyris (Cooper 1955, pl. 3, fig. 19). The thickened portion of the descending branches can be clearly seen. RemARKS. Tamarella juddi is distinguished from other species of Tamarella by its acute biconvexity, extensive interarea, laterally compressed and tapering anterior and smaller foramen. Tamarella vesta sp. n. (Pl. 1, figs. ga—-c ; Text-fig. 9) 1868 Waldheimia pseudojurensis (Leymerie) ; Walker: 405, pl. 18, figs. 8-11. 1874 Waldheimia pseudojurensis (Leymerie) : Davidson: 48, pl. 7, figs. 12-14. DiaGnosis. Tamarella about 22 mm. long, 15 mm. wide and 10 mm. thick. Shell biconvex, elongate-oval to pentagonal in outline. Anterior commissure plane. Folding ligate. Test smooth with prominent growth-lines. Umbo slightly produced, suberect. Deltidial plates conjunct, well exposed. Beak-ridges sharp. Description. Although characteristically elongate with steep flanks and slightly produced umbo, the range of variation within this species includes forms which are broader and, in some cases, more inflated than the typical. Marked gerontic thicken- ing of the shell margins tends to accentuate the anterior sulcation of each valve. The generic characters seen in transverse serial sections (Text-fig. 9) show very clearly the overlapping hinge-plates which appear to be more marked in T. vesta than in the type-species T. tamarindus (J. de C. Sowerby). HototypPe. This was originally one of four specimens in the Walker Collection, British Museum (Natural History) which were used in Walker’s description of the species under the name Waldheimia pseudojurensis Leymerie. The holotype was figured by Walker (1868, pl. 18, fig. 8) and was collected from the Upper Aptian of Upware, Cambridgeshire. It has been re-registered as BB. 42904. MATERIAL AND LOCALITY. In addition to the holotype, forty-eight specimens from Upware registered as B.25498, B.25699, B.25701, B.25703. In the Davidson Collection there are nine specimens from Upware registered as B.6754 and three from Brickhill registered as B.6755, all in the British Museum (Natural History). REMARKS. Often confused with Terebratula pseudojurensis Leymerie from the Middle Neocomian of Marolles, Aube, France, Tamarella vesta nevertheless bears a superficial resemblance to this species. It differs fundamentally in its internal SOME TOWER CREDACEOUS DEREBRATLTELLOUDE A 63 Fic.9. Twenty transverse serial sections through the umbo of Tamarella vesta sp. n. from the Upper Aptian of Upware, Cambridgeshire. BM. B.25700. x2. generic characters which can be seen in serial section (Text-fig. 9). It is distinguished from 7. bonneyi (Keeping) by its more elongate outline, steeper flanks, more produced umbo and less prominent concentric growth-lines. It differs from 7. juddi and T. tamarindus in having flatter or less convex valves and ligate anterior margin. A common fossil in the Upper Aptian beds of Upware, T. vesta occurs, though less commonly, in beds of equivalent age at Brickhill, Buckinghamshire and at Potton, Bedfordshire. 64 SOME LOWER CRETACEOUS TEREBRATELLOIDEA Tamarella elliptica (Walker) (Pl. 1, figs. 6a—c) 1868 Waldheimia mutabilis var. elliptica Walker : 400, pl. 19, fig. 4. 1870 =Waldheimia mutabilis var. elliptica Walker ; Walker : 562. 1874 Waldheimia wanklyni var. elliptica Waller ; Davidson: 51, pl. 7, figs. 22-25. Description. Large oval Tamarella approximately 35 mm. long, 26 mm. wide and 15 mm. thick. Both valves are equiconvex and the lateral commissure is almost straight. Somewhat resembling 7. bonneyi (Keeping) in general outline and convexity, T. elliptica lacks the ligation of the anterior margin noted in the former species. A large circular foramen truncates a broad, flattened umbo. Sharp beak- ridges border a fairly extensive interarea. The shell surface is smooth apart from a trace of concentric growth-lines which become more prominent at the margins. Hototyre. In the British Museum (Natural History) registered number 67600, and figured by Walker (1868, pl. 10, fig. 4) as Waldheimia mutabilis var. elliptica. It was collected from the Upper Aptian of Upware. REMARKS. Walker (1868 : 400) used the name Waldheimia mutabilis for a brachio- pod from the Lower Greensand of Upware. No specimen was figured nor was any type material indicated. Instead, two forms which Walker called “ varieties or sub- species ’’ were erected and named W. mutabilis elliptica and W. mutabilis angusta. The descriptions of these were accompanied by illustrations (1868, pl. 19, figs. 4, 5) of two specimens now in the British Museum (Natural History) and registered as 67600 and 67601 respectively. Walker’s conception of the species W. mutabilis having been established on the description of the two “‘ varieties or sub-species ’’, is, therefore, indefinable. In any case, the name Waldheimia mutabilis had already been used by Oppel (1861) for a zeilleriid brachiopod from the Lower Lias of Germany, and it was for this reason that Walker asked Davidson (1874 : 51) to correct the name to wankylni in his Cretaceous monograph. The varietal name edliptica is here raised to specific rank and referred to Tamarella gen. nov. The varietal name angusta has also been raised to specific rank and referred to Vectella gen. nov. on p. 53. Both species are common members of collec- tions of brachiopods from Upware and are easily recognizable. Genus RUGITELA Muir-Wood, 1936 Rugitela roemeri sp. n. (Pl. 2, figs. 1a—c, 3a-c, 6a—c, 8a—c, ga—-c; Text-fig. 10) 1836 Tevebratula longa Roemer : 50, pl. 2, fig. 11. 1839 Terebratula longa Roemer ; Roemer : 22, pl. 18, fig. 12. 1840 Tevebratula longa Roemer ; Roemer: 44, No. 50. 1847 Terebratula faba Sowerby ; d’Orbigny : 77, pl. 506, figs. 8, 9, I1, 12. 1864 Tevebratula (Waldheimia) faba Sowerby ; Credner : 561, pl. 21, figs. 3, 4, 5. 1868 Terebratula longa Roemer ; Quenstedt : 338, pl. 4,6 fig. 99. Driacnosis. Rugitela about 28 mm. long, 18 mm. wide and 14 mm, thick, Shell SOME LOWER CRETACEOUS TEREBRATELLEOIDEA 05 biconvex, oval to subtriangular in outline. Umbo short, massive, suberect. Exten- sive interarea bordered by sharp, mesothyrid beak-ridges. FYoramen large, circular. Deltidial plates well exposed, conjunct. Anterior margin rectimarginate to sulcate. Ho.Lotyre. British Museum (Natural History) no. BB.42912, from the Neocomian, Hilsconglomerat of Elligser Brinke, near Hanover, North Germany. According to Judd (1870 : 331) the “ Elligser Brinke Schist ’’ has its English equivalent in the Acanthodiscus speetonensis Zone at Speeton, Yorkshire and in Lincolnshire. DEscRIPTION. This species is characteristically elongate-oval in outline with marked growth-lines becoming still more prominent at the margins. Gerontic thickening of the anterior and lateral margins is a common feature. Some variants show a tendency to broad anterior sulcation of the brachial valve. Internal characters. Short, slightly curved, divergent dental lamellae are embedded in thick callus and support inwardly curving hinge-teeth. Cardinal process not developed. A short septalium develops with long, fused hinge-plates which are deflected ventrally forming a small anterior plication. A long median septum exten- ding two-thirds the length of the shell supports the hinge-plates. Inner and outer socket-ridges are extensive and well marked. Triangular crural bases give rise to thick descending branches of the brachial loop with dorsal development and short crural processes. [airly broad transverse band developed from ascending branches of the loop. A thickening of the descending branches at a point about one-third the distance from the hinge-plates has been noted in mature forms so far sectioned and may be due to resorption of the attachment branches to the median septum in the dorsal valve. No attachment of brachial loop to septum has been noted in any sections of either young or mature forms. PCC) Fic. 10. Eleven transverse serial sections through the umbo of Rugitela voemeri sp. 0. from the Neocomian of Berklingen, North Germany, showing the characteristic W-shaped hinge-plates and thickened area of the descending branches of the brachial loop. BM. B.35056. x2. Remarks. Zieten (1830: 52) was the first to use the binomen Terebratula longa for a terebratelloid brachiopod which he figured (pl. 39, fig. 7) from the “ calcaire jurassique ’ [Upper Jurassic] of Donzdorf, North Germany. Roemer’s description of Terebratula longa (1836: 50, pl. 2, fig. 11) from the Neocomian of Hanover was, therefore, invalid. 66 SOME VOW ER CRETNCEOWS WE RE Bika Dh Ey eO Disa D’Orbigny (1847 : 77, pl. 506, figs. 8-12) described and figured a brachiopod from the Neocomian of St. Dizier, Haute Marne, France, and assigned it to Terebratula faba Sowerby, a name which J. de C. Sowerby (1836 : 338) had originally used in a des- cription of a brachiopod from the Lower Greensand beds between Folkestone and Sandgate, Kent. Specimens collected from the Lower Albian beds of Folkestone in the Sedgwick Museum (B.17633-17650), have been matched with others collected from the Leymeriella tardefurcata Zone of Folkestone by Dr. R. Casey and compare favourably in size, general outline and convexity with Terebratula faba as figured by J. de C. Sowerby (1836, pl. 14, fig. 10). A duplicate specimen (Brit. Mus (N.H.) no. B.25994) has been sectioned (Text-fig. 12) and is shown to belong to the genus Modestella Owen (1961 : 573) described from the Lower Albian of Folkestone. D’Orbigny’s illustrations (1847, pl. 506, figs. 8-12) were probably an idealization of eight specimens which were collectively numbered 5159 in the d’Orbigny Collection, Laboratoire de Paléontologie, Muséum National d’Histore Naturelle, Paris. Unfor- tunately only four of these specimens can be traced. One has been identified as a specimen of Vectella celtica (Morris) from the Lower Greensand of the Isle of Wight, and another (see Pl. 2, figs. 8a—c) is almost identical with the specimen figured by Roemer (1839, pl. 18, fig. 12) from the Neocomian of Hanover. Rugitela roemeri occurs rarely in Britain, being found in the upper part of the Claxby Ironstone, Lincolnshire and in the Speeton Clay of Yorkshire. A fine, well preserved specimen (Pl. 2, figs. 1a—c) from the Claxby Ironstone of Nettleton, Lincolnshire, collected by Mr. Peter Rawson of Hull University, is the only example known from these beds which compares exactly with Roemer’s original illustration (1836, pl. 2, fig. 11). The dimensions of this specimen are: 40 mm. long, 20 mm. wide and 16 mm. thick. A form described here as Rugitela rugosa sp. n. resembles R. roemeri but is shorter, more gibbous and oval in outline with no marked sulcation of the anterior portion of the brachial valve. It is found in the same beds of the Claxby Ironstone and is probably present in the Speeton Clay, although specimens collected from beds within the range C6 to C3 of Swinnerton (1936-55) and which may belong to this species are too crushed for accurate determination. J. rugosa also occurs in the Hauterivian at Salins-les-Bains in the French Jura Mountains. A well preserved example is figured here (Pl. 2, figs. 5a—d) for comparison. The reference of this species to the genus Rugitela has been made after careful comparison with transverse serial sections of the type-species R. bullata Muir-Wood and extends the range of the genus from ? Inferior Oolite to the Lower Cretaceous. It is possible that the final range may include species from the Inferior Oolite at present under review by Dr. H. M. Muir-Wood. MATERIAL AND LOCALITY. Well preserved material from the German Neocomian is rare but there are, in addition to the holotype metioned above, nine other specimens in the Davidson Collection, Brit. Mus. (Nat. Hist.) B.6759, BB.41g911, from the “ Elligser Brinke Schist ’’ of Delligsen near Hanover. A specimen from the Neocomian of St. Dizier in the Haute-Marne is also in the Davidson Collection (B.6736). It is accompanied by the note in Davidson hand- writing which reads: “... given to me by dOrbigny . . . from the d’Orbigny Coll. yaaa SOME LOWER CRETACEOUS TEREBRATELLOIDEA 67 Seven specimens, British Museum (Natural History) (BB.42925~—31), from the Acanthodiscus speetonensis Zone of the Speeton Clay, Yorkshire. Although crushed they compare favourably with the North German forms. Rugitela rugosa sp. n. (Pl. 2, figs. 2a—c, 4a—c, 5a—c ; Text-fig. 11) 1874 Waldheinua faba d’Orbigny ; Davidson: pl. 6, figs. 12-14. DiaGnosis. Biconvex Rugitela about 29 mm. long, 17 mm. wide and 17 mm. thick. Oval to sub-circular in outline. Shell surface covered with pronounced growth- lines. Anterior and lateral margins thickened. Umbo short, slightly incurved. Interarea extensive. Deltidial plates conjunct. Beak-ridges sharp. Anterior commissure, rectimarginate. Ho.otyre. Davidson Collection, British Museum (Natural History) no. B.673 from the Claxby Ironstone (Lower Hauterivian) of Acre House, Tealby, Lincolnshire. MATERIAL. In addition to the holotype, four specimens from the Claxby Iron- NCCT Ry oh ae mate J 2 re, Fic. 11. A series of sixteen transverse sections through the umbonal part of Rugitela rugosa sp. n. from the Neocomian, Claxby Ironstone, Acre House, Tealby, Lincolnshire. BM. B.50359. 2. 68 SOME LOWER GRELNCE OMS DERE BRAD Ewe Ow Dias stone, British Museum (Natural History) nos. B.6734, B.50358—59, B.50324 and two well preserved specimens from Salins-les-Bains, nos. 33984 and BB.42913. There are two specimens from the Claxby Ironstone in the Sedgwick Museum (B.11400 and Be2278)e Description. In general outline this is a shorter, more gibbous species than 0-3 Fic. 12. Eleven transverse serial sections through the umbo of Modestella faba (J. de C. Sowerby) from the Lower Albian, Folkestone, Kent. BM. B.25994. 4. SOME LOWER CRETACEOUS TEREBRATELLOIDEA 69 Rugitela roemert. It has more acutely convex valves with an absence of any marked sulcation of the brachial valve, although some mature forms exhibit a faint depression in this part of the shell. The surface of the valves is ornamented by numerous con- centric growth-lines which give it a rugose appearance. Marginal thickening of the valves is a common feature in more mature forms. The umbo is less produced and more incurved than in R. voemeri. Well defined, sharp beak-ridges border an extensive interarea. LOCALITY AND HORIZON. This species enjoys the same lateral range as Rugitela vyoemeri, occurring in the Lower Hauterivian at Salins-les-Bains in the French Jura Mountains ; a similar horizon at Brunswick, North Germany, and in the Claxby Ironstone of Lincolnshire, as well as the Acanthodiscus speetonensis Zone of the Speeton Clay of Yorkshire. Rugitela hippopus (Roemer) (Pl. 3, figs. 1a—c, 2a—c, 3a-c, 4a—c; Text-fig. 13) 1840 Terebratula hippopus Roemer : 114, pl. 16, fig. 28. 1842 Terebratula hippopus Roemer ; Geinitz : 87. 1864 Tevrebratula (Waldheimia) hippopus Roemer ; Credner : 565, pl. 21, figs. 17, 18, 19. 1866 Tevebratula hippopus Roemer ; Schloenbach : 33. 1868 Tevebratula ? defluxa Schloenbach : 31, pl. 2, figs. 10-12. 1868 Tevebratula hippopus Roemer; Judd: 245. 1874 Waldheimia hippopus ? Roemer var. tilbyensis Davidson : 53, pl. 6, figs. 10, IT. 1884 Tevebvatula (Waldheimia) hippopus Roemer ; Weerth : 62, pl. 11, figs. 5, 6. DEscRIPTION. This well established species is characterized by its almost circular outline, sulcate brachial valve, slightly incurved beak and acutely carinate pedicle valve. It is 12 mm. long, 10 mm. wide and 8 mm. thick. Because of the similarity of external form it has often been assigned to the Liassic genus. Aulacothyris. It differs from this genus, however, in having a less elongate outline, smaller dimensions, larger foramen and longer brachial loop. Internal characters. From the transverse serial section of the species (Text-fig. 13) it is possible to compare the internal characters of R. hippopus with sections of species of Rugitela figured by Muir-Wood (1936: 124, 128, 131). While agreeing basically with sections of the type-species Rugitela bullata from the Fuller’s Earth Rock of Somerset, it appears to have closer morphological affinities with R. emarginata from the same horizon and locality. Neotyre. Dr. R. G. Thurrell of H.M. Geological Survey has kindly given me permission to publish some of the notes he prepared in a revision of this species which he included in his thesis for the degree of Ph.D of London University, 1957. The following is a quotation from p. 320 of his thesis : “ Owing to war damage, the Roemer Collection in Hildesheim Museum was partly destroyed. Dr. Friedrich Smid of the German Geological Survey has undertaken to search for the type material, but nothing has so far come to light.”’ In spite of renewed enquiry no specimens have been forthcoming. A specimen in the British Museum (Natural History) no. 32313 from the Hilscon- glomerat of Berklingen, North Germany (PI. 3, figs. 2a—c) appears to be the only 70 SOME LOWER CRETACEOUS TEREBRATELLOIDEA Fic. 13. A series of eleven transverse serial sections through the umbo of Rugitela hippopus (Roemer) from the Neocomian, Claxby Ironstone, Acre House, Tealby, Lincolnshire. BM. BB.42934. x2. available specimen from the type locality, and is here proposed as neotype. It departs from the original illustration (Roemer 1840, pl. 16, fig. 28) in having less clearly defined beak-ridges and less extensive interarea but, after examination of some 150 specimens from both English and North German localities, these features appear to vary within the species. MATERIAL AND LOCALITY. Numerous specimens from the Claxby Ironstone from Acre House, Tealby and from Nettleton, Lincolnshire. Ten specimens from the Neocomian of Saltzgitter, Hanover (B.6748, BB.42921, BB.42932), a single speci- men from the Speeton Clay of Yorkshire (BB.42923), and one specimen from Marolles, Aube, France (B.6742), all in the British Museum (Natural History). ReMARKS. Differing only in size from the North German form, the British species, originally described as Waldheimia hippopus var. tilbyensis by Davidson (1874), attains an average length of 15 mm., a width of 13 mm., and a thickness of Io mm. The specimen from the Lower Hauterivian of Marolles, France, differs slightly in beak characters. It has less prominent beak-ridges and less extensive interarea. CONCLUSION Owing to the lack of recent exposures of the Lower Greensand at Upware, Brickhill and Potton, it is impossible to draw any real conclusions about the distribution of species. All information regarding these localities has been obtained from research on museum collections and the results recorded here. Critical examination of transverse serial sections of species of Tamarella gen. nov. has revealed that they have many characters in common with those of Rugitela. For example, the divergent dental lamellae, broad, shallow septalium, anteriorly SOME LOWER CRETACEOUS TEREBRATELLOIDEA 71 arched and fused hinge-plates, triangular crural bases and thickened portion of the descending branches of the brachial loop. The dorsal median septum of Rugitela is, however, more strongly developed and considerably longer than that of Tamarella. Prozorovskaya (1962 : 111) described Gusarella from the Upper Jurassic of Turk- menistan as a subgenus of Zezlleria. Serial sections of the type species Z. (Gusarella) gusarensis have much in common with those of Tamarella particularly T. bonneyi (Text-fig. 7). The hinge-plates show the anterior arching or plication, and the crural bases, brachial loop and general outline are similar to those of Yamarella. In Gusarella, however, the dorsal median septum appears to be absent. ACKNOWLEDGMENTS Iam grateful to Dr. H. M. Muir-Wood for her kindness and generosity in many ways and help with the manuscript and also for her continued encouragement. I am also grateful to Dr. W. T. Dean for many helpful suggestions and for some of the photographs. My thanks are also due to Mr. John Ferguson for photographic assistance and helpful discussion ; to Mr. Peter Rawson of the Dept. of Geology, Hull University ; to Dr. Colin Forbes, Sedgwick Museum, Cambridge and to Dr. R. G. Thurrell of H.M. Geological Survey. REFERENCES CasEy, R. 1961. The stratigraphical palaeontology of the Lower Greensand. Palaeontology, London, 3 : 487-621, pls. 77-84. CoopEer, G. A. 1955. New Cretaceous Brachiopoda from Arizona. Smithson. Misc. Coll., Washington, 131, 4: 1-18, pls. 1-4. CREDNER, H. 1864. Die Brachiopoden der Hilsbildung im nordwestlichen Deutschland. Z. dtsch. geol. Ges., Berlin, 16 : 542-572, pls. 18-21. Davipson, T. 1852-55. A Monograph of British Cretaceous Brachiopoda, 1. 117 pp., 12 pls. {Mon. Palaeont. Soc., London.| 1874. A Monograph of the British Fossil Brachiopoda, 4, 1. Supplement to the Recent, Tertiary, and Cretaceous Species. 72 pp., 8 pls. [Mon. Palaeont. Soc., London.| Davipson, T. & Morris, J. 1847. Descriptions of some species of Brachiopoda. Ann. Mag. Nat. Hist., London (1) 20 : 250-257. Fitton, J. 1836. Observations on some of the strata between the Chalk and Oxford Oolite in south-east of England. Tyvans. Geol. Soc. Lond. (2) 4 : 335-349, pls. 11-23. Gerinitz, H. B. 1842. Ueber Versteinerungen Altenburg und Ronneburg. Mitth. Osterl. Altenburg, 6 : 86-99. Jupp, J. W. 1868. On the Speeton Clay. Quart. J. Geol. Soc. Lond., 24 : 218-250. 1870. Additional observations on the Neocomian strata of Yorkshire and Lincolnshire, with notes on their relations to the beds of the same age throughout Northern Europe. Quart. J. Geol. Soc. Lond., 26 : 326-348. KEEPING, W. 1883. The fossils and palaeontological affinities of the Neocomian deposits of Upware and Brickhill. Sedgwick Prize Essay foy 1879. xi + 167 pp., 8 pls. LANKESTER, E. R. 1863. Oncertain Cretaceous brachiopoda. Geologist, London, 6 : 414-415. Lreymerig£, A. 1841-42. Mémoire sur le terrain Crétacé du Département de l’Aube, contenant des considérations générales sur le terrain Néocomien. Mém. Soc. géol. Fr., Paris (1) 4 (1841) : 291-364, pls. 16-18, 5 (1842) : 1-34, pls. 1-18. Meyer, C. J. A. 1864. Notes on the Brachiopoda from the Pebble-bed of the Lower Green- sand of Surrey ; with descriptions of the new species, and remarks on the correlation of the Greensand Beds of Kent, Surrey and Berks., and of the Farringdon Sponge-gravel, and the Tourtia of Belgium. Geol. Mag., London (1) 1 : 249-257, pls. 11-12. 72 SOME LOWER CRETACEOUS TEREBRATELWTOIDEA Meyer, C.J. A. 1868. Notes on Cretaceous Brachiopoda and on the development of the loop and septum in Tevebratella. Geol. Mag., London (1) 5: 268-272. Mippremiss, F. 1959. English Aptian Terebratulidae. Palaeontology, London, 2 : 94-142, pls. 15-18. Morris, J. 1843. TAL WY BY RE BATE PP. 73-133 ; 21 Plates; 24 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Wolsrn Now 3 LONDON : 1965 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), «mstituted in 1949, 1s issued in five series corresponding to the Departments of the Museum, and an Historical serves. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. In 1965 a separate supplementary series of longer papers was instituted, numbered serially for each Depariment. This paper is Vol. 11, No. 3 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1965 TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) Issued 8 December, 1965 Price £4 MIDDLE JURASSIC OSTRACODA FROM THE GREY LIMESTONE SERIES, YORKSHIRE By RAYMOND HOLMES BATE CONTENTS page I INTRODUCTION AND ACKNOWLEDGEMENTS : : : : 5 9K®) II STRATIGRAPHICAL SECTIONS . : : c : - - 80 III STRATIGRAPHY . F é A : 4 : 5 z 92 IV CONCLUSIONS . ; : : : : : : : 5 OF V SYSTEMATIC DESCRIPTIONS . ; . 5 : ‘ . ‘ 98 Subclass Ostracoda Latreille . : : : : ; 5 Ge} Order Podocopida Miiller : ‘ : : ‘ ; . 98 Suborder Podocopina Sars. : c F F : . 98 Superfamily Cypridacea Baird. : : : : . 98 Family Paracyprididae Sars : ; : é - 98 Genus Paracypris Sars . . : é : . 98 Paracypris bajociana Bate . : : . 98 Superfamily Cytheracea Baird. : : : . - 99 Family Bythocytheridae Sars : é : : 5 6) Genus Monoceratina Roth . ; : - 99 Monoceratina scarboroughensis sp.nov. . 3 99 Family Progonocytheridae Sylvester-Bradley. 6 . 100 Subfamily Progonocytherinae SF iy me . 100 Genus Caytonidea nov. . : . 100 Caytonidea faveolata Sp. nov. : : . 100 Genus Cloughtonella nov. : : : : . I01 Cloughtonella rugosa sp. nov. . . c . 102 Genus Fuhrbergiella Brand & Malz ¢ 5 - 104 Subgenus Praefuhrbergiella Brand & Malz . 104 Fuhrbergiella (Praefuhrbergiella) hovrida horrida Brand & Malz . , . 104 Genus Glyptocythere Brand & Malz é é . 106 Glyptocythere costata sp. nov. 3 < . 106 Glyptocythere polita sp. nov. é o . 107 Glyptocythere scitula sp. nov. ; 0 - 108 Genus Mailzia nov. 4 : : ; . I10 Malza bicarinata sp.nov. . 5 : o | 1yuar Malzia unicarinata sp. nov. : 9 q 3003) Genus Progonocythere Sylvester-Bradley . : . II4 Progonocythere acuminata sp. nov. ' - <4 Progonocythere yonsnabensis sp.nov. . . I16 Subfamily Pleurocytherinae Mandelstam . é 5 yy Genus Pleurocythere Triebel . é : : ) FI7 Pleurocythere sp. : C 3 : qe EI, GEOL. 11, 3 76 WR st, IAI page Family Cytherideidae Sars . : ‘ o ; = D7, Subfamily Cytherideinae Sars . é : ‘ ap ty Genus Vernoniella Oertli 5 ; ; P ame GER Vernoniella bajociana sp. nov. 0 : . «118 Vernoniella ? caytonensis sp. nov. ; 35 1G) Genus Ljubimovella Malz , é : 5 7 20 Ljubimovella piriformis Malz 2 ; | 120 Family Schulerideidae Mandelstam F 6 c q erat Subfamily Schulerideinae Mandelstam 5 : 5 Genus Mesocytheridea nov. . é 2) 2 Mesocytheridea howardianensis sp. nov. 22 Genus Praeschuleridea Bate . : 1. £24 Praeschuleridea subtrigona (Jones & Sherborn) 124 Praeschuleridea subtrigona inteymedia subsp. noy. : ; : : : : = pled Family Cytheruridae Miiller. : . : : E20) Genus Eocytheropteron Alexander . é : 5 1tfo) Eocytheropteron ? sp. . : : : 3 r26 Genus Paracytheridea Miiller . . ; 27 Paracytheridea ? caytonensis sp.nov. . = A227 Family Protocytheridae Ljubimova : é : +) £28 Subfamily Kirtonellinae Bate . : : : 5 te4S) Genus Southcavea Bate . : : : 228 Southcavea microcellulosa Sp. nov. ; 2 £28 Genus Systenocythere Bate. : : : gO Systenocythere ovata sp. nov. : 6 = | 30 VI REFERENCES : 6 c . : : : : 0 ; £32 SYNOPSIS The Middle Jurassic (Bajocian) Grey Limestone Series of Yorkshire is shown to have been deposited in a broad embayment, transgressing deltaic sediments of Bajocian age, and is here divided into two ostracod faunal zones : a lower zone of Glyptocythere polita and an upper zone of G. scitula, the type section being at Hundale Point. The zone of G. polita is present only towards the centre of the basin and represents the first phase of the marine transgression. The facies changes within the Series are produced by intermittently rising sea-level and the repeated southward extension of the northern delta. Shore-line sediments are also identified. Fourteen stratigraphical sections are described in detail. The period of marine deposition is considered to have taken place approximately during sauzei to blagdeni times and is thus in part at least equivalent to the Coronaten-Schichten of N.W. Germany. The ostracod fauna of the Grey Limestone Series is described for the first time. Four genera: Caytonidea, Cloughtonella, Malia and Mesocytheridea are new. Sixteen new species and one new subspecies are also described. The palaeoecology of the ostracod fauna is briefly mentioned. I INTRODUCTION AND ACKNOWLEDGEMENTS In north-eastern England, the uppermost marine horizon of Bajocian age consists of a well developed succession of calcareous shales and sandy limestones known as the Grey Limestone Series or Scarborough Beds. The maximum thickness of this sequence occurs in the neighbourhood of Cloughton Wyke (62 feet) and Ravenscar (62 feet), the beds thinning to the north, south and west where they eventually pass MIDDLE JURASSIC OSTRACODA 77 NORTH SEA RAVENSCAR 5 CLOUGHTON WYKE 4 A SCARBOROUGH 5 PICKERING , 1 FORDON NOI MARKET WEIGHTON@ 2 SOUTH ZN HULL 01234 8B miles ——S— KIRTON LINDSEY Fic. 1. Outcrop of Middle Jurassic Strata in N.E. England, with the localities and sections (I-14) mentioned in the text. 78 R. H. BATE into arenaceous shore-line deposits. The open sea during this period lay to the east, occupying the site of the present day North Sea, whilst a delta lay to the north. Sediments equivalent in age to the Grey Limestone Series were also deposited in north-western Germany. The dating of the Grey Limestone Series is difficult owing to the almost complete absence of ammonites : those which have been recorded (Fox-Strangways 1892 : 231 ; Buckman IgII : 205-208; and Hemingway 1951 : 119)! indicate that part of the sequence at least belongs to the Teloceras blagdent Subzone. As the ammonite zones are well known in north-western Germany, comparison of the ostracod faunas of the two regions should eventually offer a solution to this problem. At the moment, however, only two ostracods, present in the Grey Limestone Series have also been, recorded from Germany ;_ these are Ljubimovella piriformis Malz (1961 : 165, pl. 2, figs. 15-25) and Fuhrbergiella (Praefuhrbergiella) horrida horrida Brand & Malz (1962 : 19, pl. 4, figs. 33-37; pl. 5, fig. 46). Ljubimovella piriformis has a range in Germany of sauzer to blagdent Zones, with the type horizon being in the sauwzei Zone. In Yorkshire this ostracod is restricted to the zone of Glyptocythere scitula, the upper zone from which the ammonites recorded from the Grey Limestone Series were most probably obtained. The ostracod Fuhr- bergiella (Praefuhrbergiella) horrida horrida is recorded from the upper part of the Coronaten-Schichten (romani to blagdeni Zones) in Germany, whilst in Yorkshire it is TABLE I Generalised table of the Middle Jurassic stata in Yorkshire, north of Market Weighton Strata Stage CORNBRASH (UPPER) CALLOVIAN UppEeR DELTAIC SERIES | BATHONIAN GREY LIMESTONE SERIES MIDDLE DELTAIC SERIES (UPPER) MILLEPORE SERIES MIDDLE DELTAIC SERIES (LOWER) BaAJOCIAN ELLER Beck BED/HyDRAULIC LIMESTONE Lower DELTAIC SERIES DOGGER —— Lias F TOARCIAN 1 The map reference for Monk’s Walk Wood, Sneaton has been given with a misprint, and should read NZ/896086 (personal communication, Prof. Sylvester-Bradley). 1 1 1 NORTH r w 0) 2 il ~ ox ty 19) ra moe || ter = ow oo os | Oz 3 fa z 4 12 6 wo | = | e = ll 1 | | Sty, AL i Hi iil Bu A SOUTH 1. Carsnoere may MIDDLE JURASSIC OSTRACODA 79 largely restricted to the G. scitula zone, being found at only one locality in the G. polita Zone. ; The Grey Limestone Series represents a period of deposition which appears to approximate with the Coronaten-Schichten of Germany although there is insufficient evidence to say how many of the German ammonite zones are represented in the Yorkshire sediments. It should be possible, however, to correlate the Grey Lime- stone Series with the German succession more precisely at a future date when the ostracod faunas are more fully known. Fox-Strangways (1892 : 223) recorded the presence of lamellibranch casts within a 15 ft. bed of sandstone at Ravenscar. This is bed 17 of Section 5 (Text-fig. 2), which is the lower part of the Moor Grit (basal Upper Deltaic Series). The period of marine deposition was thus brought to a close by the infilling of the Yorkshire basin with deltaic sediments. The next marine transgression did not occur until Callovian times, it is, therefore, evident that the end of the Bajocian and the whole of the Bathonian is represented in Yorkshire by deltaic sediments of the Upper Deltaic Series. The localities listed below are those at which the Grey Limestone Series has been examined at outcrop, the numbers corresponding with those in Text-fig. 1. Map references refer to the one-inch Ordnance Survey map of Great Britain (seventh series). Coastal Exposures 1. Gristhorpe Bay, cliff section and foreshore, map reference TA/085842. A complete section, maximum thickness 15 ft. 6 in. 2. Cayton Bay, cliff section, map reference TA/082843. A complete section, maximum thickness fro ft. 7 in. 3. White Nab, Scarborough, foreshore section, map reference TA/058865. Base below Low Water Mark, seen to 25 ft. 6 in. 4. Hundale Point, Cloughton Wyke, cliff and foreshore section, map reference TA/o024949 to TA/o27943. A complete section, maximum thickness 62 ft. 4in. 5. Ravenscar, cliff section, map reference NZ/988012. A complete section, maxi- mum thickness 62 ft. 4 in. Inland Exposures 6. Hawsker, stream section at Hawsker Bottoms, map reference NZ/937079. Top of section not seen, 22 ft. 5 in. 7. Bloody Beck, stream section, map reference SE/945981 to SE/947982. A complete section, 29 ft. 4 in. 8. May Beck, stream section, map reference NZ/8go0015 to NZ/891019. Isolated exposures, base not seen, 14 ft. 9 in. g. Ramsdale Beck, stream section NZ/925034. Upper part of Series only exposed, seen to 9 ft. 6 in. 80 Re Hy BAe to. Eller Beck, just downstream of road bridge, map reference SE/856984. Upper part of Series only, seen to 6 ft. 5 in. 11. Harland Beck, stream section, map reference SE/684914 to SE/686913. In- complete section seen to 14 ft. 4 in. 12. Bogmire Gill, stream section, map reference SE/608905 to SE/60g909. Almost a complete section, seen to 40 ft. 3 in. 13. Yearsley Moor, map reference SE/579754. Asmall quarry by the side of the Yearsley-Ampleforth road. Fauna indicates position of beds high up in the Grey Limestone Series, seen to 8 ft. 3 in. 14. Stonecliff Wood, Crambeck, map reference SE/744676, a small gully exposing 12 ft.2in. of sediment. Map reference SE/743676, section higher up the bank, above the York—Scarborough railway-line, exposing 5 ft. 8 in. of limestone. Map reference SE/740675 a 9 inch bed of limestone is exposed. All strata high up in the Grey Limestone Series. Morphological terms used in the text are those introduced in vol. Q of the American Invertebrate Paleontology Treatise (Moore 1961), Sylvester-Bradley (1956) and Bate (1963). The work embodied in this paper was largely conducted in the Geology Depart- ments of Sheffield University and Leicester University. My sincere thanks go to Professors L. R. Moore and P. C. Sylvester-Bradley respectively, for the use of their departmental facilities. Grateful thanks are also due to the Department of Industrial and Scientific Research for a grant to finance this work. Dr. H. Malz, Senckenberg Museum, Frankfurt-am-Main and Dr. H. J. Oertli, S.N.P.A., Centre de Recherches, Pau (Basses-Pyr) also assisted with the loan of type and/or comparative material, for which they are gratefully acknowledged. II STRATIGRAPHICAL SECTIONS SECTION No. I. Gristhorpe Bay (Text-fig. 2), exposure in south face of Yons Nab headland and on the foreshore at low tide. The section is complete and exposes 15 ft. 6 in. of sediment (G.L.S.). Upper Deltaic Series fit. ™. 10. Yellow-grey sandstone (Moor Grit) with coaly lenses and grey sul- phurous shale . 2 ; : : ; : : . 30-40 Oo Grey Limestone Series g. Black, sulphur stained shale 4 fo) 8. Dark grey, ironstained shale 4 9 7. Mudstone ; , ; : 0) 9 6. Black shale with indet. ostracods I 4 UPPER DELTAIC SERIES GREY LIMESTONE SERIES MIDDLE DELTAIC SERIES 7. BLOODY BECK UPPER DELTAIC SERIES UMESTONE WDveassier Moon 1 I 1 i ; 1 MIODLE DELTAIC ions 7 and 12-14. MIDDLE JURASSIC OSTRACODA 5. Fossiliferous mudstone with Glyptocythere scitula; Vernoniella bajociana and Monoceratina scarboroughensis : ; 4. Black fossiliferous shale with a compact limy band at base. Lamil- libranch Gervillia scarburgensis and Belemnitescommon. Glypto- cythere scitula ; Praeschuleridea subtrigona intermedia ; V ernontel- la bajociana ; Vernoniella ? caytonensis ; Systenocythere ovata ; Paracypris bajociana; Caytonidea faveolata; Fuhrbergiella (Praefurbergiella) horrida horrida and Pleurocythere sp. Middle Deltaic Series 3. Yellow sandstone, flaggy in part with carbonaceous bands 2. Black Shale I. Grey shale SECTION No. 2. Cayton Bay (Text-fig. 2), exposure in north face of Yons Nab headland. This section is complete, but like Section No. 1 is liable to be cut into by washouts infilled with deltaic sandstone (Moor Grit). Up to ro ft. 7 ins. of sediment (G.L.S.) exposed. Upper Deltaic Series 8. Moor Grit. Grey Limestone Series 7. Chocolate-brown to grey coloured shale with fragmentary shells 6. Fossiliferous, chocolate-brown mudstone, variable in thickness. Glyptocythere scitula; Vernoniella bajociana; Vernoniella ? caytonensts ; Systenocythere ovata ; Paracytheridea ? caytonensis ; Progonocythere acuminata ; Progonocythere yonsnabensis and Praeschuleridea subtrigona intermedia 5. Grey fossiliferous shale with: Glyptocythere scitula ; Vernontella bajociana ; Fuhrbergiella (Praefuhrbergiella) horrida horrida ; Monoceratina scarboroughensis ; Praeschuleridea subtrigona inter- media ; Progonocythere yonsnabensis and Caytonidea faveolata . 4. Brown fossiliferous mudstone with: Glyptocythere scitula and Vernoniella bajociana : : ‘ ; 3. Grey shelly shale with : Ga praythere Sona Vernoniella bajociana and Praeschuleridea subtrigona intermedia. : 4 : 2. Grey-black sulphurous shale Middle Deltaic Series I. Flaggy yellow sandstone. ft. of to mn. Io 8r 82 IR ES Aa: SECTION No. 3. White Nab (Text-fig. 2), Scarborough. Exposure largely along the foreshore, cut out as a series of steps by the sea —well exposed and possibly almost complete, lowermost beds below Low Water Mark. 25 ft. 6 in. of marine sediment exposed. r8. Upper Deltaic Series Massive deltaic sandstone (Moor Grit) forming base of high cliffs. Grey Limestone Series . Sandy shale passing into base of Moor Grit above : : . Fossiliferous, grey calcareous shale, partly ironstained. Fuhrber- giella (Praefuhrbergiella) horrida horrida . Fossiliferous shale with mudstone nodules. Internal casts, possibly of Glyptocythere scitula and Praeschuleridea subtrigona intermedia . Mudstone with Glyptocythere scitula ; Vernomella bajociana and Praeschuleridea subtrigona intermedia . Grey shale with ooliths . Grey to chocolate-brown foseilieeons Sreils wath Chance scitula ; Vernoniella bajociana and Praeschuleridea subtrigona intermedia . Ironstained mudstone crowded with the Jaymellibranch Gorvillia scarburgensis . Sandy limestone Containine ioree specimens of Gales Seinen sis Glyptocythere scitula and Praeschuleridea subtrigona intermedia . Purplish-brown calcareous shale with Glyptocythere scitula ; Ver- nontella bajociana and Praeschuleridea subtrigona intermedia . Ironstained, sandy limestone. Vernoniella bajociana ; y . Fossiliferous grey shale with Glyptocythere scitula; Vernoniella bajociana ; Monoceratina scarboroughensis and Praeschuleridea subtrigona intermedia . Calcareous shale with mndeeone hed at 408 eh varies Roun 0 in. to g in. in thickness . Massive, well bedded sandy Uarsgione, hese Pret no microfauna so far obtained . Hard calcareous shale, almost an qeedllaceous imrestes . Soft, grey fossiliferous shale. Glyptocythere scitula ; M poem scarboroughensis ; Fuhrbergiella (Praefuhrbergiella) horrida hor- vida and Praeschuleridea subtrigona intermedia . . Ironstained limestone ‘ : j : . Grey sandy shale with large ‘yeleneaiiies. Glyptocythere scitula ; Vernoniella bajociana and Praeschuleridea subtrigona intermedia seen to ft. in. Ho MIDDLE JURASSIC OSTRACODA ft SECTION No. 4. Hundale point (Text-fig. 2). A complete succession of these beds is exposed in the cliff face and along the foreshore, with a maximum thickness of 62 ft. 4 in. Upper Deltaic Series 32. Moor Grit—carbonaceous and more thinly bedded at base. Grey Limestone Series 31. Grey, sandy and micaceous shale, ironstained 30. Dark grey sulphurous shale : 2g. Fossiliferous mudstone, variable in dnicisess maximum 28. Grey shale, ironstained at base . fe) fe) Tal 5) 27. Chocolate-brown, fossiliferous shale. : : iO 26. Grey calcareous shale, almost an argillaceous famcetone Glyptocy- there scitula ; Vernonella bajociana ; Fuhrbergiella (Praefuhr- bergiella) horrida horrida ; Praeschuleridea subtrigona intermedia. 0 25. Chocolate-brown mudstone, shelly. Glyptocythere scitula ; Vernon- ella bajociana and Monoceratina scarboroughensis ‘ O 24. Purplish brown, fossiliferous shale grading into bed 25. Cyst cythere scitula; Vernontella bajociana ; Monoceratina scar- boroughensis and Praeschuleridea subtrigona intermedia f 0) 23. Grey shale becoming less fissile towards base. Guprocvihere scitula ; Vernoniella bajociana ; Monoceratina scarboroughensis and Fuhrbergiella (Praefuhrbergiella) horrida horrida . : 4 22. Grey-black, very fossiliferous shale. Ostracod fauna at top of sequence : —Glyptocythere scitula ; Monoceratina scarboroughen- sis; Fuhrbergiella (Praefuhrbergiella) horrida horrida and Praeschuleridea subtrigona intermedia. Fauna 7 ft. from base :— ostracods indet. Fauna 6 ft. from base :— Lyubimovella piriformis. Fauna 5 ft. from base :— Glyptocythere scitula and Vernoniella bajociana. Fauna 2 ft. from base:— Monoceratina scarboroughensts ; Cloughtonella rugosa and Praeschuleridea subtrigona intermedia. Fauna at base :— Glyptocythere scitula ; Ljubimovella eis mis and Praeschuleridea subtrigona intermedia . : II 21. Calcareous shale. Cloughtonella rugosa ; Fuhrbergiella oe bergiella) horrida horrida ; Praeschuleridea subtrigona intermedia ; Monoceratina scarboroughensis ; Systenocythere ovata and Lyubi- movella piriformis . : : : : : ‘ : oO to mn. NAN @ IO Io IOC 83 84 20. IQ. 18. 17. 16. 5. 4. Tey, 12. Re BAv Sandy limestone, very fossiliferous bab ee sp. sae sp. etc.). Glyptocythere scitula : : Calcareous, grey shale with macrofossils as for ied 20. Glypto- cythere scitula ; Fuhrbergiella (Praefuhrbergiella) horrida horrida ; Praeschuleridea subtrigona intermedia and Systenocythere ovata . Grey sandy shale. Systenocythere ovata Calcareous sandstone Dark grey shale extensively nerd if marine Oreanisms the burrows infilled with sand. Belemnites present—no microfossils Hard calcareous shale. Pentacrinus ossicles common . Grey sandstone with Pentacrinus ossicles. The Canora Ga. Current bedded with ripple markings along the bedding planes. Worm burrows common. This bed grades upwards into bed 15. Foraminifera but no ostracods have been obtained from this bed Grey shale Purplish-brown samdieuns orien Guy procyinere polite and Se yaesciier™ idea subtrigona intermedia . Limestone Glyptocythere polita ; M ae. dniearinata aad Prae- schuleridea subtrigona intermedia . Sandstone. : : : : : : : : : . Grey sandy shale. Glyptocythere polita ; Malzia bicarinata and Praeschuleridea subtrigona intermedia . Dark grey shale. Glyptocythere polita ; Venus be orien: Fuhrbergiella (Praefuhrbergiella) horrida horrida and Prae- schuleridea subtrigona intermedia . Greymudstone. Glyptocythere polita in this bed, as in bed 12, occurs in extremely large numbers. Other members of the ostracod fauna are :— Progonocythere acuminata ; Vernoniella bajociana and Praeschuleridea subtrigona intermedia. . Grey shale. Glyptocythere polita and Pracschuleridea subtrigona intermedia . . Grey calcareous sandstone. Glyptooythere polit’ and Glyptocythere costata . Sandy shale . Hard black shale Middle Deltaic Series . Massive sandstone . ft. ie) H nh WwW 36) in. to to Io ie) MIDDLE JURASSIC OSTRACODA I. Black shale . , : , : : 4 : seen to SEcTION No. 5. Ravenscar (Text-fig. 2). As in the previous section, a complete sequence of marine strata is exposed, although the junction between the Grey Limestone Series and the Upper Deltaic Series is obscured by talus. The beds crop out high up in the steep cliff face, the Moor Grit of the Deltaic Series above capping the cliffs. Upper Deltaic Series 17. Soft yellow, false bedded sandstone, about 15 ft. of which cap the cliffs at this point. Grey Limestone Series 16. Reddish-brown, rather soft sandstone above which there is 3-4 ft. of sediment obscured by talus : : : ; 15. Ochre coloured sandy shale 14. Ironstone with fossil casts 13. Ochre coloured sandy shale : 12. Dark grey calcareous shale, fossiliferous throughout with ‘shelly bands and nodules. Fauna 27 ft. from base :— Monoceratina scarboroughensis ; Ljubimovella piriformis ; Fuhrbergiella (Praefuhrbergiella) horrida horrida and Praeschuleridea subtrigona intermedia. Fauna 25 ft. from base :— indeterminate ostracods. Fauna 21 ft. from base :— Glyptocythere scitula ; Vernoniella bajociana and Monoceratina scarboroughensis. Fauna 1g ft. from base :— indet. internal casts. Fauna 17 ft. from base :— indet. internal casts. Fauna 15 ft. from base :— Glyptocythere scitula ; Praeschuleridea subtrigona intermedia and Fuhrbergiella (Praefuhrbergiella) horrida hornida. Fauna 13 ft. from base in shelly band :— Glyptocythere scitula. Fauna 11 ft. from base in shelly bed :— Glyptocythere scitula and Lyubimovella pirtformis. Fauna 6 ft. from base :— Glyptocythere scitula ; Glyptocythere costata?; Ljubimovella pirtformis ; Fuhrbergiella (Praefuhrbergiella) horrida horrida and Praeschuleridea subtrigona intermedia. In the shale sampled within the basal 4 ft. of shale, no microfauna has been so far obtained F F 11. Fossiliferous yellow sandstone with Beara. oetles, The Crinoid Grit ft O BOW H 30 85 in. Io of UM 86 iRe5 1ehs 1B} eV IN 1S; to. Light-grey, calcareous shale with abundant Gervillia scarburgensis. Fauna at top of bed :— Malzia unicarinata ; Praeschuleridea subtrigona intermedia ; Glyptocythere costata and Vernoniella bajociana. Fauna at base:— Malzia umnicarinata; Malua bicarinata ; Glyptocythere costata and Progonocythere acuminata ? 9. Dark-grey, fossiliferous shale. Glyptocythere polita and Mala unicarinata . 8. Fossiliferous argillaceous ices oath Genuillia SRO Glyptocythere polita ; Glyptocythere costata ; Malza unicarinata ; Mailzia bicarinata and Praeschuleridea subtrigona intermedia 7. Grey sandy shale. Fauna at top :— Glyptocythere polita ; Malza bicarinata ; Progonocythere acuminata and Praeschuleridea sub- trigona intermedia. Fauna at Base:— Glyptocythere polta ; Vernoniella bajociana and Praeschuleridea subtrigona intermedia. 6. Calcareous mudstone, almost an argillaceous limestone. Glypto- cythere polita ; Oe acuminata and Praeschuleridea subtrigona intermedia 5. Rubbly bed—same lithology as bed 4. Glyptocythere Pola sie Praeschuleridea subtrigona intermedia 4. Dark grey, massive, sandy limestone. Glyplocyther polite and Praeschuleridea subtrigona intermedia 3. Dark-grey shale extensively burrowed by marine enigassths whole bed being a mixture of shale and sandstone. Large speci- mens of Gervillia scarburgensis present. Glyptocythere polita and Praeschuleridea subtrigona intermedia 4 2. Grey-black, rather brittle shale—no microfauna . Middle Deltaic Series 1. Deltaic sandstone, massively bedded SECTION No. 6. Hawsker (Text-fig. 2.). The thickness of the Grey Limestone Series has been considerably reduced and although the top of the succession is not exposed, the observed thickness of 22 ft. 5 in. cannot be far short of the total in this area. Grey Limestone Series 13. Well bedded grey sandstone, rather coarse grained, virtually a grit, which grades down into a chocolate-brown sandstone shelly in parts. Ostracods (Glyptocythere scitula ? and Praeschuleridea subtrigona intermedia) present though somewhat decalcified seen to 574. Sandy fester conte with shells towards: the base—rather fissile on weathering. Vernoniella ? caytonensis ; Systeno- cythere ovata at base ft. in. 2 3 I 9 0) 4 2 6 it 2 ie) 7 2 0) I 6 if Io 6 IO 5 2 303, Io. dw pb of MIDDLE JURASSIC OSTRACODA Sandy and very fossiliferous chocolate-brown shale. Southcavea microcellulosa ; Praeschuleridea subtrigona intermedia ; Eocy- theropteron ? sp.; Glyptocythere scitula ? and Fuhrbergiella (Praefuhrbergiella) horvida horrida . : : : ‘ Chocolate-brown shale. Glyptocythere Sas Vernoniella bajociana and Praeschuleridea subtrigona intermedia . Grey shale crowded with fossils at top. Oy rete: eine and Vernonella bajociana at the top and Glyptocythere scitula ; Cloughtonella rugosa and Fuhrbergiella i al a horrida horrida at base : : . Grey shale with large modules of fenestone . ' . Grey fossiliferous shale with Glyptocythere scitula “Vernoniella bajociana ; Fuhrbergiella (Praefuhrbergiella) horrida hornida ; Ljubimovella piriformis ; Systenocythere ovata ; Progonocythere acuminata and Praeschuleridea subtrigona intermedia . Calcareous sandstone, fossiliferous. No microfossils obtained. Tae Crinoid Grit of elsewhere : : : . Dark-grey sandy shale, fossiliferous along bedding planes, Internal casts of Glyptocythere sp. : : Middle Deltaic Series . Light-grey shale with sandstone lenses . Grey and white laminated sandy Shale “sandstone jenses . Grey sandy shale with 1 ft. 1 in. lens of sandstone at ae Shale carbonaceous with os remains : . Grey sandstone é : : < seen to SECTION No. 7. Bloody Beck (Text: fig 3). Of the inland exposures the Grey Limestone Series, this is by far the best—it is complete and all beds are accessible along the course of the stream. 29 ft. 4 in. of marine sediment are developed here. 25. Upper Deltaic Series Moor Grit—massively bedded sandstone forming the base of the Deltaic Series—flaggy towards base, grading into bed below. Grey Limestone Series 24. Sandy, micaceous, dark-grey shale 23. 22. ai. Clay ironstone with specks of pyrite . Fossiliferous grey, sandy shale, no microfossils Light-grey mudstone, very fossiliferous. Glyptocythere seca ft OH ON in. 10 Ann 87 88 20. IQ. 18. 1s 16. ids, 14. 135 12. at aC. Io. R. H. BATE Fossiliferous grey sandy shale—internal casts of Glyptocythere ? sp. and Vernomella ? sp. Beds 24-20 have suffered extensive decalcification—fresh material if possible to obtain would probably yield a larger fauna . ; : é : é Hard, grey, calcareous shale. Ostracods fragmentary and in- determinate ‘ : ; ’ 3 : ; , Grey calcareous shale, fossiliferous. Glyptocythere scitula and Vernomella bajociana : : : ; : : : Hard, grey, fossiliferous mudstone. Glyptocythere scitula and Ver- nonella bajociana. Hard, grey, calcareous sieieecquite foscniterous aihoueH tile ostracod fauna appears to be restricted to the single species, Vernoniella bajociana Grey, rather shelly shale, cotter tan ell ad Saran eo restricted to Fuhrbergiella (Praefuhrbergiella) horrida horrida Fossiliferous calcarous shale—poor ostrocod fauna. Glyptocythere scitula and Praeschuleridea subtrigona intermedia Hard, grey, calcareous shale /mudstone with sub-rounded sie grains embedded particularly in upper part. Glyptocythere scitula : : : Sparsely fossiliferous anal with Taastene médias, No ostracods recovered Grey fossiliferous shale. Terenas at erp = dete Fauna 3 ft. 8 in. from top :— Glyptocythere scitula ; Systeno- cythere ovata and a fragment of Praeschuleridea subtrigona inter- media. Fauna 6 ft. 6 in. from top :— Glyptocythere scitula ; V ernoniella bajociana and Praeschuleridea subtrigona intermedia Grey, very fossiliferous shale. Cloughtonella rugosa and iBencaatinntee idea subtrigona intermedia . Calcareous shale grading upwards into bed 10. ‘It e roe a much more resistant bed to weathering—fossiliferous. Glypto- cythere scitula ; Progonocythere acuminata and ? Southcavea micro- cellulosa . Hard, grey shale, sparsely fossiliferous. Fuhrbergiella (Prae- fuhrbergiella) horrida horrida and indet. ostracod fragments . Grey sandy shale, fossiliferous. Mesocytheridea howardianensis ; Glyptocythere scitula and Malza bicarinata . Grey-black fossiliferous limestone becoming shaly toni pages Fauna at top in limestone :— Glyptocythere polita and Glypto- cythere scitula. Fauna in shaly beds at the base consists entirely of Glyptocythere polita which occurs in enormous numbers . ft. mn. IEE Io Io MIDDLE JURASSIC OSTRACODA ft. mn. 5. Well jointed, grey limestone. Glyptocythere polita and Praeschuler- idea subtrigona intermedia. 0) 9 4. Grey, fossiliferous shale. Cn fous c polite ; M alzia LMG : Progonocythere acuminata and Praeschuleridea subtrigona inter- media. : ; ; : : ; : ie 4 3. Grey, well jointed, limestone: Glyptocythere polita and Praeschuler- idea subtrigonaintermedia ) 8 2. Grey sandy shale. ls polita tel Ppaeschuleyiden ani trigona intermedia : : 5 . wre: 3 Middle Deltaic Series 1. Massive sandstone, flaggy in places. ‘ : : seento 8 0) SEcTION No. 8. May Beck, stream section. Exposures poor and discontinuous, the ostracod fauna being almost entirely leached out. Section at NZ/8go0015. Upper Deltaic Series 10. Moor Grit—flaggy at base : f i : : seen to 12 ) Grey Limestone Series g. Sandy limestone. Glyptocythere scitula and ? hase (Prae- fuhrbergiella) horrida horrida . : : seento 2 9 8. Rubbly sandy bed with mudstone nodules ‘ : at 2 7. Grey, calcareous sandstone. Praeschuleridea svaieene hep r in Section at NZ/8g0017, about 100 yds. downstream. 6. Grey flaggy sandstone with fossil casts : : ; seento I O Section at NZ/8g1019, about 100 yds. downstream. 5. Grey calcareous sandstone with fossil casts. This bed is probably more correctly a sandy limestone, leaching of the CaCO, reducing the bed to a sandstone. Pvaeschuleridea subtrigona intermedia 3 B 4. Purplish-red siltstone with decalcified shells. Glyptocythere scitula and Vernoniella bajociana ; : : ; : 3. Grey sandstone with fucoid markings . 2. Grey, brittle shale—ostracod internal casts . : ¢ ’ 1. Light-grey shale with plant remains . : ‘ : seen to Cy Ii ©) Ee OoONN SECTION No.9. Ramsdale Beck, stream section. Only 9 ft. 6 in. of marine shale exposed in the left bank of the stream (facing downstream). Although macrofossils are present within the shale, all the microfauna has been leached away except for the occasional indeterminate internal cast. GEOL. II, 3 7 89 90 RH See AGIEE: SECTION No. 10. Eller Beck, stream section, exposing 6 ft. 5 in. of marine sediment in the right bank of the stream, below road bridge. Grey Limestone Series 5. Dark grey shale—sandy and ironstained. Internal casts of Glyptocythere ? sp. : : : : : 4. Calcareous mudstone with macrofossils. Glyptocythere scitula and Praeschuleridea subtrigona intermedia 3. Grey, sandy shale with a few fossils : 2. Calcareous grey shale. This bed has a good sdibee es cement and has not been so extensively decalcified. Glyptocythere scitula ; Vernoniella bajociana and Praeschuleridea subtrigona intermedia. 1. Dark grey shale, internal casts of Hee ? sp., and Ver- noniella bajociana ‘ ; seen to SECTION No. 11. Harland Beck, stream section in left bank in a disused bend of the stream just above the junction of the Harland Beck with the river Dove: 14 ft. 4 in. exposed. Grey Limestone Series 7. Sandy shale. Glyptocythere scitula; Vernoniella bajociana and Praeschuleridea subtrigona intermedia ; ; ; seen to 6. Grey, sandy limestone, fossiliferous along the bedding planes. Praeschuleridea subtrigona intermedia 5. Grey sandy shale with calcareous nodules , : 4. Massive bedded sandstone almost a oak flaggy at base . 3. Sandy shale ; : ; 2. Dark-grey shale, almost a mudstones in appearance. Lamellibranch Gervillia scarburgensis common. Glyptocythere scitula and Prae- schuleridea subtrigona intermedia 1. Dark-grey mudstone. Glyptocythere scitula and Pagesehileraden subtrigona intermedia. : s 3 : é seen to SECTION No. 12. Bogmire Gill (Text-fig. 3), an almost complete stream section exposing some 40 ft. 3 in. of the Grey Limestone Series. Upper Deltaic Series 10. Moor Grit—flaggy at base : ; : : : seen to Grey Limestone Series g. Grey sandy shale . ft. oMmH N in. NQNI © MIDDLE JURASSIC OSTRACODA 8. Dark-grey ironstained shale with fossils. Glyptocythere scitula and Vernonella bajociana ‘ : ‘ : : ‘ : 7. Grey-brown mudstone. Glyptocythere scitula; Vernoniella bajo- ciana ; Progonocythere acuminata and Praeschuleridea subtrigona intermedia . ‘ ; : : : : : : 6. Dark-grey, fossiliferous orale) Glyptocythere scitula and Vernonella bajociana : : : : : : unexposed section approx. 2 ft. . Flaggy yellow sandstone with fossil casts . Grey sandy shale . Grey ironstained shale Oo BU unexposed section approx. 20 ft. 2. Calcareous sandstone, known as the Crinoid Grit. Fossiliferous and false bedded. Ostracod internal casts, Glyptocythere ? sp. . I. Steel-grey, flinty limestone, ripple marked ee bedded planes. Indeterminate ostracods : : : seen to SECTION No. 13. (Text-fig. 3). A small, disused quarry on Yearsley Moor, exposing 8 ft. 3 in. of marine sediment. Upper Deltaic Series 3. Yellow, rather soft, flaggy sandstone with plant remains along bedding planes. : : : : . seen to approx. 2. Yellow sandy bed with lenses of grey shale . Grey Limestone Series 1. Buff coloured sandy limestone, weathers to a sandstone—fossili- ferous. Fauna approx. 3 ft. 6in. from the base :— Glyptocythere scitula ; Praeschuleridea subtrigona intermedia ; Mesocytheridea howardian- ensis ; Fuhrbergiella (Praefuhrbergiella) horrida horrida and Systenocythere ovata. SECTION No. 14. (Text-fig. 3). Three isolated sections within the Grey Limestone Series, exposed in Stonecliff Wood, above the York— Scarborough railway-line. Altogether a maximum of 20 ft. 1 in. of sediment is exposed. 10 IZ to QI mn, HO 92 R. H. BATE Grey Limestone Series 8. Sandy, fossiliferous limestone exposed at map reference SE/743676. Fauna at top of section :— LEocytheropteron ? sp. ; Southcavea microcellulosa and Praeschuleridea subtrigona intermedia. Fauna 3 ft. 8 in. from base :— Southcavea microcellulosa. ; Praeschuleridea subtrigona intermedia; Glyptocythere scitula ; Mesocytheridea howardianensis and Fuhrbergiella (Praefuhrber- giella) horrida horrida, Fauna I ft. 6 in. from base includes all those listed above, with the exception of Fuhrbergiella (Praefuhrbergiella) horrida horrida : ; : ; Section at SE/744676 :— 7. Grey limestone weathering to orange-red sandstone containing fossil casts. This bed is largely decalcified at outcrop DS . Yellow sand, false bedded at top. Fossil casts : 5. Limestone, impersistent laterally, being replaced by yellow Saal Glyptocythere scitula ,; Eocytheropteron ? sp.; Praeschuleridea subtrigona intermedia and Southcavea microcellulosa 3 4. Soft yellow sandstone—easily weathers to a sand—fossil Bogs common in lower part ; 3. Shelly limestone, impersistent and replaced laterally by sand. Glyptocythere scitula and Southcavea microcellulusa : 3 Reddish-brown sand with fossil casts . i : ‘ . Coarse silver sand . : ‘ ; : : : seen to HN Section at SE/740675 :— 2. Grey limestone. Praeschuleridea subtrigona intermedia; Meso- cytheridea howardianensts ; Glyptocythere scitula and Vernoniella bajociana. This bed is possibly at a ge lower horizon than those listed in the two sections above 1. Yellow sandstone . : : ; : : ; seen to TTETS 3S RSAC GaRGAC EET Ys ft. ° 1. oo to or During Middle Jurassic times an important axis of downwarping, the Cleveland Axis, extended approximately east-west through the centre of the Yorkshire Basin, cutting the present coastline somewhere about Ravenscar (Text-fig. 1). of movement played an important role during the deposition of the Grey Limestone Series. This axis MIDDLE JURASSIC OSTRACODA 93 The Grey Limestone Series or Scarborough Beds are well exposed along the coast as far north as Whitby where the outcrop turns inland. In the northern part of the Yorkshire basin, the Cleveland Hills (Text-fig. 4), the outcrops are generally rep- resented by expanses of coarse calcareous grit known as the Crinoid Grit. Down the western part of the outcrop, exposures are sometimes good but incomplete ; the most complete inland exposures are found towards the centre of the basin in stream sections. Fox-Strangways (1892 : 236) was the first to subdivide this marine series on lithological terms. His divisions consisted of an upper shale division, a middle sandstone division and a lower limestone division. This arrangement was continued by Richardson (1911). However, the subdivision of the Grey Limestone Series in this way, although basically correct, is a simplification of what really takes place. In fact failure to recognize facies changes resulted in Arkell (1933 : 221) making the following statement “ Although the exact horizons of the ammonites recorded from the Scarborough or Grey Limestone are not known, it may be presumed that most if not all came from the lowest or limestone division’’. The impure limestones developed in the Scarborough district belong to the upper shale division. The lower so called limestone division is not exposed at Scarborough and may not, in fact, extend as far south. As many of the ammonites recorded come from Scarborough, it is clear that they must come from the upper division. The examination of the ostracod faunas brought out this fact and at the same time made it possible to correlate the various sections more precisely than would have been possible on purely lithological evidence. The division of the Grey Limestone Series on lithological evidence is here aban- doned, instead two ostracod zones are recognized: a lower zone of Glyptocythere polita and an upper zone of Glyptocythere scitula, the type section for both being taken at Hundale Point—see Section 4. Glyptocythere scitula Zone The estracod fauna associated with the index ostracod of this zone is more varied than that of the lower zone and possesses several species which are restricted in their geographical distribution. The typical faunal assemblage of this zone is as follows:— Glyptocythere scitula ; Vernontella bajociana ; Monoceratina scarboroughensis ; Fuhrbergiella (Praefuhrbergiella) horrida horrida; Ljubimovella piriformis ; Systenocythere ovata and Praeschuleridea subtrigona intermedia. Of this fauna P. subtrigona intermedia ranges throughout the entire Grey Limestone strata and is not indicative of either zone. The two ostracods Vernoniella bajociana and Fuhrbergiella (P.) horrida horrida are similarly found to occur in sediments below this zone, but they are relatively uncommon at the lower horizon and are not con- sidered to be characteristic there. A number of ostracods, restricted to this zone, have a limited geographic distribu- tion, the palaeoecological considerations of which will be dealt with later. These are :— Cloughtonella rugosa ; Cytheropteron ? yonsnabensis ; Caytonidea faveolata ; 04 R. H. BATE Mesocytheridea howardianensis ; Vernoniella ? caytonensis,; Paracytheridea ? caytonensis ; Progonocythere yonsnabensis and Southcavea microcellulosa. The ostracod Progonocythere acuminata is present in this zone, but is equally, if not more so, as common in the lower zone of G. folita and is of little value strati- graphically. Glyptocythere polita Zone The diagnostic fauna of this zone is :— G. polita, G. costata ; Malza unicarinata and M. bicarinata. As mentioned above (see also Text-fig. 24), several ostracods are present in both zones and need not be listed again. The above assemblage is charac- teristic of and restricted to this lower part of the Grey Limestone Series. In all cases the junction between the zone of G. folita and G. scitula is clear-cut. There is a very slight overlap of one or two species in some cases, but this is relatively insignificant, and where observed limited to a few feet of strata only. A few species, however, as mentioned above are not restricted to either zone, but range throughout the complete succession (Text-fig. 24). Along the coast the most southerly exposure of the Grey Limestone Series is in Gristhorpe Bay. Here the beds rapidly thicken to the observed maximum of 15 ft. 6 in. at the north of the Bay in Yons Nab headland (Section 1, Text-fig. 2). The marine sediments consist entirely of calcareous shales and mudstones both here and on the other side of the headland in Cayton Bay (Section 2, Text-fig. 2) where a maximum of 10 ft. 7 in. was observed. In this area the basal sandstone of the Upper Deltaic Series (here a rather incipient lateral equivalent of the massive Moor Grit further north) cuts down into the marine sediments as a number of well exposed washouts. These have been described and figured by Black (1928). At White Nab, Scarborough, the Series is again exposed (Section 3, Text-fig. 2) and has increased to an exposed maximum of 25 ft. 6 in. The shales are rather more strongly calcareous, and impure sandy limestones are developed. The ostracod fauna throughout the sequence, as in the previous two sections, is indicative of the G. scitula zone. It is possible that the lower zone of G. polita may be present under the sea. However this part of the section is never exposed even at low tide and must remain not proven for the present. North of Scarborough, at Hundale Point, Cloughton, the marine strata attain their maximum development of 62 ft. 4 in. (Section 4, Text-fig. 2). This is exactly the same thickness of sediment as measured for the Grey Limestone Series a few miles further north at Ravenscar (Section 5, Text-fig. 2). In both cases there is a very thick development of calcareous shale in the upper part, tending to be arenaceous at the top of Hundale, whilst at Ravenscar a bed of sandstone is developed. At Hundale there is still a remnant of the sandy limestone of the White Nab succession, present towards the lower part of the shale sequence. This limestone is not rep- resented at Ravenscar. At both localities the shale beds are followed by a thick bed of calcareous sandstone in which Pentacrinus ossicles are plentiful. This is the Crinoid Grit which is more typically developed further north. The ostracod fauna of the shale beds is indicative of the G. scitula Zone. No ostracods have been obtained MIDDLE JURASSIC OSTRACODA 95 from the calcareous sandstone (Crinoid Grit) and it is taken to belong to the G. scitula Zone. As all the sediments below this bed at Hundale and Ravenscar belong to the G. polita Zone, the bed is here, a good marker horizon. The strata belonging to the G. polita Zone here reach their maximum development and consist of calcareous and sandy shales, some impure limestones and calcareous sandstones. Some mudstones are also present and these invariably are crowded with the ostracod Glyptocythere polita. Just south of Whitby the outcrop of the Grey Limestone Series turns inland away from the coast. However, an excellent exposure has been obtained in a stream sec- tion at Hawsker, or more precisely, Hawsker Bottoms (Section 6, Text-fig. 2). According to the section described by Fox-Strangways (1892 : 237) only 6 ft. 4 in. of sediment at the top of the section is unexposed at the present time. In this section the upper shale horizon, so well developed at Hundale and Ravenscar, is much thin- ner and a thick bed of fossiliferous sandstone and a bed of sandy limestone are introduced at the top. The sandstone is almost certainly laterally equivalent to bed 16 present at the top of the shale horizon at Ravenscar, but here much more massive in character. The lateral equivalent of the Crinoid Grit (bed 6) is not very strongly developed here and beds possibly belonging to the G. folita Zone are very much reduced in thickness. The shale bed which probably represents this horizon (bed 5) has only produced extensively decalcified internal casts of Glyptocythere sp., so that conclusive evidence is at the moment lacking. Stratigraphically, however, there is good reason to consider bed 5 as representing the G. polita Zone with beds 6-13 belonging to the G. scitula Zone. To the north and west of Whitby the marine sediments of the Series are poorly exposed and appear to consist almost entirely of coarse grained fossiliferous grits in which Pentacrinus ossicles are common. This lithofacies has been named the Crinoid Grit (Richardson rg11i : 195 and 197) and is laterally equivalent not only to the calcareous sandstone containing Pentacrinus ossicles at Hundale and Ravenscar but to practically the whole of the G. scitula Zone. However, shale and limestone horizons are still to be found in this northern area but are definitely subordinate to the arenaceous facies. Towards the centre of the depositional basin the Grey Limestone Series continues to exhibit a wide variety of lithofacies. At the Bloody Beck stream section (Section 7, Text-fig. 3) calcareous shales and mudstones predominate with some impure lime- stones coming in towards the base in the zone of G. polita. The Crinoid Grit appears to be absent here and the junction of the two ostracod zones falls between beds 6 and 7. Only a few miles to the north-west the majority of the sediments exposed in the May Beck section (Section 8) are quite arenaceous. In the west at Bogmire Gill (Section 12, Text-fig. 3) a large part of the section is again arenaceous and the Crinoid Grit (bed 2) is well developed and shows strong false bedding. The sediments in the Harland Beck section (Section Ir) are also predominantly arenaceous. Apart from the Bloody Beck section where beds of G. polita age definitely occur, the only possible exposure of these beds is at Bogmire Gill where a flinty ripple marked limestone is probably to be correlated with this zone. So far no ostracods have been extracted 96 R. H. BATE from this lithology in a recognizable form. The basal limestone at Bogmire Gill was seen to be 3 feet, and is probably not very much thicker. The G. polita Zone at Bloody Beck is 6 feet in thickness. Not only does this zone diminish considerably in thickness north and south away from the Hundale/Ravenscar area but it also thins westwards. The probable area covered by sediments of the G. polita Zone is shown in Text-fig 4. The most westerly outcrop of the Grey Limestone Series runs north/south through the Howardian Hills and consists of strongly arenaceous limestones interbedded with pure sand. In many cases decalcification of the limestone has resulted in the pro- duction of beds of sand. No complete section is exposed. Section 13 (Text-fig. 3) on Yearsley Moor exposes 8 ft. 3 in. of sandy limestone overlain by yellow sandstone of the Upper Deltaic Series. Further south at Stonecliff Wood, near Whitwell a coarse grit | |) Gs \ re oa Sige SUNG) Re. * Aas \ 9 e LA "CLEVELAND HILLS |” Vien: : : o apa X<_ WHITBY NORTH SEA Zz 19 i Ww n . < 4 \ \ PICKERING SCARBOROUGH ie So e S37 c ~ ata of \ a RO eS! aie > 2 sandy Rep HOSA Sat Boe ae MALTON limestone Speke Si oie ’ = 5 SAS E : ; OS Lea = (NE 14 —==-55< - € | gunk AC | VANE * | | sand & : YORK sandy limestone | \ ~—-— —Zone of G.polita o1234 8miles \ ——— Zone of G.scitula Fic. 4. Plan view of the marine basin during the deposition of the Grey Limestone Series, showing the probable maximum extent of the two faunal zones. MIDDLE JURASSIC OSTRACODA 97 rather more complete section is exposed (Section 14, Text-fig. 3). Here sandy lime- stones are seen to be not only interbedded with false-bedded sand but are also observed to pass laterally into red sandstone or sand. This strongly arenaceous facies of the Grey Limestone Series is also present in the east, south of Gristhorpe Bay. Here, however, not exposed at the surface. The Fordon No. 1 borehole proved 5 feet of sandy limestone with belemnites (Falcon & Kent 1960: 27). This facies is, therefore, continued around the western and southern perimeters of the Yorkshire basin whilst coarse grits are present along the northern edge. The shales and impure limestones are the major development towards the east and centre of the basin. IV CONCLUSIONS The Grey Limestone Series of north-east England is considered to represent the marine deposits of a broad embayment which cut into the deltaic sediments in Middle Jurassic times. The transgression of the sea at this point was the result of downwarping along the Cleveland Axis, the lowermost beds of the Series (zone of Glyptocythere polita) being deposited only towards the centre of the basin (Text-fig. 4). The higher beds (zone of Glyptocythere scitula) extend over a wider area and in part rest directly upon deltaic sediments. The probable maximum extent of these higher beds is indicated in Text- fig. 4. That deposition of the marine sediments occurred in shallow water is evidenced by the presence of ripple markings, false bedding, worm burrows etc. The macrofauna is also generally indicative of shallow water conditions, which, together with a possible lowering of the salinity close to a delta may explain the almost complete absence of ammonites. The shore-line of the Series is indicated by the change in facies to a very sandy limestone and even to pure sand around the south and western boundaries of the outcrop. To the north, the whole series is more coarsely arenaceous, the detrital material being brought in by the delta. The present day perimeter of the Grey Limestone Series outcrop approximates closely to the original shore-line. As shown in Text-fig. 5, there are two prominent sandstone horizons which extend across the basin. Each reflects the influence of the northern delta. A third sand- stone, the deltaic Moor Grit brings the period of marine deposition to a close. This type of sedimentation with marine shales and limestones interfingered by marine deposited deltaic sandstones is suggestive of intermittently rising sea-level, see Dunbar & Rodgers (1958 : 85, text-fig. 35c). The ostracod fauna is typically shallow water benthos and in the majority of species present, appears to be largely independent of bottom facies. There is, how- ever, a typical shore-line fauna consisting of :— Southcavea microcellulosa ; Meso- cytheridea howardianensis ; Praeschuleridea subtrigona intermedia and Glyptocythere scitula. A number of other species such as Vernomella bajociana ; Systenocythere ovata and Fuhrbergiella (P.) horrida horrida also occur in this very sandy shore-line facies, although they are only poorly represented. Towards the centre of the basin in slightly deeper water, the characteristic fauna consists of (in the G. scitula Zone) :— Glyptocythere scitula; Vernoniella bajociana,; Monoceratina scarboroughensis ; 98 Tit DLS 1/01: Ljubimovella piriformis ; Praeschuleridea subtrigona intermedia ; Cloughtonella rugosa ; Systenocythere ovata and Fuhrbergiella (P.) horrida horrida. A number of other species such as Caytonidea faveolata ; Vernontella ? caytonensis ; Paracytheridea ? caytonensis and Progonocythere yonsnabensis also occur but are rather restricted in their geographical range, reflecting an environment most probably restricted to the Cayton Bay region. The ostracod Progonocythere acuminata is only sporadically present at this horizon. In the G. polita Zone the fauna appears to have lived in a FORDON 182 3 4 5 ee 45 miles == a = > Fic. 5. Section across the Yorkshire Basin showing the facies changes of the Grey Lime- stone Series. Sections 1-6 are indicated together with information from the Fordon bore-hole. Vertical scale exaggerated. similar environment, with regard to bottom sediment and depth of water, as the above and consists of the following :— Glyptocythere polita; G. costata; Malua bicarinata ; M. unicarinata ; Progonocythere acuminata and Praeschuleridea sub- trigona intermedia. Vernoniella bajociana occurs in the upper part of this zone whilst Fuhrbergiella (P.) horrida horrida has only been recorded from a single locality. No definite shore-line fauna has been identified in this lower zone, due, no doubt to lack of exposures, although the ostracod Mesocytheridea howardianensis occurs in sandy shale at the base of the G. scitula Zone in the Bloody Beck section, whilst a little higher up in the same section there is a single specimen possibly belonging to South- cavea microcellulosa. The presence of these ostracods here may indicate a shallowing of the water or a change in environment to their liking. It is not possible, however, to draw any further conclusions at this stage. V SYSTEMATIC DESCRIPTIONS Subclass OSTRACODA Latreille 1806 Order PODOCOPIDA Miller 1894 Suborder PODOCOPINA Sars 1866 Superfamily CYPRIDACEA Baird 1845 Family PARACYPRIDIDAE Sars, 1923 Genus PARACY PRIS Sars 1866 Paracypris bajociana Bate MIDDLE JURASSIC OSTRACODA 99 For complete synonymy see Bate 1964 : 9. REMARKS. Paracypris bajociana has been recorded from the Lincolnshire Lime- stone and from the Cave, Whitwell and Millepore Oolites of N.E. England (Bate 1963, 1963a and 1964). Plumhoff (1963: 18) records this species from beds of discites age and younger from Noith Germany. The only occurrence within the Grey Limestone Series is at the base of the Gris- thorpe Bay sequence where two specimens have been found. Superfamily CYTHERACEA Baird 1850 Family BYTHOCYTHERIDAE Sars 1926 Genus MONOCERATINA Roth 1928 Monoceratina scarboroughensis sp. nov. (PL. x, figs. 1-12) Diacnosis. Monoceratina, with finely punctate, subquadrate to elongate cara- pace, slightly constricted just anterior of mid-dorsal region. Hototyre. lo.1711, top of bed 22, Hundale Point, Cloughton. ParatTyPes. l[o.1712—23, from top and base of bed 22 and bed 25, Hundale Point ; bed 5, Cayton Bay and bed 12 (21 ft. and 27 ft. from base), Ravenscar. DESCRIPTION. Carvapace subquadrate in outline, slightly constricted just anterior of mid-dorsal region, the more elongate dimorphs are considered to be the males. Greatest length of carapace through mid-point; greatest height in anterior or posterior third ; greatest width in posterior third. The shallow constriction (sulcus) does not extend below mid-point, the ventro-lateral part of the carapace being noticeably swollen, particularly postero-ventrally. Dorsal margin straight with distinct, rounded cardinal angles. Ventral margin incurved medially; anterior rounded ; posterior triangular with a short, straight or slightly concave postero- dorsal slope and a long, convex postero-ventral slope. Shell surface finely orna- mented with small, round puncta, arranged in longitudinal rows in the male dimorph. Valves almost equivalve : mid-ventrally the left valve slightly overlaps the right, whilst dorsally the right valve overlaps the left, the degree of overlap increasing towards the posterior cardinal angle. Muscle scars consist of a subvertical row of four rectangular adductor scars and two antero-dorsal scars situated below mid- length and below the shallow sulcus. Hinge in the left valve consists of the slightly downset mid-dorsal edge of the valve acting as a hinge-bar ; not seen in the right valve, but presumably consisting of a simple dorsal groove for the articulation of the left valve hinge. Duplicature not clearly seen, though there appears to be a narrow anterior vestibule developed in one paratype (Io.1716). Dimensions Holotype Io.1711, female carapace (Pl. 1, figs. 1-3), length 0-61 mm.; height 0-32 mm. ; width 0-25 mm. 100 1g, Jel, IBy/NA0 IB: Io.1719, female right valve, length 0-60 mm. ; height 0-35 mm. I[o.1720, female left valve (PI. 1, fig. 12), length 0-55 mm. ; height 0-33 mm. I[o.1721, male carapace (Pl. 1, figs. g-11), length 0-70 mm.; height 0-32 mm.; width 0-27 mm. Io.1722, female carapace, length 0-52 mm.; height 0-27 mm.; width 0-22 mm. I[o0.1723, female carapace (PI. 1, figs. 4-8), length 0-57 mm. ; height 0-31 mm.; width 0-23 mm. REMARKS. Monoceratina scarboroughensis differs from all previously named species in outline (greatest length being through mid-point and not dorsal of mid- point as in the majority of cases), surface ornamentation of fine puncta coupled with the presence of dimorphism. Family PROGONOCYTHERIDAE Sylvester-Bradley 1948 Subfamily PROGONOCYTHERINAE Sylvester-Bradley 1948 Genus CAYTONIDEA nov. DiaGnosis. Progonocytherinae, oval-rectangular in outline with well rounded anterior and posterior margins. Low eye swelling situated at anterior cardinal angle. Cardinal angles prominent, broadly rounded. Hinge antimerodont. Muscle scars consist of subvertical row of four adductor scars, rounded antero-dorsal antennal scar and rounded antero-ventral mandibular scar. Radial pore canals long, straight, few in number. Left valve larger than right. TYPE SPECIES. Caytonidea faveolata sp. nov. REMARKS. Only a single species can be placed in the genus at the present time and this is typified by a strongly reticulate ornament. The possession of a well rounded oval-rectangular carapace with an antimerodont hinge, type A muscle scar arrangement and a distinct eye swelling identifies the genus Caytomzdea from all other cytheracean genera. The genus (feminine) is named after the type locality, Cayton Bay. Caytonidea faveolata sp. nov. (Pl. 1, figs. 13-14; Pl. 2, figs. 1-10; Text-figs. 6, 7) Diacnosis. Caytonidea, with strongly reticulate ornament of 5-6 sided pits. Hototyre. Io. 1831, bed 5, Cayton Bay. PARATYPES. Io.1832-35, horizon and locality as above and from bed 4, Gris- thorpe Bay. Fic. 6. Hinge of right valve, Caytonidea faveolata sp. nov. Paratype, lo.1832, approx. X190. MIDDLE JURASSIC OSTRACODA 101 DESCRIPTION. Carapace oval-subrectangular in outline with well rounded anterior and posterior margins, slightly concave mid-dorsal margin and antero-medially incurved ventral margin. The carapace is constricted slightly in the mid-dorsal region, and is noticeably swollen in the postero-dorsal region. Greatest length of carpace through mid-point, greatest height in the anterior third, greatest width in the posterior third. Shellsurface ornamented by prominent 5—6sided pits. Ventrally the reticulate ornament is somewhat subdued and is dominated by a series of longi- tudinal ridges. A low, smooth, eye swelling is situated at the anterior cardinal angle. Left valve larger than the right which it over-laps along the ventral margin and slightly we s Ky: Fic. 7. Muscle scars, Caytonidea faveolata sp. nov. Paratype, 10.1833, approx. X 250. in the region of the anterior cardinal angle. Hinge antimerodont, seen only in the right valve: 4 anterior and 5 posterior teeth observed ; median groove long and locellate. Duwplicature of moderate width, the inner margin and line of concrescence coinciding. Radial pore canals long, straight and widely spaced, 7 observed anteriorly. Muscle scars with rounded antero-dorsal antennal scar (Type A). Dimensions Holotype Io.1831, carapace (PI. 2, figs. 1-4), length 0-56 mm. ; height 0-33 mm. ; width 0-30 mm. Io.1834, carapace (PI. 1, figs. 13, 14; Pl. 2, figs. 5, 6), length 0-65 mm. ;_ height 0-37 mm. ; width 0-33 mm. _ 10.1835, carapace length 0-56 mm. ; height 0-34 mm. ; width 0-27 mm. REMARKS. Externally similar to Southcavea reticulata Bate (1964: 27, pl. Io, figs. 3-14; pl. 11, figs. 1-4), Caytomdea faveolata differs in the presence of an eye swelling, type A muscle scar arrangement as against type D, and in being less convex in dorsal view. The reticulate ornament of S. veticwlata differs markedly from that of C. faveolata in the presence of strong punctation inside the shallow pits. Genus CLOUGHTONELLA nov. Diacnosis. Progonocytherinae with subquadrate carapace, virtually parallel- sided in dorsal view. Ventero-lateral border convex projecting downwards and 102 Re CE eB AGE outwards in type species. Dorsal margin of larger left valve concave with upstanding cardinal angles. Posterior traingular; anterior rounded with narrow marginal border. Hinge antimerodont. Duplicature of moderate width, inner margin and line of concrescence coinciding. Radial pore canals straight, few in number, widely spaced. Muscle scars not observed. TYPE SPECIES. Cloughtonella rugosa sp. nov. REMARKS. Cloughtonella is very close to Aulacocythere Bate (1963: 198) from which it probably evolved. The general morphological features of these two genera suggest a very close relationship. However, Cloughtonella can be distinguished by the absence of the horseshoe-shaped swelling of Aulacocythere and does not possess an eye swelling. At present only a single species can be definitely assigned to the genus : Cloughtonella rugosa sp. nov. However, the ostracod Procytheridea hoffmani Brand (1961 : 159, pl. 1, figs. 1-8) is possibly congeneric although tending to be more oval in side view, with the dorsal margin of the left valve broadly convex, passing down to the extreme posterior without any break at the cardinal angle. The dorsal margin of the left valve, male dimorph, may be slightly concave, however. P. hoffmanni does not appear to belong to the genus Micropneumatocythere Bate (19634 : 29), to which many of the European procytherids belong, nor is it a true Procytheridea Peterson (1954 : I7I) which is a much more posteriorly acuminate genus. It is here tentatively considered to be congeneric with C. rugosa. The specimens of Procytheridea hoffmanni examined here were obtained from a sample of the vomani Schichten, South Hannover, sent to me by Dr. F. Plumhoff, Erdél A. G., Wietze krs. Celle. The known range of P. hoffmanmi is romani to blagdeni Zones, that of C. rugosa uncertain because of the almost complete absence of an ammonite fauna, but probably just below blagdeni Zone. The genus Cloughtonella (feminine) is named after the type locality Hundale Point, Cloughton Wyke. Cloughtonella rugosa sp. nov. (Pl. 3, figs. 1-13 ; Text-figs. 8, 9) Diacnosis. Cloughtonella, with small subquadrate carapace. Dorsal margin medially concave in left valve. Ornamentation consists of prominent diagonal, rather irregular median ridge extending from postero-dorsal to antero-ventral region. Weak longitudinal ridges occur on either side of diagonal ridge. Ventro- lateral border of valves project slightly outwards and downwards, with longitudinal groove above, particularly prominent in male dimorphs. Hinge antimerodont. Muscle scars and radial pore canals not seen. HoLotyPe. lIo.2118, base bed 22, Hunsdale Point, Cloughton Wyke. PARATYPES. Io.2119-36, horizon and locality as above and from horizon 2 ft. from base of bed 22, Hundale Point ; from bed 10, Bloody Beck and from base of bed 10, Hawsker. MIDDLE JURASSIC OSTRACODA 103 DESCRIPTION. Cavapace subquadrate in outline, more elongate in the male dimorphs. Greatest length of carapace extends through mid-point whilst the greatest height is in the anterior third. Greatest width in the posterior third, although there is only a slight increase in width posteriorly when compared with the width in the anterior part of the carapace. This is clearly seen in dorsal view, the carapace tending to be almost parallel-sided. Anterior broadly rounded, posterior triangular. Posterior marginal border narrow ; anterior border broad, directed obliquely back towards anterior cardinal angle. Dorsal margin concave in the left valve, convex in the right. Cardinal angles prominent in both valves. Ventral margin tends to be straight, the central part of the ventral surface being flattened, overhung on either side by the convex ventro-lateral margins, which in some specimens tend to project slightly outwards although not sufficiently developed to be termed alate. Left valve larger than the right which it overlaps along the ventral margin and strongly over- reaches along the dorsal margin. Terminally the left slightly overreaches the right with the possible exception of at the extreme posterior where there is no overreach or Fic. 8. Left side, female carapace, Cloughtonella rugosa sp. nov. Holotype, Ilo.2118, approx. x 85. overlap. Shell surface ornamented by a series of low ridges, the dominant ridge running obliquely across the carapace from the postero-dorsal region to the antero- ventral region. This oblique ridge is rather irregular in outline and is bounded on either side by short, also irregular ridges which tend to be lateral below the main diagonal ridge and vertical above. Just above the ventro-lateral margin, which in some specimens also bears lateral ridges, there appears to be a shallow groove which gives the ventro-lateral extension of the carapace a pinched-up appearance particularly noticeable in the male dimorphs. The intensity of the ornamentation varies in each specimen but generally the impression given is of a wrinkled carapace. The ventral surface is ornamented by 3-4 longitudinal ridges on each valve. Hinge antimerodont. In the left valve the terminal sockets are separated by a short median ridge above which the shell slopes upwards to the dorsal margin. Asa result there is virtually no accommodation groove developed. In the right valve only the anterior dentate element has been observed, bearing some 5-6 teeth. The median groove is poorly developed in the material examined but appears to be loculate. Inner margin and line of concrescence coincide, the duplicature being of moderate width. Radial pore canals almost imperceptible but can be made out as being short, straight and few in number. Muscle scars not observed. 104 R. H. BATE Fic. 9. Left side, male carapace, Cloughtonella vugosa sp. nov. Paratype, lo.2119, approx. x95. Dimensions Holotype, Io.2118, female carapace (Pl. 3, figs. 1-4; Text-fig. 8), length 0-53 mm. ; height 0-34 mm. ; width 0-28 mm. Io.2119, male carapace (Pl. 3, figs. 5-7; Text-fig. 9), length 0-55 mm.; _ height 0:28 mm.; width 0-26 mm. I[o.2120, male carapace (Pl. 3, figs. 12, 13), length 0-61 mm.; height 0-33 mm.; width 0:26 mm. lIo.2121, female left valve (Pl. 3, fig. 11), length 0-48 mm. ; height 0-32 mm. I[o.2134, male carapace (PI. 3, fig. 10), length 0-54 mm.; height 0-29 mm.; width 0-22 mm. [o0.2135, female carapace (Pl. 3, figs. 8, 9), length 0-51 mm.; height 0-32 mm.; width 0-26 mm. [o.2136, male right valve, length 0-60 mm. ; height 0-29 mm. REMARKS. Cloughtonella rugosa is close to the genus Aulacocythere in outline but as mentioned under remarks for the genus does not possess the generic characters of the latter. The present species is also close to Procytheridea hoffmanni Brand, which can, however, be distinguished by the strongly arched dorsal margin in the left valve ; presence of a definite accommodation groove and in the more positive ornamentation. The two species are probably congeneric however. Genus FUHRBERGIELLA Brand & Malz 1962. Subgenus PRAEFUHRBERGIELLA Brand & Malz 1962 Fuhrbergiella (Praefuhrbergiella) hovvrida Brand & Malz. RemarRKs. Brand & Malz (1962 : 19) described a new subgenus Praefuhrbergiella with Fuhrbergiella (P.) horrida as type species. Two subspecies were introduced : Fuhrbergiella (P.) horrida horrida having a range of romani to blagdent Zones and Fuhrbergiella (P.) horrida bicostata typically developed in the Sonninien-Schichten (sowerbyi to grandis Zones) but also occurring in the Coronaten-Schichten (pinguis Zone). The subspecies recorded here from the Grey Limestone Series is Fuhrbergiella (P.) horrida hornida. Fuhrbergiella (Praefuhrbergiella) horrida horrida Brand & Malz (Pl. 4, figs. 1-12) 1962. Fuhrbergiella (Praefuhrbergiella) horyvida horvida Brand & Malz: to, pl. 4, figs. 33-37; pl. 5, fig. 46. 1962. Huhrbergiella (Praefuhrbergiella) horrida horrida Brand & Malz: Simon & Bartenstein : 141, pl. 20, fig. 32; table 9. MIDDLE JURASSIC OSTRACODA 105 MATERIAL. Seventy-two specimens examined from the Grey Limestone Series, of which the following have been registered in the British Museum collections : Io.2109-17. DESCRIPTION. Cavapace subquadrate narrowing towards the posterior. Sexual dimorphism indicated by the presence of more elongate dimorphs, considered to be the males. Greatest length through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. Shell surface strongly reticulate, the reticulae in adults being extended into thin lamellae. Postero-dorsally a keel- like extension of the carapace projects above the dorsal margin of the valve bending down at about mid-point to die out close to the anterior margin below mid-length. Along the ventral surface a ridge is developed which turns upwards anteriorly towards an anterior vertical ridge which bounds the broad, flattened anterior border, to die out below the prominent eye node. The latter is situated just below the prominent, well rounded anterior cardinal angle. A short ventro-lateral ridge may be present above the ventral ridge in some specimens. Posterior marginal border also well developed. Anterior and posterior margins may possess short spines. Left valve larger than the right which it overlaps along the ventral margin and over- reaches along the antero-dorsal and postero-dorsal slopes. Hinge antimerodont with strongly loculate terminal sockets in the left valve and a long denticulate median bar. Accommodation groove poorly developed. In the right valve the hinge consists of some 8 anterior teeth and approximately 7 posterior teeth. Median groove poorly preserved in the present material. Inner margin and line of con- crescence coincide, the duplicature both anteriorly and posteriorly being quite broad. Radial pore canals long, straight and widely spaced ; 8-9 anteriorly and 4 posteriorly. The selvage forms a prominent ridge around the free margin, outside of which there is a narrow flange developed around the anterior margin and along the ventral margin. Only the 4 oval adductor scars have been seen in the present material, the anterior muscle scars not being preserved. Dimensions Io.2109, female left valve (Pl. 4, figs. 1-3), length 0-78 mm. ; height 0-45 mm. Io.2110, female carapace (Pl. 4, figs. 11, 12), length 0-68 mm. ; height 0-37 mm. ; width 0-38 mm. J[o.2111, male carapace (PI. 4, figs. 6-9), length 0-74 mm. ; height 0-37 mm.; width 0-37 mm. lIo.2115, female left valve, length 0-61 mm. ;_ height 0-35 mm. Io.2116, female right valve (Pl. 4, figs. 4, 5), length 0-75 mm. ; height 040mm. lo.2217, female right valve (PI. 4, fig. 10), length 0-73 mm. ; height 0-38 mm. ReMARKS. All the specimens of Fuhrbergiella present within the Grey Limestone Series are here referred to Fuhrbergiella (Praefuhrbergiella) horrida horrida although in some cases the presence of a short ventro-lateral ridge indicates some affinity with the subspecies F’. (P.) horrida bicostata Brand & Malz (1962 : 21, pl. 4, figs. 38-40). The variation observed is here, however, restricted to the subspecies F. (P.) horrida hornida. GEOL. II, 3 8 106 Wea debe Je} 10 18; Genus GLYPTOCYTHERE Brand & Malz 1962 Glyptocythere costata sp. nov. (Pl. 5, figs. 1-7) DiaGnosis. Glyptocythere with subquadrate, posteriorly tapered carapace. Ornamented medially by branching transverse ridges radiating outwards from dorsal margin; ventro-laterally by single longitudinal ridge and ventrally by second longitudinal ridge. HoiotyPe. 10.1775, base bed 10, Ravenscar. PARATYPES. [0.1776—82, from top and base of bed 10, and from bed 8, Ravenscar. DESCRIPTION. Carapace subquadrate, tapering towards the narrowly triangular posterior. Elongate dimorphs indicate the presence of males within the population. The shell surface is strongly ornamented by irregular, branching transverse ridges which radiate downwards from the dorsal margin of the valve and by two prominent longitudinal ridges. The uppermost ridge extends along the ventro-lateral part of the carapace whilst the lower ridge, being an increased development of one of the finer longitudinal ridges which occur on the ventro-lateral and ventral surfaces, is ventral in position and in some specimens forms a prominent ventral keel. Anterior and posterior with narrow marginal border. Left valve larger than the right which it overlaps along the ventral margin and overreaches around the anterior margin, in the region of the anterior cardinal angle and along the postero-dorsal slope. Greatest length through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. Anterior broadly rounded, posterior narrow, triangular, with short, concave, postero-dorsal slope and convex postero-ventral slope. Ventral margin incurved medially. Dorsal margin strongly convex in the right valve con- cave in the left. A low eye swelling is suggested in the right valve only, elongate in outline and situated below the anterior cardinal angle. Internal characters not observed. Dimensions Holotype, Io.1775, female carapace (Pl. 5, figs. 1-4), length 0-65 mm. ; height 0-40 mm. ; width 0:37 mm. Io.1776, female right valve (Pl. 5, fig. 7), length 0-62 mm.; height 0-35 mm. Io.1777, female carapace, length 0-60 mm.; height 0-38 mm.; width 0-36 mm. Io.1780, male carapace, length 0-80 mm. ; height 0-43 mm. ;widtho-4z2mm. _ [o.1782, male carapace (PI. 5, figs. 5, 6), length (broken) 0-59 mm. ; height 0-34 mm. ; width 0-32 mm. REMARKS. So far this species has only been found in sediments of the Grey Limestone Series as exposed at Ravenscar. Glyptocythere costata resembles Glypto- cythere dorsicostata Brand & Malz (1962a: 145, pl. 21, fig. 10, table g) although it differs in being smaller with a more pronounced posterior taper and in the absence of the dorso-median ridge which characterizes the latter. G. costata is also close to MIDDLE JURASSIC OSTRACODA 107 Progonocythere juglandica (Jones, 1884), Sylvester-Bradley (1948 : 193, pl. 12, figs. 5, 6, pl. 13, fig. 8; text-fig. 4) to which it could be ancestral. G. costata, however, differs in being smaller and in the possession of strong longitudinal ridges in the ventral and ventro-lateral regions. [It is here considered that the species Progonocythere juglandica should be assigned to the genus Glyptocythere, to which it bears greater relationship than to Progonocythere. This will, however, be discussed in more detail in a subsequent paper. | Glyptocythere polita sp. nov. (Pl. 5, figs. 8-11 ; Pl. 6, figs. 1-9) DiaGnosis. Glyptocythere with subquadrate/subtriangular carapace. Shell sur- face smooth with occasional wrinkles in dorso-median part. Downwardly projected ventro-lateral margin may be extended into thin keel, particularly in female dimorph. Horotyre. [o0.1724, bed 7, Hundale Point, Cloughton. PaRATYPES. [o0.1725—49 and Io.2210, beds, 7 and 8, Hundale Point, Cloughton ; beds 6, 7 and 9 Ravenscar and bed 6 Bloody Beck. DESCRIPTION. Cavapace subquadrate to subtriangular in outline with strong sexual dimorphism : the males being much more elongate in outline and quite strongly convex in dorsal view. Greatest length of carapace through mid-point with the greatest height in the anterior third. Greatest width just behind valve middle. Left valve larger than the right which it overlaps along the ventral margin and overreaches in the region of the antero-, and postero-dorsal slopes and slightly around the anterior margin. Shell surface smooth with a shallow median sulcus marking the position of the adductor muscle scars. Occasionally there is apparent a slight wrinkling of the dorso-median part of the carapace, but this is rarely well developed. Normal pore canals are large, circular and prominently displayed over the valve surface. The ventral surface of each valve is strongly ornamented with longitudinal ridges some of which may be bifurcate, about 4-6 per valve, the outermost being situated along the convex ventro-lateral extension of the carapace and may be developed as a thin, blade-like keel. One or two weaker longitudinal ridges may occur outside this keel on the ventro-lateral part of the carapace. The keel, when developed appears to be restricted to the female dimorphs, being little more than a well developed ridge in the males. Anterior margin of carapace broadly rounded with flattened marginal border. Posterior triangular with marginal border, concave postero- dorsal slope, almost straight in the male, and convex postero-ventral slope. Ventral margin medially incurved, sweeping upwards posteriorly in female dimorph. Ventro- lateral margin extended below ventral surface, also sweeping strongly upwards posteriorly in the female dimorph. Dorsal margin medially concave in the left valve with prominent cardinal angles ; in the right valve medially convex, although the strong median convexity noticeable here is really dorso-median in position. Hinge entomodont : left valve with terminal loculate sockets and a strongly dentate median bar, especially so antero-medially. Accommodation groove virtually absent. Right 108 RS SE AGL valve hinge complementary to the left. Hinge in juveniles antimerodont. Muscle scars consist of a subvertical row of 4 oval adductor scars with (as seen in a single specimen) a crescent-shaped antero-dorsal antennal scar. Mandibular scar not observed. This muscle scar type is placed in type A (Bate 1963 : 181) rather than type B. Inner margin and line of concrescence coincide the duplicature being of moderate width. Radial pore canals straight and widely spaced, approximately 9 anteriorly and 4 posteriorly. Anteriorly a narrow flange may be developed outside the selvage but is rarely preserved. Dimensions Holotype, Io.1724, female carapace (PI. 6, figs. 1-4), length 0-84 mm. height 0-52 mm. ; width 0-47 mm. Io.1725, male carapace (Pl. 6, figs. 5-8), length 1-18 mm.; height 0:53 mm. ; width 0-55 mm. [o.1736, female carapace (PI. 5, fig. 10), length 0-71 mm. ; height 0-46 mm.; width43 mm. Io0.1737, female carapace (PI. 6, fig. 9), length 0-85 mm. ; height 0-54 mm.; width o-45 mm. [o.1738, female left valve (Pl. 5, fig. 9), length 0-82 mm.; height 0-49 mm. [o0.1739, female right valve, length 0-77 mm. ; height 0-46 mm. l[o.1741, female left valve (Pl. 5, fig. 11), length 0-72 mm. ; height 0-46 mm. lo.1742, male carapace, length I-oo mm. ; height 0-50 mm. ; width 0-51 mm. Io.1743, female carapace (PI. 5, fig. 8), length 0-93 mm. ; height 0-56 mm.; width 0-48 mm. Io.1744, female carapace, length 0-71 mm.; height 0-45 mm.; width 0°37 mm. REMARKS. In general appearance Glyptocythere polita is similar to Progonocythere acuminata sp. nov., although it is not so strongly tapered posteriorly. The main differences between these two species are to be found in the dorsal margin, that of P. acuminata being straight and not convex as in G. polita. The dorsal margin of the left valve is also almost straight, or may be slightly concave, but not strongly so as in the present species. The much straighter dorsal margin of species of Progonocythere and the strongly convex dorsal development of the right valve in species of Glypto- cythere serve to distinguish the species of these respective genera. Glyptocythere scitula sp. nov. (Pl. 7, figs. 1-13; Pl. 8, figs. 1-9; Pl. 9, figs. 1-4; Text-fig. 10) DiaGnosis. Squat, subquadrate to elongate Glyptocythere with strong branching and anastomosing ridges, radiating from dorsal margin. Ventro-lateral and ventral surfaces ornamented with longitudinal ridges, some of which bend upwards antero- ventrally. All ridges with rounded surfaces. Internal characters as for genus. Ho.otyPe. [o0.1750, bed 5, Cayton Bay section. ParRATyPES. I[0.1751-74, from bed 5 Cayton Bay ; bed 5, Gristhorpe Bay ; beds 17 and 18, Bloody Beck ; 6 ft. from base and 15 ft. from base bed 12, Ravenscar and from bed 10, Hawsker. DESCRIPTION. Cavapace subquadrate in the female dimorph, elongate in the male. MIDDLE JURASSIC OSTRACODA 109 Greatest length through mid-point ; greatest height in the anterior third. Greatest width in the posterior third. Carapace strongly convex in dorsal view, slightly constricted medially. Anterior and posterior with flattened marginal borders. Left valve larger than the right which it overlaps most strongly mid-ventrally and over- reaches along the anterior, posterior and dorsal margins. The degree of overreach is most strongly developed mid-dorsally. Anterior margin broadly rounded ;_pos- terior triangular with concave postero-dorsal slope and convex postero-ventral slope. Ventral margin medially incurved ; dorsal margin strongly concave in the left valve convex in the right. The dorsal margin in the left valve besides projecting above that of the right valve is also very much thicker, almost keel-like. Ornamentation of carapace strongly developed and consists of transverse ridges which radiate outwards and downward from the dorsal margin, branching and anastomosing to produce a coarse reticulate pattern. The ridges are rounded and not lamellate. Ventro- lateral part of carapace strongly convex and ornamented by approximately 4 longitudinal ridges. The same number of longitudinal ridges also occur along the ventral surface of each valve. The ventro-lateral ridges tend to turn upwards antero-ventrally. In some specimens the ornament is only poorly developed, here the very large, circular normal pore canals are particularly evident, widely scattered over the shell surface. Hinge entomodont : left valve with terminal loculate sockets and a strongly dentate median bar of which the anterior part is more coarsely dentate. Accommodation groove virtually absent, the valve sloping upwards to the dorsal margin from the base of the median bar. In the right valve 6 anterior and 6 posterior teeth are developed. Median groove expanded anteriorly, strongly loculate. Muscle scars (Type A) consist of a subvertical row of 4 oval adductor scars, a small round, Fic. 10. Muscle scars, Glyptocythere scitula sp. nov. Female paratype, Io.1770, approx. x 300. antero-dorsal antennal scar and a much larger rounded antero-ventral mandibular scar which appears as a rosette of several smaller scars. Inner margin and line of concrescence coincide the duplicature being of moderate width. Radial pore canals long, straight, 9 anteriorly and 4 posteriorly. A narrow flange may be present around the anterior margin outside the selvage, whilst in the right valve a short ventral “lip” occurs just below the ventral incurvature of the valve. 110 RG dsl, IBA AIS Dimensions Holotype, Io.1750, female carapace (Pl. 8, figs. 1-4), length 0-72 mm. ; height 0-47 mm. ; width 0-43 mm. Io. 1751, male left valve (Pl. 7, figs. 11, 12), length 0-82 mm. ; height 0-45 mm. Io. 1752, female carapace (Pl. 8, figs. 6-9), length 0-77 mm. ; height 0-50 mm. ; width 0-47 mm. I[o.1753, male carapace (PI. 9, figs. 1-4), length 0-93 mm. ; height 054mm. ; widtho-52mm. [o0.1754, female left valve (PI. 8, fig. 5), length 0-75 mm. ; height 0-48 mm. Io.1756, female right valve, length 0-63 mm. ; height 0-36 mm. Io. 1760, female right valve (Pl. 7, figs. 1, 7), length (broken) 0-72 mm. ; height o-40mm. I[0.1768, female right valve (PI. 7, figs. 4, 6, 10), length 0-60 mm. ; height 0-35 mm. Io.1769, female left valve (PI. 7, figs. 5, 9), length 0-61 mm. ; height 0-37 mm. I[o.1771, female right valve (Pl. 7, figs. 3, 8), length 0-68 mm. ; height 0-37 mm. Io.1772, male right valve (PI. 7, fig. 13), length 0-82 mm. ; heighto-4omm._ [0.1773, female left valve, length 0-82 mm. ; height 0-48 mm. REMARKS. Glyptocythere scitula although smaller than Glyptocythere tuberodentina Brand & Malz (1962a : 143, pl. 21, figs. 11, 12 ; table 9) is similar in general appear- ance. The two species may, however, be distinguished by the ornamentation which in G. tuberodentina consists essentially of a reticulate ornament on the lateral surface with prominent longitudinal ridges ventrolaterally. Both the longitudinal ridges and the ridges which comprise the reticulations are thin and lamellate. In G. scitula the ornament as described consists of radiating transverse ridges which produce a coarse reticulation on branching and anastomosing. In this species the longitudinal ridges and the transverse ridges are rounded, contrasting strongly with the lamellate ridges of G. tuberodentina. The known range of G. tuberodentina is from the middle of the acvis Zone to the top of the friederici-augusti Zone of the Parkinsoni-Schichten. The range of G. scztula is more difficult to give precisely but probably does not occur higher than the blagdenz Zone of the Coronaten-Schichten. Genus MALZIA nov. Named after Dr. Heinz Malz of the Senckenberg Museum, Frankfurt-am-Main, Germany. DiaGnosis. Progonocytherinae with subquadrate carapace, tapering to posterior. Anterior and posterior with flattened marginal borders. Ventro-lateral part of carapace extended into one or two keel-like projections. Low eye swelling developed at anterior cardinal angle. Species may be dimorphic. Hingeentomodont. Radial pore canals long and straight, approximately 8 anteriorly, 3 posteriorly. Muscle scars as for subfamily (Type A). Type SpEciES. Malza bicarinata sp. nov. REMARKS. The genus Malzia (feminine) is erected here with two species: M. bicarinata having two ventro-lateral keel-like extensions and M. wnicarinata sp. nov., having but a single valvular extension. It is considered that Malzia has a position MIDDLE JURASSIC OSTRACODA III transitory between Progonocythere Sylvester-Bradley (1948 : 189) and Marslatourella Malz (1959: 19). This is suggested by the general similarity of carapace outline, muscle scars and radial pore canals present in all three. Progonocythere has a strong entomodont hinge and only a faint suggestion of an eye swelling. In Marslatourella the hinge is antimerodont and a strong eye tubercle is developed. At the same time prominent ventro-lateral outgrowths of the carapace occur. Malzia possesses a hinge which can be considered as entomodont, a low eye swelling and ventro-lateral outgrowths of the carapace. It appears, therefore, that the development of ventro- lateral outgrowths coupled with a change from an entomodont to an antimerodont hinge and the development of an eye tubercle results in the appearance in the Bathonian of the genus Marslatourella. This evolutionary series commences in the Bajocian with the genus Progonocythere an offshoot of which produces Malza. This second lineage then continues giving rise to Marslatourella. The genus Marslatourella described by Malz from the Bathonian of France (Mars-la Tour and Boulonnais) and Germany (Eichberg) is also common in the Bathonian sediments of England. Species of this genus will be described in forthcoming pub- lications. Malzia bicarinata sp. nov. (Pl. 9, figs. 5-8; Pl. xo, figs. 1-3; Text-figs. 11-14) Diacnosis. Malzia with two, short, well developed ventro-lateral keels. Internal details as for genus. Hototyre. l[o.1797, bed 9, Ravenscar section. PARATYPES. Io.1798-1800, beds 7 and 8, Ravenscar and bed 7, Bloody Beck. 12 Fic. 11. Muscle scars. Malzia bicavinata sp. nov. Paratype, lo.1800, approx. x 290. Fics. 12-13. Dorsal and left views, complete carapace Malzia bicarvinata sp. nov. Holotype, 10.1797, approx. x 85. 112 We, Jel, 1ss4\ 1018: DESCRIPTION. Carapace subquadrate in outline with straight or very slightly convex dorsal margin. Cardinal angles well developed. Anterior rounded ; posterior triangular with concave postero-dorsal slope and convex postero-ventral slope. Ventral margin medially incurved. Anterior and posterior with flattened marginal borders. Shell surface laterally smooth with two, stubby keels developed, the uppermost of which tends to project slightly outwards, particularly noticeable in dorsal view. Ventral surface may be smooth or possess two longitudinal ridges on each valve. A low eye swelling is situated just below the anterior cardinal angle, particularly noticeable in the ~ight valve. Greatest length of carapace through mid- point ; greatest height in the anterior third ; greatest width in the posterior third. Left valve larger than the right which it overlaps mid-ventrally and slightly over- reaches around the posterior and along the dorsal margin. Hinge entomodont, only seen in the left valve where the terminal loculate sockets are separated by a dentate median bar the dentition of which increases in coarseness towards the anterior. Accommodation groove narrow, elongate. Muscle scars (Type A), the antennal scar being round and antero-dorsal in position. Mandibular scar not seen. Radial pore canals straight and simple : approximately 8 anteriorly and 3 posteriorly. Duplica- ture in completely seen in present material. Selvage prominent external to which there is a narrow flange developed around the anterior margin and along the ventral margin. Fic. 14. Dorsal view, left valve hinge, Malzia bicarinata sp. noy. Paratype, 1lo.1799, approx. X 105. Dimensions Holotype, Io.1797, carapace (PI. 9, figs. 5-8 ; Text-figs. 12-13), length 0-70 mm. ; height 0-43 mm. ; width 0-41 mm. MIDDLE JURASSIC OSTRACODA 113 Io.1798, carapace (PI. 10, fig. 1), length 0-69 mm. ; height 0-43 mm. ; width 0-38 mm. Io.1799, left valve (Pl. ro, figs. 2, 3; Text-fig. 14), length 0-70 mm. ; height 0-43 mm. REMARKS. Malzia bicarinata cannot easily be confused with Marslatourella exposita Malz (1959: 20, text-figs. 1-4), lacking the prominent eye tubercle and antimerodont type hinge of the latter. It is, however, sufficiently close as to be considered ancestral to M. exposita. The present species differs from species of Progonocythere in the development of ventro-lateral outgrowths of the carapace and in the development of an eye swelling. Although some*species of Progonocythere may show the development of an eye swelling it is not so prominent as in Malzia although even here it is hardly a dominant feature of the carapace. Malzia unicarinata sp. nov. (Pl. ro, figs. 4-10 ; Pl. 11, lig. 1-4; Text-fig. 15) Diacnosis. Malza, with single ventro-lateral keel. Dimorphic. Hototyre. l[o.1801, bed 9, Ravenscar section. PARATYPES. I[o0.1802-6, beds 8 and 10, Ravenscar and bed 4, Bloody Beck. Fic. 15. Right side, female carapace, Malzia unicarinata sp. nov. Holotype, Io.1801, approx. X 105. DESCRIPTION. Carapace subquadrate in outline in the female dimorphs, elongate in the male. Ventro-lateral part of carapace convex, extended as a short, rather stubby keel. Greatest length through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. Shell surface punctate. Anterior broadly rounded ; posterior broadly triangular with concave postero-dorsal slope and convex postero-ventral slope. Ventral margin medially incurved. Dorsal margin very slightly convex in the female, straight in the male although slightly concave just behind the anterior cardinal angle. Cardinal angles prominent, the anterior angle being broadly rounded with the posterior angle tending to be more acute. Anterior and posterior margins with well defined borders. Eye swelling elongate in 114 Wee Jel, 1sj/AVIl 1B, outline situated below the anterior cardinal angle, more prominently developed in the male dimorph. Left valve larger than the right which it overlaps mid-ventrally and overreaches along the remaining part of the ventral margin and along the dorsal margin. Internal details not seen. Radial pore canals as seen from the exterior straight and widely spaced, approximately 8 anteriorly and 3 posteriorly. A narrow flange extends around the anterior margin and although not clearly seen probably also along the ventral margin. Dimensions Holotype Io.1801, female carapace (PI. 10, figs. 4-8 ; Text-fig. 15),lengtho-7Imm. ; height 0-42 mm. ; width 0-41 mm. Io.1802, female carapace (PI. 10, figs. 9, 10), length 0-73 mm. ; height 0-44 mm. ; width 0-43 mm. lIo.1804, female carapace, length 0-75 mm.; height 0-46 mm. ; width 0-43 mm. I[o.1806, male carapace (PI. 11, figs. 1-4), length 0-85 mm. ; height 0-45 mm. ; width 0-44 mm. Remarks. Malzia unicarinata is distinguished easily from M. bicarinata by the possession of but a single ventro-lateral keel. The male dimorph of M. wnicarinata is, however, similar to Progonocythere acuminata sp. nov., from which it can be identified by the mid-laterally swollen, strongly convex carapace, a feature which characterizes species of this genus, and by the possession of a short stubby keel. Genus PROGONOCYTHERE Sylvester-Bradley 1948 Progonocythere acuminata sp. nov. (Pl. 11, figs. 5-10 ; Pl. 12, figs. 1-4) DiaGnosis. Pvogonocythere with posteriorly acuminate carapace. Ventro-lateral margin convex with knife-edge keel developed in some specimens. Lateral surface smooth, though may possess faint transverse furrows medially. Low eye swelling at anterior cardinal angle. Ventral surface with longitudinal ridges in region of ventro-lateral overhang. Anterior with well developed border. Hootype. [0.1783, bed 7, Hundale Point, Cloughton. PARATYPES. I0.1784-91, bed 7, Hundale Point; bed 6, Cayton Bay; bed 7, Bogmire Gill; top bed 7, Ravenscar and bed 10, Hawsker. DESCRIPTION. Carapace elongate, tapering strongly towards thesharply acuminate posterior. Sexual dimorphism not apparent. The ventro-lateral border of the carapace is extended below the ventral surface and generally possesses a thin, knife- edge keel, developed from one of the longitudinal ridges which extend along the undersurface of the ventro-lateral part of the carapace, remainder of ventral surface smooth. The ventro-lateral margin of the carapace sweeps obliquely upwards just behind valve middle. Shell surface smooth, although weak transverse furrows may be observed in some specimens in the mid-lateral area. Normal pore canals often well developed, large and circular, widely scattered over the carapace. Greatest length MIDDLE JURASSIC OSTRACODA 115 through mid-point ; greatest height in the anterior third ; greatest width at or just behind middle. A shallow, indistinct sulcus, medially situated marks the position of the adductor muscle scavs. Anterior broadly rounded with a distinct marginal border ; posterior narrow, acuminate with a short, concave postero-dorsal slope and a convex postero-ventral slope. Ventral margin broadly concave. Dorsal margin slightly convex in the right valve slightly concave medially in the left. Cardinal angles prominent. Below the anterior cardinal angle an oblique swelling, separated off from the convex part of the carapace below by an oblique groove, is suggestive of an eye swelling. Left valve larger than the right which it overlaps along the ventral margin and overreaches along the dorsal margin and around the anterior. The left valve may also overreach the right along the postero-dorsal slope but not at the extreme posterior. Hinge poorly seen in a single individual (left valve) where the terminal sockets are separated by a median groove which can be made out as dentate. Accommodation groove elongate, deep. Duplicature appears to be of moderate width, but imperfectly seen. Other internal details not observed. Dimensions Holotype, Io.1783, carapace (Pl. 12, figs. 1-4), length 0-80 mm. ; height 0-49 mm. ; width 0-38 mm. Io.1784, left valve, length 0-71 mm.; height 0-41 mm. [o0.1786, left valve (Pl. II, figs. 7, 8), length 0-60 mm. ; height 0-36 mm. I[o.1787, carapace (PI. 11, figs. 9, 10), length 0-69 mm. ; height 0-41 mm. ;. width 0-34 mm. [o.1789, carapace (PI. II, figs. 5, 6), length 0-73 mm. ; height 0-41 mm.; width 0-36 mm. [o.1791, right valve, length 0-71 mm. ; height 0-36 mm. Remarks. The similarity of P. acuminata to the male dimorph of Malza um- cavinata has already been discussed and need not be gone into again. The male dimorph of Progonocythere yonsnabensis sp. nov. also bears some resemblance to the Fics. 16-17. Dorsal and left views, female carapace, Progonocythere yonsnabensis sp. nov. Holotype, Io. 1792, approx. x95. 116 R. H. BATE present species although it can be readily distinguished by its much smaller size and by the distinct median sulcus not present to such an extent here. Progonocythere yonsnabensis sp. nov. (Pl. 12, figs. 5-14; Pl. 13, figs. 1-4; Text-figs. 16-19) DiaGnosis. Small progonocytherid with distinct median sulcus. Eye swelling prominent. Ventro-lateral margin extended into ventral keel. Shell surface punctate. Hototyre. I[o0.1792, bed 5, Cayton Bay section. PARATYPES. [0.1793-96, horizon and locality as above. DESCRIPTION. Carapace small, subquadrate; elongate in the male dimorph. Greatest length through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. As seen in dorsal view a shallow sulcus divides the cara- pace into an anterior and a posterior half. This sulcus is more strongly developed in the female. Ventro-lateral margin strongly convex, extended ventrally into a prominent keel. Ina single male specimen there is a second, short keel dorsal to the first and terminating ina smallnode. Almost directly above this node a second node is situated dorso-medially on each valve of the carapace posterior to the median sulcus. In all other respects this specimen is identical to the others. Dorsal margin strongly convex in the left valve, slightly convex in the right. Anterior rounded with a short, obliquely convex antero-dorsal slope. Posterior triangular: postero- dorsal slope concave; postero-ventral slope convex. Ventral margin medially incurved. Cardinal angles prominent, the posterior angle being the more strongly angled of the two, anterior angle tending to be broadly rounded. A prominent oval Fics. 18-19. Left and dorsal views, male carapace, Progonocythere yvonsnabensis sp. nov., showing development of second keel and tubercles. Paratype, lo.1795, approx. X95. MIDDLE JURASSIC OSTRACODA 117 eye swelling is situated below the anterior cardinal angle. Anterior and posterior marginal borders distinct. Shell surface punctate. Left valve larger than the right which it overlaps mid-ventrally and overreaches postero-ventrally, postero-dorsally and antero-dorsally. Mid-dorsally the left valve is strongly projected above the right. There is little or no overreach around the anterior margin. Internal charac- ters not observed for this species. Dimensions Holotype, Io.1792, female carapace (Pl. 12, figs. 5-8; Text-figs. 16-17), length 0-54 mm. ; height 0-32 mm. ; width 0-30 mm. Io.1793, male carapace (PI. 12, figs. 11-14), length 0-57 mm. ; height 0-34 mm. ; width 0-31 mm. lIo.1794, female carapace (PI. 12, figs. 9, 10), length 0-60 mm. ; height 0-36 mm. ; width 0-35 mm. [o.1795, male carapace (PI. 13, figs. 1-4 ; Text- figs. 18-19), length 0-63 mm. ; height 0-36 mm. ; width 0-35 mm. REMARKS. The single specimen (I0.1795) mentioned in the description, possessing two ventro-lateral keels and two lateral nodes is considered to be simply an extreme variant of the present species. Possibly this morphological variation may have been brought about by changes in the salinity of the water. Morphologically the posses- sion of two lateral keels results in this specimen having some resemblance to Malzia bicarinata, although it can be distinguished by the possession of a prominent median sulcus and a very much smaller adult size. The species Progonocythere yonsnabensis more closely resembles Progonocythere cristata Bate (1963 : Io1, pl. 4, figs. 5-15; pl. 5, figs. 1-6) than any other ostracod. It is, however more strongly sulcate, possesses a more prominent eye swelling and is very much smaller. Subfamily PLEUROCYTHERINAE Mandelstam 1960 Genus PLEUROCYTHERE Triebel 1951 Pleurocythere sp. (PL. 13, fig. 5) Remarks. A single left valve belonging to the genus Pleuwrocythere but not readily assignable to any known species is here recorded from the Gristhorpe Bay section, bed 4. The ornamentation of the valve consists of a longitudinal ventro- lateral ridge, an oblique median ridge and a short diagonal ridge which is situated below the anterior cardinal angle. This short ridge V’s downwards, connected by a short ridge at the apex of the V to the median ridge. The dorsal end of the diagonal ridge terminates in an eye swelling. Shell surface between the ridges strongly reticulate. Dimensions. 10.1836, left valve (Pl. 13, fig. 5), length 0-80 mm. ; height 0-35 mm. Family CYTHERIDEIDAE Sars 1925 Subfamily CYTHERIDEINAE Sars 1925 Genus VERNONIELLA Oertli 1957 118 Re Ee Atsaas REMARKS. The genus as diagnosed by Oertli (1957 : 659) possesses either a hemi- merodont or antimerodont type hinge, and is considered to be without any strong ornamentation. Two species are here placed in Vernoniella: V. bajociana sp. nov., a smooth form possessing an antimerodont hinge and the strongly ornamented V. ? caytonensis sp. nov. There is, however, some uncertainty concerning the generic designation of the last named species. Vernoniella bajociana sp. nov. (Pl. 13, figs. 6-11 ; Pl. 14, figs. 1-13) Diacnosis. Vernoniella with subquadrate carapace, elongate in male dimorphs. Shell surface finely punctate. Anterior and posterior margins with narrow marginal borders. Hinge antimerodont. Radial pore canals long, straight, approximately 10-11 anteriorly, 4—5 posteriorly. HorotyrPe. Io.1807, bed 23, Hundale Point, Cloughton. PaRATYPES. Io.1808—30 and Io.2988, bed 5, Cayton Bay ; beds 7, 8, 23 and 25, Hundale Point ; bed 12, Ravenscar and beds 9 and 10, Hawsker. DESCRIPTION. Cavapace subquadrate to subtriangular in the female dimorph, elongate in the male. Shell surface punctate. Ventral surface may possess weak longitudinal ridges. In the dorso-median part of the carapace the valve is slightly constricted and here may exhibit weak, transverse furrowing. Greatest length of carapace through mid-point ; greatest height in the anterior third ; greatest width in the posterior third, although there is here only a slight increase over the width in the anterior third. Carapace appears compressed in dorsal view, almost parallel-sided. Dorsal margin medially concave in the left valve (both dimorphs), convex in the right. Cardinal angles in both valves prominent : anterior angle broadly rounded, posterior angle sharply angled and situated at extreme posterior of the carapace. Ventral margin medially incurved. Ventral surface overhung slightly by the convex ventro- lateral margin. The incurving of the ventral surface produces, particularly in the male, an enlarged anterior portion of the carapace as seen in lateral view. Anterior broadly rounded with marginal border which is separated off from the convex lateral part of the carapace by a marginal groove. Posterior broadly triangular with a steeply angled, straight or slightly convex postero-dorsal slope and a convex postero- ventral slope. Left valve larger than the right which it overlaps along the ventral and postero-ventral margins and overreaches around the anterior margin and along the postero-dorsal slope. Mid-dorsally the convex dorsal margin of the right valve projects above the left. Hinge antimerodont: left valve with terminal loculate sockets and a strong denticulate median bar. No accommodation groove. Right valve with 6 terminal teeth, dorsally bifid and an elongate, finely locellate median groove. Muscle scars (Type B?) with a subvertical row of 4 oval adductor scars, a rounded antero-ventral mandibular scar which as seen in one individual is composed of a rosette of several small scars and an antero-dorsal antennal scar which is clover- leafin shape. The antennal scar appears to be formed by the fusion of at least three MIDDLE JURASSIC OSTRACODA 119 scars which together form a clover-leaf pattern, or may appear rounded. Between the antennal scar and the dorsal adductor scar there is a large depression which is the mandibular support spot. This muscle scar arrangement is tentatively placed as Type B. Inner margin and line of concrescence coincide, the duplicature being of moderate width. Radial pore canals straight, approximately 10-11 anteriorly and 4-5 posteriorly. Outside the se/vage in the right valve a flange is developed around the anterior margin, extending back along the ventral margin, where it expands opposite the ventral incurvature into a broad “ lip ” Dimensions Holotype, Io.1807, female carapace (Pl. 13, figs. 6-9), length 0-67 mm. ; height 0-38 mm. ; width 0-30 mm. Io. 1808, female carapace (PI. 13, figs. 10, 11), length 0-66 mm. ; height 0-37 mm. width 0-26 mm. [o.1812, female left valve, length 0-69 mm.; height 0-44 mm. Io.1814, male carapace (PI. 14, figs. 1-4), length 0-84 mm. ; height 0-43 mm. ; width 0-29 mm. l[o.1816, female carapace, length 0-61 mm.; height 0-34 mm.; width 0-28 mm. l[o.1818, female left valve (Pl. 14, figs. 5-9), length 0-48 mm. ; height 0-31 mm. lIo.1819, female right valve (Pl. 14, figs. 10, 12, 13), length 0-65 mm. ; height 0-37 mm. l[o.1821, male carapace, length 0-68 mm.; height 0-37 mm. ; width 0-28. REMARKS. Vernoniella bajociana is similar in general external appearance to V. sequana Oerth (1957 : 659, pl. 3, figs. 70-85) but has an antimerodont instead of a hemimerodont hinge, and also lacks the accommodation groove which is present in V. sequana. V. bajociana has a distinct marginal border, a feature not present in Oertli’s species. Vernoniella ? caytonensis sp. nov. (PI. 15, figs. 1-9) Diacnosis. Vernoniella ? with elongate, posteriorly acuminate carapace. Shell surface strongly ornamented by 4—5 longitudinal ridges arranged in broad, inverted V. Two to three obliquely transverse ridges occur anterior to these. Anterior and posterior with distinct marginal borders. Normal pore canals prominent. Hototype. [0.1855, bed 6, Cayton Bay section. PARATYPES. Io.1856—69, bed 4, Gristhorpe Bay ; bed 6 Cayton Bay and bed 11, Hawsker. DESCRIPTION. Cavapace elongate tapering towards the narrowly rounded pos- terior. Greatest length through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. In dorsal view the carapace is slightly con- stricted medially. Anterior broadly rounded; posterior narrowly rounded with steeply angled, convex postero-dorsal slope and convex postero-ventral slope. Dorsal margin in the right valve medially convex, in the left valve thickened and medially concave. Cardinal angles prominent; the anterior angle being broadly 120 15, Jals Is /AN AID, rounded, posterior angle more acute, situated at extreme posterior of carapace. Anterior and posterior with distinct marginal borders. Ventral margin medially incurved. Shell surface strongly ornamented by 4-5 longitudinal ridges arranged in a broad inverted V, the lowermost being almost straight. These ridges are situated on the strongly convex part of the carapace which is cut off posteriorly by the posterior border and anteriorly by an oblique groove running just behind and below the anterior cardinal angle. In front of this groove there are 2-3 obliquely transverse ridges which antero-ventrally bend back to extend along the ventro- lateral and ventral surfaces where weak longitudinal ridges may be made out. Normal pore canals distinctly secn along the ventral surface where they are large and circular, elsewhere tending to be masked by lateral ornamentation. Left valve larger than the right which it overlaps along the ventral margin and overreaches in the region of the antero-dorsal and postero-dorsal slope. Hinge incompletely seen : right valve with the terminal elements consisting of approximately 6 dorsally bifid teeth. Median groove obscured by matrix. Left valve hinge not seen. Inner margin and line of concrescence appear to coincide. Fadial pore canals long, straight and widely spaced, approximately 8 anteriorly and 3-4 posteriorly. Below the ventral incurvature in the right valve an elongate “ lip’ is developed. Dimensions Holotype, Io.1855, carapace (PI. 15, figs. 1-4), length 0-63 mm. ; height 0-33 mm. ; width 0-30 mm. Io.1857, right valve (Pl. 15, figs. 8, 9), length 0-66 mm. ; height 0-33 mm. [o.1858, carapace (PI. 15, figs. 5-7), length 0-66 mm. ; height 0-37 mm. ; width 0-32 mm. REMARKS. In outline this species approaches Vernoniella and as a result has been placed tentatively in that genus. It is, however, a strongly ornamented form whose hinge and muscle scars are as yet unknown. Vernoniella ? caytonensis resembles the male dimorph of Eocytheridea carinata Bate (1964 : 18, pl. 5, figs. 5-8) primarily because of the similarity of ornament although it is easily distinguishable by the more prominent posterior cardinal angle, the not so strongly tapered carapace, and by the more strongly pronounced marginal borders. Genus LJUBIMOVELLA Malz 1961 Lijubimovella piriformis Malz (Pl. 15, figs. 10-13 ; Pl. 16, figs. 1, 2) 1949. Ostracod 96 Brand : 337, pl. 10 (fauna 2), fig. 4, pl. 14. 1961. Ljubimovella piriformis Malz : 165, pl. 2, figs. 15-25; text-figs. 2-3. 1962. Ljubimovella piriformis Malz: Brand & Fahrion : 134, pl. 20, fig. 33; table 9g. MATERIAL. Twenty-three specimens examined from the Grey Limestone Series, of which the following have been registered in the British Museum collections : Io.2103-08. MIDDLE JURASSIC OSTRACODA 121 DESCRIPTION. Cavapace piriform, very much enlarged in the anterior third, in which region it attains maximum height. Greatest length below mid-point ; greatest width at, or in front of valve centre. A strong, downwardly projected spine is developed at the postero-ventral angle whilst a much shorter spine may also be present on the lower half of the anterior margin. Shell surface smooth except for the ventral surface where a number of fine longitudinal striae occur. Normal pore canals circular and widely scattered over the surface. Dorsal margin slightly concave medially ; cardinal angles broadly rounded, although the posterior angle may be acute. Ventral margin with a strong median incurvature. Anterior broadly rounded ; posterior obliquely rounded with a relatively long, convex postero-dorsal slope and a short, almost vertical but slightly convex postero-ventral slope which terminates in the prominent posterior spine. Left valve larger than the right which it overlaps along the ventral margin. Along the antero-dorsal and postero-dorsal slopes the left valve tends to overreach the right, whilst mid-dorsally the right valve overreaches the left. Hinge hemimerodont: left valve without an accommodation groove ; right valve with smooth median groove and terminal dentate elements. Muscle scars not clearly seen. Inner margin and line of concrescence coincide. Radial pore canals short and widest at their base, 5 in number antero-ventrally and 5 postero-ventrally, of which the lowermost passes down through the centre of the posterior spine. A well developed flange extends around the anterior margin and along the ventral margin in the right valve, not observed in the left valve probably because of the preservation of the material. Dimensions Io.2103, carapace, length 0-69 mm. ; height 0-34 mm. ; widtho-31 mm. _ [o.2106, carapace (Pl. 16, figs. 1, 2), length 0-75 mm.; height 0-40 mm.; width 0-32 mm. Io.2107, juv, carapace (PI. 15, figs. Io, 11), length 0-44 mm.; height 0-25 mm. ; width 0-22 mm. [o.2108, right valve (PI. 15, figs. 12, 13), length 0-72 mm. ; height 0-37 mm. ReMARKS. The single juvenile specimen observed of this species exhibits a much enlarged anterior half of the carapace. It would appear that in subsequent instars of the species there is a proportionately greater increase in the posterior part of the carapace, a process almost certainly related to the development of the reproductive organs as the animal reaches maturity. Family SCHULERIDEIDAE Mandelstam 1959 Subfamily SCHULERIDEINAE Mandelstam 1959 Genus MESOCYTHERIDEA noy. DiacGnosis. Schulerideinae with oval/subquadrate carapace. Central part of valve strongly cut off from remainder of carapace by oblique groove below the anterior and posterior cardinal angles, particularly well developed in right valve. Dorso- median part of right valve strongly projected above dorsal margin. Hinge anti- GEOL. II, 3 9 122 IRS lele JE VN IIe merodont with anterior socket of left valve cutting back into median bar. Radial pore canals long, slightly curved, Io anteriorly, 4 posteriorly. Inner margin and line of concrescence coincident. Muscle scars (Type C) as for family. Left valve larger than right. Type species. Mesocytheridea howardianensis sp. nov. REMARKS. The genus (feminine) is close to Eocytheridea Bate (19634 : 35) from which it has probably evolved. Mesocytheridea is distinguishable from Eocytheridea by the presence of a strongly antimerodont hinge and a slight reduction in the num- ber of anterior radial pore canals (10 as against 14) which tend to be straighter. The species Eocytheridea carinata Bate (1964 : 18, pl. 4, figs. 6-11 ; pl. 5, figs. 1-8) is very close to this genus but has a hemimerodont hinge and the radial pore canals are slightly more curved. Mesocytheridea howardianensis sp. nov. (Pl. 16, figs. 3-11 ; Pl. 17, figs. 1-3; Text-figs. 20, 21) DraGnosis. Mesocytheridea with oval/subquadrate carapace, convex in dorsal view. Ornamentation consists of some 5-6 low, broad ridges arranged in an inverted V. The lowermost, ventro-lateral ridges are longitudinal, thereby forming base to V. Ornamentation never strong, may be almost completely lacking, when shell appears smooth. Shell surface with prominent, circular, normal pore canals. Hinge, muscle scars, radial pore canals as for genus. Left valve larger than right. Species dimorphic. HototypPe. I[o.1870, bed 2, Stonecliff Wood, locality SE/740675. PaRATyPES. I[o.1871-81, horizon and locality as above and 1 ft. 6 in. and 3 ft. 8 in. from base bed 8, Stonecliff Wood, locality SE/743676. Fic. 20. Right view, female carapace, Mesocytheridea howardianensis sp. nov. Holotype, Io.1870, approx. X93. DESCRIPTION. Cavapace oval/subquadrate, the male dimorphs being more elongate in outline. Greatest length through mid-point; greatest height in the anterior third ; greatest width in the posterior third. Dorsal margin medially concave in the left valve with well rounded cardinal angles. In the male dimorph the dorso-median MIDDLE JURASSIC OSTRACODA 123 part of the left valve tends to project slightly above the dorsal margin, but not so strongly as in the right valve. Cardinal angles in the right valve more acute than in the left. Ventral margin medially incurved, convex in the posterior half where the carapace appears to be noticeably swollen. Anterior rounded with oblique, straight or slightly convex antero-dorsal slope. Posterior narrowly rounded with convex postero-dorsal and postero-ventral slopes. Narrow, convex, anterior and posterior marginal borders are separated from the lateral part of the carapace by a marginal groove. Dorsal margin of the left valve noticeably thickened. Shell surface with something like 5-6 low, rather poorly developed ridges which are arranged in the form of an inverted V, the apex of which reaches the dorsal margin, more noticeably developed in the right valve. Several of the lower, ventro-lateral ridges are straight and form a base to the inverted V. A strong diagonal groove separates the central part of the valve from the terminal parts, particularly in the right valve. The an- terior groove extends diagonally below the anterior cardinal angle whilst the posterior groove extends below the posterior angle. A transverse ridge extends from the anterior cardinal angle towards the antero-ventral angle. Remainder of shell surface punctate, although a few very minor ridges may connect some of the antero- dorsal ridges. The degree with which the ornamentation is developed varies with each individual, but is often so poorly represented that the specimen appears smooth. Fic. 21. Internal view, right valve showing hinge, Mesocytheridea howavdianensis sp. nov. Paratype, Io. 1879, approx. x95. In all cases the normal pore canals which are large and circular, are prominently displayed over the shell surface. Left valve larger than the right which it overlaps along the ventral margin and overreaches practically everywhere else apart from mid- dorsally where the two valves are drawn apart. Hinge antimerodont: left valve with elongate terminal loculate sockets separated by a strong, denticulate median bar. The anterior socket, and to some extent the posterior socket also, tends to cut back into the median bar. There is virtually no accommodation groove. In the right valve the median groove is quite strongly locellate, whilst terminally there are 5 anterior and 7 posterior teeth, all dorsally bifid and well developed. Muscle scars (Type C) consist of a slightly crescentic row of 4 oval adductor scars with an antero- median, round, antennal scar. No mandibular scar observed. Inner margin and GEOL. II, 3 9§ 124 R. H. BATE line of concrescence coincide, duplicature being quite broad. Radial pore canals long and widely spaced, some straight others slightly curved ; 10 anteriorly and 4 pos- teriorly. Dimensions Holotype, Io.1870, female carapace (Pl. 16, figs. 3-6; Text-fig. 20), length 0-73 mm.; height 0-43 mm. ; width 0-38 mm. To.1871, female right valve (Pl. 16, fig. 9, Pl. 17, figs. 2, 3), length 0-71 mm. ; height 0-37 mm. Io0.1872, male left valve (Pl. 17, fig. 1), length 0-93 mm. ; height 0-47mm. I[o.1875, female left valve (Pl. 16, fig. 11), length 0-76 mm. ; height 0-44 mm. Io.1877, male right valve, length 0-71 mm. ; height 0-36 mm. Io.1879, male right valve (Pl. 16, figs. 8, 10), length 0-78 mm. ; height 0-41 mm. I[o.1881, female left valve (Pl. 16, fig. 7), length 0-71 mm. ; height 0-41 mm. RemMARKS. This is a much larger species than Eocytheridea carinata with which it bears some resemblance, although in M. howardianensis the ornamentation tends to be much weaker and the presence of a strong antimerodont hinge and a slight reduc- tion in the number of radial pore canals help to distinguish the two species. The majority of the specimens examined come from the uppermost beds of the Grey Limestone Series as exposed along the western shore-line. However a population belonging to this species has been found at the base of the Series in the Bloody Beck section. The specimens here are generally smaller, no males having been found and are very poorly ornamented. They are not, however, considered to be sufficiently distinct as to be separated specifically. Genus PRAESCHULERIDEA Bate 1963 Praeschuleridea subtrigona (Jones & Sherborn 1888) Synonymy. See Bate (1964 : 22). REMARKS. Two subspecies of P. subtrigona have been recognized: P. subtrigona subtrigona having a size range of up to 0-56 mm. in the female dimorph and 0:58 mm. in the male; and P. subtvigona magna where the range extends to 0-73 mm. for the female and 0-83 mm. for the male. Within the Grey Limestone Series, however, there is a third subspecies having a maximum size of about 0-68 mm. for the female and 0-77 mm.—o-82 mm. for the male. There are also a number of minor morphological details which help to distinguish this third subspecies. As all the specimens examined from the Grey Limestone Series fall into this intermediate range the subspecies Praeschuleridea subtrigona intermedia subsp. nov. has been erected to account for them. Praeschuleridea subtrigona intermedia subsp. nov. (Pl. 17, figs. 4-10 ; Pl. 18, figs. 1-9) Diacnosis. A subspecies of Praeschuleridea subtrigona in which adult female is of the order of 0-68 mm. and male is 0-77 mm. to 0-82 mm. Carapace subtrigonal, MIDDLE JURASSIC OSTRACODA 125 punctate. Posterior dorsal margin virtually straight, sloping strongly to posterior. Anterior dorsal margin long, obliquely convex. Posterior with steeply inclined, convex postero-ventral slope. Anterior with narrow marginal border; posterior border poorly developed. HototyPe. [0.1837, bed 7, Hundale Point, Cloughton. PaRATYPES. I[0.1838-54, bed 4, Gristhorpe Bay; bed 12, White Nab, Scar- borough ; bed 7 & base bed 22, Hundale Point and bed 2 Stonecliff Wood, locality SE/740675. DESCRIPTION. Carapace subtrigonal in outline, more elongate in the male dimorph. Shell surface finely punctate. Normal pore canals fairly large, circular, evenly scattered over carapace. A very low eye swelling may be seen on the right valve, female dimorph and slightly more strongly developed in the right valve of the male, situated below the anterior cardinal angle, not observed in the left valve of either dimorph. Greatest length passes through mid-point ; greatest height and width approximately at centre of carapace. A narrow marginal border delimited along its inner side by a marginal groove extends around the anterior margin, only poorly developed around the posterior. Left valve larger than the right which it overlaps along the ventral margin and overreaches elsewhere around the carapace. Dorsal margin “ umbonate ” the highest point being just about at valve centre, the dorsal margin sloping steeply away from this point to the anterior and posterior. Anteriorly the dorsal margin is broadly convex and passes into the anterior margin without a break. Posteriorly the dorsal margin is steeply angled, straight or very slightly convex. Posterior cardinal angle more prominently developed than the anterior angle. Posterior rounded-triangular with a short postero-dorsal slope which is convex in the left valve and straight in the right. Postero-ventral slope longer, convex and tending to be obliquely angled away from the ventral margin. Anterior uniformly rounded. Ventral margin antero-medially incurved, medially convex. Hinge paleohemimerodont ; left valve with terminal loculate sockets and a short median bar (longer in the male dimorph) across which there is a narrow groove con- necting the terminal sockets. Accommodation groove broad and shelf-like. Right valve with strongly dentate terminal elements, not clearly seen in the present material. Muscle scars (Type C) with rounded anteromedian antennal scar. Inner margin and line of concrescence coincide ; duplicature of moderate width. Anterior radial pore canals slightly curved and in some specimens appearing to thicken slightly towards the outer termination ; 12-16 observed in the present material. Dimensions Holotype, 10.1837, female carapace (Pl. 17, figs. 4-6), length 0-64 mm. ; height 0-40 mm. ; width 0:32 mm. To.1838, male carapace, length 0-82 mm.; height 0-45 mm.; width 0-37 mm. To.1839, male carapace (PI. 18, figs. 1-3), length 0-77 mm. ; height 0-43 mm. ; width 0-33 mm. Io.1840, female carapace (PI. 17, figs. 7-10), length 0-62 mm. ; height 0-41 mm.; width 0-32 mm, [0.1843, female right valve (PI. 18, fig. 9), length 0-60 126 1k, dels. IBUAIDIS, mm. ; heighto-34mm. 10.1846, female left valve (PI. 18, figs. 6, 7), length 0-61 mm. ; height 0-43 mm. REMARKS. Apart from the variations in size range between the three subspecies of Praeschulerideasubtrigona there are also a number of additional characters by which the subspecies may be distinguished. The first of these characters concerns the angularity of the carapace which in the female dimorph of intermedia is very close to that of the type subspecies, the dorsal margin sloping strongly away from the region of greatest height. If anything, however, the posterior half of the dorsal margin is more elongate and not so steeply sloping as in subtrigona subtrigona. The dorsal margin in magna appears more uniformly convex on either side of the region of greatest height, contrasting with the almost straight posterior part in the other two subspecies. Posteriorly there are slight differences in all three: triangular in subtrigona subtrigona ; rounded in magna and bluntly flattened in intermedia where the postero-ventral slope although convex appears to flatten out slightly on ap- proaching the extreme posterior. Actually the posterior margin of intermedia is closer to that of magna than to subtrigona subtrigona. The male dimorph of inter- media may be equivalent in size to that of magna—there however, it is the male of the last named species which shows the greater degree of angularity, being more noticeably ““umbonate ”’ than intermedia. The conditions pertaining in north-eastern England during Middle Jurassic times appear to have offered a number of environments inhabited by a subspecies of Praeschuleridea subtrigona each population being characterized by a variation in size range. Praeschuleridea subtrigona subtrigona inhabited the marine waters of the shallow oolitic sea which covered Lincolnshire during Bajocian times. The northern extension of this sea lapped against the Yorkshire delta and provided a changed environment in which the subspecies P. s. magna developed. Higher up in the succession, the Yorkshire delta was still influencing the sedimentation of the Bajocian and throughout the entire marine embayment which spread over N.E. Yorkshire the population that existed was of P. s. intermedia. Family CYTHERURIDAE Muller 1894 Genus EOCYTHEROPTERON Alexander 1933 Eocytheropteron ° sp. (Pl. 18, figs. ro-13 ; Pl. 19, tgs. 1-4) REMARKS. Four complete carapaces of a species externally resembling the genus Eocytheropteron have been obtained from beds high up in the Grey Limestone Series. Two male carapaces (Io.Igt0-11) have been obtained from bed 8, Stonecliff Wood section whilst a single female carapace (Io.2102) has been obtained from bed 5 of the same section. From bed 11, Hawsker, a further female carapace (Io.1909) has been found. The carapace is elongate-oval in side view with the ventro-lateral margin strongly convex and overhanging the ventral surface, especially so in the female dimorph. A short caudal process is developed and the greatest length of the cara- MIDDLE JURASSIC OSTRACODA 127 pace extends through mid-point. Shell surface with a very faint reticulation. A shallow sulcus is present at about valve centre in the female, slightly anterior to this inthe male. Left valve larger than the right. Internal details not known although some radial pore canals can be made out externally in the antero-ventral area where they appear to be straight and widely spaced. This species is close to Cytheropteron (Cytheropteron) purum Schmidt (1954 : 88, pl. 6, figs. 3-6 ; pl. 7, figs. 25-29) although somewhat smaller and may be further distinguished by the more elongate and pos- teriorly tapering carapace of the female dimorph which in C. (C.) purum is shorter in comparison. Io.1g09, female carapace (Pl. 19, figs. 1-4), length 0-47 mm. ; height 0-26 mm. ; width 0:26 mm. l[o.1911, male carapace (Pl. 18, figs. 10-13), length 0-53 mm. ; height 0-26 mm. width 0-28 mm. Genus PARACYTHERIDEA Miiller 1894 Paracytheridea? caytonensis sp. nov. (Pl. 10, figs. 5-16; Text-figs. 22, 23) Diacnosis. Paracytheridea ? with backwardly projected ala terminating in prominent node. A second node is situated dorso-medially on lateral surface of valve, just behind median sulcus. Shell surface reticulate. Anterior and posterior with flattened marginal borders. Left valve larger than right. Species dimorphic. HoLotype. [0.2137, bed 6, Cayton Bay Section. PARATYPES. 10.2138—42, horizon and locality as above. Fics. 22-23. Dorsal and left views, female carapace, Pavacytheridea ? caytonensis sp. nov. Holotype, lo.2137, approx. 105. DESCRIPTION. Cavapace subquadrate in side view with a prominent, backwardly projected ala at the tip of which is situated an oval node-like swelling. Above and slightly in front of this swelling a much larger node is situated on the lateral surface of the valve, In front of this circular node a transverse median sulcus is present 128 Re oe BATE which appears to curve under the lateral node. An oblique groove extending from the dorsal margin, below the anterior cardinal angle, extends down to the antero- ventral part of the valve where it turns back to extend along the dorsal side of the ventro-lateral ala. Shell surface rather coarsely reticulate, including the surface of the two nodes. An eye swelling is situated just below the anterior cardinal angle. Anteriorly and posteriorly the marginal borders are flattened, the convex lateral part of the carapace not extending right up to the margins. Ventral surface with approxi- mately 5 longitudinal ridges to each valve. Sexual dimorphism indicated by the presence of an elongate specimen considered to be the male. Greatest length of carapace passes through mid-point ; greatest height in the anterior third ; greatest width in the posterior third. Dorsal margin slightly concave in the left valve, convex in the right. Anterior margin rounded with oblique, convex, antero-dorsal slope. Posterior triangular, extended into a caudal process, especially in the female. Postero-dorsal slope strongly concave, postero-ventral slope strongly convex. Ventral margin medially incurved. Cardinal angles prominent; anterior angle broadly rounded ; posterior angle more acute. Left valve larger than the right which it overlaps along the ventral margin and slightly at the cardinal angles. Elsewhere the left valve overreaches the right, apart from along the dorsal margin where the valves diverge. Internal characters not seen. Dimensions Holotype, Io.2137, female carapace (Pl. 19, figs. 5-8; Text-figs. 22, 23), length 0:45 mm. ; height 0-26 mm. ; width 0-25 mm. Io.2138, male carapace (Pl. 19, figs. 13-16), length 0-54 mm. ; height 0-25 mm. ; width 0:23 mm. l[o.2140, female carapace (Pl. 10, figs. 9-12), length 0-43 mm. ; height 0-25 mm. ; width (broken) 0-22 mm. REMARKS. Only six specimens of this species have been found and these all occur in the same bed and represent a single population. Owing to the lack of knowledge concerning the internal details of this ostracod the generic designation is given with a query, although on general external morphology there is good reason to place this species into Paracytheridea. Superficially there is some resemblance between P. ? caytonensis and Cytheropteron (Cytheropteron) bispinosum crassum Schmidt (1954 : 87, pl. 7, figs. 23-24) although in the last named ostracod the ventro-lateral alar projection is not backwardly directed as in P. ? caytonensis and also lacks the characteristic nodes of that species. Family PROTOCYTHERIDAE Ljubimova 1955 Subfamily KIRTONELLINAE Bate 1963 Genus SOUTHCAVEA Bate 1964 Southcavea microcellulosa sp. nov. (Pl. 20, figs. 1-13 ; Pl. 21, figs. 1-4) Diacnosis. Southcavea with oval-subquadrate carapace and coarse reticulate MIDDLE JURASSIC OSTRACODA 129 ornament. Pits produced by reticulate ornament 4-6 sided, internally strongly punctate. Hinge antimerodont. Species dimorphic. Hototyre. Io.1882, bed 5, Stonecliff Wood section. ParRATYPES. Io0.1883-99, beds 3, 5 & 8, Stonecliff Wood and bed 11, Hawsker. DESCRIPTION. Carapace oval-subquadrate in outline, more elongate in the male dimorph. Posteriorly tapered in the female. Greatest length through mid-point ; greatest height in the anterior third; greatest width just behind valve centre. Dorsal margin in the left valve slightly concave medially with broadly rounded cardinal angles, in the right valve slightly convex, cardinal angles somewhat more acute. Anterior broadly rounded ; posterior more narrowly rounded with convex postero-dorsal and postero-ventral slopes in the left valve, and a concave postero- dorsal and convex postero-ventral slope in the right. Ventral margin medially incurved. Ventro-lateral margin convex. Anterior without a marginal border, whilst posteriorly there is a very narrow, flattened border. Left valve larger than the right which it overlaps mid-ventrally, at the cardinal angles and along the postero-dorsal slope. Antero-ventrally, postero-ventrally and antero-dorsally the left valve overreaches the right. Shell surface strongly reticulate, the 4-6 sided pits produced being strongly punctate. This degree of punctation in many cases has resulted in the pits themselves being subdivided by secondary ridges—this develop- ment appears to be more characteristic of the ventro-lateral areas of the carapace. Ventral surface strongly ornamented by longitudinal ridges. Normal pore canals few in number and widely scattered over the carapace, although because of the ornamentation only clearly seen along the ventral surface. Hinge antimerodont : left valve with terminal sockets separated by a finely denticulate median bar. The accommodation groove is elongate and shelf-like. Right valve with 5 bifid, posterior teeth, exact number anteriorly not known but appears to be more than 5. Median groove elongate and finely locellate. Inner margin and line of concrescence coincide, duplicature of moderate width. Radial pore canals straight, simple and widely spaced, approximately 8 anteriorly, at least 4 posteriorly. Muscle scars (Type D?) : adductor scars an oblique row of 4 scars with an antero-ventral mandibular scar and an antero-dorsal antennal scar which may be heart-shaped, but not so definitely V- shaped as in other species of this genus. A narrow flange has been observed in the right valve extending around the anterior margin and along the ventral margin. Dimensions Holotype, Io.1882, male carapace (PI. 20, figs. 1-4), length 0-67 mm. ; height 0-36 mm. ; width 0:38 mm. To.1883, female carapace (PI. 20, figs. 5-8), length 0-60 mm. ; height 0-36 mm. ; width 0:37 mm. Io. 1886, female right valve (PI. 20, figs. 9, 10), length (broken) 0-60 mm. ; height 0-34 mm. I[o0.1888, male carapace (Pl. 21, figs. 1-4), length 0-77 mm. ; height 0-41 mm. ; widtho-46mm. ; Io.1891, femalecarapace, lengtho-65 mm. ; height 0-36 mm.; width 0-36 mm. I[o0.1893, female carapace, length 0:54 mm. ; height 0-30 mm. ; width 0-31 mm. 130 Ise, daly 187 Able REMARKS. Southcavea microcellulosa is a much more elongate ostracod than Southcavea reticulata Bate (1964 : 27, pl. ro, figs. 3-14; pl. 11, figs. 1-4) with which it might be confused, and also has a much finer ornamentation. S. microcellulosa appears to be virtually restricted to the sandy shore-line facies along the western outcrop of the Grey Limestone Series with the exception of a single female carapace found in bed 11, Hawsker. However, even at Hawsker in the east it is doubtful whether the northern coast-line was very far away. Genus SYSTENOCY THERE Bate 1963 Systenocythere ovata sp. nov. (Pl. 21, figs. 5-12) DraGnosis. Systenocythere with ovoid carapace showing a strong reticulate ornamentation when well preserved, but generally only possessing ornamentation of longitudinal ridges on ventro-lateral and ventral surfaces. Internal characters as for genus. HototyPe. lI[o.1900, bed 7, Hawsker. PARATYPES. lIo.1g01-8, bed 4, Gristhorpe Bay ; bed 6, Cayton Bay; beds 19 and 22, Hundale Point, Cloughton ; beds 7 and 12, Hawsker and bed 1, Yearsley Moor. DESCRIPTION. Carapace ovoid in side view, tapering posteriorly. Greatest length through mid-point ; greatest height in the anterior third; greatest width in the posterior third. Dorsal margin convex in both valves; cardinal angles rounded. Anterior rounded with oblique, convex, antero-dorsal slope. Posterior narrowly rounded with convex postero-dorsal and postero-ventral slopes in the left valve and concave postero-dorsal and convex postero-ventral slopes in the right valve. Ventral margin medially incurved. Ventro-lateral margin convex, overhanging the ventral surface in side view. Anterior and posterior with flattened marginal borders. Left valve larger than the right which it overlaps most strongly mid-ventrally, from which point, along the ventral margin the left valve progressively overreaches the right. The left valve also overreaches the right along the antero- and postero-dorsal slopes. Shell surface may be either strongly reticulate or only noticeably ornamented along the ventro-lateral and ventral surfaces with longitudinal ridges. Fairly large normal pore canals may be seen, depending on preservation, widely scattered over the carapace. Hinge merodont, not fully determinable. Right valve with 5 anterior teeth and possibly 5 posterior teeth ; median groove long and very narrow, it is difficult to state precisely whether this groove is smooth or locellate because of the preservation of the material. Hinge poorly preserved in the left valve: Accom- modation groove of moderate width situated above the median bar which cannot be identified as being either smooth or denticulate. Muscle scars (Type D) consist of a subvertical row of 4 oval adductor scars with an oval anteroventral mandibular scar and a V-shaped antero-dorsal antennal scar. Inner margin and line of concrescence coincide, the duplicature being of moderate width. Approximately 8 anterior and 3 MIDDLE JURASSIC OSTRACODA 131 posterior vadial pore canals may be distinguished, these appear to be straight and widely spaced. A narrow flange widening opposite the ventral incurvature, extends GREY LIMESTONE SERIES = 9 a “ a = + =< ayes UN BRod3 o UN = Smears ve ar) (ay ta (0) aa x oe) 7a a x 225 | = = @ = = 2) SiS} aN ® on} FS clo | oe ited < GT 4 fo 5 a4 3 a3 5285 a a a a a au 3 © o ° Glyptocythere scitula 2 MEE == . = = ee eee Fuhrbergiella (P) | : _ eit horrida | a _ Praeschuleridea | subtrigona | lintermedia | Vernoniella bajociana | a = | Monoceratina scarboroughensis a Progonocythere | acuminata |Ljubimovella piriformis Cloughtonella rugosa es ovata (i Paracypris. = bajociana |Eocytheropteron ? [nS eS ore | sp. I Hlsurog ater sp. [Caytonidea ml CESS AD faveolata |Paracytheridea ? |caytonensis |Vernoniella ? jcaytonensis |Progonocythere =z | yonsnabensis Mesocytheridea howardianensis all Southcavea ma TIES microcellulosa i‘ : |Glyptocythere | polita |Glyptocythere |costata Malzia bicarinata | — | |Malzia unicarinata Fic. 24. Range table of the Grey Limestone Series Ostracoda. 132 Ise dela deat de along the ventral margin of the right valve and possibly also extends around the anterior margin. A flange has not been observed in the left valve. Dimensions Holotype, Io.1900, carapace (PI. 21, figs. 5-8), length 0-70 mm. ; height 0-44 mm. ; width 0-37 mm. To.1g901, left valve, length 0-66 mm. ; height 0-39 mm. _Io.1903, left valve (Pl. 21, fig. 12), length 0-60 mm. ; height 0-36 mm. I[o.1904, right valve, length 0-48 mm. ; height 0:28 mm. lo.1907, right valve (PI. 21, figs. 9-11), length 0-60 mm. ; height 0-34 mm. REMARKS. Systenocythere ovata is similar in outline to the female dimorph of S. exilofasciata Bate (1963 : 212, pl. 14, figs. 7-10 ; pl. 15, figs. 1-4), the type species, although it differs in not being so noticeably acuminate posteriorly and, when preservation permits, in being strongly reticulate. VI REFERENCES ARKELL, W. J. 1933. The Jurassic System in Great Britain. viii + 681 pp., 41 pls. Oxford. Bate, R. H. 1963. Middle Jurassic Ostracoda from North Lincolnshire. Bull. Brit. Mus. (Nat. Hist.) Geol., London, 8 : 173-219, pls. I-15. —— 1963a. Middle Jurassic Ostracoda from South Yorkshire. Bull. Brit. Mus. (Nat. Hist.) Geol., London, 9 : 19-46, pls. I-13. —— 1964. Middle Jurassic Ostracoda from the Millepore Series, Yorkshire. Bull. Brit. Mus. (Nat. Hist.) Geol., London, 10: 1-34, pls. 1-14. Brack, M. 1928. ‘‘ Washouts”’ in the Estuarine Series of Yorkshire. Geol. Mag. Lond., 65: 301-307. Brann, E. 1961. Jn Brann, E. & Matz, H. Ostracoden-Studien im Dogger, 3: Drei neue Procytheridea-Arten und Ljubimovella n.g. aus dem NW-deutschen Bajocien. Senck. leth., Frankfurt a.M., 42 : 157-173, pls. 1, 2. Brann, E. & Faurion, H. 1962. Dogger NW—Deutschlands : 123-158, pls. 16-21. In Simon, W. & BaRTENSTEIN, H. (editors), Leitfossilien der Mikvopaldontologie. viii + 432 pp., 59 pls. Berlin. Branp, E. & Matz, H. 1962. Ostracoden-Studien im Dogger, 4: Fuhrbergiella n.g. Senck. leth., Frankfurt a.M., 43 : 1-39, pls. 1-6. —— 1962a. In BRAND, E. & FaHrion, H. Dogger NW-Deutschlands : 123-159, pls. 16—21. In Stmon, W. & BARTENSTEIN, H. (editors), Leitfossien dey Mikropaldontologie. iti + 432 pp., 59 pls. Berlin. BuckMAN, S. S. t911. Appendix 1—Ammonites from the Scarborough Limestone. Proc. Yorks. Geol. Soc., Leeds, 17 : 205-208. DuNBAR, C. O. & RopaeErs, J. 1958. Pyrinciples of Strvatigyvaphy. viii + 356 pp., 123 text- figs. New York. Fatcoin, N. L. & Kent, P. E. 1960. Geological Results of Petroleum Exploration in Britain 1945-1957. Mem. Geol. Soc. Lond., 2 : 1-56, pls. 1-5. Fox-STRANGWAYS, C. 1892. The Jurassic Rocks of Britain, 1, Yorkshire. Mem. Geol. Suv., London. ix + 551 pp. HEMINGWAY, J. E. 1951. Report ona field meeting for 1949. Proc. Yorks. Geol. Soc., Leeds, 28 : 118-122. Matz, H. 1959. Ostracoden-Studien im Dogger, 1: Marslatourellan.g. Senck. leth., Frankfurt a.M., 40 : 19-23, 4 figs. -—— 1961. Jn BRAND, E. & Matz, H. Ostracoden—Studien im Dogger, 3 : Drei neue Procy- thevidea-Arten und Ljubimovella n.g. aus dem NW-deutschen Bajocien. Senck. leth., Frankfurt a.M., 42 : 157-172, pls. I, 2. MIDDLE JURASSIC OSTRACODA 133 Moore, R. C. 1961 (editor). Tveatise on Invertebrate Palaeontology, Pt. Q., Arthropoda, 3. xxiii + 442 pp., 334 figs. Kansas. OerRTLI, H. J. 1957. Ostracodes du Jurassique Supérieur du Bassin de Paris (Sondage Vernon 1). Rev. Inst. frang. Pétrole, Paris, 12 : 647-695, pls. 1-7. PETERSON, J. A. 1954. Jurassic Ostracoda from the “‘ Lower Sundance’”’ and Rierdon formation, Western Interior United States. J. Paleont., Tulsa, 28 : 153-176, pls. 17-109. PiumuHorrF, F. 1963. Die Ostracoden des Oberaalenium und tiefen Unterbajocium (Jura) des Gifhorner Troges, Nordwestdeutschland. Abh. Senckenb. naturf. Ges., Frankfurt a.M., 503 : 1-100, pls. I-12. RicHARDSoON, L. to911r.. The Lower Oolitic Rocks of Yorkshire. Pyvoc. Yorks. Geol. Soc., Leeds, 17 : 184-204. ScHMIDT, G. 1954. Stratigraphisch wichtige Ostracoden im “ Kimeridge”’ und tiefsten “ Portland ’’ NW-Deutschland. Paldont. Z., Stuttgart, 28 : 81-101, pls. 5-8. Simon, W. & BARTENSTEIN, H. (editors). Leztfossilien dex Mikropaldontologie. viii + 432 pp., 59 pls. Berlin. } SYLVESTER-BRADLEY, P. C. 1948. Bathonian ostracods from the Boueti Bed of Langton Herring, Dorset. Geol. Mag. Lond., 85 : 185-204, pls. 12-15. 1956. The Structure, evolution and nomenclature of the ostracod hinge. Bull. Brit. Mus. (Nat. Hist.) Geol., London, 3 : 1-21, pls. 1-4. ESE ASN AG LOIN TOR eA S) All the specimens illustrated are now in the Department of Palaeontology, British Museum (Natural History). All photographs, taken by the author, x85 unless otherwise indicated. GEOL. II, 3 10 Te AS Monoceratina scarboroughensis sp. nov. p. 99. Fics. 1-3, 12. Bed 22, Hundale Point. Fics. 4-11. Bed 5, Cayton Bay. Fics. 1-3. Right, left and ventral views, female carapace. Holotype, Io.1711. Fic. 4. Muscle scars, female carapace. Paratype, lo.1723, Xx3I0. Fics. 5-8. Right, left, dorsal and ventral views, female carapace. Paratype, 10.1723. Fics. 9-11. Left, dorsal and ventral views, male carapace. Paratype, 10.1721. Fic. 12. Internal view, female left valve. Paratype, Io.1720. Caytonidea faveolata gen. et sp. nov. __p. 100. Specimen from bed 4, Gristhorpe Bay. Fics. 13, 14. Right, left, views, carapace. Paratype, 10.1834. Bull. B.M. (N.H.) Geol. 11, 3 BEA 1 i ote 3 * 2. 7 aw PI Awe) 2 Caytonidea faveolata gen. et sp. nov. p. 100 Fics. 1-4, 7-10. Bed 5, Cayton Bay. Fics. 5,6. Bed 4, Gristhorpe Bay. Figs. 1-4. Right, left, dorsal and ventral views, carapace. Holotype, Io.1831. Fics. 5, 6. Dorsal and ventral views, carapace. Paratype, 10.1834. Fics. 7, 8. Dorsal and lateral views of right valve hinge., Paratype, 10.1832, 95. Fics. 9, 10. Internal views of right valve fragment to show muscle scars and anterior radial pore canals. Paratype, 1o.1833. Bull. B.M. (N.H.) Geol. 11, 3 PIL JN Is, W PLATE 3 Cloughtonella rugosa gen. et sp. nov. pp. 102 Fics. I-7, 11-13. Bed 22, Hundale Point, Cloughton. Fics. 8-10. Bed 10, Bloody Beck. Ties. 1-4. Right, left, dorsal and ventral views, female carapace. Holotype, Io.2118. Fics. 5-7. Ventral, right and left views, male carapace. Paratype, Io.2119. Fies. 8, 9. Left and right views, female carapace. Paratype, 1o.2135. Fic. to. Left view, showing large normal pore canals, male carapace. Paratype, 10.2134. Fic. 11. Dorsal view of hinge, female left valve. Paratype, Io.2121. Fics. 12, 13. light and left views, male carapace. Paratype, lo.2120. Bull. B.M. (N.H.) Geol. 11, 3 IL AN I, & DILATIS, a Fuhrbergiella (Praefuhrbergiella) horrida horrida Brand & Malz _p. 104 Fics. 1-3. Bed 5, Cayton Bay. Fics. 4, 5, 10. Bed 6, Cayton Bay. Fics. 6-9, 11, 12. Bed 23, Hundale Point, Cloughton. Fics. 1-3. Internal, external and dorsal views, female left valve, lo.2109. 5. External and dorsal views, female right valve, Io.2116. Fics. 6-9. Dorsal, left, right and ventral views, male carapace, [o.2111. o. External view, female right valve showing anterior and posterior radial pore canals, Fics. 11, 12. Dorsal and right views, female carapace, lo. 2110. AMIN 4: Bull. B.M. (N.H.) Geol. 11, 3 PLATE 5 Glyptocythere costata sp. nov. _p. 106 Fics. 1-4, 7. Bed 10, Ravenscar. Fics. 5, 6. Bed 8, Ravenscar. Iics. 1-4. Right, left, dorsal and ventral views, female carapace. Holotype, 1o.1775. Iies. 5, 6. Right and left views, male carapace. Paratype, lo.r782. Fic. 7. Female right valve. Paratype, lo.1776. Glyptocythere polita sp. nov. _p. 107 Fic. 8. Bed 6, Bloody Beck. Fic. 9. Bed 7, Ravenscar. Ic. 10. Bed 7, Hundale Point, Cloughton. Fig. 11. Bed 9, Ravenscar. Fic. 8. Right view, female carapace. Paratype, Io.1743. Fic. 9. Internal view, female left valve. Paratype, 10.1738, x7o. Fic. 10. Right view, female carapace, showing normal pore canals. Paratype, 10.1736, X 70. Fic. 11. Dorsal view of hinge, female left valve. Paratype, Io.1741. Leib PANIES, °) Bull. B.M. (N.H.) Geol. 11, 3 PATA EN6 Glyptocythere polita sp.nov. pp. 107 Fics. 1-8. Bed 7, Hundale Point, Cloughton. Fic. 9. Bed 8, Hundale Point, Cloughton. Fics. 1-4. Right, left, dorsal and ventral views, female carapace. Holotype, lo.1724, x70. Fics. 5-8. Right, left, dorsal and ventral views, male carapace. Paratype, lo.1725, x79. Fic. 9. Muscle scars, female left valve. Paratype, lo.1737, x 150. PIL AN IP IED (6) Bull. B.M. (N.H.) Geol. 11, 3 PLATE 7 Glyptocythere scitula sp. nov. pp. 108 Fics. 1, 7. Bed 10, Hawsker. Fics. 2-6, 8-10. Bed 12, Ravenscar. Fics. 11, 12. Bed 5, Cayton Bay. Fic. 13. Bed 5, Gristhorpe Bay. Fics. 1, 7. Internal and dorsal views, female right valve. Paratype, lo.1760. Fic. 2. Muscle scars. Note antero-ventral mandibular scar, which is a rosette of several. smaller scars. Broken female right valve. Paratype, 1lo.1774, x 150. Fics. 3, 8. Muscle scars (125) and anterior radial pore canals (x90), female right valve. Paratype, Io. 1771. Fics. 4, 6, 10. Internal view, showing radial pore canals, and two dorsal views of hinge, female right valve. Paratype, 10.1768. Fics. 5, 9. Dorsal views of hinge, female left valve. Paratype, 10.1769. Fics. 11, 12. External and internal views, male left valve. Paratype, lo.1751. Fic. 13. External view, male right valve. Paratype, lo.1772. Bull. B.M. (N.H.) Geol. 11, 3 PIL AN WB, 7) PLADE 8 Glyptocythere scitula sp. nov. p. 108 Fics. 1-5. Bed 5, Cayton Bay. Fics. 6-9. Bed 10, Hawsker. Fics. 1-4. Left, right, dorsal and ventral views, female carapace. Holotype, lo.1750. Fic. 5. External view, female left valve. Paratype, 10.1754. Fics. 6-9. Left, right, dorsal and ventral views, female carapace. Paratype, 10.1752. Bull. B.M. (N.H.) Geol. 11, 3 PLATE 8 PLATE 9 Glyptocythere scitula sp. noy. p. 108 Specimen from bed 10, Hawsker. Fics. 1-4. Left, right, dorsal and ventral views, male carapace. Paratype, lo.1753. Malzia bicarinata gen. et sp. nov. p. III Specimen from bed 9, Ravenscar. Fias. 5-8. Left, right, dorsal and ventral views. Holotype, Ilo.1797. Bull. B.M. (N.H.) Geol. 11, 3 IP IL AN WES, IP IDA IDB, G10) Malzia bicarinata gen. et sp. nov. Palast Fic. t. Bed 8, Ravenscar. Fics. 2, 3. Bed 7, Ravenscar. Fic. 1. Left side, carapace. Paratype, 10.1798. Fics, 2, 3. Internal view and dorsal view ( x 100) left valve. Paratype, 10.1799. Malzia unicarinata gen. et sp. nov. Pp. 113. Fics. 4-8. Bed 9, Ravenscar. Fics. 9, 10. Bed 8, Ravenscar. Fics. 4-7. Left, right, dorsal and ventral views, female carapace. Holotype, lo.18o01. Fic. 8. Right view of female carapace to show radial pore canals. Holotype, Io.1801. Fics. 9, 10. Left and right views, female carapace. Paratype, lo.1802. Bull. B.M. (N.H.) Geol. 11, 3 PLATE 10 FIGs. Fics. FIGs. Fics. I-4. 5, Op Fy (2s 9, 10. PALE 17 Malzia unicarinata gen. etsp. nov. Pp. I13. Specimen from bed Io, Ravenscar. Left, right, dorsal and ventral views, male carapace. Paratype, 10.1806. Progonocythere acuminata sp. nov. Pp. 114. Fics. 5,6. Bed 7, Ravenscar. Fics. 7-10. Bed 6, Cayton Bay. Left and right views, carapace. Paratype, lo.1789. External and internal views, left valve. Paratype, 10.1786. Right and left views, carapace. Paratype, Io.1787. PILATES Wl Bull. B.M. (N.H.) Geol. 11, 3 PLATE 12 Progonocythere acuminata sp. nov. p. 114. Specimen from bed 7, Hundale Point, Cloughton. Fics. 1-4. Left, right, dorsal and ventral views, carapace. Holotype, 1o.1783. Progonocythere yonsnabensis sp. noy. p. 110. All specimens from bed 5, Cayton Bay. Fics. 5-8. Left, right, dorsal and ventral views, female carapace. Holotype, lo.1792. Fics. 9, 10. Left and right views, female carapace. Paratype, Io.1794. Fics. 11-14. Left, right, dorsal and ventral views, male carapace. Paratype, 10.1793. Bull. B.M. (N.H.) Geol. 11, 3 PILA IIS, Pi A Temes Progonocythere yons ensis sp.nov. p. 110. Specimen from he Bay. Fics. 1-4. Left, right, dorsal and ventral views, male carapace, showing development of two lateral keels and nodes. Paratype, 10.1795. Pleurocythere sp. p. 117. Specimen from bed 4, Gristhorpe Bay. Fic. 5. External view, left valve, 10.1836. Vernoniella bajociana sp. nov. p. 118. Specimens from bed 23, Hundale Point, Cloughton. Fics. 6-9. Left, right, dorsal and ventral views, female carapace. Holotype, 10.1807. Fics. ro, 11. Right and left views, female carapace. Paratype, To.1808. Bull. B.M. (N.H.) Geol. 11, 3 Pa 13 PEAT BE 14 Vernoniella bajociana sp. nov. pp. 118. Fics. 1-4. Bed 7, Hundale Point, Cloughton. Fics. 5-10, 12, 13. Bed 12, Ravenscar. Fic. 11. Bed 9, Hawsker. Fics. 1-4. Right, left, dorsal and ventral views, male carapace. Paratype, 10.1814. Fics. 5-9. 5, anterior radial pore canals (140) ; 6, dorsal view of hinge ; 7, internal view to show boring of shell by marine organism ; 8, lateral view of hinge ; 9, internal view to show radial pore canals. Female left valve. Paratype Io.1818. FIGs. 10, 12, 13. Muscle scars (x 180) showing composite antennal scar, dorsal view of hinge (95) and internal view of male right valve. Paratype, Io.1819. Fic. 11. Muscle scars, fragment of female right valve. Paratype, 1o.1813, 110. Bull. B.M. (N.H.) Geol. 11, 3 Fics. Fics. FIGs. Fics. FIGS. Coa Coe iO), Wit, TA, 03, PLATE 15 Vernoniella caytonensis sp.nov. pp. 119. All specimens from bed 6, Cayton Bay. Left, right, dorsal and ventral views, carapace. Holotype, 1lo.1855. Right, left and ventral views, carapace. Paratype, 10.1858. Dorsal and internal views, right valve. Paratype, 10.1857. Ljubimovella piriformis Malz p. 120. Fics. to, 11. Bed 22, Hundale Point, Cloughton. Fics. 12, 13. Bed 12, Ravenscar. Left and right sides, juvenile carapace. Internal and external views, right valve. 10.2107. Io.2108. B.M. (N.H.) Geol. 11, 3 ILA ID, WS JP by JN AD IB, a0 (5) Ljubimovella piriformis Malz. p. 120. Specimen from bed 7, Hawsker. Fies. 1, 2. Left and right sides, adult carapace. I[0.2106. Mesocytheridea howardianensis gen. et sp. noy. _p. 122. Fics. 3-6, 9, 11. Bed 2, Stonecliff Wood. Fic. 7. Bed 7, Bloody Beck. Fics. 8, 10. Bed 8, Stonecliff Wood. Fics. 3-6. Right, left, dorsal and ventral views, female carapace. Holotype, Io.1870. Fic. 7. Dorsal view hinge, female left valve. Paratype, 10.1881. Fics. 8,10. Dorsal and internal views, male right valve. Paratype, 10.1879. Fic. 9. Dorsal view, female right valve. Paratype, Io.1871. Fic. 11. Internal view, female left valve, showing anterior socket cutting back into the median bar. Paratype, I0.1875. Bull. B.M. (N.H.) Geol. 11, 3 PIL AB, WG PAIN AIS, an Mesocytheridea howardianensis gen. etsp.noy. p. 122. Specimens from bed 2, Stonecliff Wood. Fic. 1. Internal view, showing radial pore canals, male left valve. Paratype, Io. 1872. Fics. 2,3. Internal views to show hinge and radial pore canals, female right valve. Paratype, lo.1871. Praeschuleridea subtrigona intermedia subsp. noy. pp. 124. Fics. 4-6. Bed 7, Hundale Point, Cloughton. Fics. 7-10. Bed 2, Stonecliff Wood. Fics. 4-6. Left, right and dorsal views, female carapace. Holotype, Io.1837. Fics. 7-10. Ventral, dorsal, right and left views, female carapace. Paratype, I1o.1840. Bull. B.M. (N.H.) Geol. 11, 3 PIL NIE, 7 PLATE 18 Praeschulerida subtrigona intermedia subsp. nov. p. 124. Figs. 1-3. Bed 7, Hundale Point, Cloughton. Fics. 4, 5, 8,9. Bed 2, Stonecliff Wood. Fics. 6, 7. Bed 4, Gristhorpe Bay. Fics. 1-3. Left, right and ventral views, male carapace. Paratype, 1o.1839. Fic. 4. Muscle scars from internal cast. Paratype, Io.1844, x 180. Fic. 5. Anterior radial pore canals, right valve fragment. Paratype, Io.1841, x 135. Fics. 6, 7. Dorsal and internal views to show hinge, female left valve. Paratype, Io.1846. Fic. 8. Right side to show normal and radial pore canals, female carapace. Paratype, Io. 1842. Fic. 9. Dorsal view of hinge, female right valve. Paratype, Io.1843. Eocytheropteron ? sp. p. 126. Specimen from bed 8, Stonecliff Wood. Figs. 10-13. Ventral, left, right and dorsal views, male carapace. Io.1g11. Bull. B.M. (N.H.) Geol. 11, 3 PLATE 18 FIGs. FIGs. Fics. Fics. I-4. 5-8. Q-12. 13-16. PLATE 19 Eocytheropteron ? sp. p. 1206. Specimen from bed 11, Hawsker. Right, left, dorsal and ventral views, female carapace. lo.1909. Paracytheridea ? caytonensis sp.nov. pp. 127. All specimens from bed 6, Cayton Bay. Dorsal, ventral, left and right views, female carapace. Right, left, dorsal and ventral views, female carapace. Right, left, dorsal and ventral views, male carapace. Holotype, lo.2137. Paratype, Io.2140. Paratype, 10.2138. Bull. B.M. (N.H.) Geol. 11, 3 PEATE 19 IPILINILIS, BO Southcavea microcellulosa sp. nov. _p. 128. Fics. 1-8, 11-13. Bed 5, Stonecliff Wood. Fics. 9, 10. Bed 3, Stonecliff Wood. Left, right, dorsal and ventral views, male carapace. Holotype, 10.1882. Fics. 1-4. Fics. 5-8. Left, right, ventral and dorsal views, female carapace. Paratype, Io.1883. Fics. 9, 10. Dorsal view of hinge and muscle scars (x 120), female right valve. Paratype, 10.1886. Fics. 11-13. Dorsal view of hinge and internal views showing hinge and radial pore canals, female right valve. Paratype, 10.1885. Bull. B.M. (N.H.) Geol. 11, 3 PLATE 20 PUAT E22 Southcavea microcellulosa sp. noy._ p. 128. Specimen from bed 8, Stonecliff Wood. Fics. 1-4. light, left, dorsal and ventral views, male carapace. Paratype, 10.1888. Systenocythere ovata sp. nov. pp. 130. Fics. 5-8. Bed 7, Hawsker. Figs. 9-11. Bed 6, Cayton Bay. Fic. 12. Bed 4, Cayton Bay. Fics. 5-8. light, left, dorsal and ventral views, carapace. Holotype, Io.1900. Fics. 9-11. External, internal and dorsal views, right valve. Paratype, lo.1907. Fic. 12. Muscle scars, note V-shaped antennal scar, left valve. Paratype, Io.1903, x 240. Bull. B.M. (N.H.) Geol. 11, 3 PLATE 21 aN ; Se S$ S ve 1969 4 a , WS Ht ark ntoth f; i} ADM ah eee lee 4 vi et Mh * ‘a BoP av WON wi) rd HUMAN SKELETAL MATERIAL on g FROM CEYLON, WITH AN ANALYSIS OF THE ISLAND’S oe PREHISTORIC AND _ CONTEMPORARY POPULATIONS K. A. R. KENNEDY ee BULLETIN OF E BRITISH MUSEUM (NATURAL HISTORY) LOGY Vol. 11 No. 4 Ben: LONDON: 1965 HUMAN SKELETAL MATERIAL FROM CEYLON, Witt AN ANALYSIS OF THE ISLAND’S PREHISTORIC AND CONTEMPORARY POPULATIONS BY KENNETH A. R. KENNEDY, Ph.D. (Department of Anthropology, Cornell University, Ithaca, New York) Ph. 135-213; 15 Plates; 9 Text-figures 12 Tables BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 4 LONDON : 1965 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), tmstituted im 1949, 15 issued in five series corresponding to the Departments of the Museum, and an Historical series. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. In 1965 a separate supplementary series of longer papers was instituted, numbered serially for each Department. This paper is Vol. 11, No. 4 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1965 TRUSDTERS OF THE BRITISH MUSEUM (NATURAL HISTORY) Issued 8 December, 1965 Price {2° 5s: MUMAN SKELETAL MATERIAL FROM CEYLON, WITH AN ANALYSIS OF THE ISLAND’S PREHISTORIC AND CONTEMPORARY POP UEAMOINIS By K. A. R. KENNEDY CONTENTS Page I IntTRopucTION : : : 6 ¢ : 138 The Site of Bellan Beal Palassa ; é c : . 138 The Nature and Condition of the Specimens . : : : 141 II DEscRIPTION . . : : ; 145 The Methodology of the Metrical sais j : 2 . 145 Sex and Age Determination ; 146 The Metrical and Morphological Aeeivats of the Oscars Remerine 150 The Metrical and Morphological pea of the Dentition : 165 The Biochemical Analysis . : j : : : 173 III ComPaRATIVE ANALYSIS. . : 183 The Nature of the Comparative Data Used i in the Determines of the Biological Affinities of the Balangodese with Other Populations : : : ‘ : : F 183 A Comparison with the Veddas : : : F : é 184 IV Discussion. : : é : : : 5 : ; 201 V CONCLUSIONS . ¢ 9 : : : 5 : E : 207 VI ACKNOWLEDGEMENTS : : : é : : é c 208 VII REFERENCES . : : : , ‘ : 5 j 3 209 SYNOP STS The fossilized human skeletal remains from the prehistoric site of Bellan Bandi Palassa in Sabaragamuva Province, Ceylon, are analysed anthropometrically and biochemically. Results of uranium and radiocarbon assays indicate a relatively contemporary population living at a period around 114 B.c. + 200 years, a date that confirms the archaeological evidence which ascribed the cultural associations of the site to the Bandarawelian, a regional variant of the Indian “ Mesolithic ’” or Late Stone Age. Comparative anthropometric studies of these Balan- godese fossils with other hominid specimens, both fossil and living, reveal that their closest phenotypic affinities are with the Veddas of Ceylon. Many of the physical traits regarded by earlier investigators as distinguishing the Veddas from their Ceylonese and Indian neighbours and which have been the basis for separating the Veddas into racial sub-types are apparent in the fossil “‘ pre-Vedda’’ remains from Bellan Bandi Palassa. This suggests a close genetic affinity between these Balangodese—-Vedda phenotypes at the dawn of the historic period in Ceylon. The evidence from the ethnographic and prehistoric record for this region strongly supports the view of a continuous cultural tradition with local modifications extending from Late Stone Age times to the present, a situation which lends independent but supportive evidence for the postulation of Balangodese and Vedda phenotypic affinities. The association of con- temporary hill tribes or relict populations with the manufacturers of prehistoric lithic industries has hitherto been unsubstantiated in the Indian Sub-Continent, and the anthropological problems inherent in this line of research are discussed. GEOL. II, 4 14 138 HUMAN SKELETAL MATERIAL FROM CEYLON I INTRODUCTION The Site of Bellan Bandi Palassa THE presence of palaeolithic artifacts in India and Pakistan testifies to the human occupation of the Sub-Continent during the Pleistocene, but the skeletal remains of the manufacturers of these stone implements have not been found. The most ancient human bones from this part of Asia have been recovered from Langhnaj in Northern Gujarat (Ehrhardt 1960, 1963, Karve & Kurulkar 1945, Khan & Karve 1946, Karve-Corvinus & Kennedy 1964, Kennedy 1964, Sankalia 1945, 1946, 1949, Sankalia & Karve 1944, 1945, 1949, Subbarao 1952, 1955 : 73-74, 1958 : 71-74, Zeuner 1950: 44, 1951: 7) and from the District of Mirzapur in Uttar Pradesh (Personal communication with Shri Radhakant Varma in May 1964, Deccan College, Poona) where the cultural associations are ascribed to the Indian Late Stone Age, or Mesolithic, period. For the Island of Ceylon the discovery of the makers of the lithic industries has long been awaited. The work of the Sarasins (1892-93, 1907) confirmed the claims of earlier investigators that the island possessed Stone Age tools, and from their data two problems originate: (1) To which of the prehistoric ages, as understood in terms of conventional typological classifications, can the Ceylonese stone artifacts be assigned: (2) What kinds of hominids manufactured these tools? “ Until definite stratigraphic evidence, showing a sequence of distinct cultures with perhaps associated animal and skeletal remains, have been established at several sites, the only procedure seems to be to treat the implements as pro- visionally of one culture ... Subdivisions into cultures and phases can be made when adequate evidence is discovered justifying it ... The Sarasins believed that the people who lived in the rock shelters and made stone imple- ments were ancestors of the Veddah, but without skeletal evidence there is nothing to support such a conclusion.”’ (Noone & Noone 1940: 20-21). It is with this latter problem that the present study is concerned, for the anthropo- metric analysis of the human remains from Bellan Bandi Palassa, Ceylon, indicates that the manufacturers of its Bandarawelian (Late Stone Age) industries bear striking phenotypic similarities to the surviving Vedda population of the island. The site of Bellan Bandi Palassa is situated at 6 degrees 31 minutes North longitude and 80 degrees 484 minutes East latitude between the 400 and 300 foot contours in the Balangoda District of Sabaragamuva Province. Three and a half miles to the east of the site is the Pansadara Chena near Hath Kinda where the Valave Ganga ramifies into its seven channels. Traversing the site is an intermittent stream which joins the Pusalli Ara just before its junction with the Valave south of Pansadara (Text-fig. 1). Within the palassa, or glade, of Bellan Bandi an expanse of flat crystalline limestone extends for about a quarter of a mile from the northwest to the southeast and has a width of 70 yards. At right angles to the stream bed and extending across the limestone exposure is a ridge of earth eroded by the stream to form a gap of some 60 feet across. This erosion has exposed on the left bank a kitchen midden some 30 feet wide and averaging 3 feet in height above the limestone. HUMAN SKELETAL MATERIAL FROM CEYLON 139 Some materials have been redeposited about 150 yards downstream from the midden. Erosion has produced along the left bank for a distance of about 100 yards a number of small rock shelters which have since collapsed leaving large limestone blocks which overlie the midden. The surface soil is primarily red sandy loam ranging in depth from 3 feet 6 inches to 4 feet 6 inches above the limestone outcrop. Within this soil cover were found human and animal bones, molluscan shells, chert and quartz implements and potsherds. ~\ 155, KALUGALA | _ » Soo va — _foitaviwa ~- =~) z Fic. 1. The region of Bellan Bandi Palassa in the Balangoda District of Sabaragamuva Province, Ceylon. One inch to one and a half miles. The excavation was carried out from 24th June to 6th July, 1956, then resumed on oth September, 1956. A site survey had been conducted in March of that same year as a result of reports of fossil remains to be found in the forest near the Valave. Mr. Arthur Delgoda sent to Dr. P. E. P. Deraniyagala, Director of the National Museums, some fossilized fragments of human and animal bones, and on 12th March, 1956 led a party of archaeologists under Dr. Deraniyagala to the site. Before the commencement of the excavation the site was pegged out into forty squares each I40 HUMAN SKELETAL MATERIAL FROM CEYLON 6 feet by 6 feet. The north-south base line was represented by the plots Beta, Alpha, A, B, C, D, E, F, and the series beginning with Beta in the south end. The east—west base line was represented by the plots I, 2, 3, 4, and 5, the series beginning with 5 in the west end and adjacent to square Beta. Such a grid was superimposed on the site so that the areas of limestone exposure and the left bank of the stream were approximately equal in extent, the edge of the stream forming a diagonal from square F5 in the northwest to square Beta I in the southeast. Consequently the squares which were possible to excavate were Beta 1, Alpha 1, Ar, Az, Br, B2, B3, Fic. 2. The plan of excavation at Bellan Bandi Palassa showing the relative distances of the skeletal remains from the ground surface and from the limestone bedrock. C1, C2, C3, C4, Dr, D2, D3, D4, D5, Er, Ez, E3, E4, and E5. Digging was continued in each square until the limestone bed rock was encountered. This base was humped with an apex at the C-line and declinations at A to the south and E to the north of the site. The loci of the specimens in this midden are represented in Text-fig. 2. In some of the burials the corpse was flexed and lying upon its left side (specimens BP2/17, BP4/8, BP2/21) or right side (BP2/25, BP3/27-34). Other skeletons were flexed but supine (BP 3/15a, BP2/15b). Fractional burial and bag burial cannot be excluded from consideration, although difficult to establish. The par- ticular artifacts found in direct contact with the skeletons are as follows: with BP2/17 were three unpitted hammer pebbles, of which one was discoidal, plus a bone dart. A cluster of twelve small pebbles in close proximity to one another suggest that they had once been encased in a bag and placed at the head of the corpse. In cleaning this specimen from its soil matrix, the mandible of a Macaca sinica, HUMAN SKELETAL MATERIAL FROM GEYLON 141 probably female, was discovered (Personal communication with Dr. W. C. Osman Hill in January 1961, London). With BP2/20 was uncovered the left molar of Melursus valaha. With BP2/21 were found unworked spherical pebbles some 60 mm. in diameter which perhaps were bola stones. The mandible of Hystrix leucurus leucurus lay across the right zygoma of the specimen. Specimen BP3/27-34 was associated with an unpitted pebble above the level of the skull and a block of quartz under the right humerus. During the excavation of 1956 the remains of some nine individuals were uncovered. In the decade prior to this discovery, Deraniyagala had found fragmentary skeletal remains of single specimens from the middens of neighbouring sites. The significance of these sites has been discussed by Deraniyagala (1956b, 1956c, 1956d, 19574, 19575: 8, 20, 19584, 1958) : 64-71, 1959, 19604, 1960c, 1962, 19634, 19630), Clark (1961 : 189-190), Cole (1963 : 87) and Coon (1962 : 424-425). However, no thorough laboratory examination of the bones was undertaken until 1960 and 1961 when the specimens were sent on loan to the British Museum (Natural History) in London. It was advantageous to conduct the programme of research at this institution where the present investigator had available comparative osteological material from several collections of Vedda specimens and where excellent opportunities for subjecting the specimens to various biochemical tests existed. The Nature and Condition of the Specimens An abbreviated list of the skeletal specimens described in this report is given in Table 1. These are from the collection made during the 1956 season of excavation. There are, however, two fragments included in the collection sent to the British Museum (Natural History) which cannot be assigned to any of the numbered and catalogued specimens. These are: 1. A fragment of right ischium labelled BP2/17g found in Square Bz with speci- men BP2/17. 2. A fragment of right scapula which includes the glenoid cavity and a portion of the axillary border. This fragment is unlabelled but is associated with BP3/27—34 in Square D4. Two specimens were found at Bellan Bandi Palassa which have not been described by the writer but which are noted in Deraniyagala’s report of the site. These are: I. Specimen BP3/11 from Square C2 which lay at a depth of 2 feet 6 inches below the surface and 2 feet above the limestone. Only the maxilla and femur are des- cribed (Deraniyagala 1958a : 230, 233, 236-237, table 1r). 2. Specimen BP3/15a was excavated from Square Cr at a depth of 3 feet 8 inches above the limestone. A second skeleton, specimen BP3/15b, was discovered beneath it and it has been described by Deraniyagala (1958a : 230, 233, 260, pls. II-13, table 1, 1963a: 92-97). The skull of BP3/15b (Deraniyagala 19582 : 233, pls. 11, 12) was sent to the American Museum of Natural History in New York where its restoration is now under way. This is the specimen called T-23-B by Coon (1962 : 424-425). Through the kindness of Dr. Harry L. Shapiro the writer was permitted to examine this specimen. HUMAN SKELETAL MATERIAL FROM CEYLON 142 ‘OPIS }JO] JOF T ue ‘opts FYSII oF YJ Ue AQ PoZOIPUT SI sIqUTIVAL SL OUOG B YOIYAM 10} UOJTOYS 94} JO Opis oY TL ‘uauitoads UaATS AU IOF dUOG ¥v JO BdUESeId JY} SezeOIpuT (+) snjd y : 930N - rd os 0 Ta ounryed [eseN ewoshzZ exe prousydg jerodwoay jezdio09 [eyoreq jeyuoIy tt4++4+44 4 Ta Ta . Ta u Ta u Ta ar Ta ar Ta lp Ta ap Ta ee dGadlyosaq IVINALV] IVLATANS AHL AO ISIT YVINAVYE$, V PeLICATeD I aTaVy a + a + ey +¢€ OZ-gI ZZ—OZ "PV PIO FINPV FINPY FIUPYV o£€-Sz asVy a[eula,T a[eule,y a[eUla,y o[eUla\T SIP OTe oe eTeW XaS rq ZO i) bq 1g zq zq 100) a 2010) | 91S b€—-Lz/€aq Sz/7dq 1z/tda 9/1da 3/haa 1b—-oz/Z7qq 141/74 41/74 Joquinu uewit9eds 143 HUMAN SKELETAL MATERIAL FROM CEYLON be-Lz/€ag 1 Ta $z/7qa few Ce) Iz/ td 9/Ida al Ta Ta 3/haa ° s. oe Ib—oz/ td 41/7aa uimoueds[e9 proqng Ta Ta ar syjeL AT, = aah Lt/tdq siqnq winryosy WMT, wires 4004 eInqiy eIqI,T iINWwo7 tJequinu SSS a ee aS SS COS) STATO, AjUIeI}xXe IOMOT sIqIpueyl be—Lz A at Al uw Ta "s eyIXe /€aa ASE ee it i al - ey a oa + s[9IAIaD eX $z/zad jedieg 1 sjedieorjzoyy S proydeyss eyeqideg a1qipueyl snizode1y, Al Ti ST[IXe] 1z/7dq siqrpurN Byres 9/1da We a ART TY exe s/ bag oIqIpueyl 1b—0z exe = TY ar [7d i. erqrpueyW = TY =F 41/7 Cb ‘E Ta a 1 + Jequiny PLPIXeN 41/74 puey vu[() SNIpey sniounzy vyndeog spore winuss3s qry wn (ole) uonjued snuwey sndiop sequinu SN “TOA UY uoumtoeds Ajruieiy}xe reddy, YY sTqIpuey, uuwinjoo [e1qoz10A HUNT, daaldosad IVINALVI| IVLATANS AHL AO LSIT UVINAVE, V (panuywor) I ATAV, 144 HUMAN SKELETAL MATERIAL FROM CEYLON Of the specimens examined by the present writer, two are deserving of special comment. These are: 1. The mandible of BP2/17i which was found some Io inches away from the calvarium of specimen BP2/17. This was regarded at the time of discovery as belonging to specimen BP2/17, but upon cleaning the calvarium from its matrix it became obvious that the wear patterns of the maxillary and mandibular dentitions were strikingly different for the two specimens. 2. Deraniyagala (1958a : 232) mentions the presence of another skeleton in the vicinity of the deposit where specimen BP2/21 was found. Its skull lay in Square C4 with BP2/21 and its post-cranial parts extended into Square C3. Elsewhere in his report Deraniyagala (19584 : 230, 233) notes the presence of a specimen uncovered in Square C3 at approximately the same level as BP2/21. This specimen in C3 is called BP2/28 and it is without a cranium. The question arises as to whether these post-cranial bones in C3 may not belong to specimen BP2/21 rather than to another individual. The bones of BP2/28, namely a left humerus, left talus, calcaneum and femur were not available for study by the present writer. Furthermore, the great difference in the appearance of the cranial fragments of specimen BP2/21 is such that they could be certified as belonging to a single individual only after they had been cleaned of their matrix and concretion and examined for the purpose of reconstruction. This question of the presence of one or two individuals in Squares C3 and C4 is further complicated by the fact that Deraniyagala (1958a : 232, 260, pl. 9, tables 1-2) notes the presence of certain post-cranial bones belonging to specimen BP2/21—a scapula, humerus, tibia, femur, talus, calcaneum—which again were not observed by the writer. As the bones were encountered in progress of excavation they were coated with shellac and labelled and packed in cotton. Fragile pieces were plaster-jacketed in a position exactly as they were found 7m situ. The plaster jacket for the calvarium of BP2/17 was later removed from one side and shellac dissolved in alcohol was poured into the mass of matrix and bone. Specimens that were placed into the plaster jackets were BP2/17 (calvarium and pelvis), BP2/21 (cranium), BP2/25 (cranium), BP3/15a (skeleton). Photographs were taken of the specimens 7m situ and some measurements were made upon the bones before their complete excavation. Upon the return of the archaeologists to the laboratory at Colombo these specimens were superficially cleaned. No reconstruction was attempted, save for the mending of bones which had been broken as a result of their transport from the field. Laboratory measurements carried out at the Colombo Museum were of a limited nature since the specimens were not entirely removed from their plaster jackets. The reconstruction and anthropometric analysis of the specimens was undertaken by the present writer at the British Museum (Natural History). All of the specimens required some restoration, but warping and erosional damage, rather than actual breakage of bone, have most severely altered their pristine condition. Because of the poor state of preservation of the series, a morphological description is of necessity more significant than the tabulation of the metrical data. Nevertheless, where a metrical analysis could be undertaken, a record was made of the quantitative values. HUMAN SKELETAL MATERIAL FROM CEYLON 145 The anthropometry is based upon the particular reconstruction that was considered by the writer to be of greatest accuracy for each specimen. The reconstructions are intentionally impermanent in order that other scholars may realize their own interpretations of the data. The reconstructing media are acetone cement and plaster of Paris, both of which can be flaked or washed from the specimens with ease. The osseous remains were not found in a uniform state of mineralization, but all were hard, due to the chemical nature of the soil in which they were embedded : the water acidulated by the humus had dissolved the limestone over which it flowed and formed a protecting environment for the calcium content of the bones. The bones of the lowest level were almost resting upon the limestone outcrop. The effluvia of the corpses had caused the limestone to disintegrate into yellow flakes to a depth of 5 cm., and these flakes had been subsequently impacted into a solid mass partially adhering to the bones. In addition to the osseous remains this series contains 120 permanent teeth of which 52 belong to male specimens and 68 to female specimens. II DESCRIPTION The Methodology of the Metrical Analysis The method of measurement for each of the values listed in Table 2 is to be found in Martin & Saller (1957 : 453-499, 520-595). In this compendium the standard procedures for each measurement are described and listed by number. That number relevant to each measurement in Table 2 is given in parentheses immediately after its name. Indices are based upon these measurements. Variations and additions to these procedures are the following : 1. Auricular-Vertex Height is taken according to the method devised by Ranke with the use of the Mollison craniophor, as described by Hooton (1946 : 738). 2. The Frontal Arc is the first segment of the Glabella-Opisthion Arc and differs from the Nasion-Opisthion Arc described by Martin & Saller (1957, Measurement 25) in its selection of the initial landmark. In the absence of a well defined nasion for any skull in the Bellan Bandi Palassa series, the most anterior point of the glabella in the median-sagittal plane was substituted for the conventional landmark in this arc measurement. 3. The External Palatal Arc is measured from alveolon on one side of the upper jaw to alveolon on the opposite side, the tape passing across prosthion and parallel to the anterior aspect of the dental arcade. 4. Height of the Zygomatic Bone is the distance measured with the sliding caliper from the anterior superior border of the frontal process of the malar to its anterior inferior border at the point of articulation with the maxilla. 5. Breadth of the Zygomatic Bone is the distance measured with the sliding caliper from the superior border of the maxillary process to the anterior inferior border of the frontal process. 6. Breadth of the Frontal Process of the Zygomatic Bone is taken as its greatest anterior—posterior diameter. The sliding caliper is placed at the level of the zygo- matico-facial foramen when this measurement is taken. 146 HUMAN SKELETAL MATERIAL FROM CEYLON 7. The Mental Foramen Diameter is taken with the sliding caliper, the points being placed upon the medial-—lateral borders and the inferior—superior borders of the foramen for the determination of its maximum size. 8. Sternal Head Diameter is the maximum anterior—posterior dimension of the medial extremity of the clavicle taken with the sliding caliper. g. Corocoid Process Breadth is the maximum anterior—posterior diameter of the corocoid process of the scapula, taken perpendicularly to the plane that the sliding caliper assumes in the measurement of Corocoid Process Length (Martin & Saller 1957, Measurement II). 10. Length of the Iliac Lines is measured with the sliding caliper from the anterior origin of the arcuate line on the pubic bone to its posterior termination at the sacro- iliac joint just superior to the pre-auricular sulcus. Both ilia are measured for this feature. In the discussion of the dentitions, the following abbreviations have been used : R = right side; L =—lJeft side; M3 = third molar; M2 —,second. molares Ma — first molar; PM2 =second premolar; PMi1 = first premolar; C = canine; Iz = lateral incisor ; Ir = central incisor; the placement of the number above or below the line indicates that reference is made to either the upper or lower dentition, viz. RM3 = maxillary right third molar. Sex and Age Determination Specimen BP2/17: This is a male who died between the ages of 25 and 30 years. The calvaria is the largest and most rugged in the series. The brow ridges are prominent and posterior to them is the trace of a frontal sulcus. The mastoid process is large with a supramastoid crest that exhibits a rough inferior margin. This robusticity of the mastoid is reflected also in the area of insertion for the splenius capitis and longissimus capitis. The digastric fossa is deep. The lateral border of the superior nuchal crest forms a moderately well defined ridge, but as its median portion is not represented in the fragment, its overall dimensions cannot be observed. Temporal lines are sharp at their frontal aspect and sweep posteriorly as an arc, of medium prominence on the parietals. The supramastoid crest is large. The zygomatic processes of the temporal are rough along their inferior margins, par- ticularly in the region of the anterior tubercle where some fibres of the masseter have their origin. The region of the attachments of zygomaticus major and minor on the malar shows prominent bossing. Medial to this is an elongated crest for the levator labii superioris and levator labii superioris alaeque nasi. The malar is short but massive. The maxilla exhibits very pronounced subnasal grooves. The alveolar arch of the palate is high. A basal view of the calvaria reveals further evidence of male robusticity : a deep mandibular fossa and a stout occipital condyle. The right and left innominates are characterized by narrow and deep ischiatic notches, large acetabulum, prominent sacro-iliac articulation, and a high upright ilium. A pre-auricular sulcus is present, but its dimensions are small and narrow. There is pronounced ridging of the region above the posterior superior iliac spine where gluteus maximus originates. The crest of the ilium is sharp, especially along HUMAN SKELETAL MATERIAL FROM CEYLON 147 the line of origin of obliquus abdominis externus. The ilium is thick and heavy and the pelvic basin is small. Male characteristics are exemplified in the shoulder girdle with its relatively large clavicle and large scapular glenoid fossa and in the lower extremities where the linea aspera of the femur forms a pronounced pilaster. The absence of the pubic bones precludes the possibility of aging this specimen on the basis of progressive changes of the symphysial region. The dentition of the maxilla shows that the third molars have erupted and, like the other teeth, have undergone only a slight degree of attrition. Suture closure as an age indicator is of dubious value for this specimen due to trauma it has received from erosional forces and crushing. However, a general impression of its pristine condition can be obtained in a limited number of instances. Of the coronal suture, pars bregmatica is still undergoing closure but pars pterica is advanced. Of the sagittal suture, partes verticis and lambdica are advanced. The bregmatic and obelionic parts are eroded. Pars lambdica of the occipital suture has begun closure, but pars asterica is irregular, being most patent at its inferior portion. The masto-occipital suture is half closed. The squamous portion of the temporal is patent, but the spheno-temporal is advanced. Both spheno-parietal and spheno-frontal are commencing closure. Radiography of the vault confirms visual observation that arachnoid granulations are not present. These observations suggest an age at time of death for the specimen of between 25 and 30 years. Specimen BP2/171: This is an adult male. Its mandible, in comparison with those associated with specimens sexed as females, is of greater size, weight and thickness. The symphysis is higher, and the rami form a less obtuse angle in relation to the corpus. The short broad rami have robust pterygoid attachments at their gonia, which are thick and strongly everted. The mylo-hyoid ridge is pronounced. All of the teeth have erupted, and the third molars are moderately worn. The other molars reveal pronounced attrition. Specimen BP2/20-41 : This specimen may be that of an adult male, but criteria for sexing and aging are less certain than for other specimens of the series. Subnasal grooves are moderately developed. Of more masculine appearance is the mandible which is large and heavy with a pronounced mylo-hyoid ridge, very prominent pterygoid attachments, medium-sized genial tubercles and large digastric fossae. The dentition has erupted completely, and there is a moderate degree of abrasion, except for the upper left distomolar which is unworn. Specimen BP4/8: This is an adult male characterized by robust musculature of the face and long bones. The malars are large and heavy, as is also the zygomatic process of the temporal. Subnasal grooves are pronounced, particularly in the incisive region. The palate is of moderate height with a large lump-shaped torus. The suborbital fossa is deep. Orbital and nasal borders are dull. The humeral supracondylar ridges are sharp, but the head of the humerus is reported by Deraniyagala as being comparatively small. The muscularity of the radius is pronounced as represented in the sharpness of the interosseous crest and the inferior border of the pronator quadratus attachment. The oblique line of the flexor digitorum sublimus muscle is well developed. Radial tuberosities are large. 148 HUMAN SKELETAL MATERIAL FROM CEYLON The interosseous border of the ulna is less pronounced, but the groove where the extensor pollicis longus originates is marked. Likewise apparent is the insertion for brachialis on the anterior surface of the coronoid process. The supinator crest is high. The linia aspera of the femur forms a very prominent pilaster, and other well-developed areas of this bone are the crista hypotrochanterica and the trochanter minor. The femur is heavy and massive. The tibia is characterized by a moderate degree of muscularity. Its interosseous borders are sharp. Similarly the fibula has a sharp anterior margin, and the bone is deeply fluted. The teeth of this individual have completely erupted. The third molar has under- gone slight attrition, but the other teeth reveal a moderate degree of wear. Specimen BP1/6: This is a female whose mandible exhibits signs of senile modifi- cation. The maxilla is delicate and the sub-nasal grooves are of medium develop- ment. The mandibular corpus is fragile and constricted at the molar region, but well developed, heavy and wide at the symphysial region. The chin form is median, and projection is pronounced. The mylo-hyoid ridge is of medium muscularity. The ramus is not robust, and its gonion is thin with reduced pterygoid attachments. Eversion of the gonia is medium. The teeth are erupted and, save for the slightly worn third molars, the remainder of the dentition shows a moderate to pronounced degree of wear. Specimen BP2/21 : This is a female whose age at time of death was between 18 and 20 years. The supraciliary arches are reduced in development, and the superior orbital border reveals a delicate lipping. The mastoid process is medium in size, but the bossing for the sternomastoid is prominent. The mastoid shows a slight lateral projection. The occipital crests observed from their external aspects are reduced to traces. The supramastoid crest is moderately developed, but the temporal lines are traceable only on the frontal bone, where their conformation is rounded. The zygomatic processes and malars are small and gracile. For the maxilla the subnasal grooves are of medium development and the nasal sills are sharp. The palate is medium in height. Viewed basally, the calvarium presents a shallow mandibular fossa, small postglenoid process, and a minute stylo-mastoid foramen. The mandible is of medium size. The mylo-hyoid ridge is low, and the fossa for the mandibular gland is prominent. The gonia converge, but are strikingly thick and knobby as in specimen BP2/17i, a male. The mandibular condyles are small and converging, and the coronoid process is high. The degree of muscularity of the post-cranial bones is within the moderate category. The humerus has a well-marked bicipital groove and tuberosity. The clavicle is sharply ridged in the region of the deltoid attachment. It is the low muscularity of the calvaria, rather than the features of the mandible and post-cranial bones, that suggests that the specimen is a female. The criteria for age determination are more satisfactory for this specimen. The third molars of the maxilla and mandible have not completed their eruption and lie partially embedded in their alveoli. Save for the lower incisors, attrition is negligible for all of the teeth. Further proof of the young adulthood of this specimen is adduced by the metacarpal bones of the right hand where the epiphyses of the heads HUMAN SKELETAL MATERIAL FROM CEYLON 149 and basal condyles have not completed ossification with the shaft. This fusion normally terminates in the twentieth year of life. Finally, sutural closure of the calvaria gives confirming data. The partes bregmatica and pterica of the coronal are commencing closure, but pars complicata remains patent. Of the sagittal suture, pars bregmatica is patent while the partes verticis, obelica and lambdica are commencing closure. The lambdoid suture is open at all regions save for pars media where closure has started. The masto-occipital suture is patent as are the spheno-parietal, spheno-frontal, and spheno-temporal sutures. The squamous portion of the temporal, however, is closed, which is an artifact of the degree of preservation of the specimen. Specimen BP2/25: This is a sub-adult female under 18 years of age. The sex criteria are less certain for this specimen than for the others of the series, but the cranium is small and its muscularity is reduced. Unfortunately the frontal and occipital tori can not be observed. There is a moderate degree of ridging of the squamous portion of the temporal. The supramastoid crest is of moderate develop- ment, but the zygomatic process of the temporal is very thick. Nasal sills are sharp. Subnasal grooves are pronounced. The mandible is small and has a short corpus with little muscular development. The ramus is short and narrow with negative gonial eversion. Muscularity is very much reduced on the clavicle which is smooth and gracile. The radius is moderately fluted, but the interosseous crest is high. The radial tuberosity is low. The pilaster of the femur is of medium development and is mound-shaped. The fossa hypo-trochanterica is deep. For the tibia, the anterior ridge is sharply defined but overail muscularity is reduced. Age is established on the basis of the unerupted state of the third molars of maxilla and mandible. Due to the inferior condition of the specimen the degree of suture closure on the cranium cannot be determined. Epiphyses of the long bones are complete, and this specimen can most accurately be aged as a sub-adult. Specimen BP3/27—34: This is a female whose age at time of death was between 30 and 35 years. The cranium is of medium size and exhibits a low degree of muscularity. The frontal torus is hardly discernible and glabella is very low. The occipital crests are reduced to low mounds. Inion is absent. Temporal crests are sharply defined on the frontal bone, but disappear in their progress posteriorly over the parietals. The anterior and posterior portions of the supramastoid crest are smooth. The malars are smalland smooth. The temporal fossa has a moderate degree of roughen- ing, particularly on the posterior sphenoidal surface. The alveolar region is small, but subnasal grooves are pronounced and the palate is high and moderately ridged. The palatine torus is mound-shaped. Orbital and nasal borders are sharp. The suborbital fossa is deep and massive. The basal aspect reveals a deep mandibular fossa and a thick postglenoid process of a length unusual in females. The petrous portion of the temporal is small. The mandible is medium in size and light in weight. The corpus exhibits a pronounced mental spine emerging from a sharply pointed and projecting median 150 HUMAN SKELETAL MATERIAL FROM CEYLON protuberance. Genial tubercles are well defined and of medium size, but digastric fossae are small. The mylo-hyoid ridge is low. The angle formed by the ramus is more obtuse than that for the other mandibles of the series. The ramus is moderately broad and the gonia, while thin, are markedly crested for the attachment of the pterygoid and exhibit pronounced eversion. The coronoid process is high; the condylar neck is short. The small clavicle is thin and delicate. The scapular is small, and muscular attachments are weakly developed. The humerus is likewise reduced in robusticity, although the bicipital groove is well defined and has a sharp lateral lip. The small and sinuous radius has a large ulnar notch and an extensive but low radial tuberosity. The volar aspect of the radius inferior to the anterior oblique line shows a deep hollow for the reception of the belly of flexor pollicis longus. The styloid is of medium size. The ulna has a more prominent styloid process, and its crests are sharper. An estimate of age based upon suture closure alone would place this specimen within the range of 30 to 40 years at time of death. Conditions described below as patent or advanced are unreliable, as criteria of age in this specimen, due to absorption of the majority of the sutural margins, the post-mortem separation of the sutures as a result of dessication and the consequences of the pressure of the over-burden upon sutural areas apparently complete in their closure. Hence partes bregmatica, complicata and pterica of the coronal are patent due to erosive factors working on the bone. For the sagittal, pars verticus appears to have been advanced in its closure but the degrees of fusion for pars lambdica, obelica and bregmatica are uncertain due to post-mortem separation of the parietals. The situation is the same for the lambdoid suture where the margins are preserved but unarticulated. Those sutures which have maintained their pristine condition are the spheno-parietal which is completely closed and the spheno-frontal which is advanced. The squamous portion of the temporal is fused to the sphenoid, but its relationship to the parietal cannot be accurately observed, save for a view of its superior portion on the right aspect of the vault where closure is well advanced. The dentition shows pronounced attrition. The upper third molars have erupted, but the status of the lower third molars is uncertain, for these teeth are not present. The Metrical and Morphological Analysis of the Osseous Remains THE CRANIAL SKELETON, The Calvarium (Table 2). All four calvaria of the series are dolichocranic, the female specimens being the narrower. There is less agreement in the indices of the Auricular Height of the vault in relation to Cranial Length and Cranial Breadth: the male specimen is hypsicranic and acrocranic, the female specimen is chamaecranic and tapeinocranic. However, if the Basion—Bregma Height is employed as a component of these cranial indices, the values for the male specimen fall within the same categories as do those of the female. Such a dis- crepancy in the indices of Cranial Height is due to the imperfect preservation of the basion region of specimen BP2/17. Therefore, the indices which utilize the auricular values are those which are preferable as the more reliable. HUMAN SKELETAL MATERIAL FROM CEYLON 151 Cranial capacity cannot be directly measured for the specimens, and four formulae have been employed for the estimation of endocranial size. The calculation of von Bonin, while originally devised for male crania of natives from New Britain, has been demonstrated as suitable for male Australian crania as well (Hambly 1947 : 35-39). Isserlis (1914) worked out a formula for estimating the capacity of crania from the Gaboon area of West Africa, and Hambly has noted its applicability to male and fernale crania of Vedda and Australian populations (Hambly 1947 : 57). However, when Isserlis’ formula is applied to specimens BP2/17 and BP3/27-34 of the Balangoda series, the results are 1589-72 cc. and 919-66 cc. respectively. Although there are known to exist two female Vedda crania with directly measured cranial capacities of 960 cc. (Hill 1941 : 90, 93-94), the value derived from Isserlis’ formula in the cases under consideration is questionable. The formula of Lee & Pearson (1901) using Auricular Height confirms a low cranial capacity for specimen BP3/27-34. Without a confirmation of values for specimens of the Balangoda series that can be derived from direct measurement of cranial capacity, one is cautious in favouring the formula best suited to the series. These estimates do serve to illustrate the degree of variation between the large-headed male BP2/17 and the small, nannocranic female BP3/27-34. Differences in the values for the Cranial Module confirm this degree of variation. Specimen BP2/25, which was not recon- structed, appears to have had a small cranial vault and a greater delicacy of cranial architecture than BP3/27-34. The walls of the cranial vault are uniformly thick and heavy, due in part to their impregnation with mineral matter but to a greater degree this condition reflects their original nature. Porous areas are not observable. Their uniformly dense and grainy character is best observed radiographically. The frontal and anterior portions of the parietals are somewhat thinner than the bones of the posterior region of the vault. There is no sexual dichotomy in the thickness of the bones, although the parietals and occipitals of the female BP2/21 are the least massive in the series. Observed vertically, the conformation of the vault is sphenoid for BP2/17 and BP3/27-34 and rhomboid for BP2/21. In all three crania the greatest breadth is across the posterior third of the skull. The anterior portion of the vault of BP2/21 differs from the others by the reduced proportions of its glabella and superorbital torus. Brow ridges are median in both female specimens. The male specimen exhibits a very large glabella and a heavy, divided frontal torus. The frontal development of BP3/27-34 falls between these extremes. Lateral aspects show a medium height of the frontal region but sexual differentiation in the frontal slope. The frontal bone of the male specimen inclines gradually from glabella to bregma with a curvature limited to the plane of the low frontal eminences. It has a large median boss. The females have a slightly bulbous frontal bone that curves evenly and smoothly to bregma, medium to pronounced frontal eminences, and no median crests. Specimen BP2/21 exhibits a median boss of modest size. Post-orbital constriction of the frontal is prominent in BP2/17 and BP3/27-34, but reduced in BP2/21. Frontal breadth is large for the male vault, small for the female. Supra- orbital foramina are large for both sexes. The frontal sinus is spacious in BP2/17. GEOL. II, 4 15 152 HUMAN SKELETAL MATERIAL FROM CEYLON The parietal region bears a sagittal crest of medium height in BP2/17 and BP2/2r, but this region is not elevated in BP3/27-34. All the vaults of the series exhibit small post-coronal depressions. Parietal bosses are pronounced, and these lie well back towards the wider portion of the vault. From their apices, the sides of the brain case make a steep descent to the temporal fossae. Parietal foramina are not observable. The gradual curvature of the parietal portion of the vault, viewed from the lateral aspect, takes a sudden turn in a more vertical direction at lambda. FRANKFORT HORIZON PORION Fic. 3. Left lateral contours of three Balangodese (BP2/17 , BP2/21-—-, BP3/27-34 >» Fic 4. Frontal contours of three Balangodese (BP2/17 | 1BIPD Pim = =, BP3/27-34 HUMAN SKELETAL MATERIAL FROM CEYLON 153 Lambda is moderately flat in all specimens but one, BP3/27-34. The occipital curvature is pronounced for allspecimens. The occipital torus of BP2/17 is remark- able for its large size and robusticity : the female crania exhibit either reduction or no development of this region (Text-figs. 3, 4). The temporal region is full, particularly for the male specimen, and the sphenoid depression is remarkably large. The degree of cresting of the temporal lines is never pronounced along the parietals, but is marked on the frontals of the male vault. The squamous portion of the temporal is thick for BP2/17 and BP2/21, but thin for BP3/27-34. The size of the petrous portion is greater for the male. The auditory meatus is elliptical for the male, round for the females. The tympanic plate is consistently thick for both sexes. The mastoid is large for the male. Radiography reveals in the male specimen an extensive mastoid sinus. This consists of two polycameral sinuses in the superior aspect of the process along an anterior—posterior arc which traverses the width of the process. At its apex, a third polycameral sinus is visible. Sutures of the vault are simple in conformation for all specimens. Complexity is observed only in the peripheral margins of the coronal suture at pterion and of the sagittal at lambda. Wormian bones are not observable. Metopism is absent. The form of Pterion is H. Facial indices are ascertainable for specimen BP3/27-34. Its total facial index places it at the lower limit of the leptoprosopic class. The upper face is leptene. The nose is chamaerrhine. Whereas its orbits are chamaeconch, the orbits of BP2/17 are hypsiconch. The palates of all specimens are brachystaphyline. The angle of the upper face of BP3/27-34 places it within the orthognathous category as estimated by both the von Camper and Martin methods. The protrusion is of the alveolar sort rather than subnasal or total maxillary. Facial bones of the other specimens exhibit pronounced alveolar prognathism, save for the males BP2/17 and BP2/20—41 where alveolar protrusion is medium and slight respectively. The very pronounced alveolar prognathism of BP2/25 is to some degree an artifact of post-mortem distortion of the facial bones. Mid-facial prognathism is medium in the two cases where it is observable—specimens BP4/8 and BP2/27-34. The form of the orbit is rhomboid for the males and the female BP2/21, but square for the other female, BP3/27-34. Orbital inclination is 15° for BP2/17 and 16° for BP2/21, but the inclination is only 8° in the case of BP3/27-34. There is a striking sexual differentiation in the size and robusticity of the malars and zygomatic processes. Lateral malar projection is common to the females, but of the males, BP4/8 shows pronounced anterior malar projection. The total absence of nasal bones in the series precludes any comment as to the structure of this portion of the face save insofar as the maxillary walls and floor of the piriform aperture are preserved in BP4/8. From this evidence the nose appears to have been very broad with a large nasal spine and both sharp and dull sills. The nasal notch appears to have been deep for BP2/17 and BP3/27-34. Subnasal grooves of the incisive region are consistently prominent. The palate is elliptical except in specimens BP2/20-41 and BP3/27-34 where the form is hyperparabolic, an appearance ascribable in part 154 HUMAN SKELETAL MATERIAL FROM CEYLON to the absence of third molars. The height of the palate is medium, but specimens BP2/17 and BP3/27-34 have deep palates. Palatine ridges are larger in males than in females and are mound-shaped rather than ridged for both sexes. The palates are unusual in their large overall size and in their considerable breadth (Pls. 1-7, 10-14). TABLE 2 CRANIAL MEASUREMENTS AND INDICES Males Females SS SS SSS SSS Measurements BP2/17 BP2/20-41 BP4/8 BP2/21 BP2/25 BP3/27-34 Cranial Length (1) 200 te a 183 ie 177 Cranial Breadth (8) 147 ac 6 133 ae 130 Basion-Bregma Height (17) 135 ae ae a ws ae Auricular-Vertex Height (21) 132 Ye a ce 56 93 Auricular-Bregma Height (20) 127 : 3, a 95 Minimum Frontal Diameter (9) 117 — * 110 she QI Bizygomatic Diameter (45) ae 06 ie 121 ae IIo Menton-Nasion Height (47) es tie ae i Fe IOI Prosthion-Nasion Height (48) ae ts ae ae 60 64 Nasal Height (55) ne ais ae & ot 44 Nasal Breadth (54) on 23 ne ays sy 23 Orbital Height-Right (52) 37 ne nO 36 nC ae Orbital Height-Left (52) 37 St me ah as 40 Orbital Breadth-Right (51) 4I Ke Ke a8 te é3 Orbital Breadth-Left (51) 41 ae os ate st 30 Biorbital Breadth (44) 131 ae is a Zygomatic Breadth-Right 55 48 52 40 ae ai 5 Zygomatic Height-Right 52 40 44 42 ae 47 Zygomatic Frontal Process 29 56 22 24 56 22 Breadth-Right Zygomatic Frontal Process 25 Breadth-Left Palate-External Length (60) 61 ae an 62 sa 53 Palate-External Breadth (61) 69 70 fs 65 oie 63 Palate-Internal Length (62) 52 ie sits 50 BG 45 Palate-Internal Breadth (63) 36 47 ie 42 Se 42 Palate Height at M? (64) 19 15 14 I4 Fie 18 Arc-External Palate 185 ss 152 i ok Af Arc-Transverse (24) 355 oe Ans 300 A 270 Arc-Glabella-Opisthion (25) 355 af 34 320 Sie 285 Arc-Frontal Length (25) 132 oe ae 135 ee 115 Arc-Sagittal Length (27) 146 dg at 152 Ks 128 Arc-Occipital Length (28) 77 Pie bc 53 ate 58 Chord-Frontal Length (29-1) 118 oe fe I20 oa 105 Chord-Sagittal Length (30) 130 5G ay 132 =f 112 Maximum Circumference (23) x a ai 520 Ee 495 Indices Cranial 73°50 ae 56 72:68 50 73°45 Basion-Bregma Height-Length 67-50 sit sis Se i HUMAN SKELETAL MATERIAL FROM CEYLON 155 TABLE 2 (continued) CRANIAL MEASUREMENTS AND INDICES Males Females ee SS SS Indices BP2/17 BP2/20-41 BP4/8 BP2/21 BP2/25 BP3/27-34 Basion-Bregma Height- 91-84 Se a ax 2¢ Breadth Auricular-Vertex Height- 66-00 fe gfe ate a0 52°54 Length Auricular-Vertex Height- 89-69 ee aD a0 af 71°54 Breadth Auricular-Bregma Height- 63°50 Sie Ss ae be 53°67 Length Auricular-Bregma Height- 86°53 ene ms ee = 73°08 Breadth Total Facial Ae me Ae Ale oa 91:82 Upper Facial 56 56 Si aye a 58-18 Nasal oa 52°27 Orbital-Right 90-24 ae Orbital-Left go -24 ae a¢ + “ 75°00 External Palate II3°I1 a a 104°84 a 118-87 Internal Palate 69°23 ne si 84:00 i 93°33 Cranial Module 159°67 ee : 133°33 Cranial Capacity Basion-Bregma Height : von 1448-75 Bonin (1934 : 14) Lee & Pearson (1901: 1580-35 225-264) Auricular Height: Lee & 1775-83 a or a8 ee 1098 -87 Pearson (Ibid.) Isserlis (1914 : 189) 1589:72 ae ae aa Ain 919:66 The Mandible (Table 3). In contrast to the brachystaphaline upper jaw, the lower jaw is dolichostenomandibular. The alveolar portion of the mandible conforms to the characteristic shape of the palate, but the ramus describes an obtuse angle with the corpus, and moreover, is very broad. This results in a great Condylo- Symphysial Length but a moderate Bigonial Diameter. The corpus is thick and heavy at the symphysis, but is reduced in size, particularly among the females, as it sweeps posteriorly to the gonia. The chin is prominent and median for all mandibles, save for that of BP2/21 where the chin form is bilateral. The corpora of males are thicker and higher than the corpora of females, due in part to their more developed mylo-hyoid crests. There is less sexual diversity in the development of the digastric fossae and genial tubercles. These range in size from sub-medium to large. Alveolar prognathism is small in males and medium in females. The rami are robust in the males, their pterygoid attachments are well developed and gonial eversion is pro- nounced. Female rami are delicate and the gonia may be strikingly everted or convergent. The mandibular notch is small to medium, save for specimen BP2/21 which has a deep notch due to the greater length of the coronoid process. In all 156 HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 3 MANDIBULAR MEASUREMENTS AND INDICES Males Females ae = SD Measurements BP2/17i BP2/20-41 BP1/6 BP2/2z1 BP3/27-34 Condylo-Symphysial Length (68) 106 79 a 99 93 Bicondylar Breadth (65) 118 115 a0 105 112 Symphysial Height (69) 28 20 31 31 28 Bigonial Diameter (66) 96 110 oc 99 IOI Height of Ascending Ramus—Right oe 57 70 54 49 (70) Height of Ascending Ramus—Left (70) 65 af as 48 49 Minimum Breadth of Ascending Ramus a as 30 31 31 —Right (71) Minimum Breadth of Ascending Ramus 34 35 ot 33 31 Left (71) Maximum Breadth of Ascending 56 3¢ 3¢ 36 41 Ramus—Right (71-1) Maximum Breadth of Ascending 44 a he 36 40 Ramus—Left (71-1) Bimental Diameter (67) 45 30 ae ae 41 Depth at PM?—Right (69-2) 31 27 ae ae 27 Depth at PM?—Left (69-2) 31 26 27 ac 20 Depth at M1—Right (69-2) 28 25 27 22 23 Depth at M1—Left (69-2) Si 26 26 ae 23 Depth at M?—Right (69-2) 26 25 23 23 23 Depth at M?—Left (69-2) 28 26 25 23 20 Depth at M’—Right (69-2) 26 O06 21 24 20 Depth at M’—Left (69-2) 29 27 22 24 20 Thickness at M!—Right (69-3) 16 II 14 12 ae) Thickness at M!'—Left (69-3) 16 12 15 13 9 Thickness at M?—Right (69-3) 20 12 18 15 14 Thickness at M?—Left (69-3) 19 13 17 16 13 Thickness at M*—Right (69-3) 18 12 17 17 II Thickness at M’—Left (69-3) 19 16 18 17 II Mental Foramen Diameter—Right aS 4X3°5 15 X2°5 Mental Foramen Diameter—Left a6 4X3°5 25X3 eo ae Mean Angle (79) 118 123 on 116 133 Indices Mandibular 89-83 68 - 69 are 94°28 83:03 Fronto-Gonial ae Bs 104°76 81-25 Zygo-Gonial ie i ad 81-82 Ior-81 mandibles the condylar process does not rise very high above the coronoid process and the mandibular head is small. The lingula of the mandibular fossa is uniformly small. The ramus is broad and moderately high. The Balangodese mandible is of medium size but is distinctive for its broad and elevated rami and thick corpora. Sexual differences are difficult to isolate (Pls. 8, 9). a HUMAN SKELETAL MATERIAL FROM CEYLON 157 Tue Post-CRANIAL SKELETON. The Skeleton of the Trunk. Vertebrae were available for two specimens: BP2/17 and BP2/25. Their fragmentary and badly warped condition renders a metrical description of dubious value, but measurements were feasible for the fourth and fifth lumbar vertebrae of BP2/17. The cervical vertebrae are small and delicate with cordiform bodies and triangular neural foramina. The transverse processes are short. The atlas has a thin posterior arch and a very reduced posterior tubercle. The facet for the odontoid process is proportionately large for the accommodation of the thick, bulbous axial dens. The thoracic vertebrae are small. Their vertebral foramina are round and their bodies contain broad costal facets. The transverse processes are bulbous. The spinous processes are not large. In contrast to the pre-lumbar vertebrae, the last five segments of the spine are robust and are equipped with prominent posterior and transverse processes. The kiolorachic lumbar vertebrae of BP2/17 exhibit thick centra and wide epiphysial rings. For the fifth lumbar vertebra the Vertical Ventral (1) and Vertical Dorsal (2) Diameters are 23 mm. and 27 mm. respectively : the Depth of the Body is 30 mm. for the anterior—posterior diameter (4) and 46 mm. for the lateral diameter (7). Rib fragments accompany the vertebral remnants but none of the ribs is complete. The ribs of BP2/17 are very large at their sternal ends but never exceed 6-0 mm. in thickness. The costal grooves are deep with sharp borders. The ribs of the female specimen attain their greatest size at a point some distance lateral to the sternal head. The first rib shows a well-marked groove for the subclavian vessels, but the ribs just inferior to it are smooth and their thickness averages 3:0 mm. For both sexes the neck of the rib is robust and massive. The sternum of specimen BP2/17 is lacking its manubrium, but the body measures 96 mm. in length and 32 mm. in breadth. Fusion of the sternal components is complete. Facets for six sternal articulations are well marked. Curvature is very slight. There are no sternal foramina. The Skeleton of the Upper Extremity (Table 4). The clavicle of the male specimen is large and shows a moderate degree of curvature. The sternal head is oval in form. The acromial head is moderately flat. Muscularity is reduced. In contrast to this condition are the clavicles of the female specimens which are small, delicately built, and with a medium to pronounced degree of curvature. Muscularity is much more apparent among the female clavicles which are characterized by sharp ridges for the deltoid attachment. These also show wide grooves for the subclavian vessels. Like the male clavicles, those of the females have oval sternal heads and flat acromial heads. The scapula is medium in size for the male, small for the female. For both sexes this portion of the upper extremity is robust. For the male specimen there is a pronounced obliquity in the direction of this spine in relation to a line tangent to the vertebral border. Since this border has undergone damage at the region inferior to the spine, the spino-vertebral angle can only be inferred and cannot be qualitatively or quantitatively ascertained. The axillary borders of both male and female scapulae are thick and tend to double their thickness in the vicinity of the glenoid fossa. The superior border is narrow and is lacking a scapular notch. The angle 158 HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 4 MEASUREMENTS AND INDICES OF THE BONES OF THE UPPER EXTREMITY Males Females > OF BP2/17. BP4/8 BP2/2t BP2/25 BP3/27-34 Clavicle Measurement Maximum Length—Left (1) Ee ne as Ss 123 Mid-Shaft Circumference—Right (6) Ee a a 28 32 Mid-shaft Circumference—Left (6) 39 bf 5.0 ae 33 Mid-shaft Diameter—Anterior-Posterior eh ifs ia Be Io —Right (5) Mid-shaft Diameter—Anterior-Posterior a aS 55 fe Io —Left (5) Sternal Head Diameter—Right ag 50 we 18 19 Sternal Head Diameter—Left i iS 22 Index Length—Thickness—Left 56 ae a6 36 29°46 Scapula Measurement Morphological Length—Left (2) 87 Corocoid Process Length—Left (11) 43 Corocoid Process Breadth—Right 14 Corocoid Process Breadth—Left 14 Acromion Process Length—Right (10) 53 Acromion Process Length—Left (10) 52 ae He Si us Acromion Process Breadth—Right (9) 30 te ae A 22 Acromion Process Breadth—Left (9) 26 Beg ca ae a Glenoid Fossa Height—Right (12) 36 itd ais une 30 Glenoid Fossa Height—Left (12) 38 ais oh ore a Glenoid Fossa Breadth—Right (13) 5 a Se ae 20 Glenoid Fossa Breadth—Left (13) 28 Humerus Measurement Maximum Length—Left (1) ae Bc 50 o/9 302 Bicondylar Length—Left (2) oe Aig a 38 298 Bicondylar Breadth—Right (4) 5h 53 3 : Bicondylar Breadth—Left (4) es 52 Pet nL es Maximum Head Diameter—Right (9, 10) ve ae oa fhe 35 Maximum Head Diameter—Left (9, 10) oe ae ae 40 35 Mid-Shaft Diameter—Anterior-Posterior a0 21 sys bts 12 —Right (6c) Mid-Shaft Diameter—Anterior-Posterior id 21 Bo a6 12 —Left (6c) Mid-Shaft Diameter Lateral—Right (6b) he 18 — hs 17; Mid-Shaft Diameter Lateral—Left (6b) ie 18 a wt 17 Mid-Shaft Circumference—Right (7a) ae 59 50 50 S'6 Mid-Shaft Circumference—Left (7a) or 59 ae De 48 Index Robusticity—Left as bi oe a 9:60 HUMAN SKELETAL MATERIAL FROM CEYLON 159 TABLE 4 (continued) MEASUREMENTS AND INDICES OF THE BONES OF THE UPPER EXTREMITY Males SSN BP2/17 Radius Measurement Maximum Length—Left (1) ae 248 Least Circumference—Left (3) as 38 Maximum Head Diameter—Right (4-1, fe 22 ag) Maximum Head Diameter—Left (4-1, ae 22 5-1) Mid-Shaft Diameter—Anterior-Posterior a Te —Left (5) Mid-Shaft Diameter—Lateral—Left (4) ae 16 Ulna Measurement Maximum Length—Left (1) ae 261 Least Circumference—Left (3) 28 33 Mid-Shaft Diameter—Anterior-Posterior 56 15 —Left (11) Mid-Shaft Diameter—Lateral—Left (12) $6 II Capitate Measurement Length (1) Breadth (2) Height (3) Trapezium Measurement Length (r1) Breadth (2) Height (3) Scaphoid Measurement Length (1) Breadth (2) Height (3) Metacarpals Digit I Length (2) ” IV ” BP4/8 Females ——.—_—q#{TT——— BP2/2t BP2/25 BP3/27-34 35 38 18 19 15 9 19 18 259 25 12 8 25 20 18 23 19 II 23 13 8 71 71 68 68 48 31 160 HUMAN SKELETAL MATERIAL FROM CEYLON of the glenoid fossa to the scapular body lies in a plane slightly lateral and superior to it. In form the fossa is elliptical, the inferior half being broad and the superior half narrowly constricted. Lipping of the fossa is submedium. The acromion is long and rectangular. The clavicular facet is unlipped. The corocoid process is flat and extensive. The humeral shaft is long and only slightly curved. A transverse section shows a prismatic pattern in the male specimen, a plano-convex pattern in the female. Muscularity is pronounced: the tuberosities are large, the bicipital groove is deep, the supracondylar ridges are sharp on medial and lateral aspects of the bone. The head is medium in size. The olecranon fossa is shallow and unperforated. Supra- condylar processes are absent. The radius is large for specimens BP4/8 and BP2/25, but small and sinuous for BP3/27. The shaft is straight in the male, but shows moderate anterior—posterior bowing in the female. The transverse section is oval in these bones for both sexes. The extremities of the radius are more massive for the male, but both sexes have shafts with well-marked muscular attachments. Tuberosities are extensive and low in females, prominent in males. The neck is longer for the male. Lipping of the capitulum occurs in the bones of both sexes. The styloid process is particularly well marked for males. Both sexes exhibit prominent muscular attachments on their radii. The ulna is triangular in transverse section. Curvature of the shaft is slight, occurring in the male only to a limited degree in the region of the brachialis attach- ment just inferior to the coronoid process. Muscularity is pronounced for the male, moderate for the female. Both sexes show a prominent styloid process. The interosseous space is large as a result of radial rather than ulnar curvature. The bones of the hand tend to be large. The crests of the carpals are weakly developed. The form of the capitate is square. The scaphoid assumes a dumbbell conformity, but the trapezium is prominently ridged and metacarpals irregular. The metacarpals are long and exhibit pronounced dorso-ventral curvature. This being the hand of a sub-adult individual, the heads of the shaft are ununited. The metacarpal formula is II > III > IV > V(?) > I(?). The fifth metacarpal is not available for examination. The Skeleton of the Lower Extremity (Table 5). The femur of BP2/17 is stenomeric, and femora of the other male and the female of this series are eurymeric. Transverse section pattern is either round or quadrangular for the males but plano-convex for the female. The linea aspera is marked with a prominent pilaster, particularly in the case of the males. The shaft is uniformly thick and bowing is slight to medium in its degree. Tortion is slight to medium in its degree. Tortion is within the medium category (10°-20°), save for specimen BP2/17 where the tortion is more pronounced. The crista hypotrochanterica is of medium development for BP4/8, but the fossa hypotrochanterica is deep for all specimens. Trochanters major and minor are large and smooth. The neck of the femur is short and narrow and supports a small head. The intercondylar fossa is deep and broad in its transverse plane. There isa pronounced inferior orientation of the medial condyle and a slight anterior prominence of the lateral condyle. The adductor tubercle is reduced. HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 5 MEASUREMENTS AND INDICES OF THE BONES OF THE LOWER EXTREMITY Males Females SSS SS, BP2/17 BP4/8 BP2/25 Femur Measurement Maximum Length—Right (r) 419 Bicondylar Length—Right (2) 414 Be ae Mid-Shaft Circumference—Right (8) 72 80 66 Mid-Shaft Circumference—Left (8) Je 79 71 Sub-Trochanteric Diameter—Anterior- 25 24 Posterior—Right (10) Sub-Trochanteric Diameter—Anterior- 50 26 20 Posterior—Left (10) Sub-Trochanteric Diameter—Lateral— 20 2 Right (9) Sub-Trochanteric Diameter—Lateral— is 26 23 Left (9) Mid-Shaft Diameter—Anterior-Posterior 24 25 24 Right (6) Mid-Shaft Diameter—Anterior-Posterior 25 27 25 —Left (6) Mid-Shaft Diameter—Lateral—Right (7) 20 23 18 Mid-Shaft Diameter—Lateral—Left (7) 21 23 19 Maximum Head Diameter—Right (18, 19) 35 Indices Platymeria—Right I25:00 88-89 86-96 Platymeria—Left oe 100-00 D0 Middle—Right 83-33 92:00 75°00 Middle—Left 84:00 85:18 76:00 Pilastric—Right I20:00 108:59 133°33 Pilastric—Left I19°05 117739 8131-59 Robusticity—Right 10-63 O06 =i Tibia Measurement Mid-Shaft Diameter—Anterior-Posterior 26 27 23 —Right (8) Mid-Shaft Diameter—Anterior-Posterior 25 2G, —Left (8) Mid-Shaft Diameter—Lateral—Right (9) 19 22 16 Mid-Shaft Diameter—Lateral—Left (9) 18 21 aH Nutrient Foramen Diameter—Anterior- i ate 27 Posterior—Right (8a) Nutrient Foramen Diameter—Anterior- 26 33 Posterior—Left (8a) Nutrient Foramen Diameter—Lateral— ae ee 19 Right (ga) 162 HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 5 (continued) MEASUREMENTS AND INDICES OF THE BONES OF THE LOWER EXTREMITY Males o———__—s« Females BP2/17 BP4/8 BP2/25 Nutrient Foramen Diameter—Lateral— 19 23 Left (9a) Least Circumference—Right (rob) 69 ont 61 Least Circumference—Left (1ob) 66 72 Index Middle— Right 73°07 81-48 69°56 Middle—Left F2200) | 77-78 uid Platycnemia—Right s3 50 65°52 Platycnemia—Left 73°08 69-60 ae Fibula Measurement Mid-Shaft Diameter—Anterior-Posterior Be 15 —Left (3-2) Mid-Shaft Diameter—Lateral—Left (3-1) aie 9 Least Circumference—Left (4a) He 33 Distal Breadth—Left (4-2) an 23 Talus Measurement Length—Right (1) 45 Breadth—Right (2) 32 Height—Right (3) 18 Calcaneum Measurement Length—Right (1) 67 Breadth (2) 37 Cuboid Measurement Length—Left (1) 35 Breadth—Left (2) 29 The tibiae are eurycnemic for specimen BP2/17 and mesocnemic for BP4/8 and BP2/25, although these latter two stand at the upper and central loci of the mesoc- nemic category. Tibial lengths could not be ascertained. The posterior half of the transverse section of the shaft is oval, but in BP4/8 the conformation is closer to plano-convex. The bones are heavy and thick for the males. Bowing of the shaft is very slight. Muscularity is moderate for both sexes. The fibula is represented by the single fragment belonging to BP4/8. It is the left fibula. The form of its transverse section is triangular. Whereas the anterior HUMAN SKELETAL MATERIAL FROM CEYLON 163 border is sharp, fluting of the shaft is sub-medium in development. The lateral malleolus is medium in size and its articular facet is extensive. The shaft is straight. The talus possesses a large triangular articular facet for the fibular malleolus. The groove for the flexor hallicus longus is medium in development. The neck of the talus is not constricted. The calcaneum is short and deep with a round posterior curvature of the sustentaculum tali. The cuboid exhibits pronounced curvature of the cuboid-lateral cuneiform facet. The articular facet is large, round, and its borders poorly defined. There is no lipping of the facet. The cuboidal ridge is prominent. TABLE 6 MEASUREMENTS AND INDICES OF THE BONES OF THE PELVIS Male BP2/17 Sacrum Measurement Anterior Length (2) 102 Sacral Breadth (5) 125 Lateral Diameter of Body 1 (19) 46 Anterior-Posterior Diameter of Body 1 (18) 29 Anterior Height of Body 1 (24) 29 Anterior Height of Body 2 (24) 28 Index Sacral 122°55 Innominate Measurement Innominate Length—Right (1) 207 Innominate Length—Left (1) 184 Innominate Breadth—Right (6a) 141 Innominate Breadth—Left (6a) 142 Distance from Acetabulum to Apex of Iium—Right (9) 132 Distance from Acetabulum to Apex of Ilium—Left (9) 114 Length of Iliac Lines—Right 262 Length of Iliac Lines—Left 260 Vertical Diameter of Acetabulum—Right (22) 42 Horizontal Diameter of Acetabulum—Right (22) 37 Breadth of Sciatic Notch—Right (8) 50 Distance from Anterior Superior to Posterior Superior 138 Iliac Spines—Right (12) Sacrum and Innominates Measurement Bi-iliac Diameter (2) 218 Sagittal Diameter of the Pelvic Inlet (23) 157 Transverse Diameter of the Pelvic Inlet (24) 107 Index Pelvic Inlet 146:81 Innominate Breadth-Height—Right 129°58 164 HUMAN SKELETAL MATERIAL FROM CEYLON The Skeleton of the Pelvis (Table 6). The single sacrum observable in the series is platyhieric. The sacral type is homobasal. Of the five sacral vertebrae the first, second and third articulate with the innominates. Curvature commences at the second sacral body but is not pronounced. Spina fissa is absent. The overall size of the sacrum is large. Deraniyagala considers the Stone Age population of Ceylon to have been steatopygous. He discusses this in relation to the small pelvis of the female skeleton found at the site of Alu galge (Telulla) (Deraniyagala 19550 : 301) and considers the fossils from Bellan Bandi Palassa to have exhibited the same trait (Deraniyagala 1958a@: 255). The present writer is unable to find any evidence either to confirm or reject this opinion. The innominates of this individual are thick and massive. Their size is moderate and the muscular attachments are prominently sculptured. The ilium has heavy and irregular crests along its superior rim and deep hollowing of the iliacus portion of the medial surface. The origin of gluteus maximus is well marked. There is considerable torsion between ilium and ischium. The anterior superior iliac spine is broad and blunt. The ischia are divergent in their orientation and are moderately elongated in size. The ischiatic notch presents a higher angle than is usual in male specimens. The pelvic inlet is cordiform and its index places it within the dolichopel- lic group, although the absence of the complete pubic area renders this statement subject to question. An Estimation of Stature (Table 7). Limb proportions are unfortunately impossible to establish for the Balangoda population due to the paucity of long bones complete enough for their maximum lengths to be accurately measured, and to the absence of any bones of the upper and lower extremities belonging to the same individual. Those long bone lengths which could be measured are listed in the table below, where they are employed as components of the formulae devised by Trotter & Gleser (1952, 1958) for estimation of stature from long bones. The lengths of the bones of two males, BP2/17 and BP4/8, and one female, BP3/27-34, are employed in formulae appropriate to American Whites, American Negroes, and American Mongoloids, mainly Chinese. The stature estimates for each of the three groups are given. In an effort to ascertain which of the three formulae might be best suited to the Balangodese, the writer applied them to the long bone lengths of three Veddas whose stature in life was known and whose long bones had been measured by Hill. These individuals were two males—Burunda, specimen 1949-12—7—4, and Tissahamy, specimen 1949-12-7-6, both of British Museum of Natural History (Hill 1941 : 148— 149, tables 1, 22). The former individual was from Dambane and was partly Sinhalese; the latter from Kalakoluebe near Yakkure was non-Sinhalese in racial background. The female Vedda available for comparison was not measured in life but skeletal stature is known. This is the Vedda specimen in the osteological collec- tion of the Bombay Natural History Society (Hill 1941 : 227-235). Such estimations of stature as given here are only indicative of the order of stature which is to be anticipated for the three Balangodese individuals. For the Veddas, ee HUMAN SKELETAL MATERIAL FROM CEYLON 165 the formula for Mongoloids renders a value that most closely approximates the living stature of Burunda, but for Tissahamy, the formulae for the Negroes is a nearer approximation. The Metrical and Morphological Analysts of the Dentition In their dental morphology the Balangodese have teeth which are moderate to large in size by general standards for post-Pleistocene hominid populations (Table 8). The order of size of the molars within the molar rows of each specimen show con- siderable variation. For BP2/17, the RLM! and RLM? are progressively smaller than the RLM$. This is also the case for the mandibular molar row of BP2/17i, but in the remaining male specimens the maxillary molar row shows the first molars of greater size. The mandible of BP2/20—41, however, shows a larger third molar for the row. For the female BP1/6, the maxilla has a second molar of greatest size followed in descending order by the first molar and the third, but the mandibular molar row has the third molar larger than the first and the first larger than the second. TABLE 7 AN ESTIMATION OF STATURE BASED UPON LENGTHS OF LONG BONES FOR BALANGODESE AND VEDDAS BALANGODESE Males (ae SS ee Female BP2/17 BP4/8 BP3/27-34 Length Stature Length Stature Length Stature Right Left Left White : Humerus ole oe 56 5 e 302 1654 Radius SC a 248 1734 Be a Ulna 3 ae 261 1713 259 1706 Femur 419 1627 6 oe a6 bo Tibia 50 ae Fibula Negro : Humerus S% ae ae if 302 1624 Radius 5e a 248 1678 pe ae Uina G ey 201 1663 259 1656 Femur 419 1602 é a tas oY Tibia Fibula Mongoloid : Humerus ae as Be om 302 1641 Radius = 4c 248 1698 Bf ae Ulna als ee 261 1683 259 1676 Femur 419 1626 oe O6 ¥ Tibia ee: aA Fibula 166 HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 7 (continued) AN ESTIMATION OF STATURE BASED UPON LENGTHS OF LONG BONES FOR BALANGODESE AND VEDDAS VEDDAS Males Female (SS SSS Bombay Nat. 1949-12-7-4 1949-12-7-6 Hist. Soc. Length Stature Length Stature Length Stature ee R L Right R I, R 1B, White : Humerus 300 301 1648 1650 ae as 250 262-5 1503 1540 Radius 241 239 1707 1700 Be oid 202 202 1560 1560 Ulna 26052635732 e17Zr ae Ms DP BIT 1566 1566 Femur 429 430 1650 1653 398 1579 361 362 1493 1495 Tibia 365 3608 1703 1710 328 1613 Sey, Siti 1572 1572 Fibula 358 355 1686 1678 325 1600 301 300 1538 1535 Mean = 1686 1597 Negro : Humerus 300 301 1619 1622 sc a 250 262°5 1475 1512 Radius 241 239 1654 1648 os o6 202 202 1525 1525 Ulna 266 263 1679 1669 Ke ae 22 22 1535 1535 Femur 429 430 1823 1825 398 1558 3601 362 1480 1482 Tibia 365 308 1653 1659 328 1572 aye Sita 1535 1535 Fibula 358 355 1638 1631 325 15601 301 30c 1505 1503 Mean = 1677 1564 Mongoloid : Humerus 300 301 1636 1638 oe a 250 262-5 1502 1535 Radius 241 239 1673 1667 nee as 202 202 1535 1535 Ulna 266 263 1700 1690 : 46 221 221 1543 1543 Femur 429 430 1648 1650 398 1681 361 362 1502 1504 Tibia 365 368 1687 1694 328 1598 Sinn yaaa 1558 1558 Fibula 358 355 1665 1658 325 1586 301 300 1528 1526 Mean = 1667 1622 Known (Living) 1629 (Living) 1493 (Skeletal) 1267 Stature : Specimen BP2/21 offers another variant: the maxilla shows a large first molar, a medium second molar and a smaller third molar, but in the mandible the largest tooth is the second molar followed in size by the first and then the third. Specimen BP2/27-34 has a larger maxillary second molar than the first molar, but a larger mandibular first molar. These data can be summarized as follows : Males : BP2/17 Manilla. M3 >M2>Mr1 BP2/171 Mand. M3 >M2z >Mr1 BP2/20-41 Manilla. Mit > M2 > M3 Mand. M3 >Mri > Mz2 BP4/8 Maxilla. Mi > M3 > M2 Females : BP1/6 Maxilla. Mz > Mi > M3 Mand. M3 >M1 > M2 BP2/21 Manilla. Mi >M2>M3 Mand. Mz >Mi >M3 BP3/27-34 Mawnilla. Mit > M2 Mand. M2 >Mt HUMAN SKELETAL MATERIAL FROM CEYLON 167 The molars are round in their conformation, but rectangular molars are present in the LM? of BP2/20-41 and in all molars of BP2/21 save for its RLM+ which is round. Premolars are rectangular in the maxillae of both sexes, but variability occurs in the mandibles. Specimen BP2/20-41 has RLPM! and RLPM? which are round ; another male, BP2/17i, has a triangular LPM! although its other mandi- bular premolars are round. In both these cases, the molar pattern is round, with the exception of the LM® of BP2/20—41 noted above. There is no maxillary dentition available for BP2/17i1. The female BP3/27-34 has a round RLPM! adjacent to a rectangular RLPM2 and round molars. Canines of the maxillae and mandibles are triangular for both sexes, but rectangularity appears in the RLC of the male BP2/20-41. Incisors are uniformly rectangular. Cusp and groove patterns are not observable for all of the teeth of the series due to the high incidence of attrition. For the maxilla, the males show pronounced variability. Specimen BP2/17 shows 5 cusps for the LM? and 3 cusps for RLM2. The other male, BP4/8, has 4 cusps on the RM+4, RM2, and RM&, the other molars not being observable. In contrast, the maxillary molars of the female specimens are of the 4-cusp pattern. The RLM of BP2/21 has 6 cusps. In the single male, specimen BP2/20-41, for which the cusps of the lower molars are observable, the LM3 has 5 cusps and the remaining left and right molars have 4 cusps. The three female specimens show one, BP1/6, with 5 cusps for the RLM? and RLM!, but 4 cusps for the RLM?. Specimen BP2/21 has 5 cusps for all molars, save for the LM which has 6 cusps. For specimen BP3/27—34 only the RLM? with 4 cusps is available. Of the four specimens for which both maxillae and mandibles are available, there is no case of correlation of cusp pattern except for specimen BP3/27— 34 where the second molars are alone observable. The upper and lower premolars of females are bicuspid, the buccal cusp being the larger and higher. Male premolars are also of this pattern, save for the RLPM? of BP2/20-41 where the premolars have 4 cusps due to the presence of accessory cusps on the distal inclination of the buccal cusp. Carabelli’s cusp is present in the LRM+# of one specimen, BP2/21. The cusp is small and the mesio-lingual groove is of moderate development. Accessory cusps are found on the second and third maxillary molars of the male specimen BP2/17. They are small and occupy loci on the mesial half of the buccal surface. Crenulation or excessive wrinkling of the occlusal surface of the upper third molars is shared by males—BP2/17 and BP2/20-41—and females—BP1/6 and BP2/21. It is absent in the remaining male maxillae and the third molars of the female BP3/27-34 are unavailable for examination. Its occurrence is marked by the presence of 5 cusps in the male maxillae. For the females, BP1/6 has the 4-cusp pattern and BP2/21 has 6 cusps. All cases are characterized by pronounced crenula- tion regardless of cusp number. The mandible of male BP2/20-41 shows a deeply crenulated LM? with 5-cusp pattern—a reflection of the maxillary LM® of this same specimen. The mandible of female BP1/6 shows a pronounced crenulation of the RLM3?, but the cusps of these lower third molars constitute the 5-cusp pattern instead GEOL, II, 4 16 168 HUMAN SKELETAL MATERIAL PROM CEYLON of the 4-cusp pattern of the upper third molars of the same specimen. In specimen BP2/21 six cusps are observed in both maxillary and right mandibular third molars but the RM3 has 5 cusps. For the maxilla, cusp patterns are unusual in the dentition of female specimen BP2/21 where the RLM+ and RLM? are of the +3 type. The groove patterns for the remaining maxillary dentitions of the males are of the +4 pattern. The mandible of BP2/21 shows a RLM2 of the +5 pattern and a RLM? of the Y5 type. In specimen BP1/6 the RLM? is also of Y5 type, but the RLM? is +4 and the RLM is +5. The other female specimen, BP3/27-34, has a RLM? of +4 type. The only male mandible which offers data on groove pattern is specimen BP2/20-41, whose LM? is Y5 and whose RLM? and RLM! are +4 in form. Taking the series collectively, it would appear that the basic cusp pattern for the upper molars is of the +4 type. The lower molars are characterized by the groove patterns Y5, +5, and +4. A correlation of cusp type and groove pattern reveals that the presence of 4 cusps is associated with the +4 pattern in the mandible as observed in specimens BP2/20-41, BP1/6, and BP3/27-34. However, the mandibular 5-cusp type may be associated with either the +5 type, as with the RLM? of BP1r/6 and RLM? of BP2/21, or with the Y5 type as represented by the LM? of BP2/20-41, RLM? of BP1/6, and RLM! of BP2/21. As is obvious from an examination of these cases for which a mandibular molar dentition is available, each specimen has its unique combination of cusp type and groove pattern. Nevertheless, where the M1 has 5 cusps, its groove pattern is a Y5. The M; is consistently a + pattern and in three of the specimens a +4, namely with BP2/20, BP1/6, and BP3/27-34. In one specimen, BP2/21, this is +5. Crenulation of the M3 is associated with the presence of 5 cusps which are either of Y5 pattern like BP2/20 or a +5 pattern like BPr1/6. The groove pattern of the M? of BP2/21 is unobservable, but the cusp number is 5 for the RM$ and 6 for the LM?, both molars showing crenulation. This point cannot be confirmed for the mandibular third molars of BP3/27-34. Inthe maxilla, as noted above, crenulation of the third molar is associated with 4 cusps (BP1/6), 5 cusps (BP2/17, BP2/20—41) or 6 cusps (BP2/21). Attrition of the molars of male and female dentitions show considerable variation. The maxillary molars are very slightly worn in BP2/17. All cusps of the RLM+ and the mesio-labial cusp of the LM2 exhibit abrasion of the enamel with consequent exposure of the dentin in the form of pits at the apices of the cusps. Direction of abrasion cannot be established for this specimen. In contrast to this condition are the heavily abraded molars of BP2/17i. The RLM? show the least wear for the six teeth of the molar rows, but like them a rim of enamel surrounds a lake of dentin from the centre of which protrudes a reduced cusp which is still covered with enamel in the RLM, LM2, and LM?. Wear is greatest on the labial aspects of each molar, but the enamel rim is everywhere intact. The dentin surfaces are deeply hollowed around the central reduced cusps, the sculpturing being most pronounced on the buccal sides. The maxilla of BP2/20-41 contains a fourth molar on the left side which, like the LM® adjacent to it, shows an absence of attrition. The other HUMAN SKELETAL MATERIAL FROM CEYLON 169 maxillary molars show a moderate degree of wear, the greatest degree of abrasion occurring on the buccal portions of the occlusal surface. The enamel rim is thin and brittle on the mesial portion of these molars, but thick and robust on the other portions. A similar degree of wear characterizes the mandibular molars, the LM? showing less wear than the remaining molars, the buccal sides being the most abraded and denuded of enamel. The male BP4/8 shows slight wear of the RM2 and medium wear of the RM2 and RM!. The exposed dentin forms kidney-shaped lakes over the occlusal surfaces, and the enamel borders are stout. The RLM of female specimen BP1/6 shows very slight abrasion. The adjacent molars are moderately worn with small pits of exposed dentin on their cusps. The RLM? present kidney-shaped lakes of exposed dentin on the mesio-buccal occlusal surfaces. The direction of the abrasive force is lingual for both molar rows. The mandible of this specimen reflects the maxilla in regard to attrition, save that the RLM? exhibits pronounced abrasion. The maxillary and mandibular molars of BPz2/21 show negligible wear, save for the first molars which have developed small pits of exposed dentin on the apices of their cusps. Direction of this slight abrasion appears to be buccal. As the third molars are in process of eruption, their elaborately crenulated occlusal surfaces are untouched by attrition. The dentition of BP3/27-34 is heavily abraded. The third molars have been lost ante-mortem, and the first and second molars exhibit occlusal surfaces denuded of enamel, save for an island of enamel forming a central cusp. Greatest wear occurs on the lingual surfaces. The enamel rims are narrow and brittle and enclose poorly sculptured lakes of the dis- coloured dentin. As with the molars, the premolars of females are more frequently abraded than are the premolars of males. For specimen BP2/17 the premolars are very slightly worn, although attrition of the buccal cusps has resulted in exposure of dentin on the RLPM+4 and RPM2. The mandibular premolars of BP2/17i are heavily abraded, the RPM2 exhibiting the greatest degree of attrition. The exposed dentin of these teeth lacks the central reduced cusps found on the molars. Orientation of the abra- sive force is buccal. The enamel borders of these premolars are rounded in contrast to the sharpened ridges that constitute the enamel borders of the molars. Maxillary and mandibular premolars of BP2/20—41 show a moderate degree of attrition. The force of abrasion appears not to favour either side of the maxillary or mandibular premolar row, but the distal portions have maintained the greater parts of the enamel, particularly for the RLPM+. The maxillary premolars of BP4/8 show medium attrition, the RPM? having the least amount of dentin visible. Wear is greatest for both RPM? and RPM2 on the labial portion, although the lingual cusps are marked by several small pits and lines of exposed dentin. All maxillary premolars of female BP1/6 show a moderate degree of wear, except for LPM? which is heavily worn. Unlike the other premolars whose bays of exposed dentin are limited to buccal and lingual areas separated by mesio-distal ridges of enamel, the occlusal surface of LPM? is denuded of all enamel and its rim is sharp andnarrow. This tooth is separated by a gap of 10 mm. from LPM? which is due in part to post-mortem distortion of the maxilla, but which perhaps reflects as well an 170 HUMAN SKELETAL MATERIAL FROM CEYLON abnormal orientation of this tooth in the molar-premolar row. Unfortunately the mandibular premolars of BP1/6 are missing. Direction of wear appears to have had no preference for either side of the premolar rows. Specimen BP2/21 exhibits very slight attrition of both maxillary and mandibular premolars, although the RLPM2 shows somewhat more abrasion than the others. Pits of dentin on the occlusal surface are numerous and are greatest in frequency on the labial portions of the premolars. In contrast are the maxillary and mandibular premolars of BP3/27-34, all of which are heavily worn, particularly on their labial sides where the enamel borders of the RLPM2 have broken down. Vestiges of the lingual cusps are visible on the RLPM!. Canines show slight attrition in specimen BP2/17, the apices of the cusp being marked by very small pits of dentin. Those of BP2/20—41 are moderately worn with the greatest degree of attrition on the labial aspects. The RC of BP4/8 is characterized by a lingual and labial pit on its occlusal surface, but these are reduced in size. The maxillary LC of BP1/6 has moderate wear on its labial aspect. Those of BP2/21 are marked by numerous pits of exposed dentin but the cusps are not reduced in size. The canines of BP3/27—34 are very abraded in the maxilla and in the RC of the mandible, although the enamel borders are well preserved and rounded in form. Incisors are moderately to heavily abraded for allspecimens. The RLI2 of BP2/17 present a bar-like incisal margin of exposed dentin, portions of which are pitted. The more pronounced attrition of the LI? of BP2/20—41 shows a triangular mass of exposed dentin on the cutting surface, the enamel rim being reduced in size and very thin along its mesial aspect. The lingual side is worn to the base of the crown. The RI2 and RI* of BP1/6 are moderately worn, the central incisor showing slightly greater wear. The exposed dentin is rectangular in form. The maxillary and mandibular incisors of BP2/21 are heavily abraded with their rectangular cutting surfaces. Due to post-mortem damage to the mandible, the RI? and LI? have become dislocated so that the RI2 has been pushed to the left side where it is adjacent to the LI?. The cutting edges are rectangular. Pronounced wear characterizes the maxillary LI2 which is worn to a triangular-shaped stub of exposed dentin. Deraniyagala (1958a@ : 255, 260) remarks that the lower teeth are always more heavily abraded than the upper teeth. This is the tendency for the teeth taken as a whole, although for the few incisors available in the sample the upper teeth appear to be more heavily abraded. However, if the dentitions are examined according to specimen, only two, BP1/6 and BP2/21, illustrate Deraniyagala’s claim. The other two specimens, BP2/20 and BP3/27-34, show an equal degree of wear for teeth in both jaws. There is a higher incidence of mandibular molar wear among males than females, but maxillary molar wear shows no sexual dichotomy. Females show a higher incidence of premolar attrition and maxillary canine attrition, but males have a higher incidence of wear among mandibular canines and maxillary incisors. No comparison by sex is possible for the lower incisors since these are lacking for male specimens. Shovel shaping of the incisors is lacking in the male dentitions available for HUMAN SKELETAL MATERIAL FROM CEYLON 171 examination, but is present in the RI® and RI* of female BP1/6 and in the LI? and LI? and RLI? and RI! of the female BP2/21. It is expressed to a pronounced degree only in the maxillary incisors of BP2/21, the size of the shovel shaped depres- sions of this specimen’s mandibular incisors being slightly developed. Development is medium in the BPr/6 incisors. The single case of a lingual tubercle occurs in the LI? and RLI? and RI! of BP2/21. The tubercle is prominent in the mandible, but small in the maxilla of this female specimen although a foramen cecum incisal to the tubercle marks this feature with clarity. Some features that frequently appear in the dentitions of other Recent hominid populations are not observable in the Balangoda series. Caries and dental abcesses are not observed. Ante-mortem tooth loss is suggested in the case of a single specimen—the RLM3 of female BP3/27—34. Thealveoli inthis case appear completely absorbed. The maxillary third molars of this specimen appear to have been lost post-mortem, the RM® showing a small alveolus with unabsorbed borders. Super- numerary teeth are found in the case of the LM* of the maxilla of BP2/20-41. The size of this distomolar tooth is reduced, and its form is of the compressional or rectangular type. It was never of any functional benefit. Buccomesial to the RM2 of specimen BP2/21 is a peg-shaped paramolar which is encased in enamel and is half as high as the adjacent teeth. Suppression is not a feature of the dentitions of the series. The upper and lower third molars of specimen BP2/21 were almost completely erupted at the time of death of this individual. Deraniyagala (1958a : 255) remarks that bite is of the edge-to-edge type. This cannot be ascertained with certainty for any of the specimens, although it may have been a trait for those individuals with heavily worn teeth. Prominent overbite seems to be the case for specimen BP3/27-34 and possibly for BP2/25, although distortion of the skull precludes certainty for the latter. As few of the teeth were isolated from their alveoli, examination of the root systems was approached through study of radiographic plates. The number of roots for the mandibular molars was two for all samples in the series. Maxillary molars, however, exhibit greater variation. The distomolar of BP2/20—41 has two roots although the other molars adjacent to it are triradical. The first maxillary molars of BP4/8 are double-rooted. Premolars are single-rooted in the mandibles of both sexes, but in the maxilla the first premolar is double for females BP1/6 and BP2/21. These double roots are small. The direction of the molar roots within their alveoli shows considerable variation. The roots of the M® of BP2/17 are fused and straight. Its neighbour, the M2, shows fusion of only two of its roots—the mesiobuccal and distobuccal. The M1 has divergent buccal roots with the radical apices oriented toward each other. The mandibular molars of BP2/17i are divergent, those of the M1 curving toward each other. Maxillary molars of BP2/20—41 show divergent roots for the distomolar, fused roots for the M2, and divergent roots for the remaining molars. The mandibular molars all have divergent roots. The medial roots curve distally, and the distal roots are straight. HUMAN SKELETAL MATERIAL FROM CEYLON TABLE 8—DENTAL MEASUREMENTS AND INDICES Maxilla Mandible II >1V > V> 1; the metacarpal formula for the Balangodese is Il > I>I1V> V(?) >I. The metacarpals of both series show curvature in the dorso-ventral plane, but Hill does not regard this as a feature unique among the Veddas, since it is found as well among other Ceylonese groups. The shapes of the carpal bones which are peculiar to the Veddas are also approximated in the hand of the Balangodese, vzz. the dumbbell conformation of the scaphoid as a result of central narrowness, the cuboidal form of the capitate, and the well-defined ridging of the trapezium. The Skeleton of the Lower Extremity. While the Vedda femora are platymeric with an average index of 78-50 for males and 80-40 for females, the Balangodese male BP2/17 is stenomeric and the other specimens of the fossil series, both male and female, are eurymeric for this index. With regard to the Pilastric Index, the male BP4/8 comes nearest the Vedda means of 108-80 for males and 106-90 for females. There is a discrepancy in the size of the femora of BP4/8, however. The other Balangodese have a Pilastric Index which is outside the upper range observable for the Veddas. There is close agreement, however, between the two series as regards the Index of Robusticity, that of Vedda males averaging 11-69, and the females averaging 10-80. Osteological features of greatest similarity for the femora of the two series occur more frequently at the extremities of the bone rather than at the mid-shaft region. To the differences of the shaft noted above should be added the greater torsion of the femur among Veddas. Both groups have very thick shafts and a pronounced development of the pilaster with the lips of the linea aspera separated to a marked degree. A short neck, compressed antero-posteriorly, and supporting a spherical head, is common to the proximal ends of the femora of both series. The greater trochanter is prominent and the trochanteric fossa is deep. The lesser trochanter is large and smooth. A third trochanter is absent. The crista hypotrochanterica is of variable development in the femora of both series. The distal end is marked with a broad intercondylar fossa for both Veddas and Balangodese, and a further example of similarity is afforded by the weak development of the adductor tubercle. Hill notes the great backward extent of the medial condyle and the anterior development of the lateral condyle of the Vedda femur. This trait is reminiscent of the orientation of the condyles of the Balangodese femur. The tibia is mesocnemic for the majority of Vedda and Balangodese specimens, the index for the former series averaging 65-95 for males and 68-70 for females. None of the Balangodese tibias are pathologically affected as are many of the Vedda specimens with consequent distortion of their conformation. Only BP2/17 exhibits anterior—posterior bowing of the shaft to the extent found among the greater number of Vedda tibiae. The lateral prism formed by the transverse section of the Vedda tibia is not observable in the fossil series. Muscularity is moderate for bones of both series. The Balangodese tibia is somewhat more thick and massive. Torsion of the shaft is most prominent among the Veddas. The fibular fragment of BP4/8 resembles the Vedda form in its straightness and triangular transverse section. However its articular extremities are more promin- HUMAN SKELETAL MATERIAL FROM CEYLON 193 ently developed. In muscularity it is more strongly marked, although the Vedda fibula also exhibits some hollowing for the peroneal and tibialis posticus muscles. The tali of the two series are very similar in size and muscular development. The area for the fibular malleolus is extensive. However, the neck of the talus is broader among the Balangodese and the forward extension of the superior articular facet that Hill notes as a common feature of the Vedda talus is not apparent among the fossil specimens. The calcaneus is longer and narrower for the Veddas, but the pronounced curvature of the sustentaculum tali brings it nearer to the form found in the older group. The latter have a cuboid bone that approximates that of the Vedda cuboid in its squat form. The articular facets are prominent, but the Vedda cuboid shows greater convexity of the plantar surface. The Skeleton of the Pelvis. Apart from the greater size of the Balangodese sacrum this bone falls well within the range of variation for other features observable in Vedda sacra. The platyhieric sacrum of BP2/17 can be matched with several Vedda specimens although the average for the series lie between the dolichohieric and sub- platyhieric categories. Closer similarity is seen in the uniform curvature and hyperbasal form of this bone. Four out of the eleven Vedda sacra examined by Hill showed an articulation of three sacral vertebrae with the ilium. The other Vedda sacra have only one or two segments connecting with their ilia. The various sacral aberrations that Hill finds in his Vedda specimens are not present in BP2/17, i.2., Incomplete union of sacral vertebrae, failure of closure of the neural arches, and spina fissa of the terminal segments. The pelvic bones of BP2/17 exhibit a combination of infantile and “ paedo- morphic ’”’ traits which are common for Vedda pelves plus ‘‘ gerontomorphic ”’ features that are characteristic of specimens of Australian aboriginal populations. Of the traits of the latter sort, BP2/17 shows prominent muscular attachments, a thick and massive ilium, and a size that is at the upper limit of the range of variation for the Vedda pelvic dimensions. Like the more gracile innominates of the Veddas, however, the male BP2/17 shows a sciatic notch with a fairly high angle, pronounced torsion between ilium and ischium and a deep hollowing of the gluteal fossa. The ischium of BP2/17 is narrow and long, but the Vedda form is broader and shorter. The Height—Breadth Index for the Veddas measured by Hill is 80-50 for males and 76-60 for females, indices considerably lower than that obtained by the present writer for the Balangodese. The mesopellic category of the Veddas—g5-60 for males and 95-20 for females of Osman Hill’s series—also contrasts with the higher index for the Balangoda pelvis. The neural foramina in both groups are cordiform. THE DentiTION. The mean values for the sizes of molar and premolar teeth of Balangodese and Veddas show that the higher values occur in the former series. Where the Balangodese means are lower—as with the bucco-lingual diameter of the RM2 and mesio-distal diameters of the RM2, LPM1 and RLM1—the differences are very slight. Only in the latter case are the grinding teeth of both sides of the mouth smaller in size. The Balangodese molars and premolars are larger bucco-lingually than those of the Vedda series, hence their greater Length-Breadth Index. The anterior teeth in the two series show a significant difference from this molar-premolar 194 HUMAN SKELETAL MATERIAL FROM CEYLON size relationship. The upper canines of the Balangodese are smaller than those of the Vedda, but the labio-lingual diameter retains its relatively greater size over the mesio-distal length, and the indices are not as low as they are in the Vedda ratios. The mandibular canines reflect this situation. The mesio-distal length is greater for the maxillary and mandibular lateral incisors of the Vedda series, but the labio- lingual diameter shows no difference, save for the RI* which has a higher mean value among the Balangodese. The Length—Breadth Index remains higher for the fossil series, however. Their mean index for the upper central incisors is less than that for the Veddas, due to their smaller labio-lingual diameter. The mesio-distal diameter is larger for the RI+ but the same for the LI4. The mandibular central incisors cannot be compared due to the inadequate size of the sample. The Balango- dese LI! shows the higher mean labio-lingual and mesio-distal values, but the index is below that of the Vedda series. None of the specimens of the Vedda series exhibits on the occlusal surface of the lower molars the rectangular form, such as that found in BP2/21. However, the maxillary teeth in both series show a low incidence of occlusal rectangularity—a deviation from the common rounded molar form. The triangular premolar form is absent in the maxilla of the fossil series but present among the Veddas. The man- dibles of both series have triangular premolars, but the rectangular form predominated in the Balangoda series, the round form in the Vedda. Canines are only triangular in the maxillae of the Veddas but BP2/20—41 has a RLC which is rectangular. This same specimen has a RLC of like pattern. This form is not absent in the Vedda mandibular canines although the common pattern is triangular. Some Vedda males have triangular upper incisors, but the only form found in the fossil series is rectangu- lar. This is the case for upper and lower dentitions. The Balangodese exhibit a greater variation in cusp pattern than do the Veddas, who offer no examples of molars with 5 or 6 cusps in the maxillary dentition. Cases of tricuspid molars are absent in the mandibles of both series, and 6 cusps are found in this region only among the Balangodese—the LM® of BP3/27-34. This variation in the molar cusp patterns, particularly in the maxilla, is a peculiar feature of the Balangodese dentition. It is further exemplified in the 4-cusp pattern of the RLPM of BP2/20-41. The buccal cusp(s) of the premolars is consistently the higher in both series. The Veddas offer no cases of Carabelli’s cusp nor the presence of other accessory cusps, but both are to be found in the Balangoda series. Crenulation of the third molar appears to be a common feature of both series, although it is found exclusively in the females of the Vedda group. The strict correlation of cusp number and groove pattern which is universal in the latter series does not hold true with the Balangodese. Furthermore, there are no cases of Y4 or +5 groove patterns in the Vedda series, the 4 cusp and 5 cusp patterns relating to the +4 and Y5 groove pat- terns, respectively. In both series the maxillary cusp pattern for molars is +4. The greater degree of attrition of the molars and premolars of male specimens, as is the situation in the Vedda series, is not encountered in the fossil series where the degree of wear is more pronounced among females. More of the posterior teeth of the Balangodese exhibit pronounced abrasion than do the corresponding teeth of 195 HUMAN SKELETAL MATERIAL FROM CEYLON W 6-€ 30 50 W 6-76 29 ok W 0-Sg W S-4g W $-18 W 8-89 ia ate 50 W 6-06 ie W0-z9 : 2 ote W z-96 W0-gf yw 6.65 = W $-98 W S-£9 W 9-68 xaputr xopur xepur yipeeiq y} ue] yy peeiq 4qseq-3ysreqy -7 48104 o1jemiseiq o1jeuIs91q = xa}I0A -1e[NOLINY -IepMoNy -1eynowMy -r1epnony W 9-£9 19-9 W 1-9 W 2-19 a $-£9 W 0-19 W 0-99 xopur y3sue] “ysIey X9}IOA W 211 W gor aA gor W £11 aA gor W OIL W 4z1 yys1eq oryeuse1q, -Ie[nony -1epnouny W Zfr qys1eq X9}19A W $-96 W 1-02 W ofr W $-2Z W 0-211 or W 821 W 8:44 W 0-oo1 W 9-2 W fr W 9-24 P on are W+-&Z €.601 W 4-42 Wor Wrest W I-£or W9-£4 Wier Wwe-1Z ae ‘ - W £-69 ie ate a W z-9Z a0 De W 611 as W9-01r WhlZ Worr Wo-0f W z-Zor W 9-02 W 6fr W 0-99 J 1-96 a £-92 af Ofr a 74 W 7-66 W 6-SZ W4Er ywo-94 Wz-for Wr-69 ef W 8:49 W S:+6 W 8:42 W gtr WI-42 aA F-96 a b-1Z a rer 0-2 pie 19-12 of Ser qr-SZ at W £-12 W ofr W z-bZ ze 9-12 a Ser Raa 4 <6 W814 WET Wsg-0L 8-96 Ab. SZ wl QzI ee 74 W 9-66 W £-2 Wier wo-rZ a z-+6 =i a bzr J 0-1g W 4-S6 ne Wott wr-f aA b-+6 a r-SZ J gir 19-62 On ne 38 W $-zZ W 8-66 Wo-1Z WW Of: W ?-12 Wf 0-011 Raa 74 a 6z1 a £.02 W €-10r W 6-22 W ofr W 0-22 oe Ai-zZ a 621 A z-oL AG W9-& WoO Ws-04 “s ai-SZ a 421 AS. oe W 6-€2 W SEr W $-2Z A 9-42 a ger A4-2L W 9-2 W cer W 2-12 . oa a r-€Z con m6 19-22 W S-49 WSErI ws: xopur xapur yqGsI0y xepur yy pesiq 3 8u2] OeMIseig |erueID yy s184 yy s1oy -uolseg orjemiseiq o1emse1q -uoIseg -uoIseg W Str W Or W €rr W Itt W 82r W 421 W 9€r W fr W ztr JW ofr wf ver W gtr W zr W Pbr A gf ql z£r W 9€1 a 421 W z£r a 9er W Ifr wf rer W 9&r a Szr W Iz W ofr J 411 W 821 wl S21 W Ifr a 9z1 W zr a fer W 921 J ofr wy €€r W 4hr yypeeiq W 8st W 461 a SLi W gt W €gr W 48r a Per a bdr W 7zgr a S21 W 481 J O41 W 441 A o9t W vor a 4S1 W 491 W gr a 491 IW 941 wt OL1 W Sgr a 691 W €gr A 691 W 841 a 421 A €gr W 00z qq sue, yetueiy [eIueID zz61 sloqnqg Wi 0961 T[aMyyoIg WI €-6r Ajrapun\ Wi gr6r yyIWIS Wt 6161 Py WI 6161 eM WI CrOr ITey[NINyy 2 aareyy We e€6r sroqsly (é)IW I qre6r eqn = = TTeMeS Wir 6761 wD = Teves Wt DIE6I eq % Tlees W £ a oz gz6r youl W +9 o£61 ueuayonz WI og6r IeyIes WS S61 euopeiemedel pue wesnuueyq) at 9 Zz61 yueIOW W 09 a SZ £761 pueIOyy W 791 J of ob6r Ajquieyy WW zr ad gt IZ6I UeATT[NS W ez at S61 eyes Wr bEGI-L£61 nseg ar az €€61-cE6r nseg WS aZ ZOgI suIseles W £1 ar Z6gI suISeIeS W 6 a gt rh6r [IH W Pr ae b€-Lz/Eqq ay 1z/ztqq rE 41/taq eyep xas jo 4q 901no0g Aguonbary LTAVA IVINVUD AHL ‘VIVC OINLANOdOMHINY AAILVUVdWO) II ATAV], 4efPeM qeIn JOTIO resjeL JONTEIS vueteg feuysuey sundures Cah WAS tA TeN orep-ofuayoy asoury9 OL0}sTyarlg I injjeueysipy seumyeig WeIEMST}OYNITT, suerueuse |, suelerjsny suendeg asourmepuy sTeN sueraI9Q sepunyy s[rme yy, uopéag asayeyulS Seppo asaposurleg dnoi3 10 uamitoads HUMAN SKELETAL MATERIAL FROM CEYLON 196 IW 16 aw ze1 Wy 2z WJ z6 \aeee W 6z W zr W '6 W OZ1 W ze IN zP W 68 W 26 A git A 92% HE A £8 W O11 W of W EE W 68 a eer Wie it 4] IOI a gil Wie a 9f 66 o. a of oe ee W €z1 5 W O11 WW 8tt W 96 sndioj snuiey snureyy IojyoUeIp 24} fo 3y} Fo oy} Fo gjsue yyproIq Y}proiq UI, UINUITUTY WINUITxeyy A 9% W 92 q Lz WW 6z Wit WW IE A 9% W Zz A 82 Zit a We WN 92 WW 82 Ia}oweIp 4 911 W Vii 4 zii W ZiI W 911 4 66 W 111 BL itt wy Sor W SII WW Sit yypeesq WI AS: W 9: WZ: WZ: A g6 He: Wor Wo: A £6 A 8: J 66 18 W 62 WW 901 ne WI- -Lo ¢l €g 88 gor +6 L6 gil - For €11 Le absioc)| Xopur -snUWIey SuIpusose Surpusose etuosig yetsAydurAs sejApuoorg yersAydurAs 9}e[eqG -ojApuo) HIPIGNV] GNV ALVIVd AHL ‘VWLY(] OIULANOdOUHINY AAILVAVdNO’) (panuyuor) Il AAV] ssourulepuy a ore os BIG Je WW ob W oS w% sueoIO q LE a OV W It W 6+ ors sepuny sprue W &S W 79 Be uojsa9 JN 19 W 9S a4 oseTeYUlsS As or W ¢r W oS as Seppe A “£9 a €S ve—-Lz/ faq A $9 Az 12/74 - ; 9/1da wy ol 0z/z7dq 41/7da WW 69 W 19 L1/z~qq sseposureg wypeerq = yy Suey oze[ed ayered [eurIeyxXy + [eurIeyxXq 197 HUMAN SKELETAL MATERIAL FROM CEYLON WE-€2 Wr aZ-z7g wor J -1g -. i W st A6-gL ge Wo6-gZ wor H6-h’g weet W£:-98 WO6Ef aAi-6g Se Wo-6g W9E aA1ir-46 Whe WiI-Sg WZE aAZ-Sg ASE Wo-zg W st A4-98 J9F W4-98 West A1I-6g Zoe W9:€8 wWZE aAZ-€L Zoe Wwo-rZ weet o-SL Jot Wz-06 wir xepul yyprezq TeH#IqIO [e31qI0 W £& af EE W IE ze W EE ce IW E€ LIE W zt Wit W If of WW IE a @ W ze a ze W IE Wie W ze& y ob A 9f W ZE yysIOY 1e31q10 W 8-8 4 o-0S 0-66 W 1-6z 7 Z.G¢ W9-tS 19-28 W 2-16 wf O-zS W 0-1 A 6-09 Ww Z-4+S a S-6S W £-zS w Z-€¢ W 9-0 A &-Ls W z:-bS Wy 7-7z¢ xopul [eseNn W £z A gh Uypeeiq FYsIOY jesen [esen WEe-Z¢ W6S yb-€6 J rir oP re ee oe 4 9-6 Wy 6S W9-6h Wz9 WAi-iS Wz Wz-0S W99 H£-69 4 £9 W I-69 IW 99 A 76 a A 9S W 66 W 19 ee wy 0-zS o. W 6-76 WEIl WO9-gh WI9 HE-Sg 766 wqzgr gos WZ-S8 WS80Ol WO6.gr W 9 Lo-rS g So Wz:88 WIIl Wwz-zS WwW So A 09 WZ-68 Whirl wz-€S Ww Zo A464 yoo wzi1-6b os Wr-Sg Wor wo-6rh woo Wg-16 WYior Wigs ho oe ne we 09 xopur yysIey xepur = 4ysrey Terory = yeroey =yeroey = [PLO eF Te1ol) «=Teyoy, «=toddq ~=s toddq W 68 W 7Zzl WY zz W IEL WY 71 W ££1 A O11 W 8zi a 411 W zr W 9eI A git W $zI A git W SzI a bir W 9z1 + OII W Iz1 wf OIL a izi JojoweIp oryeurosAzig aOVY AHL, ‘VLVC OlYLANOdOMHINY AALLVAVANO’) (panuyuor) II ATAV], W +6 4 56 06 W +6 16 W 96 A 06 W £6 A 68 W z6 Wy 16 W 16 W 16 AH 88 W 06 W £6 W +6 AH 88 W 16 4 16 wf O11 WW 411 JojowWeIp yequoIy UINUITUTT Lseuryerg WeLIVMSTZONY OILY SUeIUeUISe TL, suelTeljsny suendeg osourule puy e[PIN suevoIO, sepunyyy s[ruey, uopAe9 asoTeyUursS SCPPeA re—-Lz/€aq Sz/tdq 12/74 8/had L£1/7dq asoposurreg HUMAN SKELETAL MATERIAL FROM CEYLON 198 W I-22 W $-zg W ¢- 9 10-8 W 1-92 W °:-Zg XOpuI Teytq10 W oF Ww ZE a EF W rr W oF W EE J O€ W z& WIE W EE W If a Ee W EE W SE qipeeiq jYsIoy TeH1q10 TeUqIO WO: W €- W 6- IN 9: W Z WS: W 6: W 0O- a v- WW 9: Wo: 09 99 1c 19 -9S €v 6r 1¢ or LY LA xopur Tesen W of WN 82 W Zz W Sz W Sz W £z W £z W zz Sz W Sz IW Zz wpeerq [esen W os WW zv W zS W Sb W bs W 6+ W 9v is W Ss W oS VYysIoy Tesen W --26 W 6-88 1-26 W 9-06 xopul [erry 1e}0L, W Off 4 vir W OzI VYySIOY yeroey [e}0L W ?- WoO: W €- W 0: W 6: A O- -9¢ WoO WW 9- AS bS 19 ol 09 €S 9S bs Xopur yerory todd q, W €Z W ort Ww SS ad W vZ IN 9£1 W £9 W 82 W 91 WN git W Sb W rer W 18 W 9It W 7zZ W OZ1 BS W Zz1 me 4 Szi W ¢Z W ZI W oZ W 8zI yysrey = JoyoueTp Teroey oneurosAztg todd q adOVY AHL :-VLVC OINLAWOdOUHINY AAILVAVdNOD) (panuyuor) II aATaV J, W 66 W 86 W IO1 W 66 W 86 W 06 W Sor W 16 W £6 W S6 A 48 W 16 W 06 Iojourerp [eq uoIf UWINUITUTIA] yelper Yen TOTO resjel FOHTCIS eurkeg feuysue Ty sundures q ueryeyy IeN orep-ofusyoy, asoulyg OLIOFSIYOIg I InjfeueyorIpy 199 HUMAN SKELETAL MATERIAL FROM CEYLON sndio9 SnUure yy oy} Fo asue ueoyl A 82 SNUILY {NUE rT Surpusose Ssurpusose ayy Jo ayy Jo yypeerq —-Ypeerq UINUNUI, «Win wrxeyy W ¢-S6 a W gor a SII WW Iz1 A 98 A &z Wf OIr Joyoureip = soy Yypeeq 4} 3U9] jetuosig jersAydurks re,Apuoorg jersAydurAs -ojApuog AIAMIGNVJ{ GNV ALVIVG AHL ‘VLV(| OIMLANOdOUHINY AAILVUVAWOD (panuyuor) II ATAV], W 6: W 9: WZ- Wo: Wz: IN €- 1o- WZ: WW 9: a 6- IN 8: We: gl €8 L9 vg 6L 68 ro) 66 771 LL 1g cL xopur 97e[ed Ww ZE WZ? W zr IN zv W oF wy OF TW oS W £9 W rr mols W Iv WW oF YFPecsq ozered [eule}xy [euioixy wos W oS W €S yy sue] a7eyed yen IOTIOST Tes[eL qd UeIS CI IeN orep-olusyoyt asoulyy OLIO}SIYIIG I InjjeueysIpy UWIPIVMSIJOYYNIALT, LiISeUYyeIg, suerlyerjsny suendeg 18 GEOL. II, 4 200 HUMAN SKELETAL MATERIAL FROM CEYLON the Vedda series. The condition of the canines is similar in both groups, the lower canines being the more heavily worn. There are no incisor teeth in the Balangoda series whose abrasion might be described as “ slight ”’ or “‘ sub-medium ”’ as is the case for the Veddas. The condition of moderate to pronounced wear is the norm for this fossil series. Only among Vedda females do the lower molars appear more heavily worn: among the Balangodese the more severe forms of mandibular molar abrasion are represented by the males. The Vedda male specimens show greater maxillary molar abrasion, but for the Balangodese this sort of sexual differentiation does not exist. Both groups, however, show greater wear of the upper incisors among the males. The assymetrical wear pattern observable in the right and left sides of the dentitions of certain Vedda specimens is not apparent in the fossil series. The occurrence of shovel-shaped incisors, while present only in the dentitions of the females of the Balangoda series, is found in both sexes of the Veddas. Further- more the development of this feature is considerably more prominent among the Veddas. This trait appears in the lower dentition of a single Balangodese female— BP2/21. Lingual tubercles are absent from the anterior teeth of males of both series. Among females their presence may or may not be coincident with shovel-shaping. With respect to dental anomalies the Balangodese and Veddas exhibit some striking differences. Caries and evidence of abscess are absent in the former but not infrequent in the latter population. The frequency of ante-mortem tooth loss is high for Veddas, negligible in the fossil series. Both populations show cases of supernumerary teeth. Among the Balangodese these are found in the molar portions of both jaws and for the Veddas there is a single case of a supernumerary tooth in TABLE 12 COMPARATIVE ANTHROPOMETRIC DATA: THE MAXILLARY MOLAR DENTITION M! M2 M3 Balangodese BP2/17 a6 LO) 12) Li42) aoO-5 12 S1r4-2 TO) Serreaeun5\-0 BP2/20 aie LON lt eee LOLON EOE 52 seat Att 56 oe BP4/8 He LO) LL LiOwOe Oe PLT ——122'-/2) 5 10) eet h mer 277 BP1/6 56 LOS 12. LIA2 TO) a2). 20-07 oO - Serr her 0 BP2/21 de II 12) 100-0) O) 13, 130-0) 585) 131-9 BP3/27 ae @) ii I22°2 10 10:5 105-0 3 : Veddas te ess 8:5 12 150°0 I0 10:6 10774 I0 10°9 109-4 Mundas 20 a0 9:6 10:2 106:2 8:5 10:2 120°5 Oreans 6% a 9°2 10:7 116:2 8-7 10-7 122°8 Male ae a Io 0 100°0 8 10° 25-0 : Australians at a II°4 12:8 105°2 10:9 13:I I20°I I0 12-3 123-70 Sampung 36 5° Ti 2 2-8) L422 LOO 2/6 S174 ONL 2 A aeIe2 ALG Gua Cha ae Brothwell 1960 10-3 I2°I I17°4 10-2 12:2 109'6 9-3 12:0 119-0 Talgai ane Campbell 1925 12-6 13:I 103°9 I1:3 13°3 117°6 We Keilor .. Adam 1943 Tek SIZ} AXON) WYO) 163} SHE} Gp PS URBOF Niah ee IO°2 12:3 120°5 10°7 13 I121°4 I1-5 12:4 107°8 HUMAN SKELETAL MATERIAL FROM CEYLON 201 the mandibular incisive region. Suppression of teeth is not a feature in either series. Crowding is found only in the third mandibular molars in the prehistoric group. Crowding is not observable among the Veddas who show, however, a high incidence of maladjustment for other parts of the dentition. The first molar is more commonly the larger in the posterior dental rows of both series, but the presence of the third molar as the one of greater size is found more frequently in the earlier series, particularly in the mandibles. There are no cases of the first molar being the larger in the lower jaws of the Balangoda series, although the Vedda series affords several such cases. Where the third molars are absent, the second molar is larger than the first. The overbite type of occlusion, which is common for jaws of the Vedda series, appears to have been present also among the Balangodese, although only BP2/25 and BP3/27-34 afford evidence of this. IV DISCUSSION From the anthropometric and comparative analyses of the human remains from Bellan Bandi Palassa originate certain problems of interpretation which require comment. Regarding the question of the phenotypic affinities of the Balangodese, Deraniyagala has entertained several views. His earliest opinion, conceived at a time when the only prehistoric Ceylonese hominid specimen was the one from the site of Batadomba lena, was that the Veddas were a composite race of several different population elements from India which had introduced the Neolithic culture to the Island. ‘‘ The tendency of the Veddas to throw back to two physical types differing from other races in Ceylon ...” suggested this hypothesis (Anonymous 1941 : 354). Later he remarked that “... the supposedly autochthonous Vaddha of Ceylon possibly carries some proportion of the blood of Balangoda man ...”’ (Deraniya- gala 1943a@: 112), but following the discovery of the specimens from the sites of Ravan alla and Telulla he wrote, “This race may be termed ‘ Proto-Vaddoid’. The so-called Vaddahs who do not differ in culture, religion, and language from the forest villagers can only be regarded as hybrids between this extinct autochthonous Stone Age race and the more modern metal using ones of Ceylon. Supporting this view is the fact that the so-called Vaddahs display a variety of physical types that have puzzled the anthropologists who attempted to study them under the impression that they were a distinct race’ (Deraniyagala 1955a : 40). As to the Sinhalese he conjectures, “The blood of Homo sapiens balangodensis however exists in almost equal intensity in the colonies of the so-called Vaddahs of today, and among the Sinhalese so that in most cases it would be impossible to distinguish a so-called Vaddah from a Sinhalese if both were clothed alike and placed in the same village : the only recognizable autochthonous race in Ceylon is the extinct one in a lithic culture phase ...”’ (Deraniyagala 19550 : 301-302). 202 HUMAN SKELETAL MATERIAL FROM CEYLON Since the appearance of the fossils from Bellan Bandi Palassa, he has insisted that the Balangodese show a combination of Neanderthaloid and Australoid traits that he calls “‘ Proto-Australoid ”’, plus a Negroid element added at a later time (Deraniya- gala 1957a:3, 4). He suggests that the Balangodese replaced the people of the “ Ratnapura ”’ culture who were the true autochthones of Ceylon and that the Balangodese continued to practice their distinctive culture until the fifth century A.D. The Veddas are, he concludes, “ a degenerate mixed population ”’ of Balango- dese and Sinhalese racial elements (Deraniyagala 19584 : 258, 1960a : 96, 108). The only other published discussion of the racial affinities of the Balangodese known to the present writer is Coon’s (1962 : 424-425). This author agrees with Deraniyagala that there is a Negroid element in the series, but believes that the dominant strain is “ Caucasoid ”’ rather than “ Australoid”’ ; although he hastens to add that the presence of some “ Australoid ”’ features should not be surprising considering the location of the Island of Ceylon just south of a “‘ Caucasoid—Australoid zone of contact ’’. He appears to base this opinion upon the presence of reduced frontal brow ridges, a sharply pointed chin, the lack of a prominent nuchal crest and what he regards as a narrow and prominent nasal structure which are represented in the single specimen now on loan at the American Museum of Natural History in New York (Specimen T-24—B or BP3/15b). Although Dr. Coon saw six of the Balangodese specimens in Ceylon in 1957, these had not yet been cleaned, repaired or described. While the present writer, basing his analysis upon the restored specimens, agrees that Balangoda Man did not differ subspecifically from the living Veddas, as Coon states, the complete series supports an interpretation that must take into consideration the non-Caucasoid phenotypic traits of the Balangodese. Since Coon regards the living Veddas as “ Caucasoids ”’ along with their Sinhalese neighbours, it is not surprising that he has emphasized the racial criteria of this phenotype in his interpretation of the Balangodese population. Emerging from this very general comparative analysis of the Balangoda phenotype with the phenotypes of other populations and individual specimens, both prehistoric and contemporary, are two major considerations: (1) the high frequency of similar morphological traits shared by the Balangodese and the Veddas which suggests a positive genetic affinity between them; (2) the number of unique morphological traits that characterize the Balangodese which are among those traits recognized by various students of the Veddas as the phenotypic hallmarks of half-breeds or “ Vedda sub-types”’. Hill (1941 : 134) lists the main effects of miscegenation in the Vedda population as the tendency to increase of the cranial capacity above 1,300 cc., an increase in the relative height and/or breadth of the cranial vault with a consequent lessening of the steepness of the lateral walls and the formation of reduced temporal fossae, an elongation of the face, especially in the increased depth of the mandible, an absence of prognathism, a constriction of the orbits and rounding off of the outlet of the orbital fossae, a narrowing of the spheno-maxillary fissure, a reduction in the extent of the frontal bone on the inner wall of the orbit, and the presence of leptorrhiny which is accompanied by longer nasal bones, an elevated nasion and an oxycraspedote apertura pyriformis. In the majority of cases Hill can confirm the mixed HUMAN SKELETAL MATERIAL FROM CEYLON 203 ancestry of certain Vedda crania by facts relating to their history. Indeed the proportion of these atypical crania is greatest for those collected since 1886. How- ever, Hill is cautious in assigning to these crania on the basis of his observations alone any traits that would identify the non-Vedda phenotype per se, i.e., absolute traits indicative of admixture from Sinhalese, Tamils, Malays, etc. Nor does he regard ail crania well outside the modal values for the Vedda phenotype as specimens which are non-Vedda in pedigree. Rather, he notes the presence of a third type of cranium which is less typical than his “ classic Vedda”’ type but which he cannot place with the “ half-breed Veddas’’’. He suggests that these individuals may either be the result of remote miscegenation followed by a later addition of Vedda genes from unadulterated gene pools, or else representatives from within the range of variation of pure Veddas. The skulls of the second type, he notes, exhibit a number of traits described as “ primitive, even simian”’ (Hill 1941 : 124, 125, 134). This is the most recent of several schemes to account for physical subtypes among the Veddas. That of von Eickstedt (19272) allows for five types : (1) the “ Veddoid ” which corresponds to the conventional “‘ pure Vedda”’ type living the old way of life; (2) the “‘ Singhalesoid”’ or acculturated Vedda living in villages; (3) the “ Mongoloid ” type which shows admixture with Malays ; (4) the “ pseudo-Austra- loid ”’ whose criterion is a heavier facial beard than that of the Veddoid type ; (5) the Coast Vedda who has become mixed with Tamils. Two or three subtypes were mentioned by Topinard (1885) after an examination of twelve Vedda crania at the Royal College of Surgeons. Hocart (1931 : 5) suggests that there is an element in Ceylon’s population that has not as yet been identified, and Raghaven (1953 : 51), seeing Mediterranean and Australoid elements at opposite ends of the range of variation of Vedda morphology, finds a third type. This he claims is shorter in stature than the Australoid and Mediterranean and has been called Negritoid. This observation that certain physical features found in the Balangodese are among those features that anthropologists have held to be the criteria of Vedda crosses or sub-types does not weaken the claim that considerable miscegenation of the Veddas with the Sinhalese, Tamil and other ethnic groups of Ceylon has trans- pired : rather, it attempts to explain how those physical features which cannot be attributed to the modern non-Vedda populations happen to be manifested in certain proportions among the Veddas of the present day. That the presence of relatively compressed temporal fossae and the absence of total facial prognathism in Vedda populations are related to their recent genetic crossings cannot be doubted. But other features—leptoprosopy, hypsicrany, greater thickness of the cranial bones, prominent mastoid development, the peculiar morphology of the mandible—are examples of physical features shared by both the Balangodese and the Veddas. On the basis of the Balangoda specimen T-—24-B the presence cf leptorrhiny in the Balangodese gene pool cannot be excluded, although specimen BP4/8 indicates that platyrrhiny was also a feature of the group. In addition to these traits which have been selected as evidence of Vedda miscegenation or penetrance of archaic sub-types are a number of others which the Veddas share with the Balangodese but which have a very low frequency among any of the Ceylonese populations today. The most 204 HUMAN SKELETAL MATERIAL FROM CEYLON striking of these are the similarities in cranial conformation, prominence of supra- orbital tori, the depth of the nasal notch, platyrrhiny, chaemaeconchy and certain similarities of the post-cranial skeleton. Of those physical traits that distinguish Balangodese from Veddas as well as from other Ceylonese, the majority appear in the post-cranial skeleton. Apart from their greater lengths, the long bones are more robust and lack the bowing and torsion that characterizes those of the Veddas. Some cranial distinctions are reflected in their greater palate dimensions, pronounced mandibular muscularity and a dentition in which the molars are larger in size and exhibit cusp patterns of a more complex development. These would appear to be physical characteristics that were never manifested in the Vedda phenotype. But while the Balangodese may share a similar genetic background with the Veddas, they are “ pre-Vedda ”’ primarily in the chronological sense of that term. Genetically both Balangodese and Veddas appear to have been recipients of a common gene pool in the past. Affinities with other phenotypic groups, living and prehistoric, cannot be questioned, as the comparative Tables are able to illustrate to a certain degree. But of all the populations with which the Balangodese were compared by the present describer, none approached them in number and significance of mor- phological similarities to the degree represented by the Veddas of the historic period of Ceylon. Historically the Veddas have been treated as though they were a homogeneous racial or sub-racial entity which the physical anthropologist could clearly distinguish from other ethnic groups living in Ceylon. However a little investigation into the problem of how the Veddas have been biologically and culturally defined indicates that this subject has been the focus of considerable debate, as attested by the fact that certain writers have declared the Veddas to be extinct while their contemporaries have estimated that they number in the thousands. It is suggested here that the matter of how the concept of the Veddas as a distinctive phenotypic pattern within Ceylon came to be is deserving of some careful research in the light of the new osteo- logical evidence from Bellan Bandi Palassa. Conceivably such a programme of research could be approached along three avenues of investigation : (1) the problem of how the term “ Veddas”’ has been employed and the determination of what this means when applied to the population as a whole as well as to individuals within the population so defined, (2) the problem of how the Veddas have been described in scientific and popular literature and in oral traditions, (3) the analysis of the various interpretations that have been made by those writers concerned with the Veddas when applying their data to the wider fields of human evolution and cultural be- haviour. Regarding the first problem, it must be understood that the concept of the Veddas as a relict population with ancient indigenous roots in Ceylon has been a basic assumption behind all definitions of their culture and phenotype. For the Sinhalese, Tamils, Malays, Chinese, Moormen, Europeans and other ethnic groups of the Island there are historical accounts of their colonization, but for the Veddas (and Balango- dese) there is no historical documentation of their longer habitation here. Their past must be reconstructed from prehistoric archaeology, native Sinhalese and Tamil HUMAN SKELETAL MATERIAL FROM CEYLON 205 chronicles, and anthropometric data. The problem is further complicated by the fact that the term Vedda has become defined not only as a distinct physical type but on the basis of sociological criteria as well. By the close of the 19th Century three concepts regarding the physical morphology of the Veddas had been put forward (x) the Veddas constituted a single homogeneous racial population morphologically distinct from the macro-population of Ceylon ; (2) the Veddas were racially hetero- geneous save for “ true’’, ie., ““ pure Veddas’’, who followed a primitive hunting- gathering economy ; (3) the “ true Veddas’’, whatever their economic status, were definable in terms of one or more osteological specimens proclaimed as typical of the Vedda phenotype. An additional concept has been offered by Hill (1945: 202-203) who considers the geographical area rather than the degree of acculturation to be the vital factor in the recognition of a physical type definable as Vedda. In scientific and popular descriptions of the Vedda phenotype, selection has been the major stumbling block of physical anthropologists. The number of osteological specimens collected from Ceylon since 1827 and bearing the name “‘ Vedda ”’ on their labels is impressive and exceeds in size the series accessible for many other Asiatic tribal groups of higher population frequency, and the amount of anthropometric data pertaining to both osseous and living Vedda specimens is not inconsiderable. Never- theless this abundance of data has not deceived the more perceptive students of the Veddas who have troubled themselves to investigate the histories of particular so-called Vedda specimens and Vedda communities. The history of the scientific investigation of Ceylon’s aboriginal population has yet to be written, but a cursory glance at the published data indicates that evidence gleaned from small samples often has been considered applicable to the Vedda population as a whole. In most cases the fact that the specimens were from Vedda populations at all is questionable, for the majority of collectors obtained them through Sinhalese and Tamil contacts by offering to pay the latter for every skull they could procure, leaving it up to the con- tacts to decide what was or was not a Vedda. The care taken by the Sarasins and Hill to get their osseous specimens from Vedda cave sites and cemeteries, personally excavated, is a commendable but infrequent condition in the history of Vedda osteological collections. Furthermore, even those specimens which most experts would recognize as Veddas were taken from a fairly circumscribed sector of the eastern portion of the Island, although historical accounts make it clear that in the past the Veddas were more widespread if not ubiquitous. Scientific research on the Veddas, which may be said to have commenced in the early decades of the last century, has never been completely independent of the notions contained in fantastic traveller’s accounts or of the oral and written traditions concerning this population. Even in the rigour of the determination to select those Vedda specimens, osseous or living, which are “ typical’’, scientific workers have repeatedly emphasized the importance of certain phenotypic traits and thus intro- duced additional selective factors into the interpretation of What is a Vedda? Of these notions of the Vedda phenotype, the most common is that it is infantile and/or anthropoidal and hence representative of a very primitive and even atavistic kind of humanity. Some writers have barely accorded the Veddas human status. The 206 HUMAN SKELETAL MATERIAL FROM CEYLON selection of certain Vedda crania as prototypes, even though it was an attempt to define with clarity the concept of the Vedda phenotype, has tended to obscure the factor of normal variability of this population. The ghost of this tradition lingers in the works of those researchers who, even in the manipulation of large series of Vedda specimens, have not been emancipated from the employment of such terms as “ typical Veddas ’’, “‘ pure Veddas ”’ or “ classic Veddas ”’ in their writings. Such expressions may be useful when denoting specimens for which no admixture with non-Vedda phenotypes is suspected, but they are misleading when the total Vedda phenotype is the subject of consideration. Such terms are meaningless when the physical anthropologist attempts to understand the phylogeny of the Veddas since the gene pool of any population is never static. In Ceylon population shifts have been especially encouraged through the operation of frequent genetic intercourse between groups as well as by the dynamics of genetic drift and natural selection in local areas of the Island. An examination of the archaeological complex at Bellan Bandi Palassa forces the anthropologist to ask these questions: (1) How are the cultural artifacts found in association with the Balangoda skeletal series related to the prehistoric picture of Ceylon and the Indian mainland and to southern Asia as a whole? (2) What affinities, if any, can be demonstrated between the cultures of the Balangodese and the historic Veddas? These problems of the archaeological significance of the cultural assemblage at Bellan Bandi Palassa are complementary to the problems of the physi- cal anthropology of the human remains. In comparing certain elements of the Balangoda culture with that of the Veddas, attention is directed to a paper by Allchin (1959). In summarizing the evidence of the prehistoric cultures of Ceylon, Allchin has pointed out striking similarities between certain of its elements and par- ticular cultural traits of the Veddas. The present writer, although restricting himself to the cultural evidence from the single site of Bellan Bandi Palassa, finds grounds for justifying a similar comparison. To the early prehistorians who pioneered research into Ceylon’s prehistory the suggestion that the ancestors of the Veddas were the manufacturers of the ubiquitous stone tools seemed obvious. Those who questioned this apparent relationship between the prehistoric or contemporary cul- tures did so because they thought the Veddas incapable of producing a lithic industry, and not because they questioned the validity of the archaeological evidence in the Vedda caves. Now that the physical remains of the artisans of these lithic industries have been recovered, the contentions of the Sarasins and the Seligmanns must be re- evaluated along with the concepts of their opponents about the role that the ancestors of the Veddas played in Ceylon’s prehistory. As a final point in this discussion it must be mentioned that what is now required of the prehistorians of Ceylonese studies is a compilation of data concerning the Veddas from ancient native chronicles and later travellers’ accounts. Whatever the affinities of the Veddas to the prehistoric and historic populations of Ceylon might be, the invasion of the Sinhalese in the Fifth Century B.C. caused a concentra- tion of the indigenes in the eastern wilderness of the Island—the Veddarata—where they are still to be found in isolated chena villages. That they once occupied the HUMAN SKELETAL MATERIAL FROM CEYLON 207 entire Island, as has been suggested by the Sarasins (Sarasin & Sarasin 1907 : 189, F. Sarasin 1926 : 83) cannot be proved as yet, but references to their presence in other portions of Ceylon can be inferred from a cursory glance at the native Sinhalese and later European records as well as from a mapping of the Vedda place names across Ceylon. These data establish that the territory of the Veddas extended far beyond the present limits of this relict group within the recent historic period. What their area of occupation may have been during the period that Ceylon was known only along its coastline and before its interior had been penetrated by Sin- halese and Tamil colonization awaits further research. The migration of Vedda communities over the Island during the past three millenia and earlier is relevant to the problems of their physical morphology today. One wonders what selective factors, be these biological or cultural, influenced the preservation of certain Vedda populations in the relict areas of the Island while other Vedda groups became in- corporated in communities of invading peoples or failed to pass on their genes at all. Useful as Hill’s concept of Vedda geographical—-biological areas may be to the physical anthropologist desirous of obtaining “ typical Vedda”’ specimens, it must be borne in mind that these individuals are a segment of a much larger Vedda gene pool which existed in the past and the factors conducive to their survival are unknown. Furthermore we are ignorant of the genetic shifts that have taken place within these population segments since the commencement of their isolation. Such considerations are relevant to the problem of identifying the Balangodese when one considers that this stone-using prehistoric population was in existence at the dawn of the period of Sinhalese settlement and that the genetic complex that makes it distinctive as a phenotypic entity at this period must have had considerable influence upon the genetic composition of the present-day peoples of Ceylon. V CONCLUSIONS The Balangoda skeletal series from Bellan Bandi Palassa constitutes a unique phenotypic pattern present in Ceylon at the dawn of the historic period, a time within the second and third millenia B.P. While possessing certain biological features that distinguish them from other fossil hominids found thus far, the Balangodese fall within the range of polytypic variation representative of post-Pleistocene Homo sapiens of the Indian-southeast Asian—Australian gene pool. The phenotypic pattern most closely resembling that of the Balangodese is that of the Veddas of Ceylon. Similarities are not only quantitatively provocative but striking as well in terms of particular isolated physical traits that distinguish both Veddas and Balango- dese from other southern Asiatic populations. Indeed, certain of the morphological and anthropometric characteristics of the fossil series have been cited by physical anthropologists studying the Veddas as indicative of Vedda sub-types or as the evidence of their miscegenation with other living racial groups. It would seem therefore that both the Balangodese and the Veddas are biologically united through their possession in the past of a common gene pool, although it must be stressed that the former are “ pre-Veddas”’ in a chronological rather than in any direct phylogenetic kind of relationship. Both populations appear to have been subject 208 HUMAN SKELETAL MATERIAL FROM CEYLON to separate evolutionary pressures for a long period of time, but for reasons which are yet unknown the Balangodese phenotypic pattern did not persist into the historic period. It is the variety and nature of physical differences between these two populations that suggests their bifurcation from a common stem at a time several millenia prior to the occupation of Bellan Bandi Palassa. At the dawn of the historic period in Ceylon, the Veddas were in all probability distributed throughout the Island, save perhaps along the northern coastline, and as a dispersed population their phenotype reflected local variations as a consequence of isolation between the segments of their population. That the Veddas are simply the hybrid descendants of a crossing of Balangodese with Sinhalese or Tamils cannot be supported on the basis of the anthropometric and historic data, although the presence of a yet unidenti- fied phenotypic element in Ceylon in the prehistoric period must not be entirely excluded from considerations of the Balangodese-Vedda phylogeny. The cultural affinities of the Balangodese from Bellan Bandi Palassa are with the manufacturers of the lithic industries which have been assigned to the Indian Late Stone Age or its Ceylonese manifestation, the Bandarawelian. The number and kinds of similarities that exist between the prehistoric and Vedda cultures forcefully suggests the existence of a cultural continuum in Ceylon that extends into the historic period. Such traits as the use of iron and Sinhalese weapons by the Veddas of recent centuries find their parallels in the acquisition of painted pottery, ground stone tools, and perhaps other traits by the hunting-gathering and microlith-using Balangodese who may well have been introduced to these by trading peoples from coastal Ceylon and the Indian mainland. Further speculations of the biological and cultural affinities of this prehistoric population await additional research along these lines : (I) a resolution of the problem of what constitutes the Vedda phenotype apart from the traditional methods of defining this population, (2) an orientation of archaeological investigation in Ceylon that can demonstrate the role that the Vedda’s have played in the manufacturing of the Island’s prehistoric industries, (3) the map- ping of Vedda occupation sites throughout the Island with the view of establishing the migration patterns and degree of mobility of the Veddas within the period of recorded history, (4) the recovery of additional osteological material from Ceylonese sites for the purpose of broadening our knowledge of the physical anthropology of the region. This present analysis of the Balangodese is an initial attempt to extend our understanding of the evolution of human populations during that interim which is marked by the close of the Pleistocene and the dawn of the historic period in southern Asia. VI ACKNOWLEDGEMENTS I should like to thank Dr. E. I. White, F.R.S., Keeper of Paleontology, and Dr. K. P. Oakley, Deputy Keeper of the Sub-Department of Anthropology, of the British Museum (Natural History) for permission to study the collections of human osteologi- cal material in their department and for their kind assistance and encouragement. Dr. P. E. P. Deraniyagala, Director of the National Museum of Ceylon, Colombo, was responsible for the discovery of the Bellan Bandi Palassa skeletal series and a person HUMAN SKELETAL MATERIAL FROM CEYLON 209 to whom I am deeply grateful. I am also greatly indebted to these staff members of the British Museum (Natural History): Mr. A. E. Rixon, Mr. D. R. Brothwell, Mr. R. J. Parsons, Mr. G. C. Ross, Mr. M. J. Rowlands, Mrs. Elizabeth Gardiner, Mrs. Madeleine Glemser, and Miss Rosemary Powers. My thanks are due to Dr. E. A. Mourant of the Lister Institute, London, and to Dr. Virginia Carbonell. Apprecia- tion is extended to Professors T. D. McCown, S. L. Washburn and S. F. Cook of the University of California at Berkeley ; to Mrs. Audrey DeJournette of Oakland, California ; and to my wife, Dr. Mary C. Marino. The research programme was made possible through a grant from the National Science Foundation, Washington, DiC: VII REFERENCES ABELSON, P. H. 1954. The organic constituents of fossils. Yearb. Carneg. Instn., Washing- ton, 53 : 97-101, figs. I, 2. 1955. The organic constituents of fossils. Yearb. Carneg. Instn., Washington, 54: 107-109. Apam, W. 1943. The Keilor skull: palate and upper dental arch. Mem. nat. Mus., Melb., 13 : 71-77, pls. Io, II. ALLCHIN, B. 1959. The Late Stone AgeofCeylon. J. R.anthrop. Inst., London, 88 : 179-201. ANONYMOUS. 1941. Stone Age man in Ceylon. Nature, Lond., 147 : 392-393. BarBER, H. 1939. Untersuchungen tiber die chemische Veranderung von Knochen bei der Fossilization. Palaeobiologica, Wien, 7 : 217-235. Basu, P.C. 1932-1933. The racial affinities of the Mundas. Tvans. Bose Res. Inst., Calcutta, 8, 12 : 211-247, figs. 119-129. 1933-1934. The racial affinities of the Oraons. Tvans. Bose Res. Inst., Calcutta, 9, 8 : 132-176, figs. 42-56. Brack, D. 1928. A study of the Kansu and Honan Aeneolithic skulls in comparison with North China and other recent crania. Part I: on measurements and identification. Palaeont. sinica, Peking (D) 7 : 1-83, figs. I-31. Bonin, G. von. 1934. On the size of man’s brain, as indicated by skull capacity. J. comp. Neurol., Philadelphia, 59 : 1-28. Boyp, W. C. & Boyp, L. G. 1933. Blood grouping by means of preserved muscle. Science, New York, 78 : 578. 1934. An attempt to determine the blood groups of mummies. Pvoc. Soc. exp. Biol., New York, 31 : 671-674. BRoTHWELL, D.R. 1960. Upper Pleistocene human skull from Niah Cave, Sarawak. Sarawak Mus. J. (n.s.) 9, 15-16 : 323-349, pls. I-9. CAMPBELL, T. D. 1925. Dentition and palate of the Australian aboriginal. Univ. Adelaide Publ., 1 : viii + 123, pls. 1-52. CANDELA, P. B. 1936. Blood-groups reactions in ancient human skeletons. Amer. J. phys. Anthrop., Washington, 21 : 429-432. CHANMUGAM, P. K. & JAYAWARDENA, F.L.W. 1954. Skeletal remains from Tirrukketiswaram. Ceylon J. Sci., Colombo (G) 5, 2 : 65-68. CrarK, G. 1961. World Prehistory. An Outline. xiii + 284 pp., 12 pls., Cambridge. CoLe, S. 1963. Races of Man. 131 pp., 12 pls., 34 figs. British Museum (Nat. Hist.), London. Cook, S. F. 1960. Dating prehistoric bone by chemical analysis. Jn Heizer, R. F. & Cook, S.F. 1960. The Application of Quantitative Methods in Archaeology : 223-245. New York. Coon, C.S. 1962. The Origin of Races. xli + 724 + xxi pp., 32 pls. New York. DERANIYAGALA, P. E. P. 1943a. Some aspects of the prehistory of Ceylon: Part 1. Spola zeylan., Colombo, 23, 2 : 93-115, pls. 1-10. 210 HUMAN SKELETAL MATERIAL FROM CEYLON DERANTIYAGALA, P. E. P. 1943b. The Stone Age men of Ceylon: Part 2. J. R. Asiat. Soc., Ceylon Br., Colombo, 35 : 159-162. 1945. Balangoda culture phase. Collections at Museum. Ceylon Times, Colombo, March 8. 1954. Stone Age Ceylon. J. R. Asiat. Soc., Ceylon By., Colombo, 3, 2 : 113-124, pls. 1-4. 1955a. Ceylon’s Stone Age. New Lanka, Colombo, 6, 2 : 39-41. 19550. Some aspects of the prehistory of Ceylon: Part 4. Spolia zeylan., Colombo, 27, 2 : 295-303, pls. 1-8. 1956a. Land oscillation in the north-west of Ceylon. J. R. Asiat. Soc., Ceylon Br., Colombo, 4, 2 : 127-142, pls. 1-4. 1956b. A Mesolithic burial tumulus from Ceylon. Nature, Lond., 178: 1481-1482, figs. I, 2. 1956c. Some aspects of the prehistory of Ceylon: Part 5. Spolia zeylan., Colombo, 28, I : 117-120, pls. I, 2. 1956d. Some early human races of Ceylon. Ceylon Today, Colombo, 5, 10 : 15-16. 1957a. Ceylon Adm. Rep., Nat. Mus., Colombo, 1956. 1957). Races of the Stone Age and the Ferrolithic of Ceylon. J. R. Asiat. Soc., Ceylon Br., Colombo, 5, I : 1-23. 1958a. Anopen air habitation site of Homo sapiens balangodensis: Part1. Spolia zeylan., Colombo, 28, 2 : 223-261, pls. I-15, map. ——1958b. The Pleistocene of Ceylon. Ceylon Nat. Mus. nat. Hist. Ser., Colombo. 164 pp., 58 pls. 1959. The persistence of palaeolithic types among the stone artifacts of Ceylon. Pyvoc. 46th Indian Sci. Congr., Calcutta, 3, 38 : 432. 1960a. An open air habitation site of Homo sapiens balangodensis: Part 2. Spolia zeylan., Colombo, 29, 1 : 95-109. —— 1960b. Some mammals of the extinct Ratnapura fauna of Ceylon: Part 4. Spoha zeylan., Colombo, 29, 1 : 1-14, pls. 1-3. —— 1960c. Stone Age man in Ceylon. Im Cotrell, L. 1960. The Concise Encyclopedia of Archaeology : 128. London. —— 1962. The extinct Hominoidea of Ceylon. Anthrop. Ann. Vidyodaya Univ., Ceylon, Colombo, 1962 : 52-62. 1963a. An open air habitation site of Homo sapiens balangodensis: Part 3. Spolia zeylan., Colombo, 30, 1 : 87-110, pls. 1-11, figs. 1-6. 1963b. The hybridization of the Vaddas with the Sinhalese. Spolia zeylan., Colombo, 30, I : 111-147, pls. 1-7. Dvusois, D. 1922. The proto-Australoid fossil man of Wadjak, Java. Proc. K. Akad. Wetensch., Sect. Sci., Amsterdam, 23, 2 : 1013-1051. EHRHARDT, S. 1960. Theurn burials. Jn Sankalia, H. D., Deo, Z. D., Ansari, S. & Ehrhardt, S. 1960. From History to Pre-History at Nevasa. Report on the Excavations and Explora- tions in and avound Nevasa 1954-56 : 506-522, figs. 207-210. Poona. 1963. Friihmenschliche Skelette aus Langhnaj in Gujarat, Vorderindien. Zeit. Morph. Anthrop., Stuttgart, 54, 2 : 151-162, pls. 18-21. EickstepT, E. F. von. 1927. Die Indien-Expedition des Staatlichen Forschungsinstituts fiir V6lkerkunde in Leipsig. Erster Anthropologischer Bericht (Ceylon). Anthrop. Anz., Stuttgart, 4 : 208-2109, pl. 5. Ezra, H. C. & Coox, S. F. 1957. Amino acids in fossil human bone. Science, New York, 126 : 80. FLower, W. H. 1884. On the size of teeth as a character of race. J. R. anthrop. Inst., London, 14 : 183-187. Gray, M. P. 1953. Group Specific Activity of Extracts for the A, B, and O Antigens Obtained from Skeletal Material, Blood Stains and Red Blood Cells : Thesis presented to the Department of Biology and Graduate School of the University of Oregon, Degree of Master of Science. HUMAN SKELETAL MATERIAL FROM CEYLON 211 Gray, M. P. 1958. A method for reducing non-specific reactions in the typing of human skeletal material. Amer. J. phys. Anthrop., Washington (n.s.) 16, 1 : 135-139. Hamsiy, W. D. 1940. Craniometry of New Guinea. * BY PERERA YE, Php) Burmah Oil Exploration Co. Ltd., 20 Esplanade, Scarborough Pp. 215-253; 11 Plates; 5 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 5 LONDON : 1965 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), tmnstituted im 1949, 1s issued in five series corresponding to the Departments of the Museum, and an Historical serves. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar ‘year. In 1965 a@ separate supplementary series of longer papers was instituted, numbered serially for each Department. This paper is Vol. 11, No. 5 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1965 TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) Issued December, 1965 Price 42 10s. SOME NEW BRITISH ALBIAN OSTRACODA By P. KAYE CONTENTS I INTRODUCTION AND ACKNOWLEDGEMENTS 218 Il LocaATION OF SAMPLES 219 III SysTEMATIC DESCRIPTIONS 221 Suborder Cladocopina 221 Family Polycopidae 221 Genus Polycope Sars. 221 Polycope nuda sp. nov. 221 Polycope owent sp. nov. 222 Suborder Podocopina. 223 Family Bairdiidae . 223 Genus Bairdia McCoy . 223 Bairdia pseudoseptentionalis (Mertens) c 223 Genus Pontocyprella Mandelstam . 224 Pontocyprella semiquadrata sp. nov. 224 Family Cypridinae . 225 Genus Argilloecia Sars. . 225 Argilloecia valuula sp. nov. . 225 Genus Paracypris Sars. : . 226 Paracypris wrothamensis sp. nov. 226 Genus Kvrausella Ulrich 2277 Krausella sp. 227 Family Cytherideidae 228 Genus Clithrocytheridea Stephenson 228 Clithrocytheridea heslertonensis Kaye 228 Genus Schuleridea Swartz & Swain 228 Schuleridea dimorphica sp. nov. 228 Genus Habrocythere Triebel . 229 Habrocythere fragilis Triebel 229 Genus Dolocytheridea Triebel 230 Dolocytheridea ee sp. nov. 230 Family Cytheruridae 231 Genus Eucytherura Muller 3 231 Eucytherura aff. nuda Kaye 231 Genus Hemicytherura Elofson 231 Hemicytherura euglyphea sp. nov. 231 Genus Cytheropteron Sars. 5 . 232 Cytheropteron (Cytheropteron) EEE s. nov. : 5 232 Cytheropteron (Cytheropteron) milbournei sp.nov. . a ARB Cytheropteron (Cytheropteron) nanissimum nanissimum Damotte & Grosdidier 234 Cytheropteron (Cytheropteron) nanissimum fenestra subsp. nov. ‘ = BBA Cytheropteron (Gy ihereptevon) lane nom. nov. e285 Subgenus Eocytheropteron Alexander . 235 Cytheropteron (Eocytheropteron) Pietoyea sp.nov. . 235 Subgenus Infracytheropteron Kaye 230 Cytheropteron (infacytheropieron). Giecuna — nov. . 236 Geol. 11, 5 22 218 SOME NEW BRITISH ALBIAN OSTRACODA Genus Orthonotacythere Alexander . : ; : : 5 AB Orthonotacythere fovdensis sp. nov.. . é : = 8 8237 Orthonotacythere minutissima sp.nov. . : : : 239 Orthonotacvthere spinifera sp. nov. : P : 230 Family Brachycytheridae . ; : : : . 240 Genus Alatacythere Murray & heey : ‘ | 240 Alatacythere vobusta vobusta (Jones & Hance) : 5 240 Alatacythere vobusta langi subsp. nov. . ; : 2 240 Family Bythocytheridae . : : 6 é : : se ite. Genus Monoceratina Roth. . ; F : : 242 Monoceratina longispina (osquen) : ; : EZAZ Monoceratina sp. - : : : : : - 244 Family Progoncytheridae i : ‘ : F : . 244 Genus Acrocythere Neale : : : : : : - 244 Acrocythere striata sp. nov. . : : : z - 244 Genus Neocythere Mertens . : : : : : - 245 Subgenus Physocythere Kaye . : a BAK Neocythere (Physocythere) tenuis Sp. nov. : . 245 Family Protocytheridae . ; i ; ; ; : - 246 Genus Veenia Butler & Jones : : : ; : - 246 Veenia compressa Kaye : : ; : ‘ . 246 Veenia florentinensis Damotte . 5 c : 246 Family Trachyleberididae : 3 : ‘ : : ZAG Genus Cythereis Jones . : 3 : ‘ : Zar, Cythereis angulatoides nom. nov. . ‘ : * 247 Cythereis gatyensis Damotte & Grosdidier c : - 248 Cythereis glabrella Triebel . : : : : . 248 Cythereis pinhayensis sp.nov. . 3 : : - 248 Suborder Myodocopina 0 : . 5 : : : . 250 Family Conchoeciidae . : ‘ : : : ; . 250 Genus Conchoecia Dana : F ; , : : . 250 Conchoecia sp.A : : : ‘ F : - 250 Conchoecia sp.B : 3 é 0 : : . 250 IV REFERENCES : : ‘ ; ‘ . ‘ : j 2) 25 SYNOPSIS Twenty new species and subspecies of Ostracoda are described from the Gault Clay (Middle and Upper Albian) of England. Additiona information is given on nine already described species and four ostracod synonyms (Clithvocytheridea ventricola Damotte & Grosdidier, Cytheropteron punctata Kaye, Cythereis angulata Kaye and Cythereis lamplughi Kaye) are corrected. Four forms are left under open nomenclature. I INTRODUCTION AND ACKNOWLEDGEMENTS A quantitative study of the distribution of Ostracoda in the English Gault (Middle and Upper Albian) has been carried out by the collection of accurately localized samples at small vertical intervals in as many Gault exposures as are currently available. Over three hundred and fifty samples have been collected, a large propor- tion of which have already been analysed working on a 1,500 gram basic starting weight of sediment. The project as a whole is only partially complete but a certain number of undescribed species have already been found. The description of these new species together with other relevant taxonomic information is the subject of the SOME NEW BRITISH ALBIAN OSTRACODA 219 present paper. Twenty species and subspecies belonging to fifteen genera are des- cribed for the first time. Additional information is given on nine already known species and four Cretaceous ostracod synonyms are corrected. Four forms are left under open nomenclature. The known range of each described species is given and its value as an index fossil is noted where possible. A complete list of the Gault exposures sampled during the course of the overall project is given. A palaeonto- logical zonation of the Middle and Upper Albian based on ammonites is given as Text-fig. 1. All specimens described and illustrated in this paper are deposited in the collections of the British Museum of Natural History (B.M.N.H.) but considerable numbers of comparative forms are retained in the author’s private collections. This study has been carried out during the tenure of a D.S.I.R. Research Fellow- ship at the Sedimentology Research Laboratory, Dept. of Geology, Reading Univer- sity, and I am extremely grateful to Professor P. Allen for all his help and encourage- ment. I would also like to thank many of my colleagues at Reading, particularly Dr. A. W. Medd, Mr. G. H. Scott and Mr. D. B. Williams for help in field work and Mr. J. L. Watkins for the photography. I would also like to express my gratitude to Mr. H. G. Owen and Mr. R. A. Milbourne whose help concerning Gault stratigraphical problems has been of immense value. II LOCATION OF SAMPLES A list of the localities from which Gault Ostracoda have been obtained is as follows : (i) Lower and Upper Gault at the British Portland Cement Co’s pit, Small Dole near Henfield, Sussex. Grid. Ref. TQ. 218131 (ti) Lower Gault at the Honey Lane Brickworks, Selbourne, Hampshire. Grid. Ref. SU. 768342 (iii) Lower Gault at the Greatness Lane Brick pit, Sevenoaks, Kent. Grid. Ref. TQ. 536578. (tv) Lower and Upper Gault at Ford Place pit, Wrotham, Kent. Grid. Ref. TQ. 636591 (v) Upper Gault at the Rugby Portland Cement Co’s pit, Paddlesworth, near Maidstone, Kent. Grid. Ref. TQ. 695623 (vi) Upper Gault at Sandown Bay and also Blackgang, Isle of Wight (vu) Upper Gault at Pinhay Point, Devon. Grid. Ref. SV. 342928 (viii) Lower Gault at Devizes, Wiltshire. Grid. Ref. ST. 986612 (ix) Lower Gault at Culham, near Abingdon, Oxfordshire. Grid. Ref. SV. 510949 (x) Lower and Upper Gault at Mundays Hill pit, Leighton Buzzard, Bedfordshire. Grid. Ref. TL. 915978 (xi) Upper Gault at the London Brick Co’s pit, Arlesey, Bedfordshire. Grid. Ref. TL. 185352 (xu) Upper Gault at Fisons Brickpit, Burwell, Cambridgeshire. Grid. Ref. TL. 516691 (xii) Upper Gault at Eastwoods Cement pit, Barrington, Cambridgeshire. Grid. Ref. TL. 394507 MIDDLE ALBIAN UPPER ALBIAN LOWER ALBIAN SOME NEW BRITISH ALBIAN OSTRACODA ALBIAN ZONES and SUB ZONES Stoliczkaia dispar Mortoniceras inflatum Euhoplites lautus Hoplites dentatus Douvilleiceras mammillatum Leymeriella tardefurcata | | | | | Mortoniceras perinflatum Arraphoceras substuderi Mortoniceras inflatum var. aequatorialis Callihoplites auritus Hysteroceras varicosum Hysteroceras orbigny! Dipoloceras cristatum Anahoplites davies: Euhoplites nitidus Euhoplites meandrinus Dipoloceras subdelaruei Dimorphoplites doris Dimorphoplites niobe Anahoplites intermedius Hoplites spathi Hoplites benettianus Hoplites eodentatus Protohoplites puzosianus Otohoplites raulinianus Cleoniceras floridum Sonneratia kitchini Leymeriella regularis Hypacanthoplites milletioides Farnhamia farnhamensis SOME NEW BRITISH ALBIAN OSTRACODA 221 (xiv) Lower Gault at Castles Farm pit, near Ely, Cambridgeshire. Grid. Ref. TL. 600775 (xv) Lower Gault at Speeton, E. Yorkshire. Grid. Ref. TA. 150758 (xvi) Lower Gault at West Heslerton, E. Yorkshire. Grid. Ref. TA. 913759 (xvii) Lower and Upper Gault at Folkestone, Kent. Grid. Ref. TR. 242365 (xviii) Upper Gault at Ashford, Kent. Grid. Ref. TR. 058435 (xix) Gault at Swanage, Lulworth, Osmington and Black Ven near Lyme Regis, Dorset. Many other pits, particularly in the Lower Gault of the Wealden area have been found to be barren of Ostracoda. These include : (i) Lower Gault at Hassocks, Sussex. Grid. Ref. TQ. 310155 (i) Lower Gault at the Marley Tile pit, Storrington, Sussex. Grid. Ref. TQ. 094138 (iii) Lower Gault at Nyewood Brick pit, Nyewood, near Petersfield, Hampshire. Grid. Ref. SU. 800218 (iv) Lower Gault at Wrecclesham, near Farnham, Hampshire. Grid. Ref. SU. 826448 (v) Lower Gault at Arnold’s sand pit, Buckland, Kent. Grid. Ref. TQ. 231512 (vi) Lower Gault at Squerry’s pit, Westerham, Kent. Grid. Ref. TQ. 442538 (vii) Lower Gault at Uffingham, Oxfordshire. Grid. Ref. SU. 315905 (viii) Lower Gault at Badbury Wick, near Swindon, Wiltshire. Grid. Ref. SU. 192818 (ix) Lower Gault at Thame, Oxfordshire. Grid. Ref. SP. 691055 (x) Lower Gault at Coney Hill Sandpit, Oxted, Surrey. Grid. Ref. TQ. 375526 III SYSTEMATIC DESCRIPTIONS Suborder CLADOCOPINA Family POLYCOPIDAE Genus POLYCOPE Sars 1866 Polycope nuda sp. nov. (Pl. 4, figs. 1-3) DERIVATION OF NAME. nuda—referring to the lack of surface ornament. DiacGnosis. Large moderately inflated Polycope with subcircular outline and smooth valve surface. Ho.otyPe: A single left valve, B.M.N.H. Io. 2847, from the Lower Gault niobe Subzone ; Wrotham, Kent. PaRATYPES: B.M.N.H. Io. 2848-2850. Three carapaces from the same sub- zone and locality. 222 SOME NEW BRITISH ALBIAN OSTRACODA MEASUREMENTS. Length MHeight Width (mm.) (mm.) (mm.) Left valve (B.M.N.H. Io. 2847, holotype) . 0°55 0*50 0-16 Carapace (B.M.N.H. Io. 2848) . ; On5S 0-50 0-32 DEscRIPTION. Valves large, moderately inflated, subcircular in outline. There is a slight flattening dorsally along the hinge line forming two weak cardinal angles. The anterior margin is more bluntly rounded than the posterior margin. Lateral surface mainly smooth but with some weak irregular reticulation particularly antero- ventrally in certainspecimens. Overlap strong particularly posteriorly and ventrally. Hinge simple with a groove in the margin of the right valve to accommodate the sharp edge of the left valve. Duplicature not seen. Muscle scars, a triangular Fic. 2. Muscle scars of Polycope nuda sp.nov. 2500. patch composed of three scars, the apex pointing dorsally. The two ventral scars are roughly triangular in shape, the dorsal scar is diamond shaped, fitting between the points of the other two scars. REMARKS. Only one other species of the genus Polycope has been previously recorded from the Cretaceous. This species, P. bonnemaz Herrig 1964, is known from the Maastrichtian of the Isle of Rugen and a similar form Polycope sp. was described by Bonnema (1940) from the Maastrichtian of the Netherlands. P. nuda is therefore the first pre-chalk Cretaceous reference to the genus and is closely related and presumably ancestral to P. bonnemat. It is closely similar in size and ornamentation to the latter but differs in details of the shape and overlap, lacking the anterior and postero-dorsal marginal thickening of P. bonnemai. In shape the anterior and pos- terior margins are more evenly rounded in P. nuda. P. oweni described below, also from the British Albian, is much smaller and strongly ornamented. The genus is fairly common in the Jurassic where several species are known, particularly from the Liassic and Oxfordian. In general these forms are smaller, much more flattened and usually strongly ornamented. P. nuda is found consistently throughout the Gault first making its appearance in the zobe Subzone. It is never very common though its relative abundance in the nzobe Subzone is a useful indicator of that horizon. Polycope oweni sp. nov. (Pl. 4, figs. 11-15) DERIVATION OF NAME. After H. G. Owen whose work on Gault stratigraphy has been invaluable to me in my studies of the distribution of the Ostracoda in the Gault. Hototyre. A left valve, B.M.N.H. Io. 2859, from the H. orbignyi Subzone (Upper Gault) ; Wrotham, Kent. SOME NEW BRITISH ALBIAN OSTRACODA 223 ParRaATYPES. B.M.N.H. Io. 2858, 2860-63. Four valves and two carapaces from the same horizon and locality. Di1aGNosis. Small subcircular Polycope with valve reticulately ornamented. Along and at the junction of the reticulation are numerous small spines. MEASUREMENTS. Length Height Width (mm.) (mm.) (mm.) Left valve (B.M.N.H. Io. 2859, holotype) . 0-42 One7 O-1II Carapace (B.M.N.H. Io. 2858) . : 5 Ooze 0°37 0:22 DESCRIPTION. Valves small, weakly inflated, subcircular in outline. The valves have a slightly greater length than height but there is only very slight flattening along the hinge line. The valves are slightly asymmetrical anteriorly but the anterior and posterior margins are without cardinal angles. Marginal protuberances are absent but the whole of the margin is thickened. The lateral surfaces are strongly but variably ornamented ranging from reticulation to spination. The basic ornament is the reticulation covering the whole of the surface on a somewhat concentric pattern. Small spines are developed along the ridges of the reticulations giving rise to a completely spined appearance in many of the specimens. Overlap is marked all round the margin except dorsally. The hinge is simple ; the margins in both valves being flattened to form sharp marginal bars which overlap each other. Dorsally in the right valves there is a weak groove above the bar. Muscle scars, three equal-sized oval scars arranged in a triangle. One apex of the triangle points dorsally. REMARKS. PP. oweni is not abundant in the Gault and is found rarely in clays of post A. intermedius Subzone age. It differs from P. nuda in the details of the shape, hinge, muscle scars and ornamentation and is smaller. Suborder PODOCOPINA Family BAIRDIIDAE Genus BAIRDIA McCoy 1844 Bairdia pseudoseptentrionalis (Mertens) (Pl. 2, figs. 1, 3-6) 21840 Cytherina subdeltoidea Munster ; Roemer : 105, pl. 15, fig. 22. 21845 Cytherina subdeltoidea Munster ; Reuss: 16, pl. 5, fig. 38. 1849 Bairdia subdeltoidea (Munster) ; Jones: 23, pl. 5, figs. 15a-f. 21874 Batrdia subdeltoidea (Munster) ; Reuss: 140, pl. 26, figs. 5a—c. 1890 Batrdia subdeltoidea (Munster) ; Jones & Hinde: 5, pl. 2, figs. 31-34. 21927 Bairdia subdeltoidea (Munster) ; Alexander, pl. 6, figs. 2, 4. ? 1929 Bairdia subdeltoidea (Munster) ; Alexander : 61, pl. 3, fig. 5. 1956 Bairdoppilata pseudoseptentrionalis Mertens : 182, pl. 8, figs. 7-10, pl. 13, figs. 89-90. 21956 Bairdoppilata roemeri Deroo : 1509, pl. I, figs. 9-12. 1958 Bairdoppilata pseudoseptentrionalis Mertens ; Howe & Laurencich : 82. ?1958 Bairdoppilata ? voemeri Deroo ; Howe & Laurencich : 82. 224 SOME NEW BRITISH ALBIAN OSTRACODA MATERIAL. B.M.N.H. Io. 2828-2832, 4 valves and I carapace from the Gault just below the Cambridge Greensand at Arlesey, Beds. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2828) . ; : 2 32 0-87 Right valve (B.M.N.H. Io. 2829) . : : ; ea: 0°75 REMARKS. For a long time all Cretaceous species of Bairvdia were placed in B. subdeltoidea (Munster) a Tertiary form. In 1956 Deroo re-named the Cretaceous references to B. subdeltoidea as Bairdoppilata roemert. Earlier in the same year Mertens had, however, erected a new species Bairdoppilata pseudoseptentrionalis which seems to be identical with Deroo’s form. Both authors had referred their species to the genus Bairdoppilata which is said to differ from true Bairdia in having two rows of denticles at either end of the hinge line in the right valves. However, as van Morkhoven (1963) has pointed out, the presence or absence of such denticles is a characteristic of most of the genera within the Bairdiidae each genus containing species both with and without them. It therefore seems unwise to perpetuate the genus Bairdoppilata. Bairdia pseudoseptentrionalis occurs rarely in the Lower Gault but is more abun- dant in the Upper Gault. Genus PONTOCYPRELLA Mandelstam 1956 Pontocyprella semiquadrata sp. nov. (Pl. 3, figs. 1-8) DERIVATION OF NAME. Semiquadrate—alluding to the shape. Dracnosis. Pontocyprella with semiquadrate shell. Anterior margin semi- circular, posterior margin bluntly rounded. Hototyre. A left valve, B.M.N.H. Io. 2834, from the H. orbigny: Subzone, Upper Gault ; Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2835-2837. Three valves from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2834, holotype) : : 0-70 0-40 Right valve (B.M.N.H. Io. 2836) . : : : 0°73 0-40 DESCRIPTION. Valves elongate, compressed, semiquadrate in shape. Dorsal margin weakly convex, without cardinal angles ; ventral margin weakly concave. Anterior margin semicircular, posterior margin bluntly rounded forming a weak bulge at 3 height. Greatest height and width at mid-length. Lateral surface smooth. Inner lamella broad with large anterior and posterior vestibules. Dupli- SOME NEW BRITISH ALBIAN OSTRACODA 225 Ge ZB Fic. 3. Muscle scars of Pontocyprella semiquadrata sp.nov. 2500. cature narrow, crossed by numerous short straight fine radial pore canals. Normal pore canals few, small and irregularly scattered. Muscle scars a central rosette of 4 scars. Hinge simple. In the left valve there is a marginal bar with a long narrow shelf-like furrow below it. In the right valve there is a narrow marginal shelf with a high bar below it. REMARKS. Pontocyprella semiquadrata occurs rarely in the upper part of the Lower Gault and more commonly in the Upper Gault. Its overall shape and particularly the shape of the anterior and posterior margins distinguish it from other Cretaceous species of the genus P. superba Neale 1962 and P. rava Kaye 1965 have an acute posterior end whilst P. harrisiana Jones is larger, more elongate and has the anterior margin bulged dorsally and the posterior margin bulged ventrally. P. semiquadrata is closest to P. mandelstami Kaye 1965 differing in that the latter is kidney shaped rather than semiquadrate. P. maynci Oertli 1958 has the dorsal margin strongly arched. Family CYPRIDIDAE Genus ARGILLOECIA Sars 1866 Argilloecia valvula sp. nov. (Pl. 7, figs. 20-25) DERIVATION OF NAME. valvulus L. = husk. DiaGnosis. Argilloecia with arched dorsal margin and acute posterior margin. Hototyre. A right valve, B.M.N.H. Io. 2914, from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. ParaTyPEsS. B.M.N.H. Io. 2915-2920. Four valves and two carapaces from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Right valve (B.M.N.H. Io. 2914, holotype) . : 0:52 0:20 Left valve (B.M.N.H. Io. 2915). : : ; 0:48 0-16 DEscripTIon. Valves, thin, small, elongate, laterally compressed. Dorsal margin strongly arched, without cardinal angles ; ventral margin long and straight. Anterior margin forming a blunt point antero-ventrally ; posterior margin meeting ventral margin at an angle of go° ventrally. Greatest height just in front of mid- length ; greatest width just behind mid-length. Right valve larger than left, over- 226 SOME NEW BRITISH ALBIAN OSTRACODA lapping around the entire margin but particularly strongly antero-dorsally and postero-dorsally. Lateral surface smooth. Duplicature narrow, crossed by few short, straight radial pore canals. Inner margin and line of concrescence separate, forming large vestibules anteriorly and posteriorly. Normal pore canals few, ir- regularly scattered. Hinge simple consisting of a groove in the dorsal margin of the tight valves to accommodate the dorsal edge of the smaller left valves. Muscle scars a central rosette of five scars. a se Fic. 4. Muscle scars of Argilloecia valuula sp. nov. 2500. REMARKS. Argilloecia valvula occurs rarely in the Upper Gault and the upper part of the Lower Gault. It is the first record of the genus in the Lower Cretaceous. The larger right valve separating the genus from most Cyprids, a feature only otherwise occurring in Macrocypris and Pontocypris. Argilloecia differs from these latter two genera in size, hinge and marginal features. The large size of the known Albian species of Macrocypris easily differentiates them from A. valvula. M. parva Kaye 1965 from the Hauterivian and Barremian of Northern England is similar in size but differs appreciably in shape. Genus PARACYPRIS Sars 1866 Paracypris wrothamensis sp. nov. (Pl. 9 figs. 9-14) DERIVATION OF NAME. After the village of Wrotham, Kent the type locality for the species. Diacnosis. Large Paracypris with strongly drawn out acute posterior end and high strongly angular anterior cardinal angle. Hototyre. A left valve B.M.N.H. Io. 2959 from the H. orbignyt Subzone (Upper Albian) ; Wrotham Kent. PARATYPES. B.M.N.H. Io. 2955-58, 2960, 2961. Three left valves, three right valves and two carapaces from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2959, holotype) : : 0:92 0°35 Right valve (B.M.N.H. Io. 2958) . : : : 0-90 0°35 Carapace (B.M.N.H. Io. 2957) : : ‘ 0-90 0°35 SOME NEW BRITISH ALBIAN OSTRACODA 227 DESCRIPTION. Valves large, compressed, very elongate. Greatest height at 1} length, greatest width at 4 length. In lateral view elongate subtriangular with the dorsal margin straight and the ventral margin weakly concave. Anterior margin broadly rounded forming a marked cardinal angle with the strongly sloping dorsal margin. Posterior drawn out forming a strongly acute point postero-ventrally. A well marked postero-dorsal cardinal angle occurs. Lateral surface smooth. Left valve larger than right overlapping particularly dorsally and ventrally. Large crescentic anterior and large triangular posterior vestibules occur. Zone of fusion very narrow anteriorly and posteriorly, doubling in width ventrally. Crossed by few thick short radial pore canals branching distally (8-10 anteriorly). Normal pore canals small, few, irregularly scattered. Hinge simple, the dorsal edge of the right valve fitting into a narrow shelf-like groove along the dorsal margin of the left valve. An anterior prolongation of the dorsal marginal bar of the left valve forms a tooth-like extension corresponding with a slight incurving of the margin in front of the anterior cardinal angle in the right valve. Muscle scars a central rosette of five or six oval scars. REMARKS. P. wrothamensis occurs throughout the Gault in England. It first appears in the imtermedius Subzone but is rare in the Lower Gault. It becomes abundant in the Upper Gault and its abundance is a useful indicator of Upper Gault age. It is strongly related to the Lower Cretaceous species P. acuta (Cornuel) 1848. This latter species is poorly described and rather confused but is smaller and not so acute posteriorly as P. wrothamensis. P. sinuata Neale 1963 from the Hauterivian and Barremian is probably, at least in part, synonymous with P. acuta. P. sinuata differs from P. wrothamensis in being lower, having a less well marked anterior cardinal angle and being much less acute. Of other Cretaceous species P. depressa Bonnema 1940 and P. jonest Bonnema 1940 have the cardinal angles rounded and the greatest height further forward ; P. gracilis (Bosquet) and P. szliqua Jones & Hinde 1890 differ markedly in outline. Genus KRAUSELLA Ulrich 1894 Krausella sp. (Pl. 3, figs. 15, 16) MATERIAL. Two right valves B.M.N.H. Io. 2845-2846, from the upper part of the Lower Gault at Castles Farm, Ely. MEASUREMENTS. Length Height (mm.) (mm.) Right valve (B.M.N.H. Io. 2845) . : F é 0°55 0:27 REMARKS. Only one other species of this genus (K. minuta Triebel) is known from the Cretaceous. It is smaller and is not so pointed posteriorly. The strong inequality of the valves distinguishes the genus. 228 SOME NEW BRITISH ALBIAN OSTRACODA Family CYTHERIDEIDAE Genus CLITHROCYTHERIDEA Stephenson 1936 Clithrocytheridea heslertonensis Kaye (Pl. 1, figs. 8-12) 1963a Clithrocytheridea heslertonensis Kaye : 30, pl. 1, figs. 10-13. 1963 Clithrocytheridea ? ventricola Damotte & Grosdidier : 53, pl. I, figs. 1a—f. MATERIAL. B.M.N.H. Io. 2825-2827, Female left and right valves, and carapace from Bed N. 5, Middle Albian sfathi Subzone, West Heslerton, East Yorks. REMARKS. Specimens of Clithrocytheridea ? ventricola kindly sent to me by Dr. E. Grosdidier show that this form is synonymous with C. heslertonensis Kaye. Clithro- cytheridea heslertonensis is found abundantly in the H. spathi Subzone of the Lower Gault and appears to be confined to that subzone. It is a valuable zonal index and has been recorded from West Heslerton, Speeton, Devizes, Culham and Henfield in clays of that age. Genus SCHULERIDEA Swartz & Swain 1946 Schuleridea dimorphica sp. nov. (Pl. 5, figs. 1-6) DERIVATION OF NAME. Alluding to the strongly dimorphic nature of the species. DiaGNosiIs. Small strongly dimorphic Schuleridea with weak ocular sulcus and without cardinal angles. Hototyre. A male left valve, B.M.N.H. Io. 2864, from the H. orbignyt Subzone, Upper Gault ; Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2865-2870. Twelve valves and two carapaces from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Male left valve (B.M.N.H. Io. 2864, holotype) : 0-64 0-41 Female left valve (B.M.N.H. Io. 2867) . : : 0°52 0-40 Male right valve (B.M.N.H. Io. 2866) . : P 0:63 0°35 Female right valve (B.M.N.H. Io. 2868) ; : 0°50 0°34 Description. Valves relatively small, ovate, laterally compressed. Dorsal margin strongly arched in left valves, weakly arched in right valves. Cardinal angles absent from both valves. Ventral margin convex in the left valves, straight in the right. Anterior margin broadly rounded ; posterior bluntly pointed just below mid- height. Greatest height just in front of mid-length, greatest width at mid-length. Eye spot very weak with a short, shallow, oblique sulcus posterior to it. Lateral surface smooth. There is a slight ventral tumidity particularly in the right valves. Duplicature broad, crossed by numerous fine radial and pseudoradial pore canals. These canals are bent upwards anterodorsally. No marginal denticulation in either valve. Inner margin and line of concrescence coincide throughout Normal SOME NEW BRITISH ALBIAN OSTRACODA 229 pore canals abundant and irregularly scattered over the lateral surface. Hinge strongly developed, having in the left valves two terminal divided sockets, open ventrally, separated by a shelf-like furrow. Dorsal of the median furrow is a high, smooth bar and a narrow, shallow accommodation groove. In the right valves there are two bar-like denticulate terminal teeth (5 denticles in each) separated by a low, median bar. Above the hinge is a prominent median shelf. Sexual dimorphism is particularly strong. REMARKS. Schuleridea dimorphica occurs abundantly in the Upper Gault and is a valuable index form for that age. Its small size and ovoid shape distinguish it from most other species of the genus. It is closest to S. sulcata Kaye 1965a from the Aptian of the Isle of Wight, differing in the poorer development of the eye tubercle and ocular sulcus and in the shape of the dorsal and posterior margins. Smaller and differing in hingement from the only other known Albian species S. jonesiana (Bosquet) 1852, it is also flatter and lacks the cardinal angles and prominent eye tubercle of S. virginis Grosdidier 1964 and the truncate posterior end of S. bernouilensis Grosdidier 1964. In shape and dimorphic features it differs also from S. derooi Damotte & Grosdidier 1963a. Genus HABROCYTHERE Triebel 1940 Habrocythere fragilis Triebel (Pl. 6, figs. 7-13) 1940 Habrocythere fragilis Triebel : 166, pl. 1, figs. 10-13, pl. 9, fig. ror. 1956 Habrocythere fragilis Triebel; Mertens: 198, pl. 10, figs. 51-52. 1963 Habrocythere fragilis Triebel; Kaye : 33, pl. 3, figs. 8-9. MATERIAL. Four normal specimens B.M.N.H. Io. 2882, 2884, 2887, 2888, from the D. cristatum Subzone (Middle Albian) ; Wrotham, Kent. Three anomalous specimens, B.M.N.H. Io. 2883, 2885, 2886, from the same subzone and locality. Remarks. Habrocythere fragilis is common throughout the English Albian. Certain specimens have, however, been found that may throw light upon the origin of this monotypic genus. Amongst the normal assemblages of this species certain specimens occur which are identical to the species in its general accepted sense but have a large eye spot mounted on a large tubercle set antero-dorsally. They also have the ventro-lateral portion of the valves drawn out into a weak alate expansion. These specimens have to be included in H. fragilis on the basis of their other features and their close association with normal specimens of the species but are closely similar to species of the genus Euryitycythere. It is therefore, very likely that Euryitycythere known from Valanginian to Aptian strata has evolved into the genus Habrocythere with certain specimens of the latter genus showing recapitulation. The major diagnostic features of Ewryitycythere are the eye tubercle, the inflated alate ventro- lateral area, the wide flattened marginal area and the hinge. The wide flattened marginal area and associated radial canals are found in all specimens of Habrocythere whilst the eye tubercle and inflation are shown in the “ recapitulation ’’ specimens. 230 SOME NEW BRITISH ALBIAN OSTRACODA The variability of the hinge in Ostracoda and its modification during phylogeny is well known in Ostracoda and modification during the Hauterivian-Albian time interval is quite feasible. The Aptian occurrence of the genus (Kaye 1965q) is of asingle closed carapace and the hinge features are not known. Slight simplification of the hinge structure during the phylogeny can perhaps be related to the small size and light build of Habrocythere and can be used as evidence against making Euryitycythere Oertli 1958 a synonym of Habrocythere Triebel 1940. Specimens of H. fragilis showing affinities with Euryitycythere have been found from several localities. They occur mainly in the daviest and cristatum Subzones at Wrotham and in the subdelaruet Subzone at Sevenoaks. They are useful indicators for recognizing the upper part of the Lower Gault and it is possible that ecological con- ditions in the Weald at this time stimulated this diversification. Genus DOLOCYTHERIDEA Triebel 1938 Dolocytheridea typica sp. nov. (Pl. 3, figs. 9-14) DERIVATION OF NAME. Typicus L. = typical. DiaGnosis. Small Dolocytheridea, inflated posteriorly, with dorsal margin strongly arched postero-dorsally. Posterior end rather truncated. Horotyre. A left valve, B.M.N.H. Io. 2830, from the basal Upper Gault (Upper Albian) ; Pinhay, Devon. PARATYPES. B.M.N.H. Io. 2838, 2840-44. Eight valves and two carapaces, from the same horizon and locality. MEASUREMENTS Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2839 holotype) : : 0:59 0°33 Right valve (B.M.N.H. Io. 2838) . . : ; 0:57 0-28 DEscrIPTION. Valves small, elongate, laterally inflated particularly posteriorly. Dorsal margin strongly arched, without cardinal angles ; ventral margin straight. Anterior margin broadly rounded, posterior margin bluntly rounded rather truncate. Greatest height and width at 3 length. Lateral surface smooth, eye spots absent. Duplicature broad, crossed by a very large number of thin, rather sinuous, simple radial pore canals. Inner margin and line of concrescence coincide throughout. Normal pore canals fairly abundant, irregularly scattered. Muscle scars, not usually seen, appear to consist of a vertical row of four oval scars with a V-shaped scar anterior tothem. Other smaller scars probably occur above and below the main group. The hinge is simple consisting in the left valves of a curved smooth groove deepened at its ends accommodating the dorsal edge of the smaller right valve. The dorsal margin of the right valve is enlarged to form weak smooth terminal teeth. REMARKS. This species is much smaller than most other species of the genus. It differs significantly from the even smaller D. minuta Kaye inshapeand greater inflation. SOME NEW BRITISH ALBIAN OSTRACODA 231 From the D. hilseana—D. intermedia—D. bosquetiana lineage it differs in the greater inflation posteriorly and the truncation of the posterior end as well as in size. D. typica has so far only been found from the basal Upper Gault at Pinhay in Devon. Family CYTHERURIDAE Genus EUCYTHERURA Muller 1894 Eucytherura aff. nuda Kaye (Ble7; figss 17,718) MATERIAL. B.M.N.H. Io.2912-13. Two specimens from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. REMARKS. Specimens similar to the Barremian form FE. nuda Kaye 1964 occur rarely in the Upper Gault. They are comparable with true E. nuda but have the postero-ventral lobe more strongly inflated and a stronger surface reticulation. It is almost certain that extra material will show that the Gault specimens belong to a separate species or subspecies. A single valve of a related but distinct form has been found from the uppermost Lower Gault at Castles Farm, Ely. This specimen is similar in shape and size to E. nuda but has two rows of low surface nodes, one row dorsally and a more prominent one ventrally. Genus HEMICYTHERURA Elofson 1941 Hemicytherura euglyphea sp. nov. (Pl. 8, figs. 1-4) DERIVATION OF NAME. euglypheus L. = distinctly marked. Diacnosis. Sexually dimorphic Hemicytherura with strong surface ornament of longitudinal ridges. Hototype. A male left valve, B.M.N.H. Io. 2921, from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2922-2925. Five valves and one carapace from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Male left valve (B.M.N.H. Io.2921, holotype) 3 0-40 0-19 Female left valve (B.M.N.H. Io. 2924) . H ; 0°33 0:16 Male right valve (B.M.N.H. Io. 2923) . : : 0-40 0-20 Female right valve (B.M.N.H. Io. 2922) : : 0:33 0-17 DEscRIPTION. Valves small, elongate, laterally compressed. Dorsal margin well arched in the right valves, weakly arched in the left valves ; without cardinal angles. Geol. 11, 5 23 232 SOME NEW BRITISH ALBIAN OSTRACODA Ventral margin straight. Anterior margin broadly rounded ; posterior drawn out into a short acute caudal process just above mid-height. Greatest height and width at mid-length. Lateral surface strongly ornamented, tumid ventrally. Eye spot distinct, low and glassy. Ornament consists of a series of longitudinal ridges joined by weaker vertical cross ridges. The pattern of longitudinal ridges is not regular in the centro-lateral area where a flattened irregularly ridged patch occurs. The strong ornament covers the whole of the valve exterior, even anteriorly where several longitudinal ridges run strongly across the marginal area. Duplicature moderately broad, crossed by few straight, simple radial pore canals. There is a well developed ocular pit antero-dorsally in the interior of the valves. Hinge rather complex. In the right valve there are two faintly denticulate terminal bar- like teeth. Above each tooth is a shelf-like furrow accommodating the dorsal edge of the left valve. Between and in line with the teeth and furrows the valve remains open but above the general line of the hinge there is a median dorsal marginal bar. In the left valve the dorsal margin is enlarged to form a curved marginal bar, more prominent at the ends where it fits above the terminal teeth of the right valve. Below this marginal bar both anteriorly and posteriorly there are shelf-like terminal grooves to accommodate the terminal teeth of the right valve. Sexual dimorphism well marked. REMARKS. fH. euglyphea occurs rarely in the upper part of the Lower Gault and more commonly in the Upper Gault, being distinguished from most other species of related genera on account of shape, hinge and ornament. It is closest to Cytherura veticulosa (Chapman) which itself is probably a Hemicytherura, differing in being less arched dorsally, in lacking the alate spine postero-ventrally and in details of the surface ornamentation. H. euglyphea differs from other Cretaceous species such as H. unisulcata Veen, H. bisulcata Veen and H. asiculcata Veen in lacking vertical sulcation and in surface ornamentation. Genus CYTHEROPTERON Sars 1866 Cytheropteron (C.) arguta sp. nov. (PL. 8, figs. 12-17) DERIVATION OF NAME. arguta L. = distinct. Diacnosis. Cytheropteron with strongly reticulate lateral surface and ridged posteriorly pointing alate expansion. Hototyre. A left valve, B.M.N.H. Io. 2936, from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. PaRATYPES. B.M.N.H. Io. 2933-2935, 2937-2939. Six valves and one carapace from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2936 holotype) 2 : 0-42 0:26 Right valve (B.M.N.H. Io. 2933) . : b : 0°42 0°24 SOME NEW BRITISH ALBIAN OSTRACODA 233 DESCRIPTION. Valves small, ovate, strongly ornamented. Dorsal margin strongly arched, without cardinal angles ; ventral margin weakly convex. Anterior margin bluntly rounded, posterior drawn out into a caudal extension at mid-height. Greatest height and width at mid-length. A rounded ventro-lateral alate expansion occurs. Along its crest there is a prominent ridge starting anteriorly at the margin at + height and terminating posteriorly in a small posteriorly directed spine. The anterior end of the ridge forms a spine on the margin. The lateral surface is strongly reticulate. The ventral undersurface is weakly striated. Duplicature moderately broad, crossed by few straight, thick radial pore canals. Normal pore canals and muscle scars not seen. Hinge very strongly developed with two long, almost interdentate, terminal sockets separated by a short straight coarsely crenulate bar in the left valve. Above the hinge is a wide marginal shelf. In the right valve there are two terminal rows of denticles (6 or 7 in each) decreasing in height towards the median element and separated by a short interdentate furrow. Above the hinge is a narrow marginal shelf. REMARKS. Cytheropteron (C.) arguta appears to be confined to the Upper Gault. It is extremely abundant in the H. orbignyi Subzone and can be used as an index species. It has been found so far at Wrotham, Henfield and Pinhay. The strength of the ornamentation and hinge make C. (C.) avguta distinct from other known species as does the shape of the alae. It is closest to C. (C.) punctata Kaye from the Barremian of Northern England which has a similar ridged alae. The latter, however, has the ridge separated from the anterior margin and lacks the strong ornamentation. Cytheropteron (C.) milbournei sp. nov. (Pl. 7, figs. 4, 6-9) DERIVATION OF NAME. After R. H. Milbourne whose stratigraphical work on the Gault has been an invaluable assistance to my study of the distribution of Ostracoda in the Gault. DiaGnosis. Cytheropteron with drawn out postero-ventrally directed alate expansion and punctate ornament over whole lateral surface. Hototyre. A left valve, B.M.N.H. Io. 2808, from the subdelaruei Subzone, Lower Gault ; Wrotham, Kent. ParatyPes. B.M.N.H. Io. 2899, a left valve from the same horizon and locality as the holotype ; B.M.N.H. Io. 2qg00-02. Two valves and one carapace from the niobe Subzone at Sevenoaks. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2898, holotype) ‘ 2 O37; 0°22 Right valve (B.M.N.H. Io. 2900) : : : 0°35 0:22 DESCRIPTION. Valves small, elongate, laterally compressed and with a prominent ventro-lateral alate expansion. Dorsal margin weakly arched; ventral margin 234 SOME NEW BRITISH ALBIAN OSTRACODA straight. Anterior margin semicircular, posterior end drawn out into a blunt caudal extension at 3 height. Greatest height in front of mid-length ; greatest width at Zenzth. Dorsal margin without cardinal angles. Alae directed postero-ventrally and bearing a weak vertical sulcus on its crest. Whole of lateral surface covered with a series of pits. The ventral underside of the alae bears a few short longitudinal riblets. Duplicature moderately broad, crossed by few thick, straight radial pore canals (5 anteriorly). Hinge merodont consisting of a long straight denticulate median bar and two small divided terminal sockets in the left valve. Above the median element is a wide marginal shelf. REMARKS. Cytheropteron (C.) milbourne: is known from the top part of the Lower Gault at Sevenoaks and Wrotham and it is never very abundant. It differs from C. (C.) nanissimum principally in the shape of the alae and in its ornamentation. Cytheropteron (C.) nanissimum nanissimum Damotte & Grosdidier (QEAUE 7a sili 902) 50/5) 1963 Cytheropteron (C.) nanissumum Damotte & Grosdidier : 56, pl. 1, figs. 2a-f. MATERIAL. B.M.N.H. Io. 2907-08, two left valves from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. Remarks. This subspecies occurs throughout the Gault, appearing first in the niobe Subzone but not becoming abundant until the Upper Gault. Cytheropteron (C.) nanissimum fenestrata subsp. nov. (PEG, figs. 14, iO wro) DERIVATION OF NAME. fenestrata L. = window. DIAGNOSIS. Subspecies of Cytheropteron (C.) nanissimum with fenestrate ornamentation on ventral alae. Hototypre. A left valve, B.M.N.H. 2910, from the H. spathit Subzone, Lower Gault ; Henfield, Sussex. PARATYPES. B.M.N.H. Io. 2909, 2911. Two left valves from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2910, holotype) ‘ : 0°33 0-20 DEscrIPTION. Valves small, elongate with a ventro-lateral alate expansion. Dorsal and ventral margins convex, greatest height at mid-length. Anterior margin semicircular, posterior margin bluntly pointed forming a weak cardinal angle dorsally. Alae directed, posteriorly, with a small spine at its end and a weak vertical sulcus on its upper surface. Above the alae near the dorsal margin is a low node. From this node a rib runs downwards to the crest of the alae, whilst ashort longitudinal SOME NEW BRITISH ALBIAN OSTRACODA 235 rib runs across the ala at mid-height of the valves. The two ribs cross on the alae and form a “ window like”’ pattern of ornament. The rest of the lateral surface is smooth. Internal features identical with C. nanissumum s:s. REMARKS. This subspecies is closely similar to C. (C.) nanissimum s.s. differing only in the nature of the ornamentation on the alae. In C. nanissumum s.s. the dorsal node and vertical rib are much stronger than in C. (C.) nanissumum fenestrata whilst the cross rib is absent. _ The limited occurrence of C. (C.) nanissumum fenestrata in the H. spathi Subzone points to an ancestral relationship to C. nanissumum s.s. which does not appear until the A. niobe Subzone. C. (C.) reightonensis Kaye 1964 from the Barremian of Northern England is also very closely related being ornamented with rows of pits along the alae. Cytheropteron (Cytheropteron) lamplughi nom. nov. 1964 Cytheropteron (C.) punctata Kaye : 103, pl. 5, figs. 7-8. REMARKS. Professor W. A. Van den Bold of Louisana State University has kindly pointed out that the form I erected from the Lower Barremian at Speeton is in synonomy with Cytheropteron punctatum Brady (1868: 449). I have therefore renamed my form Cytheropteron (C.) lamplughi in honour of G. W. Lamplugh and his basic research on the British Lower Cretaceous. Subgenus EOCYTHEROPTERON Alexander 1933 Cytheropteron (Eocytheropteron ) protonsa sp. nov. (Pl. 6, figs. 1-6) DERIVATION OF NAME. Protonsa L. = stretched out. DiaGnosis. Large Cytheropteron with bilobed wing-like alate expansion and prominent upturned posterior caudal process. Hototyre. A left valve, B.M.N.H. Io. 2879, from the D. cristatum Subzone, Lower Gault ; Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2880-81. Two valves from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2879 holotype) : ; 0-70 0-42 Right valve (B.M.N.H. Io. 2880) . ; : ; 0-67 0-42 DEscrRIpTION. Valves large, elongate, laterally compressed. Dorsal margin strongly arched ; without cardinal angles in the left valves but with weak cardinal angles in the right valves. Ventral margin straight totally obscured by the promi- nent ventral alate expansion. Anterior margin broadly rounded ; posterior elon- 236 SOME NEW BRITISH ALBIAN OSTRACODA gated into a prominent upturned caudal process at % height. A large wing-like bilobed alate expansion occurs ventro-laterally ; the posterior lobe being more extended. The lateral surface is covered with irregular vertical riblets which are only poorly developed anteriorly but are stronger posteriorly. The ventral under- surface bears a series of longitudinal ridges. A weak eye node occurs antero-dorsally. Duplicature rather narrow, crossed by few, straight, simple radial pore canals. Inner margin and line of concrescence coincide. Normal pore canals few. Hinge strongly developed. In the left valve there is a long interdentate furrow widest and deepest at the ends. In the right valve there is a long curved row of denticles, highest at the anterior and posterior ends. REMARKS. C. (E.) protonsa occurs rarely in the upper part of the Lower Gault and in the Upper Gault. The wing-like bilobed alae, strong upturned caudal pro- cess, hinge and ornament make it quite unlike any other described species of Cytherop- teron s.l. Subgenus INFRACYTHEROPTERON Kaye 1964 Cytheropteron ? (Infracytheropteron) obscura sp. nov. (Pl. 7, figs. 10-12) DERIVATION OF NAME. obscura L. = obscure. DiaGnosis. Ovate Cytheropteron s.1. with smooth lateral surface and rounded ventro-lateral tumidity. Hinge simple. Horotyre. A left valve, B.M.N.H. Io. 2903, from the H. orbignyi Subzone, Upper Gault, Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2904-06. Three valves and two carapaces from the same locality and horizon. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2903 holotype) 2 ; 0*42 0°25 Right valve (B.M.N.H. Io. 2905) . : ; : 0-40 0°25 DESCRIPTION. Valves small, smooth, ovate witha marked ventro-lateral tumidity. Dorsal margin strongly arched, without cardinal angles. Ventral margin short and straight. Anterior margin broadly rounded, posterior drawn out into a blunt caudal process at just over mid-height. Greatest height at } length, greatest width at mid-length. Lateral surface strongly and evenly inflated but with a flat marginal shelf anteriorly, posteriorly and postero-ventrally. A low, smooth eye tubercle occurs antero-dorsally. Duplicature moderately broad, particularly anteriorly and postero-ventrally, crossed by few straight, thick radial pore canals (7—8 anteriorly). Inner margin and line of concrescence coincide. Normal pore canals small and few. Hinge simple consisting of a smooth marginal bar in the left valve which fits on a marginal shelf in the right valve. Below the marginal shelf is a narrow median bar being discontinuous anteriorly and posteriorly. SOME NEW BRITISH ALBIAN OSTRACODA 237 REMARKS. C. (J.) obscura occurs throughout the bulk of the Gault. It first appears in the miobe Subzone but only becomes common in the Upper Gault. The relationship of the species is rather problematical, and it has been placed in the genus Cytheropteron s.l. on account of its shape and size. The lack of a marked alate expansion makes it akin to the subgenus Eocytheropteron but the hinge places it in the subgenus Infracytheropteron. The species shows strong similarities to the Tertiary genera Bythocythere and Loxoconcha. It differs primarily in the hingement and lack of vestibules though it does have a wide marginal shelf anteriorly and postero-ventrally. It is closest to Loxoconcha minuta Jennings 1936 from the Maastrichtian of New Jersey which is similar in shape and size but has ornamented valves. This latter form is doubtfully referred to the genus Loxoconcha. Genus ORTHONOTACYTHERE Alexander 1934 Orthonotacythere fordensis sp. nov. (PI. 5, figs. 7-13) DERIVATION OF NAME. After Ford Place near Wrotham, Kent, the location of the holotype. Diacnosis. Ovthonotacythere with weak but distinct ribbing, low tubercles and wide deep median sulcus. The postero-lateral area is reticulate and strongly inflated. HototyPe. A male left valve, B.M.N.H. Io. 2871, from the intermedius Subzone, Lower Gault, Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2872-78. Six valves and two carapaces from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Male left valve (B.M.N.H. Io. 2871, holotype) ; 0°57 0-30 Male right valve (B.M.N.H. Io. 2872) . : : 0°55 0:28 Female left valve (B.M.N.H. Io. 2873) . : ; 0:50 0°30 DESCRIPTION. Valves elongate, compressed, divided by a wide deep vertical median sulcus. Dorsal margin straight, ventral margin convex. Anterior broadly rounded, posterior forming a blunt postero-dorsal process. The median sulcus is limited ventrally by an irregular longitudinal ridge, below which, on the ventral undersurface, lies a further straight longitidunal ridge. Faint cross riblets join the two ribs. Antero-dorsally there is a prominent glassy eye tubercle jointed by a short rib to a reticulate tubercle immediately below it. Two ridges run from this lower tubercle, one anteriorly to meet the anterior margin, the other posteriorly towards the sulcus to meet a small tubercle where it turns abruptly through go° to run vertically to connect with an irregular tubercle at the anterior end of the upper ventral longitu- dinal ridge. The area between the ridges is reticulate and irregularly ribbed. The lower ventral ridge runs anteriorly to reach the margin at + height and posteriorly to fade 238 SOME NEW BRITISH ALBIAN OSTRACODA out on the postero-ventral surface. The upper ventral ridge bears two weak tubercles ventrolaterally, the posterior one being more prominent. Faint irregular ridges run ver- tically from these tubercles merging with the general reticulation of the postero- lateral surface. This reticulation is made up of a series of weak longitudinal ridges joined by cross members. Antero-dorsally on the postero-lateral surface lies a low ridged tubercle. Internal features and hinge typical of the genus. Fic. 5. Diagrammatic representation of the distribution of the ornamentation of Ovthonotacythere fordensis sp.nov. X75 REMARKS. O. fordensis occurs fairly commonly in the A. intermedius Subzone of the Lower Gault. It seems to be restricted to that subzone and is a valuable index form. O. fordensis is closely related to O. inversa (Cornuel) and related species form- ing an evolutionary sequence in the British Lower Cretaceous. It has all the major features of the group differing in minor details of the ornamentation and particularly in the subdued nature of the costation and tuberculation. Though almost certainly derived from these Lower Cretaceous forms it does not, however, show continuation of the trends seen there. The tendency towards increased tuberculation and sim- plification of costation is not followed and the species must therefore not be on the main “Boreal” stock. The related species of Orvthonotacythere found in the “Tethyan”’ province do not follow the same trends during evolution, and ornamental patterns anomalous to that seen at Speeton have already been found (Kaye 19654) in the Aptian of the Isle of Wight. It is to these species that close ancestry of O. fordensis must be attributed. The ornamentation of these Aptian species such as O. catalaunica Damotte & Grosdidier and O. atypica Kaye is closely similar to O. fordensis having the costation fairly complex with particularly the irregular nature of the upper ventral longitudinal ridge and the nature of the antero-dorsal costation in good agreement. The absence of O. fordensis s.s. from the Lower Gault at Speeton and its replacement by fragmentary specimens of a form intermediate between O. fordensis and O. inversa tuberculata seem to bear this out particularly on consideration of the joining of the “ Boreal” and “‘ Tethyan ”’ seas in this area in Apto-Albian times. Only a single specimen belonging to this group of the genus Ovthonotacythere has so far been found from post H. spathi age Gault deposits. This specimen from the nitidus Subzone (Bed 31) at Henfield shows increased tuberculation, poor costation, poor development of the ventral longitudinal ridge and no surface reticulation. These features seem consistent with the overall trends seen at Speeton and the specimen perhaps shows a further stage of this major evolutionary pattern. SOME NEW BRITISH ALBIAN OSTRACODA 239 Orthonotacythere minutissima sp. nov. (Pl. 8, figs. 5-11) DERIVATION OF NAME. minutissimum L. = very small. Dracnosis. Very small Orthonotacythere with ventral alate longitudinal rib and prominent eye tubercle. Hototyre. A left valve, B.M.N.H. lo. 2926, from the H. varicosum Subzone, Upper Gault, Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2927-32. Five valves and one carapace from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2926, holotype) ' : 0:28 0-16 Right valve (B.M.N.H. Io. 2927) . ; ‘ é 0-28 0°15 DEscRIPTION. Valves very small, elongate, compressed. Dorsal margin long and straight ; ventral margin straight, shorter and parallel to it. Anterior margin bluntly rounded ; posterior with a weak caudal process at the postero-dorsal cardinal angle. Eye tubercle exceedingly prominent, protruding well above the dorsal margin. Greatest height at 4 length at the anterior cardinal angle ; greatest width just behind mid-length. Lateral surface divided by a vertical median sulcus which is limited ventrally by a prominent alate longitudinal ridge. This ridge increases in height posteriorly, being drawn out into a blunt laterally directed spine. The base of the sulcus does, however, notch this rib slightly, weakly dividing it into two sections. A large tubercle occurs dorsally on the postero-lateral area. Lateral surface smooth or pitted. Anterior margin denticulate (5 denticles). Duplicature moderately broad, crossed by few, straight radial pore canals. Inner margin and line of concrescence coincide. Interior eye pit prominent. Hinge simple consisting of a long, straight denticulate bar in the left valve. In the right valve there is a long, straight interdentate furrow with a high smooth bar above it. REMARKS. O. minutissima occurs rarely in the Upper Gault but has not yet been found in the Lower Gault. Its exceptionally small size makes it distinct from other species of the genus. The pattern of the main ornamentation shows no affinities to any other described form but I have found identical or closely comparable forms from the Hauterivian/Barremian of Lincolnshire. The hinge of this species is not strictly typical of the genus Ovthonotacythere, lacking well defined terminal elements, but the shape, sulcus and ornamentation are all closely comparable to that genus. It may, however, be possible to include this species in a redefined subgenus such as Stillina Laurencich. Orthonotacythere spinifera sp. nov. (PI. 7, figs. 1-3, 5) DERIVATION OF NAME. Spinifera L. = spined. DiaGnosis. Small Orthonotacythere with weak median sulcus, pronounced reticula- tion over whole of valves, and a series of large spines arranged over lateral surface. 240 SOME NEW BRITISH ALBIAN OSTRACODA Hototyre. A left valve, B.M.N.H. Io. 2895, from the A. intermedius Subzone (Bed 13), Lower Gault, Wrotham, Kent. PARATYPES. B.M.N.H. Io. 2894, a right valve from the H. orbignyi subzone (Bed 13), Upper Gault, Wrotham. B.M.N.H. Io. 2896-97, two valves from the A. intermedius Subzone, Lower Gault, Henfield, Sussex. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2895, holotype) ‘ ; 0°42 0-22 Right valve (B.M.N.H. Io. 2894) . : ; : 0+42 0-22 DESCRIPTION. Valves small, elongate, strongly laterally compressed. Dorsal margin long and straight ; ventral margin short, straight and parallel to it. Anterior margin bluntly rounded forming a well marked antero-dorsal cardinal angle ; pos- terior drawn out to form an acute postero-dorsal caudal process. Anterior and posterior margins strongly spined. Vertical median sulcus very weak or absent. Eye tubercle rounded and glassy, set at dorsal end of a short high, blade-like verti- cally elongated spine. A similar bladed spine lies at ? length on the dorsal margin with a further smaller spine at 4 length just below the dorsal margin. An arcuate row of four extremely prominent spines runs parallel to the ventral margin. Below this row is a series of smaller semi-fused spines culminating posteriorly in a larger bilobed spine. A pair of smaller spines occur posteriorly, one on the dorsal margin, the other being vertically below it at 3? height. The whole of the lateral surface is strongly reticulate. Duplicature narrow, crossed by few straight radial pore canals. Normal pore canals and muscle scars not seen. Hinge strongly developed consisting in the left valve of two short, subdivided sockets separated by a long, straight, lobed bar. Above the median bar is a narrow marginal shelf. In the right valve there are two bar-like denticulate terminal teeth (5 denticles in each) separated by a wide, strongly divided, median groove. REMARKS. Orthonotacythere spinifera is rare in most of the Gault, a few specimens having been found at a variety of levels and localities. It is more abundant in the Gault at Pinhay in Devon and is perhaps a littoral species. O. spinifera is quite unlike other described species of the genus having such characteristic features as the weak median sulcus entirely covered by the surface reticulations and the series of marginal and lateral spines. It is smaller than most other species of the genus but does show some affinities with the O. :mversa group in the arrangement of the spines. Family BRACHYCYTHERIDAE Genus ALATACYTHERE Murray & Hussy 1942 Alatacythere robusta robusta (Jones & Hinde) (Pl. ro, figs. 9, 10) 1964b Alatacythere vobusta (Jones & Hinde) ; Kaye: 51, pl. 2, fig. 18. MATERIAL. B.M.N.H. Io. 2964, 2965, two valves from the uppermost Gault immediately below the Cambridge Greensand, Arlesey Beds. SOME NEW BRITISH ALBIAN OSTRACODA 241 MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2965) ; : ‘ I-05 0°51 Alatacythere robusta langi subsp. nov. (Pl. ro, figs. 1-4) DERIVATION OF NAME. After W. D. Lang in appreciation of his stratigraphical and palaeontological work on the Cretaceous in Devon and Dorset. DiaGnosis. Subspecies of Alatacythere robusta (Jones & Hinde) 1890 having ventro-lateral alae less drawn out and dorsal longitudinal ridge more pronounced. Hototyre. A male left valve, B.M.N.H. Io. 2940, from the basal Upper Gault ; Pinhay, Devon. PARATYPES. B.M.N.H. Io. 2941-44. Six valves from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Male left valve (B.M.N.H. Io. 2940, holotype) . : I-02 0°54 Male right valve (B.M.N.H. Io. 2943) ; ; 0:98 0°53 Female Right valve (B.M.N.H. Io. 2942). ; : 0:92 0°50 Female left valve (B.M.N.H.Io.2941) . : : 0°95 0-50 DESCRIPTION. Valves large, elongate, laterally compressed. Dorsal and ventral margins straight, converging slightly posteriorly. There are well marked antero- dorsal and postero-dorsal cardinal angles. Anterior margin broadly rounded ; posterior margin weakly pointed at mid-height, both strongly denticulate. Each marginal tubercle corresponds to the extremity of a radial pore canal. Greatest height at } length; greatest width at ? length. Eye tubercle large and glassy, joined to a prominent anterior marginal ridge which follows the entire anterior margin to be continued along the crest of the high extended ventral alate expansion. This alate expansion obscures the whole of the ventral margin and strongly increases in height posteriorly where it terminates in a short ventro-laterally directed spine. The ventral undersurface lacks marked costation. A short, high longitudinal ridge occurs obscuring the posterior and central part of the dorsal margin. It is not con- tinued to join the eye tubercle. The lateral surface is smooth. Duplicature moder- ately broad, crossed by numerous straight radial and pseudoradial pore canals. Inner margin and line of concrescence coincide throughout. Normal pore canals scarce over the bulk of the lateral surface but abundant along the ridged crest of the alae. Hinge strong, amphidont, consisting in the left valve of two terminal sockets, open ventrally and separated by a long, straight, denticulate bar. Below the anterior end of the median bar there is a high, smooth, circular tooth, whilst above the 242 SOME NEW BRITISH ALBIAN OSTRACODA whole of it there is a narrow marginal shelf but no accommodation groove. In the right valve there is a high “ boss-like’’ anterior terminal tooth and a triangular, elongate, divided posterior tooth separated by a long straight locellate furrow. Anteriorly the median groove opens into a deep smooth circular socket. Above and below the median groove are narrow bars. REMARKS. A robusta langi has so far been found from the basal Upper Gault, H, orbignyi Subzone at Pinhay, Devon, at Swanage, Dorset and in the Isle of Wight. It occurs earlier than A. robusta s.s. which is found in the topmost Gault of the Wealden Area and East Anglia. A. vobustas.s. is distinguished from A. robusta langi by the greater lateral elongation of the alae and the weak development of the dorsal ridge which does not obscure the margin. The greater elongation of the alae makes the valves higher. A. vobusta langi is the earliest occurrence of the genus in the Cretaceous and specimens of Cythereis reticulata s.1. Jones & Hinde from the top Lower Gault have been found at Henfield showing weak ventral alation and typical Alatacythere shape though retaining the reticulation. Such forms are possibly intermediates between the two genera and indicate the origin of the genus A/latacythere. Family BYTHOCYTHERIDAE Genus MONOCERATINA Roth 1928 Monoceratina longispina (Bosquet) (Pl. 1, figs. 3-7) 1854 Cythere longispina Bosquet: 86. pl. 6. figs 7a d 1941 Monoceratina longispina (Bosquet); Bonnima: 40, pl. 6 fihs 67-76 1964b Monoceratina longispina (Bosquet); Kaye: 53, pl. 3 fig 1 1964 Monoceratina longispina (Bosquet); Szczechura: 388, pl, 3, fig 5 pl. 11, fig. 1: MATERIAL. B.M.N.H. Io. 2820-23 from the H. orbignyi Subzone, Upper Gault ; Wrotham, Kent. B.M.N.H. Io. 2824 from the FE. doris Subzone, Lower Gault ; Wrotham, Kent. MEASUREMENTS. Length Height (mm.) (mm.) Adult left valve (B.M.N.H. Io. 2820) : : : 0-82 0-42 Adult right valve (B.M.N.H.Io. 2821). : : 0-80 0-38 Juvenile carapace (B.M.N.H. Io. 2822) . ; : 0:65 0°35 Juvenile left valve (B.M.N.H. Io. 2823) . : : 0-56 0°22 DeEscripTion. The finding of additional material allows a fuller description of this species than that given by Kaye (19640). Valves elongate, inflated laterally. Dorsal and ventral margins straight and parallel. Cardinal angles well developed. Anterior margin broadly rounded ; posterior triangular, angled postero-dorsally. Lateral surface inflated with a prominent vertical median sulcus. Below the sulcus is a broad based conical, SOME NEW BRITISH ALBIAN OSTRACODA 243 laterally directed spine. This spine and associated swelling limit the sulcus strongly. Apart from a few rows of faint concentric reticulation along the crest of the alae the lateral surface is smooth. Some faint reticulation does, however, occur on the swollen antero-lateral part of the valves in certain specimens. The greatest width is just posterior to the mid-point of the valves. The marginal parts of the valves are rather flattened particularly anteriorly and ventrally. The hinge consists of a long narrow groove in the right valve which accommodates the long, straight, smooth marginal bar of the left valve. Weak false sockets are sometimes developed, particularly anteriorly in the left valve. The porcellaneous preservation of the specimens obtained make the radial pore canals and muscle scars invisible, normal pore canals are, howeve1, rather small and irregularly scattered over the lateral surface. Juveniles are rather dissimilar to the adults. In general they have a rounded alate expansion and no spine. This difference is, however, seen in the adults of certain species of Monoceratina (particularly from the British Oxfordian) and is possibly a dimorphic feature. By no means all the juveniles of MW. Jongispina are just alate and a few specimens with spines have been found. The occurrence of specimens without spines is not common to all species of Monoceratina and no specimens without spines belonging to M. wmbonata Williamson have been foundin the Gault even though the species is much more abundant than M. longispina. The juveniles of M. longis- pina are themselves variable in ornamentation and two instars are illustrated. Penul- timate instars have a wide marginal flattened area. The inflated alate expansion is strongly reticulate and a large node covered with reticulation occurs anterior to the sulcus. A narrow ridge crosses the sulcus and runs on to the postero-lateral area. Smaller instars are much less inflated and lack the flattened marginal area. They are completely smooth and have the sulcus wider and more open. Besides being strongly limited ventrally the sulcus is weakly limited dorsally giving the form of a wide median depression. Remarks. This species is not common but occurs consistently throughout the Gault. The original description was from the Senonian and I have recorded it from the Cambridge Greensand at Barrington. It has so far been found in the Gault of the Wealden area at Folkestone, Wrotham, Sevenoaks and Henfield where specimens occur in clays ranging in age from A. intermedius Subzone to H. varicosum Subzone. It has not been found in the sfatht Subzone assemblages but probably occurs in the higher subzones of the Upper Gault which have not yet been studied. The closest species of M. longispina seems to be M. parallela Alexander 1934 from the Santonian of Texas which is a little smaller and more strongly inflated. Good details of M. parallela are lacking and it is possible that the two forms are synonymous. WM. longispina differs from M. acanthoptera (Marsson) 1880 in that the latter is much smaller and has the spine set much further back on the carapace. The form recorded as M. acanthoptera (Marsson) by Alexander (1934) probably belongs to M. parallela. As shown earlier MM. wmbonata acanthoptera (Jones & Hinde) is a separate species viz. M. umbonatoides Kaye 1964). 244 SOME NEW BRITISH ALBIAN OSTRACODA Monoceratina sp. (Pl. 11, figs. g—10) MATERIAL. A right valve, B.M.N.H. Io. 2945, from the H. spathi Subzone, Middle Albian ; Devizes, Wiltshire. MEASUREMENTS. Length Height (mm.) (mm.) Right valve (B.M.N.H. Io. 2945) . ' : : 0°55 0:25 DESCRIPTION. Valves elongate, with straight parallel dorsal and ventral margins. Lateral surface inflated with a vertical median sulcus, limited ventrally by an alate expansion. The crest of the ala bears a short broad based conical spine. This spine is set just behind the ventral end of the sulcus and is laterally directed. A large smooth semicircular swelling occurs dorsally on the antero-lateral surface whilst a short slightly curved ridge runs near to the dorsal margin on the postero-lateral surface. A weak anterior marginal rib occurs which is not joined to the ala ventrally. The ventral and posterior marginal areas are flattened. The lateral surface is smooth. REMARKS. Only a single specimen of this species has so far been found, which seems to bear no relationship to other species of the genus found in the Albian and occurs earlier than them (in the H. spatht Subzone). Its closest relative is M. bonnemai Kaye 19646 from which it differs in size and ornamentation and in having only one laterally directed spine. Family PROGONOCYTHERIDAE Genus ACROCYTHERE Neale 1960 Acrocythere striata sp. nov. (Pl. 4, figs. 4-10) DERIVATION OF NAME. Striatus L. = striate. Diacnosis. Acvocythere with ornamentation of numerous longitudinal striate ridges. Hototyre. A left valve, B.M.N.H. Io. 2853, from the basal Upper Gault (Upper Albian) ; Pinhay, Devon. PARATYPES. B.M.N.H. Io. 2851-52, 2854-57. Eight valves and two carapaces from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2853, holotype) ‘ : 0:58 0:28 Right valve (B.M.N.H. Io. 2852) . : ‘ ‘ 0:58 0°23 SOME NEW BRITISH ALBIAN OSTRACODA 245 DESCRIPTION. Valves small, elongate, strongly compressed laterally. Dorsal and ventral margins long and straight ; tapering slightly posteriorly in the left valves but parallel in the right valves. Dorsal margin without marked cardinal angles. Anterior margin broadly rounded ; posterior margin pointed at mid-height. Greatest height at }length ; greatest width at mid-length. Lateral surface ornamented by numerous longitudinal striate ridges. The ridges continue over the whole of the lateral surface being usually ten in number. Anteriorly they tend to swing slightly ventrally and certain of them coalesce. Vertical cross ribbing is absent, the area between the ridges being finely pitted. Eye spots are absent. Duplicature fairly broad crossed by few, very fine, simple radial pore canals (8—10 anteriorly). Normal pore canals rather few, confined to the crests of the ridges. Inner margin and line of concrescence coincide throughout. Muscle scars not seen. Hinge strong merodont consisting in the left valves of two deep strongly divided terminal sockets separated by a long, straight, coarsely denticulate bar. In the right valve there are two triangular terminal teeth, highest away from the centre of the valve and each divided into four large denticles. Between the terminal teeth there is a long, straight coarsely crenulate (almost interdentate) furrow. REMARKS. A. striata differs considerably from the other members of the genus in ornamentation and lack of even a rudimentary median sulcus and eye spots. Its dissimilarities are such that it could possibly belong to a new genus. The hinge and basic longitudinal striate ornamentation are strongly reminiscent of Plewrocythere and Lophocythere. Genus NEOCYTHERE Mertens 1956 Subgenus PHY SOC Y THERE Kaye 1963 Neocythere (Physocythere) tenuis sp. nov. (Pl. 6, figs. 14-17) DERIVATION OF NAME. Tenuis L. = thin. Diacnosis. Small Neocythere with thin shell and weakly developed hinge. Lateral surface devoid of ornamentation. Hototyre. A left valve, B.M.N.H. Io. 2889, from Bed 1, A. intermedius Subzone, Lower Gault ; Small Dole, Henfield, Sussex. PaRATYPES. B.M.N.H. Io. 2890-93 from the same horizon and locality. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2889 holotype). : : 0+50 0+30 Right valve (B.M.N.H. Io. 2890) . : : : 0°50 0:28 DEscripTion. Valves small, ovate, thin shelled. Dorsal margin straight ; ventral margin weakly convex and subparallel to it. Anterior and posterior margins broadly rounded. Lateral surface smooth, inflated. Greatest height at 4 length ; 246 SOME NEW BRITISH ALBIAN OSTRACODA greatest width at mid-length. In dorsal view valves convex with acute anterior and posteriorends. A slight ventro-lateral tumidity occurs. Duplicature rather narrow, crossed by few (7-8 anteriorly) thick, straight radial pore canals. Inner margin and line of concrescence coincide throughout. Normal pore canalsrather large, irregularly scattered over most of the lateral surface, but forming two concentric rows along the crest of the ventro-lateral expansion. Hinge weak having, in the left valves, two faintly divided sockets separated by a weakly denticulate bar. Above the median bar is a prominent marginal shelf but no accommodation groove. The terminal teeth in the right valve are low, triangular in shape and weakly divided. Muscle scars a slightly postero-dorsally inclined row of four oval scars with a U-shaped scar antero-dorsally and a small oval scar antero-ventrally of them. REMARKS. This species has been recorded from the Lower Gault A. intermedius Subzone at Henfield but is known to occur abundantly in the Upper Gault in East Anglia. It differs from other species of Neocythere s.1. in its shape, thinness of shell, weak hinge and lack of ornamentation. In shape and size it is nearest to N. (P.) pustulosa Kaye 1965a from the Upper Aptian of the Isle of Wight but lacks the strong ornamentation. Certain related undescribed specimens occur in the Aptian but they are larger and of unequal inflation. Family PROTOCYTHERIDAE Genus VEENIA Butler & Jones 1957 Veenia compressa Kaye (Pl. 11, figs. 13-15) 1965a Veenia compressa Kaye: 44, pl. 7, figs. 6, 7. MATERIAL. Two valves and one carapace, B.M.N.H. Io. 2868-70, from the basal Upper Gault ; Pinhay, Devon. REMARKS. This species previously recorded from the Upper Aptian of the Isle of Wight has now been found rather more abundantly and somewhat better preserved in the basal Upper Aptian at Pinhay in Devon. These latter specimens are identical although somewhat smaller (0-58 mm.). Veenia florentinensis Damotte (PI. x1, figs. 1-8) 1961 Veenia (Protoveenia) floventinensis Damotte : 102, pl. 1, figs. 1-3, pl. 2, figs. 1-6. MATERIAL. B.M.N.H. Io. 2974-75, two valves from the H. spatht Subzone Lower Gault ; Henfield. B.M.N.H. Io. 2971-73, 2976-79, ten valves and two carapaces from the H. orbignyt Subzone Upper Gault ; Pinhay, Devon. SOME NEW BRITISH ALBIAN OSTRACODA 247 MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H. Io. 2975) 0*50 0:28 Right valve (B.M.N.H. Io. 2974) 0°49 0°25 Left valve (B.M.N.H. Io. 2978) 0:70 0:36 Left valve (B.M.N.H. Io. 2971) 0:65 0°35 Right valve (B.M.N.H. Io. 2976) 0°58 0:28 Right valve (B.M.N.H. Io. 2977) 0°54 0:28 Left valve (B.M.N.H. Io. 2972) 0°50 030 Left valve (B.M.N.H. Io. 2979) 0-48 0°27 Right valve (B.M.N.H. Io. 2973) 0-48 0°25 REMARKS. Damotte erected this species in 1961 placing it within a new subgenus Protoveenia. This subgenus was differentiated primarily on a basis of having very few radial pore canals. Damotte’s specimens came from the Lower Gault and comparable specimens occur in the H. spathi and A. intermedius Subzones in Britain. Other specimens of this species occur at higher horizons particularly in Devon, Dorset and the Isle of Wight in clays from the H. orbignyi Subzone, Upper Gault and these specimens have a much larger size range than those from the Lower Gault. The Lower Gault forms have a length in the range of 0:40-0:50 mm., whilst specimens from the Upper Gault reach as much as 0-70 mm. in length. The largest specimens differ slightly from the smaller ones but specimens of typical Lower Gault size and features and all intermediates occur together with them in the same sample. The larger specimens are more inflated, have the longitudinal ribs less keel-like than the small ones and have the dorsal ridge and antero-dorsal hinge ear less well separated. They also have more radial pore canals and therefore make the subgenus Protoveenia unusable and its postulated ancestry to Veenta improbable. The smaller specimens in the Upper Gault are identical with Damotte’s Lower Gault forms and it appears that the species is therefore not only of variable size but that the number of radial pore canals is a direct function of the size. Sexual dimorphism is not known from the Lower Gault specimens but is well marked in the Upper Gault forms. Family TRACHYLEBERIDIDAE Genus CYTHEREIS Jones 1849 Cythereis angulatoides nom. nov. 1964a Cytheveis angulata Kaye : 327, pl. 54, fig. II. REMARKS. Professor W. A. Van den Bold of Louisiana State University has kindly pointed out that the form erected as C. angulata from the Upper Aptian of Surrey has the name preocupied by Sars (1866: 40). I have therefore renamed my form C. angulatoides. Geol. 11 5 24 248 SOME NEW BRITISH ALBIAN OSTRACODA Cythereis gatyensis Damotte & Grosdidier (oll, 1G, HES, 1st, 1) 1963 Cythereis ? gatyensis Damotte & Grosdidier : 58, pl. 3, figs. 8a—g. 1963b Cythereis lamplughi Kaye : 236, pl. 19, figs. 14-16. 1965a Cythereis lamplughi Kaye; Kaye: 46, pl. 7, figs. 14, 15. MATERIAL. B.M.N.H. Io. 2966-67. Two specimens from the H. spathi Subzone, Lower Gault ; Culham, Oxfordshire. REMARKS. Specimens kindly sent to me by Dr. E. Grosdidier show that Cytherers lamplught Kaye 19630 is synonymous with Cythereis gatyensis Damotte & Grosdidier 1963, the latter having two months’ priority. C. gatyensis is abundant but restricted to the H. spatht Subzone in the Gault. It is also known from the Upper Aptian of the Isle of Wight. It is a valuable index fossil for the H. spatht Subzone having been found in clays of that age at Speeton, West Heslerton, Culham and Henfield. Cythereis glabrella Triebel (Pl. ro, figs. 5-8) 1940 Cythereis glabrella Triebel : 196, pl. 6, figs. 60-62. MATERIAL. B.M.N.H. Io. 2962-63. Two valves from the basal Upper Gault at Pinhay, Devon. MEASUREMENTS. Length Height (mm.) (mm.) Left valve (B.M.N.H.Io.2962) : 0-90 0°55 Right valve (B.M.N.H. Io. 2963). ; : 0-90 0°50 Remarks. This smooth, inflated species is not common in the British Gault having been found so far only at Pinhay and in the top Red Chalk at Speeton, Yorks. It differs from the much more abundant C. folkstonensis Kaye 1964) in its less angular appearance, smaller size and in having a smooth rather than spined median rib. It is similar to C. nuda Jones & Hinde 1890 differing in being larger, more inflated and in having the median rib well developed. Cythereis pinhayensis sp. nov. (Pl. 9, figs. 1-8) DERIVATION OF NAME. After Pinhay Point, Devon, the only known occurrence of the species. DiAGNosis. Small, reticulate but not spined Cythereis with prominent muscle node and weakly convergent long margins. Horotype. 2.4 JAN 1966 — Coa = a S < a “ep, oe ‘BULLETIN OF BRITISH WEALDEN SHARKS BY COLIN PATTERSON, PhD: Pp. 281-350; 5 Plates; 31 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 7 LONDON : 1966 Issued THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), «instituted in 1949, 1s issued in five series corresponding to the Departments of the Museum, and an Historical series. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. In 1965 a separate supplementary series of longer papers was instituted, numbered serially for each Department. This paper is Vol. 11, No. 7 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1966 TRUSDEES OF THE BRITISH MUSEUM (NATURAL HISTORY II January 1966 ) Price L2ee2s. BRITISH WEALDEN SHARKS By COLIN PATTERSON CONTENTS Page I INTRODUCTION . é 5 - 5 : : : 5 Ai II Locatitirs : : : ° : 2 5 ; . 285 III SYSTEMATIC DESCRIPTIONS . : : 3 : ; 5 o Asi7 Class Selachii . : ‘ : ; : ‘ : 5 ely Order Hybodontiformes : : : é ; : 5 Bey Family Hybodontidae : : é : ‘ , 5 AS Hybodus basanus Egerton : : : 4 ; . 288 Hybodus ensis Smith Woodward : : : : . 292 Hybodus parvidens Smith Woodward . 5 : 2 5 2Ri8 Hybodus brevicostatus sp.nov. . F : ; ; . 300 Fin spines of Hybodus : ; 3 : ‘ : 3 LO Remarks on hybodont teeth . : “ : : 5 Byer Lonchidion breve breve sp. et ssp. nov. : : : pee iis} Lonchidion breve crenulatum ssp. nov. 2 ° 5 - 316 Lonchidion breve pustulatum ssp. nov. : f : a ER, Lonchidion striatum sp.nov. . : ; ; : 5 340 Lonchidion rhizion sp. nov. : ; : ; : “ie 322 Lonchidion heterodon sp.nov. . : : - 326 Fin spines and cephalic spines of onehidion 6 : . 328 The Affinities of Lonchidion : : E 5 : 2330 Family Ptychodontidae : : é : 2 382 Hylaeobatis ornata (Smith Woodward) : é ; a BER The Affinities of Hylaeobatis : : ; : é sso The origin of Hylaeobatis : 2 2 : - 340 Relationships within the Pechedontidac : é : Me 342 IV ECOLOGY AND RELATIONSHIPS OF THE FAUNA . : ; : . 346 V REFERENCES F ; : : : : : 6 6 . 348 SYNOPSIS New material, consisting mainly of abundant isolated teeth from bone-beds, allows more detailed and precise definition of the known sharks of the British Wealden and Purbeck, and contains five new species: Hybodus brevicostatus, Lonchidion breve (with three new subspecies), L. striatum, L. rhizion and L. heterodon. Samples from successive horizons allow the inter- relationships of the various species to be worked out in some detail. Hylaeobatis is shown to belong in the family Ptychodontidae and to lie near the ancestry of Ptychodus: the Ptycho- dontidae are probably derived from the hybodontid genus Lonchidion. The Wealden shark fauna is unusual in being from fresh water and in containing only hybodontoids : it is suggested that the hybodonts were able to escape from competition with more advanced selachians by entering fresh water: the radiation which they underwent there parallels their marine radiation at their first appearance, and explains the similarity between the shark fauna of the Wealden and of the marine Triassic. Certain of the more specialized Wealden forms seem to have returned to the sea and given rise to the Upper Cretaceous hybodonts and ptychodonts. GEOL. II, 7 27 284 BRITISH WEALDEN SHARKS I INTRODUCTION Tuis paper has been prompted by the arrival in the British Museum (Natural History) of two sets of new Wealden material. Between 1960 and 1962 Drs. W. A. Clemens and K. A. Kermack of University College, London, were searching the bone-beds of the British Wealden for mammalian remains. In the course of this work large quantities of bone-bed were broken down with formic acid, and the bone separated by bromoform flotation. The mammalian finds resulting from the work have already been described (Clemens 1960, 1963; Kermack, Lees & Mussett 1965), and Dr. Kermack and his colleagues were kind enough to present the residue of the treated bone-beds to this museum. This material is rich in fish remains, though the majority are rolled and water-worn. In 1961 Mr. J. F. Wyley of Richmond, Surrey, discovered a bone-bed in the Weald Clay in the Henfield Brick Company’s pit at Henfield, Sussex, and in many visits to the pit he has collected a quantity of fish material which he has generously presented to this museum. The material from Henfield is normally excellently preserved. Knowledge of British Wealden sharks is due almost entirely to Smith Woodward. Agassiz (1837) described several hybodont fin spines under various names, and Egerton (1845) described Hybodus basanus from the Weald Clay of the Isle of Wight and (1854) a fin spine from Tilgate as Astevacanthus granulosus. In 1889 Smith Woodward briefly redescribed these species, referred several teeth to Hybodus, and described Acrodus ornatus from the Upper Wealden. Later (Smith Woodward 1916), in his monograph on the fishes of the British Wealden and Purbeck, he gave more detailed descriptions of all these species and added two new species of Hybodus, H. ensis and H. parvidens, and a new genus and species, Hylaeobatis problematica, all based on isolated teeth. No new material of any of these species has since been described (H. basanus has been incorrectly recorded from the Cretaceous of Japan by Yabe & Obata 1930). In this paper a new species of Hybodus and four new species, one with three sub- species, of the hybodont genus Lonchidion Estes (1964) are described, and Acrodus ornatus and Hylaeobatis problematica are shown to be synonymous. The total list of British Wealden sharks known at present is therefore : Hybodus basanus Egerton *H. ensis Smith Woodward *H. parvidens Smith Woodward H. brevicostatus sp. nov. TH. striatulus Agassiz *Lonchidion breve sp. nov. *L. striatum sp. nov. *L. ylizion sp. nov. *L. heterodon sp. nov. +Asteracanthus granulosus Egerton (doubtfully Asteracanthus, see p. 310) *Hylaeobatis ornata (Smith Woodward) * Species known only by isolated teeth. + Species known only by isolated fin spines. BRITISH WEALDEN SHARKS 285 LE OCAE LETS The bulk of the new material described here is from four horizons, the Cliff End bone-bed, the Telham bone-bed, the Paddockhurst bone-bed and the Weald Clay of Henfield, Sussex. (a) Cliff End Bone-bed The Cliff End bone-bed, exposed on the foreshore at Cliff End, near Hastings, Sussex, the source of the mammal teeth discovered early in the century by Teilhard de Chardin and Pelletier, has recently been described by Allen (1960 : 11) and Clemens (1963 : 58), the latter also describing the method used to concentrate the vertebrate remains. The bone-bed is within the Ashdown Beds (of Valanginian age according to Hughes 1958), but its precise horizon is not yet established*. The Ashdown Beds are interpreted by Allen (1959) as a deltaic deposit, the delta flowing into a fresh- water lake. The great majority of the vertebrate remains are strongly rolled and abraded. Of the recognizable fragments about 45°% are shark teeth, 50% are teeth and fragments of dermal bones of actinopterygians (mainly Lepfidotes), and 5% or less are reptilian. The shark fauna includes : Hybodus ensis (common) H. parvidens (common) H. brevicostatus (rare) Lonchidion breve breve (moderately common) L. rhizion (ancommon) L. heterodon (rare) (o) Telham Bone-bed The Telham bone-bed is exposed at a number of localities in the south-eastern Weald (Allen 1949 : 279, text-fig. 45). The sample described here was collected by Mr. P. J. Whybrow from the exposure at Teigh Farm, Stone, Kent (GR TQ 937268). The bone-bed lies in the Wadhurst Clay, near the base (Upper Valanginian according to Hughes 1958), and is interpreted by Allen as the result of river water flooding over a delta plain. The vertebrate remains are in much the same condition as at Cliff End, rolled and abraded. Of the recognizable fragments, about 309% are shark teeth and the remaining 70° are almost entirely actinopterygian, mainly Lepidotes : reptiles account for only about 1% of the sample. The shark fauna includes : Hybodus ensis (rare) H. parvidens (common) Lonchidion breve breve (moderately common) L. rhixon (rare) This locality is referred to in the text as Telham bone-bed, Stone, Kent. * In a paper published while this work was in press, Kermack, Lees & Mussett (1965: 536) give further information on the Cliff End bone-bed, pointing out that the bed cannot now be located in the cliff or on the foreshore, and that the scattered blocks found on the beach probably come from an off-shore reef, which would place the bone-bed ‘‘ well down in the Fairlight Clays’’, near the base of the Wealden. Kermack, Lees & Mussett also publish a comment on the Cliff End fauna which I wrote on first seeing the material. I would not now infer brackish conditions from the abundant hybodont selachians: I think it probable that all the species present were capable of life in fresh water. 286 BRITISH WEALDEN SHARKS (c) Paddockhurst Bone-bed Clemens (1960, 1963 : 63) has given this name to a bone-bed in the Grinstead Clay (Hauterivian according to Hughes 1958) at Paddockhurst Park, near Turner’s Hill, Sussex. Allen (1959) interprets the Grinstead Clay as a fresh-water lake deposit. The vertebrate remains in the Paddockhurst bone-bed are neither so broken up nor so badly rolled as they are at Cliff End, and sharks are less common: shark teeth make up about 15% of the recognizable elements, reptiles about 109%, and the remaining 75% is actinopterygian, mainly scales and teeth of Lepzdotes. The shark fauna includes : Hybodus ensis (common) H. parvidens (common) H. brevicostatus (rare) Lonchidion breve breve (moderately common) L. breve crenulatum (moderately common) L. rhizion (rare) (d) Henfield The Henfield Brick Company’s pit (GR TQ/218143) at Henfield, Sussex, is a large pit in the Weald Clay which is being actively worked (Milbourne 1961 : 135). The beds dip northwards at about 5 degrees (Reeves 1947: 83). The highest beds exposed, in the north wall of the pit, are red and yellow clays without obvious fossils. These beds pass down into brown and grey clays. About 3 ft. below the red and yellow clays there is a band of ‘ Paludina’ limestone, 3-8 in. thick. Fishes occur in the clay immediately above the limestone. Below the limestone there is Io ft. of brown sandy clay, and below this about 25 ft. of grey clay in which six minor cyclo- thems, 3-4 ft. thick, are recognizable by increasing sand content giving a brownish tinge towards the top of each. Each of these cyclothems contains fish fragments, often concentrated into more or less well marked bone-beds which are occasionally cemented into a phosphatic limestone (Pl. 1, fig. 1). Below these cyclothems the floor of the pit is occupied by grey clay containing [gwanodon, lignite and occasional tree trunks, passing down in the south wall of the pit into barren red and yellow clays and sands, the lowest beds exposed. This sequence lies in Reeve’s Group II (1958 : 11), the middle division of the Weald Clay (Barremian according to Allen 1955 and Hughes 1958), probably near the top: Dr. F. W. Anderson has examined samples of ostracods collected from the various fish horizons and places the ‘ Paludina’ limestone at about 550 ft. below the top of the Weald Clay. Associated with the fishes there are abundant ostracods, all non-marine according to Dr. Anderson, and occasional charophyte oogonia, another indication of fresh water conditions. The fish material, although consisting almost entirely of dissociated fragments, is well preserved. There is no significant difference in the fish fauna of the various fish horizons, but in the minor cyclothems the fauna varies widely within the bone-beds: the block of bone-bed shown in PI. 1, fig. 1 shows abundant shark teeth of five species (teeth of Lonchidion breve and L. striatum are too small to show on the photograph), but the other side of the block, which is only 14 mm. BRITISH WEALDEN SHARKS 287 thick, shows only two or three teeth of Lonchidion, and is made up almost entirely of teleostean remains. Sharks make up less than 10% of the material, the bulk of which is teleostean. Actinopterygians identified include : Lepidotes mantelli Agassiz Coelodus mantelli Agassiz Caturus tenuidens Smith Woodward (PI. 1, fig. 2: not previously recorded above the Purbeck) Pachythrissops sp. Clupavus sp. The shark fauna includes : Hybodus basanus (common) H. parvidens (rare) H. brevicostatus (moderately common) Lonchidion breve breve (common) Lonchidion striatum (common) Hylaeobatis ornata (common) (e) Other localities Small but valuable samples from two other localities are also described here. The first was collected by Mr. I. M. West of the University of Southampton from a limestone above the Broken Shell Limestone in the Upper Purbeck exposed 100 yards east of Friar Waddon farm, near Upwey, Dorset (GR SY/643858) (see Anderson 1958 : 119, 129, text-figs. 21, 22). Mr. West thinks this horizon is equivalent to the Umio Bed of the type Purbeck. The second sample was collected by Mr. P. J. Whybrow from a limestone in the Wadhurst Clay exposed in a cutting on the east side of the road 200 yards north of Homeland, Ashurstwood, Sussex (GR TQ/419363). These two localities are referred to in the text as Friar Waddon and Ashurstwood respectively. III SYSTEMATIC DESCRIPTIONS Class SELACHITI Order HYBODONTIFORMES Family HYBODONTIDAE Owen 1846 Diacnosis. See Berg (1955 : 61) Genus HYBODUS Agassiz 1837 : 41 Diacnosis. See Smith Woodward (1916: 3), but delete ‘ palatoquadrate not articulated with the postorbital region of the skull’. TYPE spEcIES. Hybodus reticulatus Agassiz. H., basanus Egerton, H. ensis Smith Woodward and H. parvidens Smith Woodward, the three species of the genus known by teeth in the British Wealden, are easily distinguished in the type material described by Smith Woodward but appear to be linked by intermediate forms in the new material. H. basanus is the only species of the three in which the complete dentition is known, and will be described first. 288 BRITISH WEALDEN SHARKS Hybodus basanus Egerton (Pl. 1, fig. 1; Text-figs. 1-3) 1845 Hybodus basanus Egerton : 197, pl. 4. 1889 Hybodus basanus Egerton; Smith Woodward : 273, pl. 12, figs. 1-5. 1891 Hybodus basanus Egerton ; Smith Woodward : 63, pl. 1, pl. 2, fig. 1. 1898 Orthybodus basanus (Egerton) Jaekel : 139. 1916 Hybodus basanus Egerton ; Smith Woodward : 5, pl. 1, figs. 1, 2; pl. 2, fig. 1; text-figs. 1919 Hy bodus basanus Egerton ; Smith Woodward : 139, pl. 26, fig. 3. AMENDED DIAGNOSIS. Hybodus known by skull, dentition, fin spines and frag- ments of postcranial skeleton ; nine or ten files of teeth in each ramus of the jaws, symphysial file present in lower jaw, dentition weakly heterodont ; teeth reaching I5 mm. in length, central cusp high, slender (ratio of height above root/crown junction to length 1-0-2:0) and arched lingually ; three or (rarely) four pairs of lateral cusps ; labial face of crown with rather numerous fine striae, often bifurcated basally, reaching tips of lateral cusps and covering from one-third to two-thirds of central cusp ; lingual face of crown with similar but longer striae almost reaching tip of central cusp ; teeth without accessory cusps on the labial or lingual margins ; root low, bent lingually ; a single pair of cephalic spines with terminal barb; fin spines slender, compressed, not much arched, reaching about 20 cm. in length, lateral faces with 8-12 fine, sharp ridges, posterior denticles in two series, small and closely set. HoLotyPE. GSM No. 27973, Geological Survey & Museum, London, from the Weald Clay of Atherfield, Isle of Wight. MatTeErRIAL. In addition to the holotype, about twenty skulls, fragments of vertebral column, fin spines and isolated teeth in the British Museum (Natural History) and the Geological Survey & Museum. HorIZON AND LOCALITIES. Weald Clay: Atherfield, Isle of Wight ; Pevensey Bay, Sussex ; Henfield, Sussex ; Bexhill, Sussex. DEscrRIPTION. Hybodus basanus is the only species of the genus in which the skull is well known. Smith Woodward (1916) has given a good description of the skull based on several well preserved heads in the British Museum (Natural History). Of the palato-quadrate he writes ‘it can scarcely have articulated with the post- orbital prominence of the neurocranium’. Smith (1942: 701) has questioned this, noting that in the Liassic H. hauffianus Fraas the skull is said to be amphistylic by Jaekel, and that it is amphistylic in Synechodus. Berg (1955 : 62) has also suggested that Smith Woodward’s observation needs checking. I have examined all of the skulls of H. basanus in the British Museum (Natural History). The palatoquadrate and hyomandibular are best shown in P.2082a and P.6103. There can be no doubt that Smith Woodward’s reconstruction (1916, text-fig. 3) is accurate in showing the hyomandibular as being large and much higher than the postorbital part of the palatoquadrate. In this H. basanus differs from typical amphistylic sharks such as Hexanchus, where the hyomandibular is slender and no higher than the otic process BRITISH WEALDEN SHARKS 289 Fic. 1. Hybodus basanus Egerton. Tooth from the first or second file of the upper jaw, right side, in labial (a), lingual (B) and medial (c) view. P.11871, Weald Clay ; Pevensey Bay, Sussex. of the palatoquadrate, and from Heterodontus and Hybodus hauffianus (Jaekel 1906) in which the hyomandibular is stout but equal in height to the otic process of the palatoquadrate. I do not think it is possible, however, to decide whether the palatoquadrate of H. basanus had a post-orbital articulation or not. The otic process was certainly tucked well up below the postorbital process of the neuro- crahium, but there is no sign of an articular facet or condyle on the otic process. 290 BRITISH WEALDEN SHARKS It is possible that the suspension was amphistylic, but the long, large hyomandibular shows that the species tends towards the hyostylic condition. Smith Woodward describes the dentition of H. basanus as consisting of ten or eleven paired files of teeth in each jaw, with a symphysial file in the mandible. The teeth (Text-fig. 1) have a high, slender central cusp and three pairs of lateral cusps which are well separated from the central cusp. The height of the central cusp (above the root /crown junction) is only slightly less than the length of the tooth in anterior teeth, but the height decreases posteriorly. The central cusp curves lingually rather strongly, but the tip curves labially again, giving a weakly sigmoid outline to the cusp in medial view (Text-fig. 1B). Of the three pairs of lateral cusps, the innermost is sharply pointed, striated to its tip, and normally about one-third as high as the central cusp. The second pair of lateral cusps is shorter but similarly ornamented. The outermost cusps are normally very small and almost smooth. A minute fourth lateral cusp is occasionally present. Fic. 2. Hybodus basanus Egerton. A. Tooth from the first file of the upper jaw, right side, in labial view. P.2082. Bs. Tooth from the sixth file of the lower jaw, left side, in labial view. P.6356. c. Tooth from the eighth file of the lower jaw, left side, in labial view. P.2082b. All from the Weald Clay of Pevensey Bay, Sussex. BRITISH WEALDEN SHARKS 291 The base of the labial face of the crown is ornamented with moderately fine, parallel or sub-parallel vertical striae, often bifurcating basally, which extend to the tips of the lateral cusps and to about one-third or half of the height of the central cusp. There is a good deal of variation in the strength, number and length of these striae: they are closely packed, weak and short in some fish (P.20826, Text-fig. 2c), and coarse, well spaced and long, covering more than half the labial face of the central cusp, in others (P.11871, Text-fig. ra). On the lingual face of the crown the ridges are stronger, and extend almost to the tip of the central cusp (P.2082a, P.11871, Text-fig. 1B). The root is low, and is turned lingually, with the labial face of the root lying almost at right angles to the axis of the central cusp. This description is based on the teeth of the complete skulls from the Weald Clay of Pevensey Bay, Sussex, and Atherfield, Isle of Wight. It is difficult to match these teeth exactly in material from any other Wealden horizon or locality. In the new material from the Weald Clay of Henfield, which is approximately contemporary with the type material of H. basanus, there is a number of teeth of the type shown in PI. 1, fig. 1 and Text-fig. 3. In these teeth the crown is lower than in those from the complete skulls of H. basanus, with a ratio of height (above the root/crown junction) to length of little less than 2-0, the striae are less numerous and are longer, reaching well over half the height of the central 2mm Fic. 3. Hybodus basanus Egerton. Lateral tooth in lingual (a), distal (B) and labial (c) view. P.46921, Weald Clay ; Henfield, Sussex. 292 BRITISH WEALDEN SHARKS cusp on the labial face (the tooth shown in Text-fig. 3 has shorter striae than most teeth from Henfield) and to the tip of the cusp on the lingual face, and both the central cusp and the root are not turned lingually so strongly as they are in typical H. basanus (cf. Text-figs. 1c, 3c). In all these features the teeth from Henfield are intermediate between typical H. basanus from Atherfield and Pevensey, and typical teeth of H. parvidens from the Ashdown Beds and Wadhurst Clay (Text-figs. 6, 7). The Henfield teeth are referred to H. basanus rather than H. parvidens because they never show the knob or accessory cusp at the base of the labial face of the central cusp which is common in H. parvidens, because the striae are almost as numerous as they are in H. basanus and only rarely reach the tip of the central cusp, and because there is a difference in the extent of the striae on the labial and lingual faces of the central cusp (in H. parvidens the length of the striae is similar on the two faces of the central cusp). No teeth referable to H. basanus are known from horizons above or below the Weald Clay. The tooth from the Lower Cretaceous of Japan referred to H. basanus by Yabe & Obata (1930 : 4, pl. 2, fig. 3) differs from this species in the longer and more slender central cusp and in the almost complete lack of striae on the labial face : it is possibly a Synechodus. Leriche (1911 : 457, pl. 6, fig. 2) has referred to H. basanus a fragment of fin spine from the Lower Neocomian of the Paris Basin, but this identification is very doubtful: the fragment could as well belong to another species. AFFINITIES. Smith Woodward (1916 : 10) considered that teeth and spines from horizons in the lower part of the Wealden showed resemblances to H. basanus without being referable to the species. These forms (see p. 294) are here included in H. ensis which in the Wealden seems to trend towards H. basanus in the form of the teeth, but is probably not related. Teeth of H. basanus from Henfield are intermediate in a number of characters between more typical teeth of the species from Pevensey and the Isle of Wight, and teeth of H. parvidens from lower horizons. This, together with the trend towards H. basanus shown by H. parvidens in the Grinstead Clay (see p. 299), leaves little doubt that H. basanus is a species confined to the Weald Clay (probably to the upper part alone) which evolved directly from H. parvidens by increase in size, in crown height and in the number of striae. Hybodus ensis Smith Woodward (Text-figs. 4, 5) 1889 Hybodus sp. inc. (? stvictus Agassiz) Smith Woodward : 275. 1889 Hybodus sp. inc. (? striatulus Agassiz) Smith Woodward : 276, pl. 11, figs. 14, 15. 1916 Hybodus ensis Smith Woodward : 11, pl. 2, figs. 2-6; non fig. 7. AMENDED DIAGNOSIS. Hybodus known only by isolated teeth: teeth large, up to 2 cm. in length, central cusp high (ratio of height above root/crown junction to length 1-0-1:5), broad at base, evenly tapering and moderately compressed, not much arched lingually ; two pairs of lateral cusps, inner lateral cusps moderately high, slender, pointed, well marked off from central cusp but close to it ; labial face of BRITISH WEALDEN SHARKS 293 crown with many fine, parallel striae reaching tips of lateral cusps but covering only basal quarter or fifth of central cusp; lingual face of crown with coarser striae extending about half way up central cusp; root low, bent lingually. Hototyre. BMNH No. 21349 (Smith Woodward, 1916, pl. 2, fig. 6), tooth without root, Middle Purbeck ; Swanage, Dorset. MATERIAL. In addition to the holotype, about one hundred isolated and frag- mentary teeth. HoRrIzONS AND LOCALITIES. Middle Purbeck: Swanage, Dorset. Upper Purbeck : Friar Waddon, Dorset. Ashdown Beds: Cliff End, Sussex. Wadhurst Clay: Teigh Farm, Stone, Kent; Hastings, Sussex. Grinstead Clay: Paddockhurst Park, Sussex ; Tilgate Forest, Sussex. DESCRIPTION. Smith Woodward based this species on incomplete teeth from the Middle Purbeck of Dorset, but also recorded it from the Wealden of Tilgate Forest (Grinstead Clay—see p. 294). All the specimens labelled as H. ensis by Smith Woodward are from the Purbeck. Fic. 4. Hybodus ensis Smith Woodward. Tooth in labial (a), lingual (B) and medial (c) view. 21349c, Middle Purbeck ; Swanage, Dorset. 294 BRITISH WEALDEN SHARKS The species is poorly known, the Purbeckian type material consisting of teeth without roots exposed in labial view. One of these teeth, 21349c, has been removed from the matrix to expose the lingual face (Text-fig. 4). In the Purbeck material of H. ensis the crown is about as high as in H. basanus, with the ratio of length to height (above the root/crown junction) about 1-0, but the central cusp is much broader, the ratio of the breadth at its middle point to its height ranging from 2-8 to 3:0, compared with 4-0 to 5-0 in H. basanus. In no specimen of H. ensis are there more than two pairs of lateral cusps (21349), Smith Woodward 1916, pl. 2, fig. 7, has three pairs of lateral cusps but does not belong in H. ensis—see H. parvidens, p- 297). In H. ensis the labial face of the crown (Text-fig. 4A) bears fine, close packed, parallel striae which extend almost to the tips of the inner lateral cusps but cover only the basal one-fifth or quarter of the central cusp. On the lingual face of the crown (Text-fig. 4B) the striae are less numerous, a little coarser and cover almost one-third of the height of the central cusp. The central cusp curves lingually less strongly than it does in H. basanus (cf. Text-figs. 1c, 4c). Other apparent differ- ences between H. ensis and H. basanus suggested in Smith Woodward’s original description—that the inner lateral cusps are higher and closer to the central cusp in H. ensis, and that the ratio of height to length of the tooth is greater in this species— are not confirmed by measurement of the teeth. In 1889 (p. 276) Smith Woodward catalogued as “ Hybodus sp. inc. (? striatulus Agassiz) ’ about thirty teeth from the Wealden of Tilgate Forest,! noting the high, moderately broad central cusp, the high, slender inner lateral cusps, and the striae “rarely extending more than half the height of the median cone’. In rg16 (p. 10) he referred to these teeth as ‘ of the same general type as those of H. basanus .. . but not sufficiently similar to be referred with certainty to this species’. Most of these teeth are more or less rolled and waterworn, and many are quite indeterminable. Some, such as 48377 (Tilgate Forest) and P.6353 (Hastings, unknown horizon) agree exactly with the Purbeck specimens of H. ensis, and leave no doubt that this species extended into the Wealden. But the majority of these teeth do not agree exactly with the type material of H. ensis : 2693 (Tilgate Forest, Text-fig. 5, Smith Woodward 1889, pl. 11, fig. 14) is the best preserved of these, and is typical in shape. These teeth differ from H. ensis in the more slender, sharply pointed central cusp (ratio of breadth measured at the middle point to height c. 3-5), but they agree with H. ensis and differ from H. basanus in their large size, weakly arched central cusp (Text-fig. 5B), two pairs of lateral cusps, in the short, fine striae on the labial face of the crown, rarely reaching more than one-third of the height of the central cusp, and in the short striae on the lingual face of the crown. The striae on the lingual face of the central cusp extend about as far as they do in typical H. ensis, and are charac- teristically longer near the margins of the cusp than they are in the centre. In my opinion these teeth should be included in H. ensis. Some of these Wealden teeth (26026 ; Smith Woodward 1889, pl. 11, fig. 14: 26024, both from Tilgate Forest) 1 Mantell’s horizon at Tilgate Forest, previously thought to lie in the Upper Tunbridge Wells Sand, directly below the Weald Clay (Topley 1875: 92), has recently been re-identified as in the Cuckfield Stone, in the middle part of the Grinstead Clay (Stubblefield 1963 : 37). BRITISH WEALDEN SHARKS 295 2mm Fic. 5. Hybodus ensis Smith Woodward. Tooth in labial (a) and medial (B) view. 2693, Grinstead Clay ; Tilgate Forest, Sussex. tend towards H. basanus in having a more slender central cusp with rather coarse striae covering almost half of the labial face, but they have only two pairs of lateral cusps : 26024 is particularly like the typical Weald Clay H. basanus, but it seems probable that it is not an early example of this species but a tooth of H. ensis in which the evident trend towards narrowing of the central cusp and extension and coarsening of the ornament has gone farther than in other examples. Among the new material from the Cliff End and Paddockhurst bone-beds there are no complete large teeth, but isolated central cusps and imperfect crowns are very common. All of these large teeth can be referred to H. ensis: they show a range in breadth of the central cusp from stout forms like the Purbeck H. ensis to slender forms approaching the proportions of H. basanus, but the striae on the labial face are always short, fine and closely packed, and on the lingual face they are always shorter than in H. basanus and are longer at the margins of the central cusp than in the centre. No examples of H. ensis are known from horizons above the Grinstead Clay. AFFINITIES. H. ensis is a species which ranges from the Middle Purbeck to the Grinstead Clay (Paddockhurst bone-bed and Tilgate Forest). In the Wealden the teeth show a trend towards narrowing of the central cusp, sometimes accompanied by coarsening and lengthening of the striae which may produce teeth closely similar to those of H. basanus in shape. 296 BRITISH WEALDEN SHARKS Smith Woodward (1916: 11) noted the similarity between the Purbeck forms of H. ensis and the Middle and Upper Jurassic H. gvossiconus Agassiz. Extremely similar to H. ensis is H. songaensis Saint-Seine (1962 : 4, pl. 6, fig. 6), a species based on a single tooth from the Songa Beds of the Congo, probably marine and of Kim- meridgian age, which could well be synonymous with H. ensis. H. ensts is possibly derived directly from these marine Jurassic forms. Hybodus parvidens Smith Woodward (Text-figs. 6-9) 1889 Hybodus sp. inc. Smith Woodward : 276, pl. 11, fig. 16. 1916 Hybodus ensis Smith Woodward : pl. 2, fig. 7 (evvore). 1916 Hybodus parvidens Smith Woodward : 12, pl. 2, figs. 8-14. 1949 Hybodus parvidens Smith Woodward ; Allen: 277 (name only). AMENDED DIAGNOSIS. Hybodus known only by isolated teeth: teeth small, less than 10 mm. in length ; central cusp moderately high, not much compressed, lower and broader in posterior teeth ; ratio of height of central cusp (above root/crown junction) to length of tooth between 1:5 and 3:0 ; three or four pairs of lateral cusps ; labial face of crown with few coarse striae, often bifurcating basally, which reach the tips of the lateral cusps and commonly reach the tip of the central cusp except in some high anterior teeth, striae on lingual face similar ; knob or accessory cusp frequently present at base of labial surface of central cusp, no other accessory cusps ; root moderately deep, turned lingually a little. Horotyre. BMNH P.11877, tooth without root, Wadhurst Clay, Hastings. MATERIAL. About two hundred isolated teeth. HorIZONS AND LOCALITIES. Middle Purbeck : Swanage, Dorset. Upper Purbeck : Friar Waddon, Dorset. Ashdown Beds: Cliff End, Sussex; Fairlight, Sussex. Wadhurst Clay: Teigh Farm, Stone, Kent; Hastings, Sussex; Rye, Sussex. Grinstead Clay : Paddockhurst Park, Sussex ; Tilgate Forest, Sussex. Weald Clay: Henfield, Sussex. DESCRIPTION. Smith Woodward based this species on incomplete teeth from the Wadhurst Clay, in none of which was the root preserved. The holotype and two of the paratypes are shown in Text-fig. 6. Among the new material from the Cliff End bone-bed there are several more or less complete and unworn teeth (Text-fig. 7). The high-crowned teeth (Text-figs. 6, 7A) are presumably anterior, the low-crowned (Text-fig. 7B) posterior. In this typical material from the Ashdown Beds and Wadhurst Clay, the central cusp is moderately high in anterior teeth, with a ratio of height of cusp to length of tooth of about 1-8-2:0. In posterior teeth the central cusp becomes much lower, the ratio of height to length reaching about 3-0. The central cusp is not compressed as it is in H. ensis and H. basanus, and does not curve lingually. There are always three pairs of lateral cusps, and often four. On the labial face of the crown the striae are coarse, sparse, and often bifurcated basally. They reach the tips of the lateral cusps BRITISH WEALDEN SHARKS 297 2mm Fic. 6. Hybodus pavvidens Smith Woodward. Teeth from the Wadhurst Clay of Hastings, Sussex, in labial view. A. P.11877, the holotype. 8B. P.11878 (paratype). c. P.11880 (paratype). and commonly reach the tip of the central cusp: in a sample of 37 teeth from Cliff End, the central cusp is striated to its tip in 22 ; it is striated to the tip in 10 of 13 posterior teeth and in 12 of 24 anterior teeth. On the lingual face of the crown the ornament is very similar. It is interesting to note that in low-crowned posterior teeth, the striae on the lingual face of the crown tend to anastomose basally and form a reticular pattern (Text-fig. 78) approaching the pattern found in this region in low-crowned species like H. delabecher and H. brevicostatus (Text-fig. 10). At the base of the labial face of the central cusp a knob or accessory cusp is very commonly present. This accessory cusp is larger and more sharply defined in posterior teeth than in anterior (cf. Text-fig. 7A, B). In the Cliff End material this accessory cusp is present in more than half the teeth (in 28 of a sample of 40), and is commoner in posterior teeth (present in 13 of 16 posterior teeth and in 15 of 24 anterior teeth). The root in H. parvidens is typically hybodont, with foramina irregularly distributed on both surfaces, is moderately deep, and is turned lingually a little (Text-fig. 7). The above description is based on teeth from the Lower Wealden Ashdown Beds and Wadhurst Clay. Smith Woodward (1916: 12) also recorded the species from the Weald Clay of Berwick, Sussex, but I have been unable to trace the specimen on which this record was based: it was probably an example of H. brevicostatus. H. parvidens also occurs in the Purbeck, as is shown by a sample of sixteen incomplete teeth from Friar Waddon, Dorset, and by 21349) and 213409d, two teeth from the Middle Purbeck of Swanage, the first figured as H. ensis by Smith Woodward (1916, pl. 2, fig. 7). The picture presented by these Purbeck teeth (Text-fig. 8) is rather different from that in the Wealden material described above. The Purbeck teeth agree with the Wealden forms in size, but in most of them the crown is rather high, and where the ratio of crown height to tooth length is measurable (21349), d) it is GEOL, 11,7 28 298 BREDISH WAT DEIN SiRARECS 2mm Fic. 7. Hybodus parvidens Smith Woodward. An anterior (A) and a posterior (B) tooth in labial (1), lingual (2) and medial (3) view. P.46930—-31, Ashdown Beds, Cliff End Bone-bed ; Cliff End, Sussex. lower (I:5-I-7) than is usual in Wealden teeth. The striae on the labial face of the central cusp reach the tip in only four of the eighteen teeth: in six they cover between two-thirds and three-quarters of the cusp, in four about half the cusp, and in four less than half. Also, there is no accessory cusp at the base of the central cusp in any of the teeth, though there is an incipient cusp in two of them (Text-fig. 83). But in spite of these differences, I have little doubt in referring these teeth to H. parvidens : the examples with lower crowns exactly match those typical H. parvidens in which the accessory cusp is absent. | fj 2mm Fic. 8. Hybodus parvidens Smith Woodward. Fragmentary teeth in labial view. a. 213409d, Purbeck ; Swanage, Dorset. B. P.46959, Upper Purbeck ; Friar Waddon, Dorset. BRITISH WEALDEN SHARKS 299 Among the material from the Paddockhurst bone-bed (Grinstead Clay) complete teeth are rare, but small central cusps and incomplete teeth (Text-fig. gA) are com- mon. Teeth of H. parvidens which agree with the typical forms from the Ashdown Beds and Wadhurst Clay occur, but are rather rare. In a sample of 37 teeth, four are typical low-crowned posterior teeth of H. parvidens, the other 33 are high- crowned forms. Of these 33 teeth, only three are striated to the tip of the central cusp (cf. 12 out of 24 in high-crowned teeth from Cliff End), and only three have an accessory cusp (cf. 15 out of 24 at Cliff End). The striae in these teeth also tend to be finer and more numerous than in typical H. parvidens. , ' , A 7mm Fic. 9. Hybodus parvidens Smith Woodward. Teeth in labial view. a. P.46944, Grinstead Clay, Paddockhurst Bone-bed; Paddockhurst Park, Sussex. B. P.46971, Weald Clay ; Henfield, Sussex. Among the material from Tilgate Forest (Grinstead Clay) small teeth are very rare, but there are two examples of H. parvidens, 2850 and 3146. Among the material from the Weald Clay of Henfield I can find only two teeth which are referable to H. parvidens (Text-fig. 9B) ; these are typical examples, very like teeth from the Grinstead Clay. Apart from these two specimens, all the high- crowned teeth from Henfield are of the type shown in Text-fig. 3, which are assigned to H. basanus for the reasons given on p. 292. The smaller low-crowned teeth (Text- fig. 13) are placed in H. brevicostatus for the reasons given on p. 306. H. parvidens is not known from any other locality in the Weald Clay or from any higher horizon. AFFINITIES. The various samples of teeth of H. parvidens described above suggest that the history of the species was as follows. The species first appeared in the 300 BRITISH WEALDEN SHARKS Middle Purbeck. Purbeckian teeth are rather high-crowned, with no accessory cusp and with rather short, fine striae. In the Lower Wealden (Ashdown Beds and Wad- hurst Clay) the species reached its typical and most distinctive form, with a crown which is of moderate height and has both striae to its tip and an accessory cusp in the majority of teeth. In the Grinstead Clay the species seems to trend towards the Purbeck form again, with an increase in crown height, a reduction in the height and an increase in the number of the striae on the central cusp, and a reduction in the incidence of the accessory cusp. Only two undoubted examples of H. parvidens are known above the Grinstead Clay. It seems very probable that the species did not become extinct, but evolved into H. basanus by a further increase in crown height and in the number of striae, coupled with an increase in size. The increase in size which marks the transition to H. basanus in the Weald Clay is perhaps correlated with the apparent extinction of the large-toothed H. ensis shortly before the Weald Clay was deposited. It is shown below (p. 308) that the Wealden (Ashdown Beds to Weald Clay) species H. brevicostatus is also probably an offshoot of H. parvidens which originated near the base of the Wealden. Smith Woodward (1916 : 12) suggested a possible relationship between H. parvidens and the Upper Jurassic H. obtusus, a species in which the crown is moderately high, coarsely striated, and in which accessory cusps are very common. But the form of the earliest examples of H. pavvidens in the Purbeck, where there is no accessory cusp and where the crown is higher and less strongly striated than it is in the typical forms, suggests that the species did not originate from H. obtusus but from some unknown small, high-crowned species in the Upper Jurassic. Hybodus brevicostatus sp. nov. (Pl. 1, fig. 3; Pl. 2; Pl. 3, figs. 1-3: Text-figs. 10-13) Diacnosis. Hybodus known by almost complete dentition and fin spines: nine paired files and a median symphysial file probably present in each jaw, dentition moderately heterodont ; teeth ranging (in one individual) from 8-0-18-5 mm. in length ; crown low and long, ratio of crown height (above the root/crown junction) to tooth length 3:25-6:25, crown not deeper than the root ; central cusp low, lateral cusps four or more, sometimes not recognizable in posterior and lateral teeth ; several accessory cusps often present on both labial and lingual margins of the crown, especially in the lower jaw ; crown with longitudinal occlusal crest and many coarse striae, often bifurcating basally, which reach the tips of the cusps and are interspersed in larger teeth with finer striae which fail to reach the tips of the cusps ; at the base of the lingual face of the crown the striae anastomose in a reticular pattern ; root deep, not turned lingually. Fin spines reaching about 18 cm. in length ; anterior edge keeled, lateral faces with 8-10 narrow, well spaced, discon- tinuous striae, posterior face with raised band bearing recurved denticles which lie in a single irregular series with occasional doubling, denticles extending proximally beyond the limit of the striae on the lateral surfaces. BRITISH WEALDEN SHARKS 301 HototyPe. BMNH no. P.46973 (Pl. 2; PI. 3, figs. 1, 3), sixty-six teeth, fragments of calcified cartilage and an incomplete dorsal fin spine, the remains of a single in- dividual, from the Weald Clay of Henfield, Sussex. MATERIAL. In addition to the holotype, thirty isolated teeth and a fin spine. HORIZONS AND LOCALITIES. Ashdown Beds: Cliff End bone-bed, Sussex. Wad- hurst Clay : Hastings, Sussex. Grinstead Clay: Paddockhurst bone-bed, Sussex ; Tilgate Forest, Sussex. Weald Clay: Henfield, Sussex ; Bookhurst, Surrey. Wealden Shales ; Atherfield Point and Cowleaze Chine, Isle of Wight. DEscripTion. The holotype was found in 1962 by Mr. J. F. Wyley immediately above the ‘ Paludina’ limestone in the north-east corner of the Henfield pit. The teeth, calcified cartilage and fin spine were collected from an area of about two square feet : the bulk of the material was collected at one time but a few teeth have been found at the same point by Mr. Wyley in subsequent visits to the pit. There can be no doubt that this material represents the remains of a single individual. Since this is the first specimen of Hybodus in which it is possible accurately to re- construct the dentition with teeth free from matrix, and in which the upper and lower teeth can be distinguished, it will be described in some detail. t. The Dentition The teeth of the holotype were not found in natural association, but it has proved possible to reconstruct at least the anterior parts of the dentition with some accuracy. The teeth can be sorted into ‘ left’ and ‘right’ (treating all the teeth as if they are from one jaw) by the asymmetry of the cusps, which point away from the symphysis in Hybodus, and by the asymmetry of the root. Many of the teeth can be matched with others to reconstruct successional files, and these files can be placed approxi- mately in their position in the mouth by the degree of asymmetry of the crown, and by the relative height and length of the teeth. When this had been done, of the fifty-seven more complete teeth four were median, twenty-three ‘left’ and thirty “right ’, and there appeared to be fourteen successional files, containing from two to eight teeth. Fourteen teeth could not be matched with any others. In these fourteen “series ’, the measurements length of crown, length of root, depth of tooth, depth of crown, depth of root, breadth of crown and breadth of root were taken on each tooth (Table I), and the range of each measurement in each series was plotted against the inferred position in the mouth. The fourteen unmatched teeth were then measured and allocated their position in the mouth by reference to the plot. When this had been done it was clear that in each of the better represented tooth series there were two types of teeth: these represent the upper and lower jaws (see p. 303). General Features of the Dentition The dentition of H. brevicostatus (Text-fig. 10, Pl. 2) consists of a symphysial file of teeth in each jaw and probably nine paired files in each jaw. The largest teeth are the fifth paired file. In general features, the teeth are very like those of the Lower 302 BRITISH WEALDEN SHARKS 5 mm Gr | Fic. 10. Hybodus brevicostatus sp. nov. Restoration of the dentition of the left side in labi’ VII and VIII in the upper jaw and III, VII and IX in the lower jaw are reversed drawin Liassic H. delabechet Charlesworth (Smith Woodward 1889, pl. Io, figs. 1-5), with | which they share a low crown, hardly exceeding the root in depth, a central cusp and . four or more less conspicuous pairs of lateral cusps, and the ornament of very numer- ous prominent striae. The central cusp lies at or near the centre of the tooth in all but the most posterior files (Text-fig. 10, VIII and IX; PI. 2, figs. 5, 6), where it may | be strongly eccentric or unrecognizable. The number of lateral cusps is normally four on each side, but in the lateral and posterior teeth the lateral cusps may be | more numerous or not recognizable. Accessory cusps occur very commonly on both the lingual and labial margins of the crown, particularly on the labial face of | the mandibular teeth (see p. 305). There is a sharp occlusal crest running the length of the tooth from which strong striae pass to the base of the crown. These main striae often bifurcate basally, and between them there are weaker striae which fail to reach the occlusal crest. There is no significant difference between the strength of the striae on the labial and lingual faces of the crown: in some teeth they are stronger on the lingual face, in others on the labial. At the base of the lingual face of the crown the ridges anastomose in a reticular pattern. There is a good deal of variation in the ornament of the crown from one tooth to another, but this variation seems to have no significance between one jaw and the other or between tooth files. The roots of the teeth are typically hybodontoid (Casier 1947a: 9). In all the teeth the root is moderately compressed, approximately equal to the crown both in depth and breadth, with a concave labial face and a flat or weakly convex lingual face. In all the teeth the root is without specialized foramina but has a large number of irregular foramina on both the lingual and labial faces. In histological structure (PI. 3, fig. 3) the teeth are again typical of Hybodus (Jaekel 1889, pl. 7, figs. 1, 4). The crown is covered by a well marked and rather thick layer of enamel which contains fine fibrils in its basal part. Below the enamel there is a BRITISH WEALDEN SHARKS | view. “s’, symphysial file, I-IX, paired files. All teeth drawn from the holotype; V, VI, of teeth from the right side. P.46973, Weald Clay; Henfield, Sussex. layer of very regular pallial dentine which makes up about one-fifth of the thickness of the crown. Below the pallial dentine the tooth consists of normal osteodentine. In the crown the osteodentine is dense, with many slender vascular canals: in the root it is much more spongy, with large, anastomosing vascular canals. Distinction between Upper and Lower Teeth As noted above, when the teeth of the holotype had been assigned to their positions in the jaws, among the better represented series (the symphysial and first five paired series) teeth of two types could be recognized. These clearly represent the upper and lower jaws. The upper and lower teeth can normally be distinguished by features of both the root and the crown, but the most reliable characters are in the root. In one type of tooth (Text-fig. 11a) the labial face of the root is strongly concave, the base of the root is rounded, and there is normally a well marked furrow between root and crown on both labial and lingual faces of the tooth. In the second type of tooth (Text-fig. 11B) the labial face of the root is less strongly concave, the base of the root is flattened in a well marked oblique shelf, and there is normally a very weak furrow at the junction of root and crown. The flattening of the base of the root is Fic. 11. Hybodus brevicostatus sp. nov. Diagrammatic transverse sections of teeth from the lower (A) and upper (B) jaws. x3. BRITISH WEALDEN SHARKS 304 b.€ 0-% 9:% €.z €.z I-Z 6.1 g-€ g-€ z-L g:Z g-Z I IOMOT z-9 0-€ gz 0.z 0-% I-Z 0.z z.¥ z+ Sol \G-er ¢.Z1 I soddn QUIN, £.9 0-€ ZZ 6.1 6-1 0.z 0-% o-b o.b O-II 0-21 0-71 I IOMOT L.G Lie +.z ¥-z 6.Z-1-Z L.z ¢.z I-S 6-S-S.p SE.€r g9.€1 6.€1-L.€1 S toddn WVysiy €.¢ Q-z I-€ Z-€ z-€ I-€ o.€ 0-9 0-9 €.9r 6.91 6.901 i TOMO] 9-¥ €.z cr.€ EE ¢.€-1-€ SCr.€ ZE 6S.9 1-4-0.9 S6.br z-Si z-S1 Zz aoddn Y}UIAIS €.¥ €.z C6.€ c1-v €-t-o0. z-v 66.€ 0-8 o-g Sr-Zr $6.41 1.g1-g.L1 z IOMOT 6.4 C.z o-+ Cg.€ 6.€-g-€ g-€ 6-€ Z-L g.l-9-L €.g1t 0-61 0-61 Zz aoddn Y}XIS 6-€ NG 9-+ gb gh 1-4 9-4 9-8 9:8 €-21 9-91 9-81 I IOMOT 9-4 €-z o.v 6.€ €.p-1.€ 6.€ So.t g-4 0-g-9-L£ L£.91 0.91 €-.g1-L.Lr ¢ soddn UM Lae 0-% o.¥ ce.r 9-b-1.4 o.v C.F o-g Z£.9-9-4 4.41 0.91 ¥-gr-Z.S1 V4 IOMOT €.7 6.1 6.€ ge 6.€-9-€ z.V +.V G.g 0-6-0.8 G€.Gr Sz.o1 g-g1-L.S1 € ioddn yqaIno.y 9-€ g-I ¢.¥ z-v 9-b-0.b 1-4 cz.+ z-g G.g-9-4 Gg.€1 So.Sr z.S1-9.h1 ¢ IOMOT o-F L.1 So.+ 9-€ 9-€ o-+ +. GE.g S.g-z~.g GL.€r Sz.br L.¥1-g.€1 z seddn parry €.€ L- Co.+ Cz. G.¥-1-b S6.€ zh S1-g €-.g-0.g So-€r1 z-bi g-.h1-6.€1 + I9MOT g-£ g-1 So.b L.€ o-.-z~-.€ So.F o.v 6.2 €.9-$.L 9.721 I-br €.b1-g.€1 9 soddn puosas I-€ (oj So.+ Co.f €.b-g.€ G6.€ g-€ OL ~e-Q=9.L) 2-0r Quen G.St—nter ZL IOMOT c.€ Q-I cr.v L.€ 6.€-¢.€ o-b Cg-€ 6-4 o0-g-6-4£ g.11 So.€1 g.€1-1.z1 4 ioddn qSINT L.@ 9-1 Cg.€ ce.y L.v-0.4b GG.€ CL.E ¥-L G.f-€.£ Gt.or Sg.1r o.z1-Z.11 Zz I9MOT SuSE @)ou z-b cL.€ o.b-¢.€ g-€ cc.€ €-£ L-£-6.9 g-o1 So-zr 1-zI-0-z1 CG roddn yersAydwihs UAOID jo yydeap uvdyL ues osueyy ueo{ UeoyY Uva, osuey Urey URoyy os8uey podArasord TL wypesqg yeoy -JOO1 -uMOIO -UMOID -jooOI -uMoI0 -yjdep -yydep -3001 -uMoIO -uMOIO y3004 “UMOID JO UMOID JO jo jo jo jo jo es Oue EO jo jo yo jo wsueyT Suey VPP WPreIg Yprerg yydeq yideq yysueT YysueT YyWsue]T Isquiny X9SSNS “ployuapxy ‘ALTD pean, “Aou ds smynjsoo1adq snpogdyy yo adAzoyoy ayy ‘EZ6gb'g Jo Y}00} oy} JO (‘UW Ul) SUOISUStUIp oY], I a1avy BRITISH WEALDEN SHARKS 305 the most reliable method of distinguishing between the two types of teeth (Pl. 2, cf. figs. Ia, 2a, and 3a, 4a). In the crown, those teeth with a rounded base to the root have significantly more accessory cusps on the margin of both labial and lingual faces. Taking only the teeth of the symphysial and first six paired files, accessory cusps are present on the labial face of all the teeth with a rounded base to the root. The number of accessory cusps on this face ranges from one to six, the mean being 4:1. On the lingual face of these teeth accessory cusps are present in seventeen teeth out of twenty-six (65°), with a range of 1-3 anda meanofi-1. In teeth with a flattened base to the root, accessory cusps are present on the labial face in sixteen out of nineteen teeth (84%), with a range of 1-6 cusps and a mean of 1-7. On the lingual face of these teeth accessory cusps are present in six out of nineteen teeth (32%), with a range of 1-2 and a mean of 0-6. The presence of an accessory cusp at the base of the labial face of the central cusp is particularly characteristic of teeth with a rounded base to the root, and there is very rarely an accessory cusp in this position in teeth with a flattened base to the root. This is reflected in the maximum breadth of the crown (Table I) which is the only dimension in which there is a significant difference between the two types of teeth. There is no absolutely reliable criterion by which these two types of teeth can be assigned to the upper or lower jaw, but comparison with associated dentitions of Hybodus obtusus from the Oxford Clay and with various species of Hybodus and Acrodus from the Lower Lias suggests that the teeth with the more numerous accessory cusps and the rounded base to the root are mandibular. The roots are visible in very few of the teeth in these associated dentitions, but comparison with isolated Jurassic teeth shows that the characters discussed above will serve to assign most teeth of Acrodus and low-crowned species of Hybodus to the appropriate jaw. Variation in Teeth with Position in the Jaws The main variations in the teeth from different parts of the jaws are clear from Text-fig. 10, Pl.2and TableI. The length and depth of both root and crown increase to a maximum in the middle of each jaw ramus, the length reaching a maximum at the fifth paired file in the lower jaw and the sixth in the upper, the depth at the fourth file in the upper jaw and the fifth in the mandible. The depth of the root exceeds that of the crown in the symphysial, first, seventh, eighth and ninth files. The maximum breadth of the crown exceeds that of the root in all but the last two files in the mandible, while in the upper jaw the breadth of the root is always greater than that of the crown. Other Material Teeth of H. brevicostatus first appear in the Ashdown Beds, and range through to the Wealden Shales, at the extreme top of the Wealden, but the species is rare through- out this time. Among the material from the Cliff End bone-bed (Ashdown Beds) there are six more or less complete teeth of the species, the best preserved of which is shown in 306 BRITISH WEALDEN SHARKS 2mm Fic. 12. Hybodus brevicostatus sp. nov. ‘Tooth in labial (above) and lingual view. P.11801, Ashdown Beds, Cliff End Bone-bed ; Cliff End, Sussex. Text-fig. 12. All these teeth are smaller than those of the holotype, with a maximum length of about 10 mm., and they have fewer striae and fewer accessory cusps. In the Wadhurst Clay H. brevicostatus is represented by a fragment of a small tooth (P.11875) and by P.11876 (Pl. 1, fig. 3), a large tooth, 15 mm. in length, probably a parasymphysial from the upper jaw, in which the crown is unusually high. In the Grinstead Clay there are thirteen teeth of H. brevicostatus among the material from the Paddockhurst bone-bed. These are all small (less than 9 mm. in length), low-crowned forms. 26027 from Tilgate Forest is a larger (II mm.) example, again a low-crowned posterior tooth. Among the material from the Weald Clay of Henfield there is a number of teeth of H. brevicostatus in addition to the holotype. Most of these are normal teeth, as large as or a little smaller than those of the holotype, but two are very small examples (Text-fig. 13), P.46984, less than 6 mm. in length and P.46989, 4 mm. long. These small teeth agree with those of the holotype in shape and in the numerous accessory cusps, but in the few, coarse striae on the crown and in the clearly marked lateral cusps they resemble H. parvvidens. One anterior tooth of H. brevicostatus (P.12812) is also known from the Weald Clay of Bookhurst, Surrey. The latest occurrence of H. brevicostatus teeth is P.13341, an incomplete para- symphysial tooth from the Wealden Shales of Atherfield Point, Isle of Wight. BRITISH WEALDEN SHARKS 307 Fic. 13. Hybodus brevicostatus sp. nov. ? Juvenile teeth from the Weald Clay of Henfield, Sussex. A. P.46984 in labial (above) and occlusal view. B. P.46989 in labial view. 2. Calcified Cartilage Together with the teeth of the holotype Mr. Wyley collected a number of fragments of heavily calcified cartilage. These include recognizable parts of the jaws and branchial arches, but they show nothing worthy of description. 3. Fin Spines Found with the teeth and calcified cartilage of the holotype was a single incomplete fin spine (Pl. 3, fig. 1). The British Museum (Natural History) contains one spine which agrees with this, P.13268 from the Wealden Shales overlying the Hypsilophodon Bed, Cowleaze Chine, Isle of Wight (PI. 3, fig. 2). These two spines agree almost exactly in size ; each must have had a total length of about 20 cm. Although they agree well in most characters, the angle of insertion of the Isle of Wight spine (in- ferred from the extent of the enamel ridges on the surface and the obliquity of the growth lines) was probably less than that of the Henfield spine. This suggests that the Isle of Wight specimen is an anterior fin spine and the Henfield is posterior, since in those hybodonts known by complete specimens the anterior fin spine lies more obliquely than the posterior (Brough 1935). The most striking feature of these fin spines, and the one which differentiates them from all other species of Hybodus, is that the enamel ridges on the distal part of the spine are very short, ending at or above the level of the lowermost denticles on the posterior face of the spine, and well above the apex of the groove on the posterior face of the proximal part of the spine which housed the basal cartilage of the fin. The trivial name of the species refers to these short ribs on the fin spine. The spines are not strongly arched, the anterior edge curving through about 35°, the posterior through 25°, and are strongly compressed. The anterior edge is keeled and there is a well marked angle between the lateral and posterior faces. On the lateral face the ridges of enamel are well spaced. In both specimens a median 308 BRITISH WEALDEN SHARKS enamel ridge forms the keel on the anterior edge. In the Henfield spine (?posterior) there are ten ridges on each side, five reaching the tip of the spine and five being confined to the proximal part. In the Isle of Wight spine there are eleven ridges on each side, seven of which extend to the tip. In the Henfield spine all the ridges end above the level of the lowermost denticle on the posterior face, the longest ribs being the first and second on the right side and the third on the left side. In the Isle of Wight spine the longest ridges are the first on each side, and they alone extend beyond the level of the lowest denticle on the posterior face. The denticles on the posterior face of the spine are in a single median series. This single series 1s clearly the result of unequal development of paired denticles, for occasionally the denticles are double (Pl. 3, figs. 1a, 2a), and in other cases at the base of a fully developed denticle there is the rudiment of a second. The side on which these rudimentary denticles occurs is variable, showing that the single series of denticles is not the result of suppression of either the left or right series of denticles, but of apparently random suppression of one of each pair. In the Isle of Wight spine, which is complete to the tip, there is a total of 27 denticles, of which two show doubl- ing and eight show rudiments of a second denticle, six on the right side and two on the left of a fully developed denticle. In the Henfield spine sixteen denticles are pre- served, one is doubled, and one has a rudiment on its left side. The denticles are very variable in shape: some are strongly hooked, either to the left or the right, some are straight, and while the majority are smooth, some, particularly towards the base of the spine, bear striae like those on the teeth. Lack of material at present prevents the preparation of thin sections of the spines, but broken surfaces show that in structure the spine agrees with H. aschersont Stromer (1927 : 20, pl. 3, fig. 9, text-fig. 13) from the Cenomanian of Libya, in that in the exserted part of the spine the pulp cavity is surrounded by a zone of lamellar tissue which makes up about half the thickness of the wall of the spine, the outer part of the wall consisting of osteodentine. Passing back towards the base of the spine the zone of lamellar tissue becomes thinner until it disappears below the apex of the groove in the hind edge of the spine. Below this level the spine consists of osteo- dentine alone. AFFINITIES. Teeth of H. brevicostatus from horizons below the Weald Clay suggest that in the history of the species between the Ashdown Beds and the Weald Clay there was an increase in size accompanied by an increase in the number of striae on the crown and in the number of accessory cusps. Small teeth from the Weald Clay of Henfield show that in the ontogeny of the species increase in size was accompanied by increase in the number of striae on the crown and reduction in the individuality of the lateral cusps. Both these facts suggest that the ancestor of H. brevicostatus was a rather small, low-crowned species, extant during or before the deposition of the Ashdown Beds, with sparse, coarse striae on the crown, three or four pairs of lateral cusps and a tendency to develop accessory cusps. These conditions are met by H. parvidens, and it seems very probable that H. brevicostatus is an offshoot of this species, probably originating near the base of the Wealden. The increase in ornamentation and reduction in the individuality of the lateral BRITISH WEALDEN SHARKS 309 cusps which occur with increased size is an interesting feature of this species. In the holotype it results in some of the teeth, especially the more posterior ones, having a very Acrodus-like crown, sometimes without distinguishable lateral cusps or with very numerous small ones. These conditions are just those found in some Upper Cretaceous species such as Synechodus illingworthti (Dixon) (Smith Woodward Igi1 : 220, pl. 46, figs. 5-7) from the Cenomanian zones of the English Chalk ; Hybodus brabanticus Leriche (1929 : 225, text-figs. 4, 5; 1930: 105) and Acrodus dollot Leriche (1929 : 228, text-figs. 6, 7) from the Lower Senonian of Belgium ; Hybodus grewingki Dalinkevicius (1935 : 256, pl. 1, figs. 36-38) and Acrodus guedroyii Dalinkevicius (1935 : 256, pl. 1, figs. 34, 35) from the Cenomanian of Lithuania ; Acrodus affinis Reuss (1845: 1, pl. 2, figs. 3, 4), Hybodus bronnit Reuss (1845 : 97, pl. 24, fig. 26, pl. 42, fig. 7) and Hybodus cristatus Reuss (1845: 2, pl. 2, fig. 20), all from the Turonian Planerkalk of Bohemia ; and the teeth from the Upper Sono- nian of Aachen, Germany, described as Hybodus ? sp. and Acrodus ? sp. by Albers & Weiler (1964: 4, 5, text-figs. 3, 8, 9, 42). These species are also the only marine hybodonts known by teeth in the Upper Cretaceous. The best known of them is Acrodus illingworthi Dixon, a species transferred to the heterodontid genus Synechodus by Smith Woodward (1891 : 66). But there is no real evidence that this species is not a hybodont : all the teeth have typical hybodontoid roots, without the develop- ment of enlarged central foramina which is characteristic of Synechodus and other heterodontids (Casier 1947): 2). The absence of hybodont fin spines in the English Chalk is not evidence against the species being hybodont, for only about twenty teeth are known, and judging by the conditions in H. brevicostatus an individual would produce well over two hundred teeth in its life but only two spines. It is also quite possible that these Upper Cretaceous hybodonts should have lost their fin spines. The syntypes and other material of A. illingworthi in the British Museum (Natural History) show a strong resemblance to H. brevicostatus, and with Leriche (1929 : 227) I believe the species should be named Hybodus tllingworthi (Dixon). H. brabanticus Leriche is almost certainly a synonym of H. illingworthi, for Leriche included in his species some specimens figured by Smith Woodward (ro11, pl. 46, fig. 7) as S. tllingwortht, but these teeth (43511) occur in association with other teeth which are undoubtedly H. ilingwortht. There seems little to distinguish H. grewinghi Dalinkevicius from H. illingworthi, and Dalinkevicius noted the resemblance between his material and Leriche’s H. brabanticus. TReuss’s species from the Turonian of Bohemia are large teeth of the same type as H. illingwortht and H. brevicostatus. The Upper Senonian Hybodus teeth described by Albers & Weiler are compared by them with A. idlingwortt and H. brabanticus, which they closely resemble. Acrodus dolloi Leriche and A. guedvoyii Dalinkevicius are based on large teeth which resemble Lower Lias species such as A. curtus (=A. anningiae) and A. nobilis; the Acrodus teeth described by Albers & Weiler are very like A. dolloit but smaller. There is no other evidence of Acrodus of this type in the Cretaceous, but these teeth, in size, shape and ornamentation, bear considerable resemblance to the large postero-lateral teeth of H. brevicostatus, and could be derived from this species by further increase in size, in the number of striae and in the loss of individuality of the lateral cusps. 310 BRITISH WEALDEN SHARKS Fin Spines of Hybodus Only in H. basanus and H. brevicostatus among Wealden and Purbeck sharks have fin spines been found in association with teeth. Smith Woodward (1916) suggested relationships between the various spines and teeth known from the British Wealden and Purbeck. His conclusions may be summarized as follows : Tooth Horizons Spine H. basanus Weald Clay ? H. sulcatus Agassiz “teeth of the same general type as H. basanus’ Wadhurst Clay Hf. subcarinatus Agassiz Grinstead Clay HZ, ensis Middle Purbeck ‘almost identical with H. dorsalis Agassiz ’ H. parvidens Wadhurst Clay ? ? Grinstead Clay H. striatulus Agassiz ? Middle Purbeck H. strictus Agassiz Of the spines which Smith Woodward did not associate with species based on teeth, he compared H. strviatulus with the Purbeck spines assigned to H. ensis, and H. strictus with H. subcarinatus and H. basanus. We now have more complete information on the ranges and relationships of the Wealden and Purbeck teeth of Hybodus. H. parvidens is a species which ranged from the Middle Purbeck to the Weald Clay, giving rise to H. basanus in the Weald Clay, the evolution of H. basanus involving an increase in size of the teeth. We should therefore expect the fin spines of H. parvidens to be similar to those of H. basanus but smaller, and to range from the Middle Purbeck to the Lower Weald Clay. These conditions are met by H. strictus Agassiz in the Purbeck (12-13 cm. long) and H. subcarinatus Agassiz in the Wadhurst and Grinstead Clay (14-17 cm. long). The ‘ teeth of the same general type as H. basanus’ with which Smith Woodward associated H. subcarinatus Agassiz are shown above (p. 294) to belong in H. ensis : one would therefore expect them to be associated with spines of the same type as H. dorsalis Agassiz—this condition is met by H. striatulus Agassiz. A revised scheme of the (hypothetical) relationships between hybodont teeth and fin spines in the Wealden and Purbeck is as follows : Tooth Horizons Spine HI. ensis Middle Purbeck— cf. H. dorsalis (Purbeck) Grinstead Clay H. striatulus (Wealden) Hi. parvidens Middle Purbeck— H. strictus (Purbeck) Weald Clay H. subcarinatus (Wealden) H. basanus Weald Clay ? HZ. sulcatus H. brevicostatus Ashdown Beds—Weald Clay JH. brevicostatus In the Wealden and Purbeck there occur tuberculated fin spines which have been described as Asteracanthus (A. verrucosus Egerton 1854, Middle Purbeck, Swanage ; A. semiverrucosus Egerton 1854, Middle Purbeck, Swanage ; A. granulosus Egerton, 1854, Wealden, Sussex). A spine from the Lower Neocomian of the Paris Basin is described as A. cf. acutus Agassiz by Leriche (1911 : 456, pl. 6, fig. 1), and A. granulo- sus has also been recorded from the Lower Neocomian of Switzerland, where it occurs BRITISH WEALDEN SHARKS 311 in the same beds as teeth of ‘ Stvophodus’ (=Asteracanthus) which according to Peyer (1946 : 54) are not derived fossils. But although Astevacanthus occurs in the marine Lower Cretaceous, it seems very unlikely that the British spines are true Asteracanthus—that is, that they are from a fish with ‘ Stvophodus’ teeth, like the better known Jurassic species, for the large and conspicuous teeth could hardly have escaped notice, particularly since about a dozen spines of this type are known and in any individual there will be at least fifty teeth to each spine. The preservation of the spines is such that they can hardly be derived from older formations. The answer is probably that these spines are Hybodus in which the normal enamel ridges have broken up into separate tubercles as a stage in the reduction of the spine. This argument is supported by the presence of normal ridges on the distal (and first formed) parts of the spines, by the Wealden H. striatulus Agassiz (suggested above to belong to H. ensis) where the ridges are partially broken up into tubercles, and by the tendency in many Wealden Hybodus spines for the ridges to become fragmented proximally. Remarks on Hybodont Teeth This study has shown that where abundant isolated teeth of Hybodus from succes- sive horizons are available, the distinctions between species become very difficult to maintain. A great deal of parallelism and convergence is bound to occur in the evolution of sharks’ teeth, and further uncertainty arises from the degree of hetero- donty of species (usually unknown) and from variability within a species. It is clear that the erection of new species of hybodont sharks on one or two isolated teeth is an extremely hazardous procedure. The changes which occur in the teeth of Wealden hybodont sharks from species to species and horizon to horizon allow one to draw some general conclusions about the correlation between various dental characters in hybodonts. These can be sum- marized as follows : (a2) Characters which increase with increasing height of crown and decrease with decreasing height of crown. I. lingual curvature of crown 2. compression of crown 3. lingual curvature of root 4. difference between ornament on labial and lingual faces of crown. (b) Characters which decrease with increasing height of crown and increase with decreasing height of crown. degree of heterodonty number of lateral cusps number and size of accessory cusps depth of root height of striae on crown bifurcation of and anastomosis between striae Oo a Na 312 BRITISH WEALDEN SHARKS (c) Characters which increase with increasing size of tooth. I. number of striae on crown. This list of characters showing correlation with the height of the crown or the size of the tooth includes almost all those commonly used to distinguish hybodont species. If all these characters are correlated with an obviously adaptive feature like crown height it is clear that it will often be impossible to distinguish on teeth alone between relationship, parallelism and convergence. There is therefore little likelihood of any worthwhile subdivision of the genus Hybodus, now so large both in number of species and in range, on characters of the dentition. Jaekel’s (1889, 1898) attempt at a subdivision was largely based on features of the teeth which are obviously adaptive or are correlated with adaptive features : of the four genera which he proposed, only Polyacrodus, distinct on histological characters (thickness of pallial dentine), has been generally accepted. We are forced to look for other characters on which species of Hybodus may be grouped, but if the earliest (Lower Lias—H. delabechei etc.) and latest (Upper Wealden —H. basanus) species in which the anatomy is well known are compared, the only differences seem to be reduction of the cephalic spines (present in males only) from two pairs to one, the appearance of symphysial teeth, and slight changes in the fin spines (increase in thickness of the lamellar tissue, reduction in the strength of the enamel ridges, reduction in the distance between the paired denticles on the hind edge, culminating in suppression of one of each pair in H. brevicostatus). The first two of these differences are suitable for generic subdivision, but until their distribution is known in many more species no useful purpose is served by such division. Genus LONCHIDION Estes 1964 : 7 AMENDED DIAGNOSIS. Small fresh-water hybodonts, weakly or strongly hetero- dont, known only by isolated teeth, cephalic spines and fin spines. Teeth low- crowned, elongated, crown not much deeper than root; tricuspid anterior teeth present in some forms; crown smooth or with sparse vertical striae, large labial projection or buttress below central cusp, lateral cusps absent or small and irregular ; root hybodontoid except in supposed tricuspid anterior teeth of advanced forms, where it is squatinoid ; pallial dentine of crown very thick, as in Polyacrodus. Fin spines with or without enamel ridges on lateral faces, with a single series of posterior denticles. Cephalic spines of normal hybodont type, without terminal barb. TYPE SPECIES. Lonchidion selachos Estes (1964 : 7, text-figs. 1-4), Lance Formation, Wyoming. Lonchidion breve sp. nov. (Pl. 5, fig. 3; Text-figs. 14-20, 29E) Driacnosis. Lonchidion known only by isolated teeth, 3-5 mm. or less in length ; no tricuspid anterior teeth, dentition weakly heterodont ; crowns of teeth rather short, with maximum breadth between half and one-quarter of the length ; crown smooth or very feebly striated, lateral and accessory cusps very small and irregular when present, labial process strong. BRITISH WEALDEN SHARKS 313 Horotype. P.47024 (Text-fig. 14), complete tooth from the Paddockhurst bone- bed (Grinstead Clay) ; Sussex. MATERIAL. About one hundred and sixty teeth, mostly without roots. HORIZONS AND LOCALITIES. Ashdown Beds: Cliff End bone-bed, Cliff End, Sussex. Wadhurst Clay : Telham bone-bed, Stone, Kent ; top of cliff near Hastings harbour, Sussex ; Ashurstwood, Sussex. Grinstead Clay: Paddockhurst bone-bed, Sussex. Weald Clay: Henfield, Sussex. Atherfield Clay: Perna Bed, Sandown Bay, Isle of Wight (probably derived from Wealden Shales—see p. 319). The teeth which are assigned to this species fall into three types which intergrade both morphologically and stratigraphically : they will be described separately as three subspecies. OG 71mm Fic. 14. Lonchidion breve treve sp. et ssp. nov. P.47024, the holotype, a lateral tooth in labial (above), occlusal (centre) and lingual view. Grinstead Clay, Paddockhurst Bone-bed ; Paddockhurst Park, Sussex. Lonchidion breve breve sp. et ssp. nov. (Pl. 5, fig. 3; Text-figs. 14-16, 29E) Dracnosis. Lonchidion breve in which the crown of the tooth is normally smooth, without lateral or accessory cusps or striations. HototyPe. P.47024, complete tooth (Text-fig. 14) from the Paddockhurst bone- bed. MaTeERIAL_ In addition to the holotype, about 120 teeth, mostly without roots. HorRIzONS AND LOCALITIES. Ashdown Beds: Cliff End bone-bed. Wadhurst Clay: Telham bone-bed, Stone, Kent; Hastings, Sussex; Ashurstwood, Sussex. GEOL. II, 7 29 314 BRITISH WEALDEN SHARKS Grinstead Clay: Paddockhurst bone-bed. Weald Clay: Henfield. Atherfield Clay : Perna Bed, Sandown Bay, Isle of Wight. DESCRIPTION. This subspecies, the typical and commonest form of the species, occurs throughout the Wealden. In the majority of specimens only the crown is preserved, and in many examples, especially those from the Cliff End and Paddock- hurst bone-beds, the teeth have undergone a good deal of post mortem abrasion. None of the teeth exceeds 3-5 mm. in length. The crown is normally completely smooth, without ornament or any indication of lateral cusps. There is a single, very low cusp which lies at or near the centre of the crown. The occlusal edge of the tooth is compressed into a more or less sharp ridge. Below the central cusp there is a strong rounded process or buttress on the labial face of the crown. The root plays no part in the formation of this process, and is overhung by it. A ridge nor- mally runs on to the labial process from the central cusp. On the lingual face of the crown there is a gentle swelling opposite the central cusp and the labial process. The occlusal surface of the crown is much longer than its base, so that the tooth is strongly waisted at the junction of root and crown. The root is preserved in few specimens, and is always somewhat worn. It is typically hybodontoid (Casier 19474 : 9), with no specialized vascular foramina. On the lingual face of the root there is a row of vertically elongated foramina and on the labial face irregularly scattered foramina (Text-fig. 14). The root is almost as deep as the crown and only slightly shorter. In the type species of Lonchidion, the Late Cretaceous L. selachos Estes (see p. 330), A B € 7mm Fic. 15. Lonchidion breve breve sp. et ssp. nov. Three teeth from the Ashdown Beds, Cliff End Bone-bed; Cliff End, Sussex, in labial (above), occlusal (centre) and lingual view. A. P.46993; B. P.46995; c. P.47005. BRITISH WEALDEN SHARKS 315 the histological structure of the crown is described (Estes 1964 : 8, text-fig. 2D) as a fan-shaped radiation of dentine tubules from one central cavity which originated along a central longitudinal axis. The tooth crowns of L. breve breve (P1.5, fig. 3; Text-fig. 29E) agree with this description. Below a rather thick layer of enamel the crown is made up of pallial dentine containing long, subparallel, much branched tubules which arise from the tips of vascular canals which end shortly after entering the centre of the base of the crown. Estes compares this structure with that seen in teeth of the Triassic hybodont Palaeobates (Jaekel 1889, pl. ro, fig. 2; Seilacher 1943, text-figs. 22, 29, 34), especially with P. nodosus Seilacher. But in Palaeobates (Pl. 5, fig. 2) the dentine tubules are longer, parallel and more regularly branched than they are in Lonchidion, and they arise not from the tips of vascular canals but from a single open pulp cavity running the length of the tooth. The pallial dentine of Lonchidion is much more like that in the Triassic Polyacrodus, particularly in small species like P. minimus (Seilacher 1943, text-figs. 7-10; Pl. 5 fig. 1), where there are just the same wavy, irregularly branched dentine tubules arising from short vascular canals near the base of the crown. In Polyacrodus the pallial dentine is a little thinner than it is in Lonchidion, perhaps because of the greater size of the teeth, but in other respects the correspondence is exact. The worn condition and absence of roots in most examples of this sub-species make it difficult to arrive at any clear idea of the variation in shape. The length of the crown ranges from I-4—3-4 mm., the maximum breadth of the crown (at the labial process) from 0-4-1-2 mm. In five complete teeth the depth ranges from I-0-1I°35 mm., the ratio of depth to length from 1-5—2-5. The crown is normally broader than deep, the ratio of maximum breadth to length ranging from 1-9-3°8, and of maximum depth to length from 2-7—4:2. The shorter, deeper teeth, in which the labial process is large (Text-figs. 15A, 16A) are probably anterior ; longer teeth, still with a strong labial process (Text-fig. 158), antero-lateral; the longest teeth 3 3 Fic. 16. Lonchidion breve breve sp. et ssp. nov. Teeth in labial (above), occlusal (centre) and lingual view. a, B. P.47324, P.47047 from the Weald Clay of Henfield, Sussex. c, D. P.39015, P.39011 from the Perna Bed, Atherfield Clay ; Sandown Bay, Isle of Wight. 316 BRITISH WEALDEN SHARKS (Text-figs. 14, 15C) lateral, and shallow teeth with a weak labial process are probably posterior (Text-fig. 168). At lower horizons there is no difficulty in distinguishing between L. breve breve and the other subspecies of the species, but at higher horizons, the Weald Clay of Henfield and the Peyna Bed in the Atherfield Clay, examples occur in which there are weak accessory cusps (Text-fig. 16c) or lateral cusps (Text-fig. 164, D). These specimens tend towards L. breve pustulatum, to which most of the Perna Bed teeth belong, and also towards L. selachos Estes, from the Upper Cretaceous of America (see p. 330). Lonchidion breve crenulatum sp. et ssp. nov. (Text-figs. 17, 18) Diacnosis. Lonchidion breve in which the occlusal margin of the tooth crown is irregularly crenulate, and in which the lingual and labial faces of the crown are weakly and sparsely striated. Hototyre. P.47060 (Text-fig. 178), tooth without root from the Paddockhurst bone-bed. MATERIAL. In addition to the holotype, twenty-five teeth, all without roots. HORIZONS AND LOCALITIES. Wadhurst Clay: Ashurstwood, Sussex. Grinstead Clay : Paddockhurst bone-bed. GD CED coli> Ss AE? = 7mm Fic. 17. Lonchidion breve crenulatum sp. et ssp. nov. Three teeth from the Grinstead Clay, Paddockhurst Bone-bed; Paddockhurst Park, Sussex, in labial (above), occlusal (centre) and lingual view. A. P.47059. 8B. P.47060, the holotype. c. P.47066. BRITISH WEALDEN SHARKS 317 Description. In this subspecies the form and proportions of the teeth do not differ significantly from those in L. breve breve. The largest tooth (Text-fig. 17C) is 3°5 mm. long, similar in size to the largest L. breve breve. The occlusal crest of the teeth is produced into a number of small and irregular lateral cusps or beads. Both the lingual and labial faces of the crown bear a number of weak, parallel, vertical striae, the striae being strongest at the centralcusp. Both the beading of the occlusal crest and the striation of the crown are weakest in the short, deep anterior teeth (Text-fig. 174), strongest in the longer, lower lateral and posterior teeth. REMARKS. This subspecies is commonest in the Paddockhurst bone-bed, where it accounts for about half of the specimens of Lonchidion breve, the other half being L. breve breve. The fact that teeth of L. b. crenulatum from this horizon exhibit the same range of size and shape as those of L. b. breve (cf. Text-figs. 15, 17) and the ab- Wa 1mm Fic. 18. Lonchidion breve crenulatum sp. et ssp. nov. P.47081, a tooth from the Wadhurst Clay; Ashurstwood, Sussex, in labial (above), occlusal (centre) and lingual view. sence of L. b. crenulatum at Cliff End and in the Telham bone-bed show that the two subspecies are distinct forms, not teeth from different parts of the mouth of the same fish. L. b. crenulatum is not known above the Grinstead Clay. Ina sample of four teeth of Lonchidion from the Wadhurst Clay at Ashurstwood, Sussex, three are L. b. crenulatum (Text-fig. 18) and one is L. b. breve. L. b. crenulatum does not occur in the large sample of Lonciidion from the Cliff End bone-bed in the Ashdown Beds or the Telham bone-bed at the base of the Wadhurst Clay, nor is it known in the Purbeck, although the posterior teeth of L. heterodon (Text-fig. 25c) are very like it. In spite of this resemblance, it seems probable that L. b. crenulatum is a sub- species which evolved from L. b. breve, probably early in Wadhurst Clay time. Lonchidion breve pustulatum sp. et ssp. nov. (Text-fig. 19) Diacnosis. Lonchidion breve in which the occlusal crest of the crown is weakly, irregularly and finely crenulate, and in which the labial face of the crown bears a number of minute accessory cusps but no striations. 318 BRITISH WEALDEN SHARKS HoLotyrPe. P.47085, (Text-fig. 198), tooth without root from the Perna Bed, Atherfield Clay, Sandown Bay, Isle of Wight (? derived from the Wealden Shales—see below). MATERIAL. In addition to the holotype, fourteen teeth, all without roots. HORIZON AND LOCALITY. Perna Bed, Atherfield Clay: Sandown Bay, Isle of Wight. DESCRIPTION. In this subspecies the form and proportions are again very similar to those in L. breve breve. The teeth range in length from 1-8-3-6 mm., in breadth from 0-6-1I-:3 mm., and in the ratio of breadth to length from 2-0-3-6. The occlusal 7 mm Fic. 19. Lonchidion breve pustulatum sp. et ssp. nov. Teeth from the Perna Bed, Atherfield Clay ; Sandown Bay, Isle of Wight, in labial (above), occlusal (centre) and lingual view. A. P.47088. 8B. P.47085, the holotype. c. P.47089. crest of the teeth is produced into irregular lateral cusps or beads, and these are more numerous than they are in L. breve crenulatum. The labial and lingual faces of the crown are without striae. The labial process, below the central cusp, is not so well marked off from the body of the crown as it is in the other subspecies (Text-fig. IQB, C), and the ridge on its occlusal surface bears one or sometimes two small cusps. Level with the more labially placed of these cusps on the labial process there is a series of very small accessory cusps on the labial surface of the crown, which extends as a Shallow shelf at this level. The strength and number of these accessory cusps seem to be correlated with the breadth of the tooth (Text-fig. 19), but not with its length, and no real distinction between anterior and posterior or lateral teeth can be made. BRITISH WEALDEN SHARKS 319 REMARKS. This subspecies is known only in the Perna Bed, at the base of the Lower Greensand. Most of the fish teeth from this horizon are derived from the underlying Wealden Shales (Casey 1961 : 505), and this is probably true of the speci- mens of Lonchidion, a genus otherwise unknown in marine beds. The teeth are not much rolled or waterworn, but their small size will probably account for this. Together with L. b. pustulatum in the Perna Bed there occur teeth which are referred to L. b. breve but which show faint traces of the lateral and accessory cusps which characterize L. b. pustulatum (Text-fig. 16c, D): Similar examples also occur at Henfield (Text-fig. 164). These forms leave little doubt that L. b. pustulatum has evolved directly from L. b. breve by extension of the labial face of the crown into a shelf bearing accessory cusps. Teeth of L. b. pustulatum are very like the smallest posterior teeth of the ptychodont genus Hylaeobatis, suggesting a possible relationship which is discussed on p. 341. The Arrangement of the Teeth Many of the teeth of Lonchidion breve have a strong wear facet on the occlusal margin of the crown, the occlusal crest and central cusp being planed off into a level surface. This indicates, as one would deduce from the form of the teeth, that L. breve was a durophagous form, grinding its food by rubbing the upper and lower teeth across one another. In many specimens of L. breve there is also a vertical groove or scar in the centre of the lingual face of the crown: this is a pressure scar formed by the tip of the labial process of the succeeding tooth, which shows that successive teeth in each file were in contact or nearly so. If the tip of the labial process of each tooth touched the centre of the lingual face of the crown of the preced- ing tooth, there would be a space between the tapering lateral parts of the teeth— such spaces are disadvantageous in durophagous forms, and they must have been filled by the lateral parts of the teeth of the adjacent file (Text-fig. 20). It follows that in L. breve there was overlap between the files of teeth, and that there must therefore have been regular alternation between the teeth in adjacent files: this is an advanced character recalling the rays, which does not normally occur in hybodonts —— fae ain es eee Fic. 20. Lonchidion breve sp. nov. Diagram showing the probable arrangement of adjacent files of teeth in occlusal view, X15. Based on the holotype (Text-fig. 14). 320 BRITISH WEAL DEN SEPARIKS and heterodontids (Casier 1953 : 40), although it is present in the Triassic fresh-water form Lissodus africanus (Brough 1935, pl. 2, figs. 2, 3), at least in the anterior and lateral parts of the jaw. Lonchidion striatum sp. nov. (Text-figs. 21, 22) Diacnosis. Lonchidion known only by isolated teeth, 4-5 mm. or less in length ; no tricuspid anterior teeth, dentition weakly heterodont ; crowns of teeth elongated, maximum breadth rarely less than one-quarter of the length ; occlusal crest of teeth strong, lateral cusps very weak and irregular; both labial and lingual faces of crown with strong vertical striae, occasionally bifurcated basally, which diverge from the occlusal crest but do not reach the base of the crown ; labial process small and weak. HoLotyrPe. P.47103 (Text-fig. 21c), a complete tooth from the Weald Clay of Henfield, Sussex. MATERIAL. In addition to the holotype, one hundred teeth, of which nine are complete. HORIZON AND LocALITy. Weald Clay: Henfield, Sussex. DEscriPTION. The teeth of this species from Henfield are all well preserved, with little or no wear, but the root is present in only nine of the one hundred examples. The teeth range in length from 1-1 to 4-2 mm. In shape and proportions the teeth are similar to those of L. breve breve and L. b. crenulatum except that they are more elongated. Only in two teeth out of fifty does the ratio of maximum breadth to length of the crown fall below 4-0, the mean ratio being 5:9 (compared with about 2-7 in L. b. breve). The longitudinal occlusal crest of the crown is strongly marked, D SSNS CANN, SORTA AS aes MO S\siw eka aay Coen RIAL Fic. 21. Lonchidion striatum sp. nov. Teeth from the Weald Clay of Henfield, Sussex, in labial (above), occlusal (centre) and lingual view. a. P.47094. B. P.47095. c. P.47103, the holotype. pb. P.47106. BRITISH WEALDEN SHARKS 321 and there is a low central cusp. The occlusal crest is roughened and jagged, as in L. b. crenulatum, but distinct lateral cusps are not recognizable. Diverging from the occlusal crest there are well marked vertical striae on both labial and lingual faces of the crown, the striae being stronger and more numerous than in L. b. crenulatum. The striae are occasionally bifurcated basally, especially at the central cusp, where they are longest and strongest. The striae do not reach the base of the crown. The labial process is weaker than in L. breve, particularly in the more elongated teeth, where it may almost disappear (Text-fig. 214). The striae are weakest in the smaller and shorter teeth (Text-fig. 224). The root is about as deep as the crown, and is hybodontoid, without specialized foramina, just as in L. breve. In histological structure the crown agrees with that of L. breve in being made up almost entirely of pallial dentine. oo kes be PLATE 5 Fic. 1. Polyacrodus minimus (Agassiz). Vertical section of tooth crown. P.47271, Rhaetic ; Holwell, Frome, Somerset. x 30. Fic. 2. Palaeobates angustissimus (Agassiz). Vertical section of tooth, P.47272, Mus- chelkalk ; Crailsheim, Germany. X20. Fic. 3. Lonchidion breve breve sp. & ssp. nov. Vertical section of tooth crown cut through the labial process. P.47275, Ashdown Beds, Cliff End bone-bed ; Cliff End, Sussex. x50. Fics. 4, 5. Hylaeobatis ovnata (Smith Woodward). Vertical sections of crowns of a para- symphysial tooth (Fig. 4, P.47276, 20) and a posterior tooth (Fig. 5, P.47277, x40), Weald Clay ; Henfield, Sussex. Fics. 6, 7. Hylaeobatis ornata (Smith Woodward). Teeth in labial (a), occlusal (b) and lingual (c) view, Weald Clay; Henfield, Sussex. Fig. 6. Posterior tooth (probably eighth paired file), P.47216, x10. Fig. 7. Postero-lateral tooth (probably sixth paired file), P.47215, 5: Bull. B.M. (N.H.) Geol. 11, 7 PIL AIS, & Ca) ss 3 . 1966 2.4 JAN an oe a PRINTED I BY ADLARD - BARTHCLOME REPRESENTATIVES OF THE FAMILY PHOLIDOPHORIDAE S. STR. O. NYBELIN ~ ‘ BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) : 3 Vol. 11 No. 8 LONDON: 1966 ON CERTAIN TRIASSIC AND LIASSIC REPRESENTATIVES OF, THE FAMILY PHORIDOPHORIDAE S. STR. BY ORVAR NYBELIN , (Professor, Natural History Museum, Gothenburg, Sweden) Ph. 351-432 ; 15 Plates ; 16 Text-figures BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY) GEOLOGY Vol. 11 No. 8 LONDON : 1966 THE BULLETIN OF THE BRITISH MUSEUM (NATURAL HISTORY), instituted im 1949, 1s issued in five series corresponding to the Departments of the Museum, and an Historical series. Parts will appear at irregular intervals as they become veady. Volumes will contain about three or four hundred pages, and will not necessarily be completed within one calendar year. In 1965 a separate supplementary series of longer papers was instituted, numbered serially for each Department. This paper is Vol. 11, No. 8 of the Geological (Palaeontological) series. The abbreviated titles of periodicals cited follow those of the World List of Scientific Periodicals. © Trustees of the British Museum (Natural History) 1966 TRUSTEES OF THE BRITISH MUSEUM (NATURAL HISTORY) Issued 21 January, 1966 Price £3 ON CERTAIN TRIASSIC AND LIASSIC REPRESEN EAFEIVES: OF THE BAMIELY PHOLMIDORTHIORID AE. Sis SueR: By ORVAR NYBELIN Page I. INTRODUCTION : F ; j F ; : ‘ : 354 II. SySTEMATIC DESCRIPTIONS : ‘ : : F : F 350 Genus Pholidophorus Agassiz : . é 5 c : 350 Pholidophorus becher Agassiz : F : : : 357 Pholidophorus latiusculus Agassiz . ‘ . : : 368 Pholidophorus (?) caffit Airaghi ; : : ; : 376 Pholidophorus cf. pusillus 4 aa 4 : ; : : 382 Genus Pholidolepis gen. nov. : : ; : : 387 Pholidolepis dorsetensis gen. et sp. nov. . é : : 387 Genus Pholidophoroides Woodward : ; é : 5 392 Pholidophoroides crenulata (Egerton) ; 5 ; : 393 Pholidophoroides limbata (Agassiz) . : : : ; 404. Genus Pholidophoropsis gen. nov. ; : ; : 411 Pholidophoropsis caudalis (Woodward) i : ; : 411 Pholidophoropsis maculata sp. nov. ; : ; : 416 III. Discussion. 423 (a) The eeonoune leon Beare the Bene and eee within the family Pholidophoridae s. str. and a preliminary diagnosis of the family . , 423 (b) Some phylogenetic aspects of the oneal aad ates ane the family Pholidophoridae s. str. . . 426 (c) Some brief comments on the derivation ot the fevastine Pholidophoridae s. str. . : : : é 6 6 429 IV. ACKNOWLEDGEMENTS F ; é p : : é : 431 V. REFERENCES . : : : : : c é ¢ : 431 SYNOPSIS This paper is a revision, based mainly on the exoskeleton of the head and trunk, of a number of Upper Triassic and Liassic species formerly included in the genus Pholidophorus Agassiz. Special attention is paid to the shape of the preoperculum, the preopercular sensory canal and the squamation. Four species of Pholidophorus, including the type species, Ph. bechei, are redescribed. In the genus Pholidophoroides WWoodward two species are redescribed. A new genus Pholidophoropsis is made for Pholidophorus caudalis Woodward and Pholidophoropsis maculata sp. nov. is described. A new genus and species, Pholidolepis dorsetensis, is made for part of the material previously included in Ph. caudalis. Relationships between the various species and genera of Pholidophoridae s. str., and between the Pholidophoridae and the Para- semionotidae and Leptolepidae are discussed. GEOL. 11, 8 32 354 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE I. INTRODUCTION TuHE fundamental requisite for any discussion regarding the origin of the teleostean fishes is a knowledge of the species belonging to the family Pholidophoridae. Because of the vague definition of the genus Pholidophorus Agassiz 1832 many varied species have been attributed to it with the result that it has become one of the most extensive holostean genera, obviously containing many heterogeneous elements of dubious affinity. Consequently a revision of the genus Pholidophorus has long been highly desirable. In 1941 Woodward took up the question. Besides the genus Pholidophorus s. str., with Ph. bechei Agassiz as type species, he created three new genera : Pholidophoroides with Ph. crenulatus Egerton as type species and with Ph. caudalis Woodward as probably belonging to the same genus, Pholidophoristion with Ph. ovatus Agassiz as type species and Ph. micronyx Agassiz as second member, and Ichthyokentema for the two species Ph. purbeckensis Davies and Ph. brevis Davies. The last named genus has recently been treated in an excellent manner by Griffith & Patterson (1963), and its differences from Pholidophorus have proved to be so striking that a new family, Ichthyokentemidae, has been erected for it. The genera Pholidophoroides and Pholidophoristion have, however, not yet been thoroughly investigated and their validity consequently not proved. A rather large number of specimens referred to the species Ph. caudalis Woodward and belonging to the British Museum (Natural History) attracted my attention be- cause some of them showed a striking similarity to members of the genus Lepfolepis. Because of this I have tried to study this material in more detail and have arrived at the conclusion that the specimens labelled as Ph. caudalis represent numerous different species. As Ph. caudalis was considered by Woodward as probably belong- ing to his new genus Pholidophoroides I have also found it necessary to take the type species Ph. crenulatus into consideration, and have tried to make a redescription of this species based on the excellent material in the British Museum (Natural History). From Woodward’s (1895) description of Pholidophorus limbatus Agassiz it seemed to me not unlikely that the species could have some relationship to the genus Pholido- phoroides. It is therefore included here. A definition of the genus Pholidophoroides required a comparison with the type species of the genus Pholidophorus. It is, however, not obvious which species should be considered as the type species of this genus. The genus Pholidophorus was erected by Agassiz (1832) for the two species Ph. latiusculus and Ph. pusillus from the Upper Trias of Seefeld, Tyrol. The diagnoses for the genus and for the two species are rather scanty and meaningless: “‘ Pholidophorus Ag. Haringsgestalt. Grosse rautenformige Schuppen. Schwanzflosse ziemlich gleichlappig, indessen ziehen sich die Schuppen noch an den obern Lappen hinauf. Rtickenflosse den Bauchflossen gegeniiber. Afterflosse sehr klein. “zr. Ph. latiusculus Ag. Gréssere Schuppen. Im Verhaltniss breiter als der folgende. “2. Ph. pusillus Ag. Beide von Seefeld in Tyrol. In der Sammlung meines Freundes Dr. Alex. Braun, und letztere auch im Museum in Carlsruhe.”’ CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 355 In his later work Agassiz (1833, 2:9) mentions the genus Pholidophorus for the first time, with the following five species : “t. Pholidophorus limbatus Agass. Ecailles frangées a leur bord postérieur. Corps trés-allongé. Lias: Lyme Regis. “2. Pholidophorus dorsalis Agass. Caractérisé par de longs chevrons sur le bord du premier rayon de la dorsale. Lias: Seefeld. “3, Pholidophorus latiusculus Agass. Plus court; écailles plus grandes. See- feld. “4. Pholidophorus pusillus Agass. Ecailles trés-petites. Seefeld. “5. Pholidophorus nucrops Agass. Téte petite ; écailles en scie fine a leur bord postérieur, plus hautes que larges. Sohlenhofen.”’ A more exhaustive description of the genus and its twenty species follows on p. 271 of the same volume (1844) ; the first one treated is Ph. bechet Agassiz. Ph. latiusculus is only mentioned on p. 287 as the second species (after Ph. dorsalis) among those which were not figured for want of space but which the author intended to describe later on. The short note on PA. latiusculus runs as follows: “ Du lias de Seefeld et de Lyme Regis. Espece trés-voisine de la précédente, mais plus petite, ayant la dorsale moins reculée ; elle n’a guere que deux a trois pouces de long.” The first species ascribed to the genus Pholidophorus is thus Ph. latiusculus, but Agassiz never named a type species of the genus nor a holotype of Jatiusculus, and never gave a figure of it; the diagnoses cited above are too meagre and meaningless to allow an exact identification of the species. In his treatment of the fossil fishes from the Upper Trias of Seefeld, Kner (1866) tried to identify the three Pholidophorus species dorsalis, latiusculus, and pusillus, but says regarding the proposed identifica- tion: “ Ich hoffe hiedurch wenigstens anderen Paldontologen festere Anhaltspunkte zur Unterscheidung der Arten zu bieten und ihnen anschaulich zu machen, welche Formen mindestens mir den drei Arten von Agassiz zu entsprechen scheinen ; ob meine Deutung die richtige sei, dariiber mogen sie selbst dann entscheiden.’” Wood- ward (1895) obviously accepted the identification proposed by Kner and I cannot find any objection to this. Regarding the type species of the genus Pholidophorus Woodward did not choose the first named Jatiusculus but bechet, the first species described and figured by Agassiz (1837, 2, pl. 39, figs. 1-4; 1844, 2: 272). Woodward (1895 : 450-451) gives a good diagnosis of the latter species, which has more recently been investigated by Miss Rayner (1941, 1948), who was, however, principally interested in the study of the endocranium ; her description of the exoskeletal cranial bones and the accom- panying text-figures are thus rather schematic and do not allow a detailed comparison with other, closely related species. Consequently I have found it necessary to attempt a redescription and a new reconstruction of the exoskeletal cranial bones, based on some specimens belonging to the Department of Palaeontology, British Museum (Natural History). The descriptions of Ph. latiusculus given by Kner (1866) and Woodward (1895) are also quite insufficient today. During a visit to Innsbruck in November 1963 I had 356 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE the opportunity to see in the Univ.-Institut fiir Geologie und Paldontologie some specimens from Seefeld, identified as Ph. latiusculus, among them a rather well preserved specimen figured by Kner (1866, pl. 3, fig. 3). As Ph. latiusculus seems to be closely related to Ph. bechei I find it convenient to append a short description of this species also, partly based on the specimens loaned from Innsbruck, partly on specimens belonging to the British Museum (Natural History), as well as of another interesting species, Pholidophorus caffi Airaghi, the holotype of which has been kindly placed at my disposal by the Director of the Museo Civico di Scienze Naturali “FE. Caffi’’, Bergamo, Italy. The description of a specimen probably belonging to Pholidophorus pusillus has also been included. Unless otherwise stated all the registered numbers given in the text refer to speci- mens in the British Museum (Natural History) collections. i -S¥sTEMATIC DE SCRTPRTO NS Genus PHOLIDOPHORUS Agassiz 1832 Pholidophorus Agassiz : 145. PRELIMINARY DIAGNOSIS. Pholidophoridae of small to medium size. Exoskeletal cranial bones and scales with ganoin covering. Nasal well developed and well separated from its antimere by the frontals. Two well-developed supraorbitals. Maxillary not markedly stout and deep, posterior margin evenly rounded. Two supramaxillaries, overlapping dorsal margin of maxillary ; supramaxillary 2 without a marked process at antero-dorsal corner. Antorbital rather small; five infra- orbitals. Preoperculum with preopercular sensory canal running nearer to anterior than to posterior margin. Lower jaw not markedly deep, its greatest depth being in the posterior third of its length ; dentary with a smooth dental part, separated from ornamented splenial part by strong ridge. Dorsal fin above base of ventral fins. Fulcra present along anterior margin of at least dorsal, pectoral and ventral fins ; caudal fin hemi-heterocercal with dorsal and ventral margins fulcrated. Scales rather thick with articulating pegs and with their posterior margin smooth ; anterior lateral line scales much deeper than broad. TYPE SPECIES. Pholidophorus bechet Agassiz. REMARKS. The diagnosis given above for the genus Pholidophorus sensu stricto must at present be regarded as a preliminary one, and this mainly for two reasons. Firstly, it is based on two species only, the type species of the genus and Ph. latiusculus Agassiz ; only after a thorough study of the remaining species earlier ascribed to Pholidophorus or at least of a good number of them, can the limits of this genus be made out and the common features of all its known species be established. Secondly, the present investigation only deals with external features, mainly the exoskeletal cranial bones ; an investigation of the endocranium and the visceral skeleton will undoubtedly reveal further characteristics of value for a definitive diagnosis. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 357 Pholidophorus bechei Agassiz (Pishx, 2) 3 > DText-figs. 1;'2) 1837 Pholidophorus bechei Agassiz, 2, pl. 39, figs. 1-4. 1837 Pholidophorus onychius Agassiz, 2, pl. 39, figs. 5-7. 1844 Pholidophorus bechei Agassiz, 2, 1 : 272. 1844 Pholidophorus onychius Agassiz, 2, I : 274. 1895 Pholidophorus bechet Agassiz ; Woodward : 450, pl. 12, figs. I, 2. 1941 Pholidophorus Rayner : 230, text-fig. Io. 1948 Pholidophorus bechei Agassiz; Rayner : 318, pl. 21, fig. 45, text-figs. 24-29. 1963 LPholidophorus bechei Agassiz ; Griffith & Patterson : 31. PRELIMINARY DIAGNOSIS. Pholidophorus of medium size, up to about 200 mm. in total length. Nasal very large, anteriorly rounded and reaching beyond the anterior tip of the frontal. Posterior margin of preoperculum deeply notched. Pre- opercular sensory canal with about 17—19 tubules. Hototype. The first known specimen of this species is the one described and figured by de la Beche (1822) who, however, did not propose a scientific name for it ; according to Agassiz (1844, Vol. 2, pt. I: 273) this specimen was preserved in the collection of the Geological Society, London, but the specimen could not be found (March, 1965) in the collection of the Geological Survey Museum, London, where the British part of the Geological Society’s collection is now housed. The specimen figured by Agassiz (1837, pl. 39, fig. 1) belonged at that time to Miss Philpot, Lyme Regis. MATERIAL. The specimens used for the following description are Nos. 25276, 38107, 38109, 39859, P.154, P.1051, P.1052d, P.3586c, and P.3589a, all belonging to the British Museum (Natural History), London. Unfortunately they are all more or less defective, especially in the ethmoidal region. DESCRIPTION. This species may attain a total length of about 200 mm. according to Woodward (1895 : 450), who also states that the length of the head with opercular apparatus is somewhat less than the maximum depth of the trunk and occupies one-fifth of the total length of the fish. As the main purpose of my investigation is to study the exoskeletal cranial bones of the type species of Pholidophorus, I have not examined the whole material in the British Museum (Natural History) but only some specimens showing these details. The largest of them, 25276, has a total length of about 164 mm., a standard length of about 140 mm. and a length of the head of about 37 mm., thus somewhat more than one-fifth of the total length (ca. 22:5%) and about one quarter of the standard length (ca. 26°%). These values correspond rather well with those given by Woodward. Of the other larger specimens P.1051 is defective but seems to be a little larger than 25276, while P.1052d is a little smaller than that specimen. 38107 and 38109 have a standard length of 113 mm. and about 125 mm. respectively. The remaining three specimens are, however, much smaller with a standard length of about 80 mm. (P.3589a), 76 mm. (P.154), and 64 mm. (39859), respectively. Regarding the exoskeletal cranial bones no differences of taxonomic value can be observed between the larger and the smaller specimens, as far as their preservation 358 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE allows a comparison. There exists, however, a marked difference regarding the development of the ganoin layer on the cranial bones as well as on the scales. In the largest specimens (Pl. 2, fig. 3) there is a continuous, thick layer of ganoin, more or less richly ornamented, on all exposed parts of the exoskeletal cranial bones and all the scales are thick. In the smaller specimens (64-80 mm. standard length ; Pl. 2, figs. 1, 2) the ganoin covering is much thinner, on the exoskeletal cranial bones appearing as streaks, spots or small tuberculations, and the ganoin covering on the body scales seems to be remarkably thin in the anterior part of the body, gradually becoming thicker posteriorly. Even in 38107 (standard length 113 mm.) the scales on the anterior part of the body look thinner than in the posterior part. Perhaps this may be an indication of the beginning of the reduction of the ganoin covering which must have taken place in the phylogenetic development from the holostean to the teleostean stages of evolution. \ OrP, poc orp Fic. 1. Pholidophorus bechet Agassiz. Attempted restoration of head in lateral view. x 3.3. Ang, angular; Ant, antorbital; De. Spl, dentary; Dpt, dermopterotic; Dsph, dermo- sphenotic ; Ext, extrascapular; Fy, frontal; Ifo,—Ifo;, infraorbitals 1 to 5; Jop, interoper- culum; Myx, maxillary; Na, nasal; Op, operculum; Pa, parietal; Pmzx, premaxillary ; Pop, preoperculum ; fo, rostral ; Sbo, suborbital ; Smx,, Smx,, anterior and posterior supra- maxillaries; So,, So,, anterior and posterior supraorbitals; Sop, suboperculum; Sse, suprascapula ; ap, anterior pit-line; hc,, anterior division of supramaxillary pit-line ; zfc, infraorbital sensory canal; ifc. com, ethmoidal commissure; mp, middle pit-line; orp, postmaxillary pit-line; ovp,, oral pit-line; oc, preopercular sensory canal; s. com, supratemporal commissure ; soc, supraorbital sensory canal. / Ang CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 359 Exoskeletal skull roof The premauilary (Pmx, Pl. x; Pl. 2, figs. 1, 27 Pl. 3, figs. 3,4; Text-fig. 1) has the form of an equilateral triangle with the dorsal margin almost straight, the postero- dorsal margin weakly S-shaped to fit the margin of the maxillary, and the antero- ventral margin a little convex and carrying a single row of 15 or more curved teeth. The slightly bulging outer surface carries a few ganoin tuberculations. The rostral (Ro, Pl. 2, figs. 2, 3; Text-figs. 1, 2) is best preserved in the small specimen P.3589a. Its median, bulging part is ornamented with a few ganoin spots, its ventro-lateral parts are, however, very defective and nothing can be made out with accuracy regarding the outline of the bone. In P.1052d parts of the rostral can also be observed. The nasal (Na, Pl. 2, figs. 1-3; Text-figs. 1, 2) is comparatively very large. Anteriorly it is evenly rounded ; in its posterior part it tapers gradually backwards. In the smallest specimen investigated, 39859, its lateral margin has an almost sem1- circular notch for the posterior nasal opening (PI. 2, fig. 1), in the largest specimen P.1052d, this opening is entirely surrounded by the bone (PI. 2, fig. 2) just as figured by Miss Rayner (1941, text-fig. 10). In these two specimens as well as in P.3589a the nasals of both sides are well separated by the anterior tip of the frontals ; nothing indicates that the nasals meet in the mid-line as figured by Miss Rayner. In the smallest specimen the dorsal surface of the nasal is quite smooth but in the two other specimens mentioned it is provided with a few ganoin spots, principally on the ridge indicating the course of the supraorbital sensory canal. The frontal (Fr, Pl. 2, figs. 1-3; Text-figs. I, 2) is pointed in its anterior part, mesial to the nasal, and from this region backwards it becomes progressively broader and has its broadest part posteriorly. Above the orbit the frontal margin is slightly concave, posterior to it the margin is almost straight and posteriorly directed. The postero-lateral corner of the frontal is rounded off and its posterior margin seems to be slightly wavy. The suture between the frontals of both sides is straight anteriorly but in the middle of its course it is irregularly sinuous, its path varying from specimen to specimen. The dorsal surface of the frontal is practically smooth in the smallest specimen (Pl. 2, fig. 1) ; in P.3589a@ there are smaller and larger, partly confluent ganoin spots (Pl. 2, fig. 2) and in the large specimen, P.1052d, the whole dorsal surface of the frontal is covered with ganoin forming radiating ridges of tubercula- tions on the lateral part of the anterior half of the bone (PI. 2, fig. 3). There are two well-developed supraorbitals. The anterior one, supraorbital 1 (So,, Pl. 2, figs. 1-3; Text-figs. 1, 2), is situated in the angle between nasal and frontal ; its posterior half is rather broad and posteriorly rounded off, antero- laterally it tapers considerably. In the smallest specimen, 39859, its surface is smooth, in P.3589a@ and especially in the large specimen, P.1052d, the surface is ornamented with ganoin tuberculations. Posterior to supraorbital 1 there is an elongate supraorbital 2 in P.3589a (Sop, Pl. 2, fig. 3; Text-figs. 1, 2); its mesial margin is slightly convex, fitting the concave margin of the frontal, anteriorly its margin is concave, fitting the posteriorly rounded supraorbital x and posteriorly it tapers gradually. Its dorso-lateral surface is covered with a rather thick ganoin 360 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Tic. 2. Pholidophorus bechei Agassiz. Attempted restoration of head in dorsal view. x 3-3. Lettering as in Fig. 1. layer. In P.1052d there is posterior to supraorbital 1 a small, rounded supraorbital 2; it seems not to be broken off posteriorly ; some bone fragments posterior to it may be other parts of the same bone, perhaps fragmented during an early stage in the ontogenetic development (So, Pl. 2, fig. 2). The dermosphenotic (Dsph, Pl. 1; Pl. 2, fig. 2; Text-figs. 1, 2) is missing in most specimens, and in P.1052d only fragments of it are visible. In 38107, however, a rather well-preserved, roughly triangular dermosphenotic is present with its dorso- posterior margin convex and with its slightly concave and thickened anterior margin constituting the postero-dorsal border of the orbit. The shape of the exposed surface of the parietal (Pa, Text-figs. 1, 2) is difficult to determine because of the poor preservation of this part of the skull roof, but it seems to be roughly square. The dermopterotic (Dpt, Pl. 2, figs. 1-3; Text-figs. 1, 2) is comparatively narrow in its anterior part lateral to the posterior part of the frontal, but it broadens posterior to that bone on the dorsal surface of the skull, where it meets the parietal. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 361 The extrascapular (Ext, Pl. 1; Pl. 2, fig. 2; Text-figs. 1, 2) is heavily crushed in all specimens investigated by me, but it seems to have the usual triangular shape. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pl.1; Pl 2, figs. 1,2; Pl. 3, figs. 3,4; Text-fig. 1) is anteriorly pointed but it deepens ventrally, posterior to the premaxillary, so that a notch is formed for the premaxillary. The maxillary then becomes deeper with its greatest depth near its evenly rounded posterior margin ; its ventral margin is evenly convex. The dorsal margin, on the whole slightly concave, is overlapped by infraorbital 1 (lachrymal) and the two supramaxillaries; anteriorly, at the deepening of the maxillary, its dorsal margin shows a shallow but marked concavity for the antero- ventral part of infraorbital 1. The lateral surface of the maxillary is ornamented with a longitudinal striation which is parallel to the dorsal and ventral margins in the larger specimens ; however, in the smallest specimen, 39859, this ornamentation is only feebly marked (PI. 2, fig. 1), in the somewhat larger specimens, P.154 and P.3589a, the ganoin streaks are broader and more irregularly arranged. Regarding the dentition along the ventral margin of the maxillary nothing can be said with accuracy because of the defective state of preservation; in 38107 and 38109 the posterior part of the ventral margin shows, however, some markings which suggest a feeble dentition, and in P.3589a the same part carries a delicate, comb-like dentition as in Leptolepis. The two supramaxillanes (Smx,, Smx, Pl. 1; Pl. 2, fig. 1; Pl. 3, fig. 4; Text- fig. 1) are situated dorsal to the posterior part of the maxillary, partly overlapping its dorsal margin. The anterior one, supramaxillary I, is comparatively small and roughly semicircular with its ventral margin straight. Its lateral surface has a ganoin ornamentation of irregular patches in the smaller specimens ; in the larger specimens they are confluent. In 38109 and P.1051 the anterior tip of the bone carries a well marked <-shaped crest parallel to its margins (PI. 3, figs. 4, 5). The posteriorly situated supramaxillary 2 is much larger than supramaxillary 1 and almost triangular in shape with its greatest depth anteriorly ; its ventral margin is straight, its dorsal margin convex, and its anterior margin concave to fit the convex margin of supramaxillary 1. Its antero-dorsal tip is slightly produced, and is apparently more pronounced in the smaller specimens. The lateral surface of supramaxillary 2 is ornamented with irregular ganoin spots in the smaller specimens: in the larger ones the entire surface is covered by a layer of ganoin with feebly marked tuberculations. The six bones of the infraorbital series (antorbital—five infraorbitals) are rather well preserved. The antorbital (Ant, Pl. 2, fig. 1; Pl. 3, fig. 2; Text-figs. 1, 2), well exposed in 25276, 39859 and P.3586c, is comparatively small, almost triangular in shape with the angles rounded off. Its anterior and ventral margins are slightly convex and its posterior margin is straight or moderately concave. Infraorbital 1 (lachrymal) (Ifo,, Pl. 2, fig. 1; Pl. 3, figs. 3-5; Text-fig. 1) is well preserved in 38109. It is elongate and deepest anteriorly. Its convex ventral 362 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE margin has anteriorly a shallow notch, its postero-dorsal margin is not entirely visible but seems to be slightly concave. Infraorbital 2 (Ifo, Pl. 1; Text-fig. 1), clearly visible in 38107, has the usual elongate shape with the dorsal margin slightly concave. The ventral margin is slightly convex. Infraorbital 3 (Ifo,, Pl. 1; Pl. 2, fig. 1; Text-fig. 1) is by far the largest bone in the series, stretching from the orbit to the anterior margin of the preoperculum, which it slightly overlaps. Its antero-dorsal and antero-ventral margins are concave, its dorsal margin is straight and its posterior margin is slightly convex. Infraorbitals 4 and 5 (Ifo,, Ifo;, Pl. 1; Pl. 2, fig. 1; Text-fig. 1) are almost square with their anterior margins slightly concave and their posterior margins slightly convex. Infraorbital 5 is not entirely exposed in any of the specimens investigated but it seems to be a little smaller than infraorbital 4. A comparatively large suborbital (Sbo, Pl. 1; Text-fig. 1) is situated posterior to infraorbitals 4 and 5 and seems to overlap the antero-dorsal part of the operculum. It is, however, generally crushed in the specimens available and in consequence of this, its outline can only be made out in part. The preoperculum (Pop, Pl. 1; Pl. 2, figs. 1, 2; Pl. 3, figs. 4,5; Text-fig. 1) has a very characteristic shape. The dorsal, almost vertical limb broadens continuously ventrally. The antero-ventrally directed limb is remarkably broad. The posterior margin of the bone is deeply notched below the angle between the two limbs, the postero-ventral corner projects backwards but is rounded off, not pointed as in Miss Rayner’s reconstruction, the ventral margin is almost straight or slightly convex and the anterior margin shows a rather marked convexity below the middle of the bone. The operculum (Op, Pl. 1; Pl. 2, fig. 1; Text-fig. 1) is roughly triangular; the straight anterior, vertical margin and the straight postero-ventral margin meet at a low angle, the dorsal margin is evenly rounded. The anterior margin seems to be a little thickened. The suboperculum (Sop, Pl. 1; Pl. 3, figs. 4, 5; Text-fig. 1) is comparatively large, but it is not entirely preserved in any of the specimens investigated and con- sequently its outline cannot be made out with accuracy. The antero-dorsal process anterior to the ventral tip of the operculum is comparatively strong. The posterior part of the interoperculum (Iop, Pl. 1; Pl. 3, figs. 4,5; Text-fig. 1) is well exposed in 38107, 38109, and P.1051 ; its anterior part is not visible in any of the specimens investigated. Branchiostegal rays and gular plate Branchiostegal rays are exposed in 38107, 38109, and P.1051. In 38107 there are at least 12 branchiostegal rays on the right side (R. By, Pl. 1), in P.1051 at least 13 rays may be distinguished, the six anterior ones still attached to the large ceratohyal 2 (Chg, Pl. 3, fig. 5), which is unfenestrated but provided with a deep groove for the hyoid artery. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 363 An elongate, anteriorly keeled, detached but not entirely exposed bone in P.3589a may be interpreted as the gular plate (Gu, Pl. 3, fig. 1). A fragment of a gular plate is probably visible also in P.154. Lower jaw The lower jaw is more or less exposed in most specimens ; in P.1051 it is almost entirely free from overlying bones (PI. 3, fig. 5) ; it is a little defective but the outline of the defective parts can be followed through the impression in the rock. The dental and splenial parts of the dentary (De. Spi, Pl. 1; Pl. 2, figs. 1,2; Pl. 3, figs. 4,5; Text-fig. 1) are separated by a prominent ridge on the lateral surface of the bone ; the anterior part of its dorsal margin is only gently ascending ; at about the limit between the first and second third of the length of the lower jaw the dorsal margin of the dentary is abruptly thickened and ascends postero-dorsally. The boundary between the dentary and the angulo-supra-angular cannot be made out. The lateral surface of the splenial part is ornamented with thick striations and rugo- sities, whereas that of the dental part is quite smooth. In 38109 there are, as already stated by Miss Rayner (1948 : 319), some delicate teeth a little distance from the anterior tip of the dentary (PI. 3, fig. 4) ; as far as I can see there are seven. The angulo-supra-angular (Ang, Pl. 1; Pl. 2, fig. 1; Pl. 3, figs. 4, 5) constitutes the posterior part of the lower jaw; its depth equals about one-third of the length of the jaw. The lateral surface of the angulo-supra-angular has ventrally a con- spicuous ganoin ornamentation. No true dermarticular can be seen. The articular (Art, Pl.1; Pl. 2, fig. 1; Pl. 3, figs. 4, 5) is partly visible in 38109. Sensory canal system of head The sensory canal system of the head is well developed, but because of the bad state of preservation of certain bones it cannot be described in every detail; in particular the lack of a specimen with the bones of the snout in natural position makes the interpretation uncertain. The supraorbital sensory canal (soc, Pl. 2, fig. 1; Text-figs. 1, 2) pierces the nasal, the frontal and the parietal. The nasal part of the canal is clearly visible in 39859, P.1052d (PI. 2, fig. 3), and P.3589a (Pl. 2, fig. 2) ; it runs in a gentle arch from the antero-lateral to the postero- mesial margin of the bone, where it enters the frontal. In the large specimen, P.1052d, there is a single pore visible on the dorso-mesial side of the canal a little anterior to the fenestra for the nostril. In the smaller specimen, P.358ga, a corres- ponding pore may be traced on the dorso-lateral side of the canal; in the smallest specimen no pores can be observed. The frontal part of the supraorbital sensory canal is very difficult to see because of the crushed state of the frontal in all specimens at my disposal. In the anterior part of the bone it runs, however, at first almost directly backwards, then it curves postero-laterally parallel to the lateral margin of the frontal and finally continues postero-mesially to the posterior margin of the bone. Pores or short tubules can 364 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE only be observed in P.1052d, and there the state of preservation does not allow any certain conclusions. According to Rayner (1948) there is great variability regarding the number and position of tubules and pores in this part of the supraorbital sensory canal; my interpretation of the arrangement in P.1052d is given in Text-figs. I, 2. On the left side where the canal can be followed practically without interruption, except for its hindmost part, I have observed only a single short, laterally directed tubule in the posterior part of the anterior straight course of the canal ; at the curve there issue two somewhat longer, postero-mesially directed tubules from the mesial side of the canal and behind them two very short tubules on the same side of the canal; on the postero-mesially directed part of the canal there are five very short tubules or pores. On the right side, where the canal is more defective, the tubules and pores seem to have on the whole the same arrangement, as far as they are discernible. The parietal part of the supraorbital sensory canal is visible on the left parietal ; the canal itself is very short, apparently without tubules or pores. It continues backwards as a well-marked groove, the anterior pit-line (ap, Pl. 2, fig. 2; Text-figs. 2) The infraorbital sensory canals (ifc, Pl. 3, fig. 4; Text-figs. 1, 2) of both sides are joined anteriorly by a slightly arched ethmoidal commissure in the rostral. In P.1052d a short, posteriorly directed tubule can be seen on each side of the mid-line of the bone. The small antorbital contains in its ventral half the curved anterior part of the infraorbital sensory canal. From its convex dorsal side the canal gives off a short antorbital branch, ending with a comparatively large pore ; no more pores can be observed on this part of the canal except in 25276, where a pore seems to be present at the anterior angle between the canal and the antorbital branch (Pl. 3, fig. 2). According to Rayner (1941, text-fig. 10B ; 1948, text-fig. 25) the antorbital branch joins the supraorbital sensory canal “‘to form a closed circuit around the eye”’. I cannot, however, share this opinion. As already mentioned above no specimen seen by me has the exoskeletal bones of the snout preserved in natural articulation, but when trying to reconstruct this part of the skull I have not been able to arrive at a solution other than that given in Text-fig. 1. Only much better preserved material regarding the mutual position of the bones in question can decide which solution is correct. Passing into infraorbital 1 (lachrymal) the infraorbital sensory canal pierces this bone on the whole parallel to its convex anterior and ventral margins. Judging from 38109 (PI. 3, fig. 4) the canal gives off from its anterior side three short, antero- ventrally directed tubules and from its ventral side at least seven ventrally to postero- ventrally directed tubules, the middle ones being the largest ; as the bone is apparently not complete in its posteriormost part there might, consequently, exist one or more tubules in infraorbital 1 in addition to the ten observed by me. In infraorbital 2, only preserved in 38107, the sensory canal is not visible. In infraorbital 3 the sensory canal runs parallel to the concave anterior margin of the bone. In P.154 it gives off two postero-ventrally directed tubules, the anterior CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 365 one rather short, and two posteriorly directed tubules. In 38107 there seem to be only three tubules, only one of them posteriorly directed ; in the other specimens the state of preservation does not allow any observations regarding the tubules. In infraorbitals 4 and 5 the sensory canal gives off a single, posteriorly directed tubule in each bone. In the dermosphenotic, well preserved in 38107, the infraorbital sensory canal curves posteriorly ; from its convex dorsal side the canal gives off a rather wide, dorsally directed tubule, ending in a pore ; this pore is also visible on the defective dermosphenotic in P.1052d. From the dermosphenotic the canal enters the dermopterotic, piercing it along its lateral margin. To what extent the canal in the dermopterotic belongs to the sensory canal system of the head and to the cephalic division of the main lateral line, res- pectively, must be left unanswered. Judging from P.3589a the posterior part of the canal gives off two dorsally directed short tubules, and the preopercular sensory canal seems to issue at the postero-lateral corner of the bone. Cephalic division of main lateral line A well-marked middle pit-line is visible in P.1052d on the left parietal posterior to the anterior pit-line ; it extends laterally over the broad posterior part of the dermopterotic. The cephalic division of the main lateral line passes over from the dermopterotic into the extrascapular, where it gives off mesially the supratemporal commissure, and continues posteriorly into the suprascapular. Because of the preservation of the extrascapular the canal and its tubules can be only partly observed ; in P.1052d two rather long tubules are given off from the lateral side of the canal, and at least four short tubules issue from the posterior side of the supratemporal commissure. In 38107 also, four short tubules are given off from the posterior side of the supra- temporal commissure decreasing in length mesially, but in this specimen only one large tubule can be observed with accuracy lateral to the main canal. The preopercular sensory canal runs on the whole parallel to the curved anterior margin of the preoperculum, about midway between its anterior and posterior mar- gins in the dorsal limb but decidedly nearer to its anterior margin in its antero- ventrallimb. From its posterior side the canal gives off a series of tubules the number and arrangement of which are difficult to determine with precision in the largest specimens because of the thickness of the bone, especially in the dorsal limb of the bone ; in the small specimens the preoperculum is too defective to allow any detailed analysis of number and position of the tubules, but their arrangement seems to be generally the same as in the larger specimens. In the antero-ventral limb of the preoperculum there are in 25276, 38107, and 38109 thirteen postero-ventrally directed, curved tubules increasing in length from the anterior small one to the twelfth ; the thirteenth tubule is shorter and does not reach the postero-lateral margin of the bone. In P.1051 there is the same characteristic arrangement of the tubules, but their number cannot be counted with accuracy. Dorsal to this 366 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE group of tubules the tubules are shorter and seem to vary a little in number and posi- tion. 25276 clearly shows six rather short tubules, in 38109 only five tubules can be observed, but the dorsalmost part of the bone is defective ; in 38107 and P.154 at least 4 short tubules are present, but also here their exact number is difficult to determine. In P.1051, in which the preoperculum is remarkably thick and, more- over, broken in its dorsal part, only three short tubules can be seen. The real number of small tubules in the dorsal part of the preoperculum may be estimated as five or six and the total number of tubules belonging to the preopercular sensory canal may consequently be given as 18-19. On the lateral surface of the preoperculum at about its middle and anterior to the preopercular sensory canal two short grooves are clearly visible in most specimens, the dorsal one more or less horizontal, the ventral one on the whole vertical. It seems most likely that the horizontal groove represents the posterior portion of the anterior division of the supramaxillary line and the vertical one the post-maxillary line accord- ing to the nomenclature proposed by Stensi6 (1947). In P.154, however, no supra- maxillary line can be observed on the preoperculum, but on infraorbital 3 an oblique groove is present, which may be interpreted as the anterior portion of the anterior division of the supramaxillary line. This difference in the development of the supra- maxillary line may merely be an individual variation rather than a feature of taxono- mic importance. The mandibular sensory canal and its tubules are not visible, only the openings of some separate tubules may be observed, but their total number cannot be deter- mined. In the largest specimen investigated, 25276, as well as in the small specimens 39859 and P.154, there is a small vertical groove on the lateral surface of the angular near its ventral margin ; this groove obviously represents the oral line according to Stensi6 (1947). Exoskeletal shoulder girdle and squamation The suprascapula (Ssc, Pl. 1) is partly visible in 38107 and P.1052d, but in neither case can its outline be determined ; it is, however, a rather large bone. In 38107 a smaller, reniform bone postero-ventral to the suprascapula may be the supracleithrum, but no sensory canal can be observed in it. The cleithrum is only to a small extent exposed in the material studied by me and no description can be given. The scales are, according to Woodward (1895), arranged in approximately 40 transverse rows ; in none of the specimens seen by me is the squamation uninterrupted but in 38107 there may be about 45 transverse rows reckoned from the posterior margin of the operculum to the middle of the caudal fin. On the body there are about four longitudinal rows of scales which are deeper than broad, the lateral line scales being the deepest. All scales have an even posterior margin. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 367 Lateral line The parts of the lateral line piercing the suprascapula and the supracleithrum are not visible in any of the specimens exhibiting these bones, and in the supra- cleithrum no tubules can be observed. In the suprascapula two rather wide tubules are clearly visible, the anterior one mesially, the posterior one postero-mesially directed. The lateral line runs almost straight along the sides of body and tail, ending at the base of the middle caudal rays. Each lateral line scale has a pore at about the middle of its lateral surface. Paired and unpaired fins The fins are as a rule more or less defective in all specimens investigated by me and consequently little can be added in this respect to the facts given by Woodward (1895 : 451). The pectoral fin seems to be of moderate size and with a moderate number of lepidotrichia, about 18 according to my counts, i.e. the same number as given by Woodward for the pectoral. The first lepidotrichium is rather stout and is provided with some fulcra. The ventral fin has a rather high number of lepidotrichia, about 14 or 15 (14 accord- ing to Woodward), the innermost ones rather delicate and branched almost to their bases. The first two lepidotrichia are comparatively short and undivided, the third one (the first branched lepidotrichium) carries some fulcra. The dorsal fin has, according to Woodward, about 12 lepidotrichia ; in my material no dorsal fin is complete. In P.154 the anterior part of the fin is rather well preserved and shows anteriorly three undivided lepidotrichia followed by a long divided lepido- trichium with few but rather long fulcra. Regarding the anal fin Woodward gives no number of lepidotrichia, and in my material all anal fins are damaged and incomplete. On none of them have I observed any traces of fulcra, but as Woodward mentions the presence of fulcra on all fins, he must have seen some on the anal fin also. The caudal fin is hemi-heterocercal and provided with densely set fulcra on its dorsal as well as on its ventral margin. REMARKS. Woodward (1895 : 450) considers Ph. onychius Agassiz to be identical with Ph. bechei. I have had no opportunity to see the holotype of Ph. onychius, belonging to the Oxford Museum, but 38109, originally labelled as Ph. onychius is, as far as I can see, in all respects a quite typical Ph. bechet, which supports Wood- ward’s opinion. HorRIzON AND Locality. Lower Lias; Lyme Regis, Dorset. GEOL. II, 8 33 368 CERTAIN DRIASSIC AND LIASSTIG PHOLTIDOPHORI DAE Pholidophorus latiusculus Agassiz (Pleas Blige eles iigss 271077 oailiextties: 2.04) 1832 Pholidophorus latiusculus Agassiz : 145. 1833 Pholidophorus latiusculus Agassiz; Agassiz, 2: 9. 1844 Pholidophorus latiusculus Agassiz; Agassiz, 2, I : 271, 287. 1866 Pholidophorus latiusculus Agassiz; Kner : 328, pl. 3, figs. 2, 3. 1867 Pholidophorus latiusculus Agassiz; Kner : 903, pl. 2, fig. 1. 1895 Pholidophorus latiusculus Agassiz ; (partim?) Woodward : 454, ? pl. 14, fig. 3. PRELIMINARY DIAGNOSIS. Pholidophorus of small size, up to about 85 mm. in total length. Nasal not very large, anteriorly pointed. Posterior margin of pre- operculum with a shallow notch. Preopercular sensory canal with about 14-15 tubules. Neotypre. As already mentioned in the introduction, Agassiz did not name a holotype for this species. A species with this name is, however, described and figured by Kner (1866) ; the best preserved of his specimens, figured on Pl. 15, fig. 2, is still present in the collections of the Univ. Institut fiir Geologie und Palaon- tologie, Innsbruck, Austria, and may conveniently be chosen as neotype of Pholido- phorus latiusculus. The correctness of the identification is beyond doubt as can be seen by a comparison between the figure given by Kner (represented on PI. 15, fig. 1) and a photograph in natural size of the same specimen (PI. 15, fig. 2). MATERIAL. The description given below is based on the following specimens : Three specimens, F.123 (neotype), Lit. F and No. 1028, all belonging to the Univ. Institut fiir Geologie und Paldontologie, Innsbruck, and two specimens, 33987 (in counterpart) and P.1063, both belonging to the Department of Palaeontology, British Museum (Natural History); further, a rather disarticulated specimen, P.11780, showing a number of scales and some isolated cranial bones may also belong to this species or perhaps to Ph. cf. pusillus. DeEscripTION. All the specimens, except P.11780, show the same characteristic shape, with the head and forepart of the body bent dorsally. The three Innsbruck specimens are of about the same size with a length without caudal (standard length) of ca. 75 mm. ; in the neotype the caudal fin is defective, in the two other specimens the caudal fin has a length of ca. 10 mm. ; the total length may consequently attain ca.85mm. Specimens 33987 and P.1063 are smaller ; the first with a total length of ca. 75 mm. and a standard length of ca. 65 mm., P.1063 with a standard length of ca. 57mm. The greatest depth of the body cannot be given as all specimens are more or less flattened through pressure. The length of the head seems to be roughly one quarter of the standard length. Exoskeletal skull roof The snout region is defective in all specimens available and the premaxillary and rostral bones are missing. The nasal (Na, Pl. 4; Pl. 5, fig. 1; Text-figs. 3, 4) is preserved only in the neotype and in P.1063 ; it is comparatively large but its anterior part is pointed, not broadly CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 369 Ley, LY, Met y yyy, ge ss st" Fic. 3. Pholidophorus latiusculus Agassiz. Attempted restoration of head in lateral view. X7 approx. Ang, angular; De. Spl, dentary ; Dpt, dermopterotic ; Dsph, dermosphenotic ; Ext, extra- scapular; Fy, frontal; Ifo,—Ifo;, infraorbitals 1 to 5 ; lop, interoperculum ; Myx, maxillary ; Na, nasal; Op, operculum; fa, parietal; Pop, preoperculum; Sbo, suborbital; Smx,, Smx,, anterior and posterior supramaxillaries ; So,, So,, anterior and posterior supraorbitals ; Sop, suboperculum ; ap, anterior pit-line ; fc,, anterior division of supramaxillary pit-line ; tfc, infraorbital sensory canal; mp, middle pit-line; ovp, postmaxillary pit-line; orp,, oral pit-line ; poc, preopercular sensory canal; soc, supraorbital sensory canal. rounded as in Ph. bechei and the bone is not pierced by the posterior nasal opening. Its dorso-lateral surface is ornamented with irregular streaks and tuberculations. The frontal (Fr, Pl. 4; Pl. 5, figs. 1, 2; Text-figs. 3, 4) has on the whole the same shape as in Ph. bechei, but its anterior part, mesial to the nasal, seems to be compara- tively more slender. As in that species the suture between the frontals of both sides is straight anteriorly but in the middle of its course somewhat sinuous ; in P.1063 (Pl. 5, fig. 1) it curves at first a little to the left, then to the right and then again to the left. The entire dorsal surface of the frontal carries small ganoin rugosities which form low radiating ridges in its antero-lateral part. The two supraorbitals (So,, So,, Pl. 4; Pl. 5, fig. 1; Text-figs. 3, 4) are of about the same size. The anterior one, supraorbital 1, is situated posterior to the nasal ; in P.1063 its anterior part is hidden below the posterior part of the nasal, but judging from the neotype its anterior part is rounded off, not drawn out into an antero-lateral 370 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE projection as in Ph. bechet. The mesial margin of supraorbital 2 slightly overlaps the lateral margin of the frontal ; posteriorly its lateral margin is concave, fitting the antero-dorsal margin of the dermosphenotic. Both supraorbitals are ornamented with low ganoin ridges and rugosities. The dermosphenotic (Dsph, Pl. 5, fig. 1; Text-figs. 3, 4) is, judging from P.1063, a rather small bone, pointed anteriorly and rounded posteriorly with its dorsal margin a little convex and its antero-ventral margin a little concave. The parietal (Pa, Pl. 5, fig. 1; Text-figs. 3, 4) and the dermopterotic (Dpt, Pl. 4; Pl. 5, figs. 1, 2; Text-figs. 3, 4) cannot be described in detail because of the bad state of preservation which does not allow a determination of their boundaries. Only the lateral part of the extrascapular is preserved in P.1063 (Ext, Pl. 5, fig. 1). Apparently it is pushed over the posterior part of the dermopterotic, and conse- quently nothing can be made out with regard to its general shape. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pl. 4; Pl. 5, fig. 1; Text-fig. 3) is defective in all specimens but to judge from 33987 it is pointed anteriorly, rather as in Ph. bechei, and then deepens with its deepest part at its posterior rounded end. Its dorsal margin is slightly concave and is overlapped by infraorbital 1 (lachrymal) and the two supra- maxillaries. The lateral surface of the maxillary is provided with a dense longitudinal striation. Along the whole slightly convex ventral margin there is a feeble dentition like that in Leptolepis. The two supramaxillaries (Smx,, Smx,, Pl. 4; PI. 5, fig. 1; Text-fig. 3) overlap the posterior part of the dorsal margin of the maxillary. The anterior one, supra- maxillary 1, best preserved in the neotype (PI. 4), is comparatively small, semicircular to almost triangular in shape ; its lateral surface is ornamented with well-marked concentric ridges or streaks. Among the few cranial bones preserved in P.11780 there is a small semi-circular bone with a marked striation, which agrees rather well with supramaxillary 1 in the neotype. Supramaxillary 2 is fairly large and elongate with its dorsal and ventral margins slightly convex ; posteriorly it is rounded off ; its antero-ventral margin is concave, fitting the postero-dorsal margin of supra- maxillary 1. The antero-dorsal tip of supramaxillary 2 is slightly thickened but not markedly produced. Its lateral surface is ornamented with numerous well- marked striations running parallel to its dorsal and ventral margins ; they are partly confluent in the middle of the bone. Unfortunately, of the bones of the infraorbital series, the antorbital is missing in all specimens investigated. Infraorbital 1 (lachrymal) (Ifo,, Text-fig. 3) is present only in 33987 and there only as a fragment of its anterior part ; nothing therefore can be made out with regard to its general shape. Infraorbital 2 (Ifo, Pl. 5, fig. 1; Text-fig. 3) is preserved in P.1063 ; it is, as usual, a small elongate bone which is situated below the orbit ; its posterior margin is postero-ventrally directed. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 371 \ ee Sue Fic. 4. Pholidophorus latiusculus Agassiz. Attempted restoration of head in dorsal view (possibly somewhat flattened). x7 approx. Lettering as in Fig. 3. Infraorbital 3 (Ifo;, Pl. 4; PI. 5, figs. 1, 3; Text-fig. 3) stretches from the orbit to the anterior margin of the preoperculum, which it overlaps as in Ph. bechei ; its general shape is also almost as in that species. Infraorbital 4 (Ifo,, Pl. 4; Pl. 5, figs. 1, 3; Text-fig. 3) has its anterior, dorsal and ventral margins almost straight whereas its posterior margin is slightly convex ; it is about half as broad as infraorbital 3. Infraorbital 5 (Ifo;, Pl. 4; Pl. 5, figs. 1, 3; Text-fig. 3) can only be seen in part in the neotype, in P.1063 and, in impression, in Innsbruck Lit. F ; it seems to have about the same general shape as infraorbital 4. The suborbital (Sbo, Pl. 4; Pl. 5, fig. 1; Text-fig. 3) seems to be a little broader thanin Ph. becheix. Judging from the neotype its anterior margin lies almost vertical, whereas its posterior margin is slightly curved in the antero-ventral direction. The preoperculum (Pop, Pl. 4; Pl. 5, figs. 1, 3; Text-fig. 3) is best preserved in the neotype, where it is partly hidden below infraorbital 3 and the suborbital, in P.1063 and, in impression, in Innsbruck Lit. F, where its postero-ventral part is lost. In 33987 it is rather defective. Its general shape resembles that of the pre- operculum in Ph. bechei, but its dorsal limb is broader and its postero-ventral part 372 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE does not protrude posteriorly as in that species ; the notch in its posterior margin is consequently much shallower. It is not possible to determine the course of the anterior margin of the preoperculum because of the overlying bones, but judging from the impression in Innsbruck Lit. F, the anterior margin has a slight convexity ventral to the middle of its length (Pl. 5, fig. 3). In P.11780 there is a fragmentary bone which I interpret as the left preoperculum ; its anterior margin is rather well preserved and shows the same convexity as in the specimen just mentioned. The operculum (Op, Pl. 4; Pl. 5, fig. 1; Text-fig. 3) has, on the whole, the same general shape as in Ph. bechei, but its dorsal part seems to be broader and its ventral tip is more obtuse ; its anterior margin is a little thickened except ventrally where the margin is slightly notched for the antero-dorsal process of the suboperculum. The suboperculum (Sop, Pl. 4; Pl. 5, fig. 1; Text-fig. 3) seems to be proportionately as large as in Ph. bechei but its posterior margin is defective in all specimens except P.11780, possibly belonging to this species; in this specimen its posterior, well- preserved part shows a shallow concavity high on the posterior margin. Of the interoperculum (Iop, Pl. 4; Pl. 5, fig. 1; Text-fig. 3) only the posterior part is visible in the neotype, in P.11780 and, in impression, in Innsbruck Lit. F. Branchiostegal rays and gular plate The branchiostegal rays and the gular plate cannot be seen in any of the available specimens. Lower jaw The lower jaw is only partly exposed in the neotype and in 33987 and P.1063 ; in Innsbruck Lit. F. the left lower jaw is exposed but very defective (Pl. 15, fig. 6) and in P.11780, possibly belonging to this species, an isolated right lower jaw is present but partly crushed (PI. 15, fig. 7). These specimens suggest that the lower jaw in Ph. latiusculus has about the same shape as in Ph. bechei (Pl. 13, fig. 5). In all the specimens of Ph. latiusculus the ventral, splenial part of the dentary is separated from the dorsal, dental part by a well-marked ridge and its lateral surface is ornamented with thick ganoin rugosities and irregular striations. The dental part is anteriorly only gently ascending but further back it bends dorsally to form a small process, and then runs postero-dorsally ; its lateral surface is without orna- mentation. No teeth can be observed. The limit between the dentary and the angulo-supra-angular (Ang, Pl. 15, fig. 7) is not clearly visible. The postero-ventral part of the lateral surface of the last named bone carries a well developed ornamentation. Sensory canal system of head The sensory canal system of the head seems to follow the same general pattern as in Ph, bechet, as far as the preservation of the material allows a comparison. The supraorbital sensory canal (soc, Pl. 5, fig. 1; Text-figs. 3, 4) pierces the nasal, the frontal and the parietal. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 373 In the nasal the canal runs almost straight from the antero-lateral to the postero- mesial part of the bone in P.1063; no tubules or pores can be observed. In the neotype this part of the canal is not clearly visible. In the frontal the supraorbital sensory canal is rather well-marked as a slight ridge on the dorsal surface of the bone. Its course is on the whole the same as in Ph. becheit : an anterior part running almost straight posteriorly is followed by a curved part corresponding to the broadening of the frontal; the posterior part runs at first in postero-mesial direction, then straight back to the middle of the posterior margin of the bone. P.1063 has the following arrangement of tubules and pores (Text-figs. 3, 4): from its anterior straight part the left canal gives off laterally two short tubules ; from its curved part three short tubules are given off from the mesial side ; the straight postero-mesially directed part has on its lateral side three very short tubules or pores ; the posteriormost part of the sensory canal is damaged and the presence of pores or tubules in that part cannot be made out. On the right side corresponding tubules are given off from the anterior and the curved parts of the canal, but further back only a single pore or tubule on the mesial side of the canal can be observed. In Innsbruck Lit. F. (Pl. 5, fig. 2) there seem to be three laterally directed tubules issuing from the anterior straight part of the canal but only two pores from the curved part can be observed ; along the lateral side of the posterior, postero-mesially directed part of the canal there are four pores. The parietal part of the supraorbital sensory canal is well exposed in P.1063 (PI. 5, fig. 1). It is comparatively much longer than in Ph. bechei and on the right parietal there is a small shallow concavity on the dorsal surface of the canal which may be a pore, but its interpretation as a true pore is questionable. The sensory canal con- tinues posteriorly as a well-marked groove, the anterior pit-line (ap, Pl. 5, fig. 1; Text-figs. 3, 4). The infraorbital sensory canal (ifc, Text-fig. 3) is quite unknown in its anterior parts because of the absence of the rostral and antorbital bones. Only in 33987 is infraorbital 1 (lachrymal) present, and there only in its antero- ventral part ; it shows the anterior curved part of the canal in this bone with two antero-ventrally directed, pore-like tubules. For the rest the sensory canal and its tubules in infraorbital I are unknown. In infraorbital 2 the sensory canal is not visible. In infraorbital 3 the sensory canal pierces the bone parallel to its concave anterior margin. From its morphologically posterior side it gives off some tubules, the number of which is not easy to determine because of the thickness of the bone. In the neotype (Pl. 4) there seem, however, to be three rather long posteriorly directed tubules situated close together ; antero-ventral to these tubules a short postero- ventrally directed tubule is present. In P.1063 (PI. 5, fig. 1) only two long, posteriorly directed tubules can be seen with certainty ; a short postero-ventrally directed tubule in the anterior part of the bone seems to be present but cannot be observed in detail. In infraorbital 4 the sensory canal gives off posteriorly a single tubule in the neotype. In infraorbital 5, which is only partly visible, no tubule can be observed with accuracy. 374 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Judging from 33987, in which the posterior part of the dermosphenotic is preserved, the posteriorly curving sensory canal pierces this bone in its posterior part ; from the anterior side of the curve a rather wide tubule is given off in antero-dorsal direction. In P.1063 the canal itself cannot be followed exactly, but the tubule is partly visible and seems to end in a pore at the dorsal margin of the bone. The sensory canal passes over into the dermopterotic and runs along its lateral margin. Judging from specimen Innsbruck Lit. F it seems to give off dorsally two short tubules. As in Ph. bechei the preopercular sensory canal seems to be given off near the postero-lateral corner of the bone. Cephalic division of main lateral line The middle pit-line (mp, Pl. 5, figs. 1, 2; Text-figs. 3, 4) is well developed, stretch- ing from the postero-mesial part of the parietal to near the lateral margin of the dermopterotic. The cephalic division of the main lateral line passes over from the dermopterotic to the extrascapular, but as this bone is very badly preserved or missing nothing can be made out regarding this part of the canal and its tubules. The preopercular sensory canal runs, as in Ph. bechei, on the whole parallel to the anterior margin of the preoperculum. The tubules given off from its morpho- logically posterior side show the same general pattern as in Ph. bechet. In the neotype, the best preserved specimen in this respect, there seem to be three short, almost straight tubules dorsal to the angle on the posterior margin of the bone, the dorsalmost difficult to observe with accuracy as it is hidden below the suborbital, and ventral to the angle two shorter, slightly curved tubules; these five dorsal tubules obviously correspond to the four to six shorter dorsal tubules in Ph. bechev. Ventral to these shorter tubules there is, as in Ph. bechet, a rather long tubule which, however, does not reach the posterior margin of the bone. Anteriorly it is followed by nine curved tubules which decrease in length antero-ventrally ; there are thus ten openings belonging to this group of tubules, but it seems as if the fourth tubule reckoned from the antero-ventral tip of the preoperculum is only a branch of the following tubule. The ten antero-ventral tubules in question obviously correspond to the thirteen antero-ventral tubules in Ph. bechet. In P.1063 the shorter dorsal tubules are likewise five in number but the ventral part of the preoperculum is too damaged to allow a count of the tubules. The preopercular impression in Innsbruck Lit. F shows five dorsally situated shorter tubules and ventral to them about nine tubules. The total number of tubules belonging to the preopercular sensory canal in Ph. latiusculus is consequently 14-15 against 17-19 in Ph. bechet. In P.1063 (PI. 5, fig. 1) the lateral surface of the preoperculum shows one almost horizontal (hc,) and one vertical (orp) groove, obviously the posterior portion of the anterior division of the supramaxillary line and the postmaxillary line respectively as described above in Ph. bechet. An oblique, posteriorly directed groove on the lateral surface of infraorbital 3 may represent the anterior portion of the anterior division of the supramaxillary line. In the neotype (Pl. 4) the postmaxillary line CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 375 is hidden below infraorbital 3, but the other two lines are clearly visible, that on infraorbital 3, however, is not continuous but divided into two parts. In Innsbruck Lit. F. (Pl. 5, fig. 3) all the three lines mentioned can be traced on the impression. As compared with the corresponding lines in Ph. becher the two lines on the pre- operculum are proportionately much longer and the line on infraorbital 3 seems to occur normally in all specimens. The mandibular sensory canal can partly be seen in the neotype and in P.1063 ; the openings of separate tubules, marked by ganoin thickenings, are visible here and there ventral to the ridge separating the dental and splenial parts of the dentary ; their total number cannot be made out. In 33987, the only one in which the lateral surface of the angular is clearly exposed, a small vertical groove representing the oral line is visible on the angular near its ventral margin. Exoskeletal shoulder-girdle and squamation A suprascapula cannot be identified in any of the specimens investigated. In the neotype an almost triangular bone with strongly marked ganoin ridges on its lateral surface may be a supracleithrum (Scl, Pl. 4). In the smaller specimen, 33987, the same bone is visible but the ganoin ridges are not as marked as in the neotype ; in P.1063 (Pl. 5, fig. 1) a fragment of the same bone is present. In the last named specimen much of the cleithrum (Cl, Pl. 5, fig. 1) is exposed, and shows a large, deep notch above the insertion of the pectoral fin. The squamation (PI. 4) is only partly preserved, but in the neotype about 38 transverse scale rows can be observed, reckoned from the posterior margin of the operculum to the middle of the caudal fin base. The lateral line scales are the deepest, but also the rows immediately above and below the lateral line scales seem to be deeper than broad. All the scales are comparatively thick and have an even posterior margin. Lateral line Nothing can be made out regarding that part of the lateral line piercing the supra- scapula and the supracleithrum. The lateral line scales are arranged in an almost straight row along the side of the body (./, Pl. 4; cf. also Pl. 15, fig. 2). Paired and unpaired fins Very little can be said regarding the fins in this species. The pectoral fin is partly preserved in all specimens investigated but only in P.1063 can the number of lepidotrichia be counted with accuracy (PI. 5, fig. 1) ; there seem to be about 19 lepidotrichia, the first one carrying some small fulcra. Only a few lepidotrichia belonging to the ventral fin are preserved in specimen 33987, in others the ventral fin is missing. The dorsal and anal fins are lacking in all specimens seen by me, but according to Kner (1867, pl. 2, fig. 1) the dorsal fin is situated approximately above the ventral fins as in Ph. bechei. 376 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE The caudal fin is hemi-heterocercal and carries well-developed fulcra on its dorsal and ventral margins. ReMARKS. Apart from small differences, mainly in the ornamentation of the exoskeletal cranial bones, all the specimens mentioned above except P.11780 are so similar that they may be considered to belong to one and the same species. Whether the specimens from Italian localities attributed to Ph. latiusculus by authors such as de Alessandri, Bassani, Costa, Mariani etc., are identical with the specimens described above is a question which at present must be left open, as I have had no opportunity to see them. HORIZON AND LOCALITY. Upper Trias; Seefeld, Tyrol, Austria. Pholidophorus (?) caffii Airaghi (Pl. 6, fig. 1; Text-fig. 5) 1908 Pholidophorus Caffi Airaghi : 3, text-fig. 2. 1914 Pholidophorus latiusculus Agassiz ; (partim) Bassani : 379. 1920 Pholidophorus latiusculus Agassiz ; (partim) Alessandri : 96. 1937. Pholidophorus latiusculus Agassiz; Boni: 132, pl. 4, fig. 4, pl. 5, fig. 3, text-figs. 9, Io. DiaGnosis. Pholidophorus (?) of smallsize, up to 58 mm.intotallength. Greatest depth of body about one quarter of the standard length or slightly more. Depth of caudal peduncle about one-tenth of the standard length. Maxillary and lower jaw rather short. Infraorbitals 3 and 4 deeper than broad. Preoperculum broad, its posterior margin without a marked notch. Preopercular sensory canal running nearer to posterior than to anterior margin of bone in dorsal limb, in ventral limb at about the middle, and with about eight tubules. Scales thick, not serrated posteriorly, arranged in about 35 transverse rows counted from posterior margin of operculum to middle of caudal fin base. A few longitudinal rows of body scales deeper than broad. HoLotyPeE. Specimen originally described by Airaghi (1908) and belonging to the Museo Civico di Scienze Naturale di Bergamo, Italy. The only specimen. DeEscripTIon. The total length of the holotype is 58 mm., the length from the tip of the snout to the hindmost lateral line scale (standard length) is 49 mm. The estimated depth of the body is slightly more than one quarter (26:5°%) of the standard length and the depth of the caudal peduncle is about one-tenth of that length. The length of the head is estimated to be slightly less than one quarter of the standard length. The maxillary and lower jaw are, comparatively, a little shorter than in Ph. bechet and Ph. latiusculus. The position of the paired and unpaired fins is clearly visible. The base of the ventral fin is situated a little nearer to the hindmost lateral line scale than to the tip of the snout, and the distance between the base of the ventral and the origin of the anal is about 75% of the distance between the bases of the pectoral and ventral fins. The dorsal fin begins above the ventral fin base. A good figure of the entire fish is given by Boni (1937, pl. 5, fig. 3). CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 377 The cranial bones are, unfortunately, partly defective ; most bones belonging to the exoskeletal skull roof are lacking and many of the other cranial bones are poorly preserved. Consequently the following description is rather incomplete. Exoskeletal skull roof The premaxillary, rostral, nasal, frontal, supraorbital, dermosphenotic and dermopterotic bones are missing or represented by insignificant fragments. Parts of the parietal and the extrascapular seem to be present (Pa?, Ext?, Pl. 6, fig. 1), but their shape cannot be made out. Dermal bones of cheek and opercular apparatus Only the posteriorly directed lateral part of the comparatively short maxillary is present (Mx + Sm, Pl. 6, fig. 1; Text-fig. 5). It is, however, very defective, almost the entire ventral part is missing and consequently nothing can be made out Fic. 5. Pholidophorus (?) caffii Airaghi. Holotype. Attempted restoration of head in lateral view. XI10°5. Ang, angular; Ant, antorbital; De. Spl, dentary; IJfo,—Ifo;, infraorbitals 1 to 5; Top, interoperculum; Mz + Smx,, maxillary and anterior supramaxillary; Op, operculum ; Pop, preoperculum ; Sho, suborbital; Smvx,, posterior supramaxillary ; Sop, suboperculum ; hc, anterior division of supramaxillary pit-line ; ifc, infraorbital sensory canal; mdc, mandibular sensory canal; orp, postmaxillary pit-line; ovp,, oral pit-line ; poc, preopercular sensory canal. 378 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE regarding the depth of the maxillary or of the dentition. The lateral surface of the preserved part is, however, ornamented with longitudinal streaks and, posteriorly, with some short oblique ridges. The posterior end of the maxillary is broken off, and thus its shape, if rounded or notched, cannot be stated. According to Boni (1937, text-fig. 9) the posterior part of the maxillary is wedged between infraorbital 3 and the antero-ventral limb of the preoperculum, but the part in question is, as far as I can see, the antero-dorsal part of the ventral limb of the preoperculum, and not part of the maxillary. Boni’s text-fig. 9 shows only a single supramaxillary, obviously corresponding to supramaxillary 2 in other species ; anterior to it the figure shows, however, a roughly triangular or semicircular elevation on the dorsal margin of the maxillary, corres- ponding to supramaxillary I in other forms. At first I was inclined to consider this eminence as not belonging to that bone, but as soon as I had the opportunity of examining the holotype I could confirm that there was no division between this elevation and the rest of the maxillary. Whether this is a secondary, perhaps individual, fusion of the two bones or a primitive stage of development, characteristic of Ph. caffit, is a question which must be left open until more specimens of this species are available for study. Supramaxillary 2 (Smx, Pl. 6, fig. 1; Text-fig. 5) is rather well preserved ; it is almost rectangular, its antero-dorsal corner forms a blunt short process, and its lateral surface is ornamented with rather dense, horizontal striations. All the bones of the infraorbital series, an antorbital and five infraorbitals, are present. The antorbital (Ant, Pl. 6, fig. 1 ; Text-fig. 5) is rather well preserved but its outline cannot be made out in detail; it is large in comparison with the antorbital in Ph. bechei and tapers postero-dorsally. Infraorbital 1 (lachrymal) (Ifo,, Pl. 6, fig. 1; Text-fig. 5) was exposed by prepara- tion ; it is rounded anteriorly, its ventral margin is convex, its dorsal margin concave and the exposed dorso-lateral surface is slightly concave. Infraorbital 2 (Ifo,, Pl. 6, fig. 1 ; Text-fig. 5) is well preserved but broken into two pieces ; it has a generally elongate shape with the dorsal margin slightly concave, the ventral margin slightly convex. Infraorbital 3 (Ifo, Pl. 6, fig. 1; Text-fig. 5) is likewise well preserved except that its postero-dorsal corner is broken off; it is markedly deeper than broad and its dorsal margin constitutes an obtuse angle so as to fit the ventral margins of infraorbital 4 and the suborbital. Infraorbital 4 (Lfo,, Pl. 6, fig. 1 ; Text-fig. 5) is partly incomplete but seems to be a little deeper than broad. Infraorbital 5 (Ifo;, Pl. 6, fig. 1; Text-fig. 5) is better preserved and seems to be almost square and only a little smaller than infraorbital 4. Posterior to infraorbitals 4 and 5, the ventral part of the swborbital (Sbo, Pl. 6, fig. 1; Text-fig. 5) is clearly visible but it is broken into two pieces on to the lateral crest of the hyomandibular ; regarding its outline it can only be said that its posterior CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 379 margin seems to be straight and vertical, thus probably not overlapping the anterior margin of the operculum, and that its postero-ventral margin is a little convex. The preoperculum (Pop, Pl. 6, fig. 1; Text-fig. 5) is, comparatively, very broad, especially in its antero-ventral limb, the antero-dorsal margin of which shows a rather pronounced convexity ; the postero-ventral margin of the preoperculum is damaged but it seems to have been almost straight or only slightly concave and thus without the notch characteristic of the preoperculum in Ph. latiusculus and especially in Ph. bechei. The uppermost end of the preoperculum is defective. The operculum (Op, Pl. 6, fig. 1; Text-fig. 5) seems to have the usual almost triangular shape, but as its anterior and dorsal margins are damaged its shape cannot be made out exactly. Its ventral tip is rounded, about as in Ph. latiusculus, and as in that species the anterior margin of the operculum is slightly notched ventrally to receive the antero-dorsal process of the suboperculum. The suboperculum (Sop, Pl. 6, fig. 1; Text-fig. 5) is only partly preserved ; its antero-dorsal process is well developed. The interoperculum (op, Pl. 6, fig. 1; Text-fig. 5) seems to be roughly triangular with the posterior margin slightly convex ; its lateral surface carries a few striations parallel to its margins. Branchiostegal rays and gular plate The posteriorly situated, rather short but broad branchiostegal rays (R. Br, Pl. 6, fig. 1) are visible ventral to the preoperculum and the interoperculum. The gular plate is not visible. Lower jaw The lower jaw is, like the maxillary, comparatively short. Anteriorly the dorsal margin of the dental part of the dentary (De. Spl, Pl. 6, fig. 1; Text-fig. 5) is only gently ascending, but as most of the dorsal margin of the lower jaw is covered by overlying bones, its outline cannot be followed. No teeth could be observed on the exposed dorsal margin of the dentary. The lateral surface of the dental part is smooth and that of the splenial part is only feebly ornamented but shows clearly the tubules of the mandibular sensory canal; the ridge separating the two parts of the dentary seems to be comparatively weak. The ventral part of the angular (Ang, Pl. 6, fig. 1; Text-fig. 5) is well exposed ; its lateral surface is practically smooth. Sensory canal system of head Regarding the sensory canal system of the head, almost nothing of the supra- orbital sensory canal can be determined because the nasal and frontal bones are lacking. On the fragment thought to belong to the parietal a single short canal is visible. This might be interpreted as the posteriormost part of the supraorbital sensory canal, but as it is not followed by a pit-line this interpretation is rather uncertain. No canals, tubules or pores can be observed in the bone fragment con- sidered as part of the extrascapular. 380 CERTAIN DRLIASSIC AND LLASSIC PHOLIDORP HORT DAE The infraorbital sensory canal (ifc, Text-fig. 5) can be followed for almost its entire length ; only the ethmoidal commissure and the part in the dermosphenotic and dermopterotic are missing. In the antorbital the canal describes an arch in its ventral part and gives off postero-dorsally an antorbital branch ending in a pore ; at the point where the antorbital branch arises there seems to be a rather large pore. In infraorbital 1 (lachrymal) the sensory canal can be followed for most of its length but its state of preservation allows no statements regarding the number or disposition of its tubules. In infraorbital 2 the sensory canal is partly visible ; an opening at the posterior end of the bone is not a real pore but an artifact due to damage. In infraorbital 3 the sensory canal gives off three postero-ventrally directed and short but strongly marked and widely separated tubules, the antero-ventral one being the shortest, the dorsal one the longest. The anterior part of infraorbital 4 is obviously pierced by the sensory canal, but no tubule can be seen with certainty. In infra- orbital 5 the sensory canal can be traced but no tubule is visible. Cephalic division of main lateral line The preopercular sensory canal seems to pierce the dorsal limb of the preoperculum nearer to its posterior than to its anterior margin ; in the ventral limb it runs at about the middle of its length, not nearer to its anterior margin as in Ph. bechei and Ph. latiusculus. The tubules given off from the morphologically posterior side of the canal are short but rather wide. Only eight tubules seem to be present, two in the dorsal limb of the preoperculum dorsal to and at the posterior angle of the bone respectively, one short tubule farther ventrally and five tubules in the ventral part of the ventral limb of the bone. Consequently, the general arrangement of the tubules is not unlike that in Ph. becher and Ph. latiusculus, but the tubules in the ventral group are not long and narrow as in the two former species and the total number of tubules is only about 8 as against 17-19 in Ph. bechei and 14-15 in Ph. latiusculus. On the lateral surface of the preoperculum there are two grooves, one horizontal and one vertical, (hc,, orp, Pl. 6, fig. 1; Text-fig. 5), meeting at a right angle; the horizontal groove represents the posterior portion of the anterior division of the supramaxillary line and the vertical one the postmaxillary line according to the nomenclature proposed by Stensi6 (1947). The mandibular sensory canal (mdc, Pl. 6, fig. 1; Text-fig. 5) is exposed to a small extent in the anterior, broken portion of the splenial part of the dentary; it is impossible to say how many tubules the canal gives off in that part of the bone, but posterior to the exposed part of the canal five short tubules are visible. Near the ventral margin of the angular there is a pore belonging to that part of the mandibular sensory canal which pierces the angular, and dorsal to this pore there is a short, vertical groove, representing the oral line (o7p,, Pl. 6, fig. 1; Text-fig. 5). Exoskeletal shoulder-girdle and squamation Dorsal to the operculum there is a large suprascapula (Ssc, Pl. 6, fig. 1) but it is crushed and partly defective and consequently its outline cannot be determined CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 381 exactly. Postero-ventral to it there is a supracleithrum (Scl, Pl. 6, fig. 1) of about the same size and shape as the suprascapula, but this bone also is crushed and defective. Almost the entire cleithrwm is missing but its impression suggests a rather robust bone. Posterior to this impression there is a deep bony plate which may be inter- preted as a postcleithrum. The scales are arranged in about 35 transverse rows counted from the posterior margin of the supracleithrum to the middle of the caudal fin base (—lateral line scales). On the body the lateral line scales and those in the longitudinal rows immediately dorsal and ventral to them, three rows in all, are deeper than broad. All the scales are comparatively thick and, as far as can be judged, with a regular posterior margin. Lateral line Those parts of the lateral line piercing the suprascapula and the supracleithrum are not visible; in the suprascapula, however, a short but rather wide, mesially directed tubule can be seen ; in the supracleithrum no tubules are visible. The lateral line scales (/.1, Pl. 6, fig. 1) are arranged in a straight row from the second scale to the last one at the base of the middle caudal rays. Paired and unpaired fins All fins are present but at least in part more or less defective. The pectoral fin is of moderate size. Boni (1937) gives the number of its lepido- trichia as II, a remarkably low number. There are, indeed, 11 rather stout lepido- trichia clearly exposed, but ventral to the postcleithrum and easily overlooked there is a cluster of broken, more slender lepidotrichia ; these may represent the posterior rays of the pectoral fin and their number may be at least 7-8, and conse- quently the total number of pectoral lepidotrichia may be estimated as about 18 or about the same number as in Ph. becher and Ph. latiusculus. The first lepido- trichium carries some very small fulcra (Fw, Pl. 6, fig. 1). The ventral fin is rather defective. For this fin Boni gives the number of lepido- trichia as 5, which is too low; as far as I can see there are at least two small, un- divided lepidotrichia anteriorly, followed by 6 divided and branched lepidotrichia, the anterior one fulcrated. Posterior to these lepidotrichia the fin is damaged and the total number of ventral fin lepidotrichia thus remains unknown. The dorsal fin is well preserved but the number of lepidotrichia is difficult to interpret. Anteriorly there are two undivided lepidotrichia, the anterior one rather short and almost scale-like ; there follows a third, rather long, apparently undivided lepidotrichium with a comparatively long and slender rod (fulcrum?) in front of it ; posterior to these three there are 8 divided and branched lepidotrichia, the anteriormost of which is richly fulcrated. Consequently, the total number of dorsal fin lepidotrichia is at least II as compared with 9 (10?) as stated by Boni. The anal fin is not as well preserved as the dorsal fin and the number of its lepido- trichia cannot be stated with accuracy. There are, however, two short, undivided 382 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE lepidotrichia anteriorly followed by at least 7 divided and branched ones, the anteriormost of which with a few, long fulcra ; thus there are at least 9 lepidotrichia as against the 5-6 as stated by Boni. The caudal fin is hemi-heterocercal and its dorsal as well as its ventral margin is densely set with fulcra. REMARKS. Pholidophorus cafii, described as a new species by Airaghi (1908), has been considered merely a synonym of Ph. latiusculus by Bassani (1914), de Alessandri (1920) and Boni (1937). It is, however, beyond doubt that the specimen on which Airaghi founded his species, redescribed above, does represent a species clearly distinguishable from Ph. latiusculus in the sense of Kner. Differences in the relative length of maxillary and lower jaw, the general shape of the preoperculum, the course of the preopercular sensory canal and the number of its tubules clearly separate Airaghi’s Pholidophorus caffii from Kner’s Ph. latiusculus. In my opinion differences in the preoperculum and the course of the preopercular sensory canal are of such a degree that the placing of caffiz in the genus Pholidophorus s. str.may even be questionable. This problem will be discussed in more detail later on (p. 423). HORIZON AND LOCALITY. Upper Trias; Viciarola, near S. Pellegrino (Val Brem- bana), Italy. Pholidophorus cf. pusillus Agassiz (PING ties 2 lexi) 1895 Pholidophorus latiusculus Agassiz ; Woodward : 455 (partim). PRELIMINARY DIAGNOSIS. Pholidophorus of small size, up to 65 mm. in total length, similar to Ph. (?) caffw but differing from that species 7.4. in the following features: Maxillary and lower jaw not markedly short. Infraorbitals 3 and 4 considerably broader than deep. Preoperculum similar to that of caffir but pre- opercular sensory canal with about 11 tubules, longer than those in caffiz. Scales arranged in about 38—40 transverse rows counted from posterior margin of operculum to middle of caudal fin base. MATERIAL. Specimen No. P.4418 in the Department of Palaeontology, British Museum (Natural History). DESCRIPTION. The total length of the specimen is about 65 mm., the standard length is estimated as 56 mm. The greatest depth of the body cannot be deter- mined. The depth of the caudal peduncle is about one-tenth of the standard length. The length of the head is estimated to be slightly less than one quarter of the standard length. The maxillary and lower jaw are relatively longer than in Ph. becher and Ph. latiusculus, and even more so than in Ph. (?) caffit. The fins are not preserved except for some lepidotrichia of the anal fin and the entire caudal fin ; the position of the dorsal fin in relation to the ventral fins is thus unknown. The cranial bones are very defective ; all bones belonging to the exoskeletal skull roof are lacking and many of the other cranial bones are fragmentary. Thus only the following rather incomplete description can be given. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 383, Dermal bones of cheek and opercular apparatus Only the posteriorly directed part of the comparatively long maxillary (Mx, Pl. 6, fig. 2; Text-fig. 6) is preserved ; apart from its relatively greater length it seems to have the same general shape as in Ph. bechei and Ph. latiusculus, but its ventral margin is defective and consequently nothing can be said about its real depth or about the dentition. Neither can it be determined with any accuracy whether its posterior margin is rounded or slightly notched ; judging from an impression on the preoperculum behind the broken end of the maxillary it looks, however, as if its posterior margin was slightly notched. The lateral surface of the maxillary is richly ornamented with ganoin spots and irregular longitudinal streaks and ridges except along its postero-dorsal margin, indicating that supramaxillary 2 may overlap the maxillary a little. Of the supramaxillaries, supramaxillary 1 (Smx,, Pl. 6, fig. 2; Text-fig. 6) is almost semicircular ; it is well separated from the maxillary, but seems not to overlap the dorsal margin of that bone ; its lateral surface carries a striation approximately parallel to the convex dorsal margin of the bone, similar to the striation on the dorsal V De.Sppl Fic. 6. Pholidophorus cf. pusillus Agassiz. Attempted restoration of head in lateral view. B.M. No. P.4418. X1Io. Ang, angular ; De. Spl, dentary ; Ifo, Ifo, Ifo, infraorbitals 1 and 3-4 ; op, interoperculum ; Mx, maxillary ; Op, operculum ; Pop, preoperculum ; Sbo, suborbital; Smx,, Smx,, anterior and posterior supramaxillaries ; Sop, suboperculum ; poc, preopercular sensory canal. GEOL. II, 8 34 384 CERTAIN TRIASSIC AND LIASSTC PHOLIDOPHORTDAE elevation of the maxillary in Ph. (?) caffii. Supramaxillary 2 (Smx», Pl. 6, fig. 2 ; Text-fig. 6) has about the same, almost rectangular shape as in that species but it is comparatively longer antero-posteriorly ; its antero-dorsal corner is broken off but may have been of about the same shape as in Ph. (?) caffii. The lateral surface of supramaxillary 2 shows a horizontal striation, but this striation is by no means as dense as in Ph. (?) caffii or in Ph. latiusculus. The bones of the infraorbital series are only partly preserved. An isolated bone fragment in the snout region looks like an antorbital (Ant?, PI. 6, fig. 2), but this interpretation is rather uncertain. Another bone fragment dorsal to the anteriormost preserved part of the maxillary may belong to imfraorbital 1 (lachrymal) (Ifo, Pl. 6, fig. 2) but it provides no informa- tion about the outline of this bone. Infraorbital 2 is missing. Infraorbital 3 (Ifos, Pl. 6, fig. 2; Text-fig. 6) is remarkably broad, about twice as broad as deep, its postero-dorsal part overlaps the preoperculum and its dorsal margin is slightly concave ; its shape is thus quite different from that of the corres- ponding bone in Ph. (?) caffir. Infraorlital 4 (Lfo4, Pl. 6, fig. 2 ; Text-fig. 6) is almost complete, nearly rectangular and considerably broader than deep. Some bone fragments dorsal to infraorbital 4 indicate the presence of an infra- orbital 5. Between the visible part of the hyomandibular and the anterior margin of the operculum there is a rather large bony plate which is obviously a suborbital (Sbo, Pl. 6, fig. 2; Text-fig. 6) ; its margins are, however, only partly visible. The preoperculum (Pop, Pl. 6, fig. 2; Text-fig. 6) is well preserved, only its upper- most end is defective. Its shape is very similar to that of the preoperculum in Ph. (2) caffii. A shallow concavity is present ventral to the ill-defined posterior angle between the upper and lower limbs of the bone, but there is no real notch as in Ph. bechet and Ph. latiusculus. The operculum (Op, Pl. 6, fig. 2 ; Text-fig. 6) seems to be remarkably broad, much broader than in the other species treated here, but this may be partly due to pressure as the bone is somewhat dislocated. The suboperculum (Sop, Pl. 6, fig. 2; Text-fig. 6), likewise dislocated, is only partly preserved ; its postero-dorsal half is missing. Of the interoperculum (Iop, Pl. 6, fig. 2; Text-fig. 6) only the posterior part is exposed ; it seems to have the same general shape as in Ph. (?) caffiz, but lacks the concentric striations. Branchiostegal rays and gular plate As in Ph. (?) caffii only some of the posterior branchiostegal rays (R. Br, PI. 6, fig. 2) are exposed ; they are, however, relatively much longer than in that species. The gular plate is not visible. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 385 Lower jaw The lower jaw is exposed to a certain extent, but its anterior part is lacking and its dorsal margin is covered by the maxillary bones and cannot be described. It is, however, obvious that the lower jaw is relatively much longer and stouter than in Ph. (?) caffii. The lateral surface of the dental part of the dentary (De. S#l, Pl. 6, fig. 2; Text-fig. 6) is quite smooth, whereas the splenial part, which seems to be deeper than in the three species described above, has a rich ganoin ornamentation obscuring the tubules of the mandibular sensory line; the ganoin-covered ridge separating these two parts of the dentary is very pronounced. No dentition could be observed. The angular part of the angulo-swpra-angular is partly visible but its posterior margin is indistinct because of damage ; its lateral surface is ornamented ventrally with ganoin patches. Sensory canal system of head Very little of the sensory canal system of the head can be made out. All the bones containing the supraorbital sensory canal are absent and only some parts of the infraorbital sensory canal are preserved (Text-fig. 6). In infraorbital 3 the sensory canal can be traced but the tubules are difficult to observe ; there are, however, traces of at least two rather long, posteriorly directed tubules, and a third, very short tubule is given off from the anterior part of the canal. In infraorbital 4 the canal itself cannot be seen with accuracy, but a single large, posteriorly directed tubule is clearly visible. Cephalic division of main lateral line Of this part of the sensory canal system only the preopercular sensory canal is visible (Text-fig. 6). The canal itself pierces the preoperculum at about the middle of its length and, on the whole, lies parallel to its anterior and posterior margins, thus not nearer to its posterior margin in the dorsal limb as it is in Ph. (?) caffii and not distinctly nearer to its anterior margin as in Ph. becher and Ph. latiusculus. The number of tubules given off from the posterior side of the canal is difficult to deter- mine, but two tubules are clearly visible in the dorsal limb of the bone, dorsal to and at its posterior angle respectively ; ventral to this angle there is a rather short tubule followed more ventrally by about eight wide tubules in the antero-ventral third of the bone. Between the second and the third tubules reckoned from above there is a structure resembling a postero-dorsally directed tubule or rather, a sensory line ; its position posterior to the preopercular sensory line may be an argument against the interpretation of it being a supramaxillary line, but as it cannot be followed anteriorly because of the damaged condition of the lateral surface of the middle part of the preoperculum the question of its real nature must be left open. The same is true of the postmaxillary line, the ventralmost part of which may perhaps be traced ventral to the damaged middle part of the bone (orp?, Pl. 6, fig. 2). The total number of tubules belonging to the preopercular sensory canal may 386 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE consequently be eleven, perhaps twelve, that is, only a little higher than in Ph. (?) caffiu ; their arrangement is virtually identical in the two species but those of the ventral group are considerably longer, almost reaching the ventral margin of the preoperculum ; they are, however, not as long and as numerous as in Ph. bechei or Ph. latiusculus. The mandibular sensory canal is not visible and only two tubular openings can be traced posteriorly on the splenial part of the dentary. A short vertical groove on the lateral surface of the angular is probably an oral pit-line (orp, Pl. 6, fig. 2). Exoskeletal shoulder-girdle and squamation No bones belonging to the exoskeletal shoulder-girdle are visible. The scales are comparatively thick with even posterior margins. There seem to be about 38-40 transverse rows of scales, counted from the posterior margin of the operculum to the middle of the caudal fin base, but as the squamation is in part rather badly damaged, this count is somewhat uncertain. On the anterior part of the body there are some longitudinal rows of scales which are deeper than broad. The dorsal margin of each body scale is drawn out into a peg (Sc, PI. 6, fig. 2). Paired and unpaired fins The pectoral, ventral and dorsal fins are missing and only a few lepidotrichia of the anal fin are partly visible. The caudal fin is hemi-heterocercal ; its dorsal margin carries a dense series of rather large fulcra ; along the ventral margin of the caudal fin the fulcra are smaller. REMARKS. The specimen just described, identified by Woodward (1895) as Pholidophorus latiusculus, cannot belong to that species, or to Ph. (?) caffit, as is evident from the descriptions given above. On the contrary it seems to me rather probable that it may be identical with the second Pholidophorus species originally mentioned by Agassiz (1832) from Seefeld, Ph. pusillus, the material of which belonged to Dr. Alex. Braun and the Karlsruhe Museum. Dr. Erwin Jorg, Landessammlungen fiir Naturkunde, Karlsruhe, has kindly informed me that no material of Ph. pusillus or of Ph. latiusculus now exists in this collection ; it was probably destroyed in 1942. The original material of Ph. pusillus must also be considered lost. The identification of Ph. pusillus proposed by Kner (1866) refers to some small fish specimens from Seefeld ; in the Univ.-Institut fiir Palaontologie und Geologie, Innsbruck, some material exists which can be identified as the two upper specimens figured by him on pl. 6, fig. 2 (in reality the specimen 0 on this figure is composed of two specimens!). Their preservation is, however, rather poor, and they show no details (except some branchiostegal rays) which can be directly compared with those preserved in P.4418. But there are some facts favouring the identification of this specimen with Ph. pusillus. The size and slender shape of the body is about the same and according to Kner the number of transverse scale-rows amounts to 38-40, just as I have found in CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 387 the specimen in question. Because of this it seems to me rather improbable that two different Pholidophorus species of about the same size and shape should exist in the Seefeld deposits. Specimen P.4418 is most probably an unusually well- preserved, large specimen of Ph. pusillus. A final decision as to its relationship with that species must, however, be postponed until more extensive and better pre- served material from Seefeld can be thoroughly investigated. Therefore I have only identified the specimen in question as Ph. cf. pusillus Agassiz. HoRIZON AND LocALiTy. Upper Trias ; Seefeld, Tyrol, Austria. Genus PHOLIDOLEPIS nov. PRELIMINARY DIAGNOSIS. Pholidophoridae of rather small size, as far as known similar to Pholidophorus but differing in the following features. Exoskeletal cranial bones with very feeble ganoin covering. Supramaxillary 2 with rather well-marked process at antero-dorsal corner. Preoperculum with preopercular sensory canal running near to anterior margin. Fulcra as far as known absent on all fins except along dorsal margin of caudal fin. Scales thin and cycloid, with fine concentric striations. The generic name Pholidolepis is a composite one derived from Pholidophorus and Leptolepis, so as to indicate the probable intermediate position of the new genus between the two other genera. Type spECcIES. Pholidolepis dorsetensis sp. nov. Pholidolepis dorsetensis gen. et sp. nov. (Pitz 3 Rext=fisss7,.8) 1895 Pholidophorus caudalis Woodward : 458 (partim). PRELIMINARY DIAGNOSIS. Pholidolepis of rather small size, up to about 110 mm. in total length. Body elongate, depth about one-fifth of the length without caudal (standard length). Head less than one quarter of standard length. Preopercular sensory canal with about 15 tubules. Hototyere. British Museum (Nat. Hist.) No. 38164. MaTeERIAL. In addition to the holotype, Nos. 35725, 38536, P.3662, P.44708, P.44709 of the British Museum (Natural History) and P.259, of the D.M.S. Watson Collection, Cambridge, have been used for the following description, all more or less badly preserved. Besides these specimens there is, however, in the British Museum (Natural History) a number of still more defective specimens, probably belonging to Ph. dorsetensis : Nos. 35562, 38537, 39862, 43007, P.9g39, P9390, P.g39c, P.4370, and P.44707. Description. The holotype has a total length of about 110 mm. ; length from tip of snout to posterior margins of hypurals (standard length) about 95 mm. Specimen P.44708 has a standard length of about 98 mm., P.3662 of about 80 mm., which lengths constitute the range for the specimens investigated. 388 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE The body is decidedly more elongate than in Ph. caudalis, to which species the specimens in question have been referred by Woodward, but as all specimens are crushed the true depth of the body cannot be given with accuracy ; it may be esti- mated as about one-fifth of the standard length. The length of the skull seems to be less than one quarter of the standard length. The dermal bones of the head are comparatively thin ; as far as can be seen only the bones of the jaws show a surface ornamentation. Exoskeletal skull roof Only very little can be made out regarding the bones belonging to the exoskeletal skull roof because of the bad state of preservation. The premaxillary is not visible in any of the specimens investigated. The vostval may be present in Watson Colln. No. P.259, but only in a defective state. The nasal (Na, P|. 7, fig. 3 ; Text-figs. 7, 8) is rather well preserved in the holotype. It is an elongate, comparatively large bone reminiscent of the corresponding bone in Ph. bechei and situated lateral to the foremost part of the lateral margin of the frontal ; Fr | | \ Zn Fic. 7. Pholidolepis dorsetensis gen. et sp. nov. Attempted restoration of head in lateral view. x 6:3. Ang, angular; De. Spi, dentary ; Fy, frontal; Jop, interoperculum; Mz, maxillary; Na, nasal; Op, operculum; Pop, preoperculum; Smzx,, Smx,, anterior and posterior supra- maxillaries ; Sop, suboperculum ; foc, preopercular sensory canal; soc, supraorbital sensory canal. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 389 anteriorly it is rounded and reaches about as far forwards as the anterior tip of the frontal. Its lateral outline cannot, however, be determined exactly because of damage and consequently it is impossible to say whether a notch or fenestra for the posterior nostril is present as in Ph. bechet. The frontal (Fr, Pl. 7, figs. 1, 3; Text-figs. 7, 8) is preserved in the holotype, in 38536 and in Watson Colln. No. P.259, but in all these specimens it is more or less defective because of crushing. Its anterior half is, judging from the holotype, rather broad but tapers towards the anterior rounded tip at the level of the nasal. The posterior half broadens considerably but its posterior margin cannot be accurately defined in anyspecimens. There are no traces of surface ornamentation on the frontal. No supraorbitals can be seen clearly in any specimens but some small bone frag- ments lateral to the lateral margins of the frontal and nasal in the holotype (So, Pl. 7, fig. 3) may belong to an anterior supraorbital similar to that bone in Ph. bechet. Only the sensory canal of the dermosphenotic seems to be present in the holotype, the rest of the bone is missing. In the other specimens investigated this bone is not visible. The parietal, the dermopterotic and the extrascapular are, in all specimens, too damaged for description. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pl. 7, figs. 1-3 ; Text-fig. 7) is more or less completely preserved in the holotype as well as in 38536 and P.44708, and Watson Colln. No. P.259, but in none of them is it undamaged and its anterior, antero-mesially directed tip is not visible. Its posteriorly directed, lateral part is, as usual, evenly curved with the convexity directed ventrally. Its lateral surface is provided with a weak ornamenta- tion, consisting of more or less oblique striations as can be seen in the holotype and particularly in P.44708 (Pl. 7, fig. 2). Inthe latter the ventral border of the maxillary carries a dentition similar to that in Leptolepis. The supramaxillaries are only visible in P.44708 and Watson Colln. No. P.259. The anterior one, supramaxillary 1 (Smx,, Pl. 7, fig. 2), is partly preserved in the first named specimen and seems to be much longer antero-posteriorly than the corresponding bone in Pholidophorus, thus more like that of Leptolepis, but as its anterior part is damaged its outline cannot be made out with accuracy. If there exists any surface ornamentation on supramaxillary 1 it must be only weakly developed. Supramaxillary 2 (Smx,, Pl. 7, fig. 2) is also best preserved in P.44708 ; its antero-dorsal process is decidedly more pronounced than in Pholidophorus but not nearly so long as in Leptolepis. The lateral surface of supramaxillary 2 carries a weak longitudinal striation somewhat recalling that in Ph. latiusculus but much weaker. Both supramaxillaries seem to overlap the dorsal margin of the maxillary but only to a small degree. The bones of the infraorbital series as well as the suborbital are practically missing in all specimens. The preoperculum (Pop, Pl. 7, figs. 1, 3; Text-fig. 7) is present in most specimens but in none is it undamaged. It is best preserved in 38536 and has the same general 390 CERTAIN TRIASSIC AND ELASSIC PHOLIDOPHORIDAE Fic. 8. Pholidolepis dorsetensis gen. et sp. nov. Attempted restoration of head in dorsal view. 6-3. Lettering as in Fig. 7. shape as in Leptolepis, the anterior margins of the dorsal and the antero-ventral limbs meeting to enclose an angle, but it has a deep notch on its posterior margin as in Ph. bechet. There is no process on its anterior margin at the angle as there is, for example, in Leptolepis normandica. The operculum (Op, Text-fig. 7) is very damaged by crushing or is lacking in all specimens except Watson Colln. No. P.259. Judging from this specimen its general shape is about the same as in Pholidophorus. The suboperculum (Sop, Text-fig. 7) is also only preserved in Watson Colln. No. P.259. It seems to be comparatively deep, and the process at its antero-dorsal corner is longer and more pointed than in Ph. bechev. Only the posterior part of the imteroperculum (Iop, Text-fig. 7) is exposed in the specimen just named and in 38536, and nothing can be said about its general shape. Lower jaw The lower jaw is partly visible in the holotype and in 38536 and P.44708, and Watson Colln. No. P.259. Its outline cannot be determined, but in all specimens mentioned the anterior part of the dentary (De. Spl, Pl. 7, figs. 1-3; Text-fig. 7) ascends rather gently as in Pholidophorus, not abruptly as in Leptolepis ; from this it may be concluded that the general shape of the lower jaw is also almost as in Pholidophorus. The ridge separating the dental and the comparatively low splenial parts of the dentary is clearly visible. In specimen P.44708 (PI. 7, fig. 2) the lateral CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 391 surface of the angular (Ang, Pl. 7, fig. 2) is provided with a surface ornamentation of thin longitudinal ridges or streaks similar to those on the maxillary. A gular plate is not visible in any of the specimens investigated. Sensory canal system of head _ ye sensory canal system of the head is only partly visible and the description must therefore be rather incomplete. Fart of the supraorbital sensory canal (soc, Pl. 7, figs. 1, 3; Text-figs. 7, 8) can be seen in the holotype, in 38536 and in Watson Colln. No. P.259. In its foremost part, piercing the nasal bone, the canal is straight and continues straight backwards in the anterior, narrower part of the frontal. As the frontal broadens the canal bends laterally and then runs obliquely postero-mesially ; its posterior part in the parietal cannot be followed. No tubules or pores can be observed with accuracy except in Watson Colln. No. P.259, where three postero-mesially directed tubules are given off from the mesial side just before and at the bend of the canal. Almost nothing can be made out regarding the infraorbital sensory canal. In the holotype only a small part of a sensory canal is present, probably the canal in the dermosphenotic (Dsph, Pl. 7, fig. 3), with the backward bend of the main canal and the antero-dorsally directed tubule. Cephalic division of main lateral line The cephalic division of the main lateral line can only be followed in the pre- operculum. The preopercular sensory canal (poc, Pl. 7, fig. 3) and its tubules follow the same general pattern as in Ph. bechei, but the number of tubules is lower. In the ventral limb of the preoperculum there are in 38536 eleven tubules, obviously corresponding to the 13 ventrally situated tubules in Ph. bechei ; the posterior ones are, as in that species, long, narrow and curved with the convexity anteriorly directed. Dorsal to these tubules there are four shorter tubules, corresponding to the five to six shorter tubules in Ph. bechetz. The total number of tubules is thus 15 in this specimen ; in the others investigated the preoperculum is too damaged to allow a count of the tubules, but as far as can be observed the general pattern is the same as that just described for 38536. Exoskeletal shoulder girdle and squamation Parts of the cleithrum are exposed in many of the specimens but in no case can it be studied in any detail. No other bones belonging to the exoskeletal shoulder-girdle can be identified. The squamation consists of thin cycloid scales with delicate concentric striations. The lateral line No lateral line can be seen with certainty in any of the specimens available. 392 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Axial skeleton and paired and unpaired fins The vertebral centra are thin bony cylinders which are, however, crushed or dis- placed in all specimens investigated ; their undamaged shape cannot be described nor can their number be determined with any degree of accuracy. The neural and haemal spines are comparatively long. The fins are in part rather well preserved but in the dorsal and anal fins the lepido- trichia are poorly preserved. No fulcra can be observed on the pectoral, ventral, dorsal or anal fins, but with regard to the two last named fins this statement needs confirmation on better preserved material. The pectoral fin has a rather high number of lepidotrichia ; in P.3662 and P.44709, and Watson Colln. No. P.259 I have counted 19 rays, in P.44708 20 rays. In the ventral fin the number of lepidotrichia is also rather high ; in P.44708 and P.44709 and Watson Colln. No. P.259 their number is 15 or about as in Ph. bechev. The dorsal fin is poorly preserved in all specimens, but in P.44709 the number of dorsal radials can be determined as 12, the anterior one as usual being composed of two elements. The anal fin has 8 radials in the holotype and in 38536 and P.44709. The caudal fin carries along its dorsal margin rather long and stout fulcra ; no fulcra have been observed on the ventral margin of the fin. REMARKS. There is no doubt that the specimens just described were erroneously identified as Pholidophorus caudalis Woodward, as can be seen from a comparison with the following description (see p. 411), and I have been unable to find another described species to which they can be referred. Consequently I have assigned them to a new species for which I propose the name Pholidolepis dorsetensis. Unfor- tunately the material on which the new species is based is rather defective, and I am not fully convinced that all specimens here referred to Ph. dorsetensis really belong to that species. In particular I am in doubt regarding Watson Colln. No. P.259, which shows very few details directly comparable with those exposed in the other specimens ; it is especially unfortunate that almost nothing of the characteristic preoperculum is preserved. On the other hand I have not found any differences justifying its separation from the other specimens. More complete and better preserved material is necessary for a fuller description of Ph. dorsetensis. HORIZON AND LOCALITY. Lower Lias; Lyme Regis, Dorset. Genus PHOLIDOPHOROIDES Woodward 1941 Pholidophoroides Woodward : 89. EMENDED DIAGNOSIS. Pholidophoridae of medium size. Exoskeletal cranial bones and scales with ganoin covering. Nasal well developed and well separated from its antimere by the frontals, anteriorly reaching level of anterior margin of frontals. A single, well-developed supraorbital. Maxillary rather stout and deep with more or less pronounced convexity on dorsal margin ventral to infraorbital 1 (lachrymal) and supramaxillary 1; posterior margin notched. Two supramaxil- laries, not markedly overlapping dorsal margin of maxillary, supramaxillary 2 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 393 without marked process at antero-dorsal corner. Antorbital well developed, more than five (seven?) infraorbitals. Preoperculum with preopercular sensory canal running along middle of bone, about equidistant from anterior and posterior margins. Lower jaw not markedly deep, greatest depth in posterior third of its length ; dentary with smooth dental part, separated from splenial part by strong ridge. Dorsal fin above interspace between ventral and anal fins. Fulcra present along anterior margin of paired as well as unpaired fins. Caudal fin hemi-heterocercal with dorsal and ventral margins fulcrated. Scales moderately thick ; some longitudinal series of flank-scales higher than broad. Type spEcIES. Pholidophoroides crenulata (Egerton). REMARKS. When creating the new genus Pholidophoroides Woodward (1941 : 89) diagnosed it in the following manner: “ One species from the Lower Lias, P. crenu- latus Egerton, is distinguished from Pholidophorus by its smaller scales, which are less deepened on the flank. The scales are arranged in from 50 to 60 transverse rows, and those of the anterior part of the lateral line are less than twice as deep as broad. The scales are also comparatively thin, often displaying their concentric lines of growth. The fins are small like those of Pholidophorus. The maxilla seems to be comparatively stout. Ring vertebrae are conspicuous in the abdominal region. P. crenulatus may therefore be regarded as the type-species of a distinct genus, Pholidophoroides. . .. The Lower Liassic species, P. caudalis A. S. Woodw., probably also belongs to Pholidophorotdes.”’ The diagnosis given above is mainly based on the taxonomic characteristics of the redescribed type species, but also taking into consideration the other species here ascribed to the genus Pholidophoroides. Pholidophoroides crenulata (Egerton) (Pls. 8,9, 10; Pl. 15, figs. 3, 4,5, 8; Text-figs. 9-11) 1843 Pholidophorus crenulatus Egerton : 184. 1852 Pholidophorus crenulatus Egerton ; Egerton, pl. 5. 1887 Isopholis crenulatus (Egerton) Zittel : 216. 1890 Pholidophorus crenulatus Egerton ; Woodward & Sherborn : 146. 1895 Pholidophorus crenulatus Egerton ; Woodward : 463, pl. 12, fig. 6. 1941 Pholidophoroides crenulatus (Egerton) Woodward : 89. Diacnosis. Pholidophoroides of medium size, at least up to 150 mm. in total length. Greatest depth of the body about one quarter of the standard length or slightly more (ca. 25-30% of that length). Depth of the caudal peduncle less than one-tenth (ca. 8-9°%) of the standard length. Maxillary deep. Preopercular sensory canal with 16 or 17 tubules. Scales rather thin, in about 45-50 transverse rows reckoned from the hind margin of operculum to the middle of the caudal fin ; about four longitudinal rows of body scales somewhat deeper than broad and with the hind margin crenulated. LectotyPe. British Museum (Natural History) Nos. P.572 and P.573 are the two specimens originally described and figured by Egerton (1843, 1852) and designated 394 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE as ‘‘ The two type specimens ’’ by Woodward (1895 : 463). P.573 is the more com- plete specimen (standard length about 82 mm., height of body about 21 mm., height of caudal peduncle 7-5 mm.) and is therefore selected as lectotype. MATERIAL. In addition to the lectotype, the specimens mainly used for the following description are P.153, P.421, P.572, P.1046a, 6, P.4415, 38110, 38739, 38738 ; they agree in all comparable features with P.573. DeEscrRIPTION. The total length of the largest specimen, P.1046a, is 150 mm. ; its length from the tip of the snout to the hindmost scale in the lateral line (standard length) is about 125 mm. The other specimens in the British Museum (Natural History) have a standard length from 120 to 74mm. The greatest depth of the body is about one quarter of the standard length or slightly more (ca. 25-30°% of this length) ; the depth of the caudal peduncle is somewhat less than one-tenth (8-9%) of the standard length. The length of the head is about one quarter of the standard length, generally slightly less (ca. 22-25%). The ventral fins are situated approximately midway between the tip of the snout and the hindmost lateral line scale and the distance between the base of the ventrals and the origin of the anal is about 75-85% of the distance between the bases of the pectoral and ventral fins. A good figure of the entire fish is given by Woodward (1895, pl. 12, fig. 6). The specimens most thoroughly investigated are rather well preserved with the bones of the skull exposed to a great extent, and consequently they allow the follow- ing description to be made. Exoskeletal skull roof The premaxillary (Pmx, Pl. to, figs. 1, 2; Text-fig. 9) is visible in P.1046a, }, and to some extent also in P.4415 and 38738. It is comparatively stout with the postero-dorsal margin weakly S-shaped. The straight antero-ventral margin is serrated, forming a single row of teeth. The lateral surface of the premaxillary is densely set with ganoin tuberculations. The vostral (Ro, Pl. 9; Text-figs. 9, 10) is best preserved in 38110, but cannot be studied in any detail. The part pierced by the ethmoidal commissure is a little swollen dorsally ; posteriorly its lateral parts are concave and rather thin at the margin. The nasal (Na, Pls. 8,9; Text-figs. 9, 10) is also best exposed in 38110. It isan oblong, rather broad bone, situated posterior to the lateral parts of the rostral. It is, on the whole, oval but its antero-lateral and postero-lateral margins (lateral to the posterior opening for the supraorbital sensory canal) are concave, probably marking the postero-mesial and antero-mesial limits of the anterior and posterior nasal openings respectively. The nasals are separated by the narrow anterior part of the frontals. The frontal (Fr, Pls. 8,9; Pl. ro, fig. 2; Pl. 15, fig. 4; Text-figs. 9, 10) is rather narrow in its anterior part, mesial to the nasal, but it broadens considerably posterior to the nasal; from there its lateral margin runs at first almost straight posteriorly, but behind the orbit the bone widens again and has its greatest breadth in its posterior CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 395 Dpt fxt Ssc Pa | are ' \ Fic. 9. Pholidophoroides crenulata (Egerton). Attempted restoration of head in lateral view. x 5 approx. Ang, angular; Ant, antorbital; De. Spl, dentary; Dpt, dermopterotic; Dsph, dermo- sphenotic ; Evi, extrascapular; Fy, frontal; Ifo,—Ifo,, infraorbitals 1 to 7; Jop, interoper- culum; Myx, maxillary; Na, nasal; Op, operculum; Pa, parietal; Pmx, premaxillary ; Pop, preoperculum; Ro, rostral; Sbo, suborbital; “ Sbo’’, ‘“accessory suborbital’’; Scl, supracleithrum ; Smx,, Smx,, anterior and posterior supramaxillaries ; So, supraorbital ; Sop, suboperculum ; Ssc, suprascapula; ant. by, antorbital branch of infraorbital sensory canal; ap, anterior pit-line ; zfc, infraorbital sensory canal ; ifc. com, ethmoidal commissure ; mdc, mandibular sensory canal; mp, middle pit-line ; foc. preopercular sensory canal; pp, posterior pit-line; s. com, supratemporal commissure; soc, supraorbital sensory canal. part. Its postero-lateral corner is rounded off and its posterior margin is irregularly wavy. The suture between the frontals is practically straight anteriorly ; posteriorly its course is difficult to follow, but it does not seem to be markedly sinuous. A strong supraorbital (So, Pl. 9; Text-figs. 9, 10) is situated behind the posterior nasal opening and lateral to the broadened part of the frontal posterior to the nasal. Anteriorly its dorsal surface bulges a little and is ornamented with some small elevations. In dorsal aspect the lateral border of the supraorbital is almost straight, ending anteriorly in a point. A very small bone between the lateral margin of the frontal and the anterior tip of the dermosphenotic in 38739 (So?, Pl. 8; Pl. 15, fig. 4) may perhaps be a small second supraorbital ; in P.10460 a similar small bone may be traced, but as I have 396 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE not found a corresponding element in any of the other specimens investigated, it may perhaps be only a fragment of the anterior part of the dermosphenotic or of the lateral frontal margin. The dermosphenotic (Dsph, Pls. 8,9; Pl. 10, fig. 2; Pl. 15, fig. 4; Text-figs. 9, 10) is mostly more or less crushed on the underlying autosphenotic. Its anterior part, situated along the postero-lateral margin of the frontal, is pointed and a little curved and constitutes the postero-dorsal border of the orbit. Its posterior margin cannot be determined with accuracy, but according to P.1046b it seems to be rounded off. The margins of the parietal (Pa, Pl. 10, fig. 2; Pl. 15, fig. 4; Text-figs. 9, 10) are difficult to determine, as the skull roof of the single specimen showing the cranial bones in dorsal view, 38110, is much crushed in its posterior part, but judging from 38739 it has a rather wide extension laterally, thus being much broader than long. The dermopterotic (Dpt, Pls. 8,9; Pl. to, fig. 2; Pl. 15, fig. 4; Text-figs. 9, 10) is comparatively narrow in its anterior part between the frontal and the dermo- sphenotic but broadens continuously backwards, attaining its greatest breadth at its straight posterior margin. The shape of the extvascapular (Ext, Pl. 9; Pl. 10, fig. 2; Text-figs. 9, 10) cannot be determined ; its lateral part is, however, almost triangular with the postero- lateral margin a little convex and the postero-mesial margin slightly concave. The mesial part of the extrascapular is not preserved in any of the specimens investigated. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pls. 8,9; Pl. 10, figs. 1, 2; Text-fig. 9) is strikingly large and stout. Its anterior, mesially directed part is only partly visible in the specimens investigated, but seems to be rather low; its dorsal margin has an arched, thickened edge with the convexity directed ventrally, fitting the concavity on the postero-dorsal margin of the premaxillary ; the surface covered by the premaxillary is smooth. The lateral, posteriorly directed part of the maxillary increases rapidly in height ; its ventral margin is evenly convex, but its dorsal margin, on the whole concave, has a marked convexity below infraorbital 1 and supramaxillary 1. The posterior margin of the maxillary is rather deeply notched. The anterior half of the lateral surface has, nearest to the premaxillary, an ornamentation of densely set tubercula- tions like those on the premaxillary ; posteriorly the ornamentation becomes a longitudinal striation parallel to the margins. The posterior half of the lateral margin is smooth except for some longitudinal striations along the dorsal margin ; in P.4415 there are, also, some longitudinal ganoin streaks parallel to the ventral margin. A delicate comb-like dentition, about as in Leptolepis, is present along the posterior half of the ventral margin, the teeth decreasing in length anteriorly. No teeth can be observed along the anterior half of the ventral margin, but whether this is the actual condition or only dependent upon the state of preservation, must remain undecided. Two supramaxillaries are situated dorsal to the maxillary, not markedly over- lapping its dorsal margin. The anterior one, supramaxillary 1 (Smx,, Pls. 8, 9; Pl. ro, figs. 1, 2; Text-fig. 9), is rather small, almost semicircular, and has a well- CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 397 Fic. 10. Pholidophoroides crenulata (Egerton). Attempted restoration of head in dorsal view, the snout flattened, with (right) the probable mutual positions of the bones of the snout. X5 approx. Lettering as in Fig. 9. marked concentric striation on its lateral surface. The posterior one, supramaxillary 2 (Smx_, Pl. 8; Pl. io, figs. 1, 2; Text-fig. 9), is considerably larger ; its anterior margin is concave, fitting the convexity of supramaxillary 1, but its antero-dorsal corner is not markedly produced into a process. The dorsal and ventral margins of supramaxillary 2 are evenly convergent towards the posterior tip of the bone. Supramaxillary 2 also has a well-marked concentric striation on its lateral surface. The bones of the infraorbital series are generally well preserved, only infraorbital 2 is lacking in all specimens investigated. The antorbital (Ant, Pls. 8,9; Pl. 10, fig. 2; Text-figs. 9, 10) is an almost L-shaped bone situated ventro-lateral to the rostral and the nasal; its posterior tip reaches the antero-lateral tip of the supraorbital. Ventral to the antorbital and between it and the anterior part of the maxillary and supramaxillary 1 there is a large imfraorbital 1 or lachrymal (Ifo,, Pl. 8; Pl. xo, figs. 1,2; Text-fig.9). Its anterior margin is semicircular and its ventral margin is slightly convex; its dorsal margin is not clearly visible but is probably slightly concave. 398 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Infraorbital 3 (Ifo;, Pls. 8,9; Pl. 10, figs. 1, 2; Text-fig. 9) is comparatively large, stretching from the orbit to the anterior margin of the preoperculum ; its antero- ventral margin is slightly concave. The following bones in the series, dorsal to infraorbital 3, are crushed, displaced or lacking in most specimens, but judging from 38110 and P.1046a and 6 there are about four small, square infraorbitals between infraorbital 3 and the dermosphenotic (Ifo,, [fo;, [fog, Ifo,, Pl.9 ; Text-figs. 9, 11) ; in 38110 there seems to exist one more element in the series on the right side (Ifo,?, Pl. 9 ; Text-fig. 11), but here the arrange- ment is rather obscure. Fic. 11. Pholidophoroides crenulata (Egerton). Infraorbital 3, postorbital infraorbitals and suborbitals (same specimen as on Pl.g). Lettering as in Fig.9; Aright side, Bleftside. B.M. 38110. X5 approx. A large suborbital (Sbo, Pls. 8, 9; Pl. 10, fig. 2; Text-fig. 9) bone is situated posterior to infraorbitals 4-7, covering the hyomandibular ; as a rule it is crushed in the middle on to the strong lateral crest of this latter bone. Dorsal to the sub- orbital there are in 38110 (“ Sbo’’, Pl. 9; Text-figs. 9, 11) two small bones on the right side and a single similar bone on the left side, tapering posteriorly and, like the suborbital, with a conspicuous concentric surface striation. A single similar bone is also seen in 38738. It is difficult to interpret these bones ; they may be homologous to the supraspiracular plates described by Stensi6 (1932, text-fig. 67) in Perleidus stoschiensis Stensid and by Lehman (1952, text-figs. 85, 86) in P. madagascariensis Piveteau. But they are more likely to be “accessory suborbitals”’. In many holosteans there is a series of suborbitals posterior to the infraorbitals which may fuse or remain separate in various ways. Thus, for example the posterior part of infraorbital 3 in Pholidophoroides as well as in Leptolepis is probably the originally ventralmost suborbital, secondarily fused with infraorbital 3 (cf. Stensid 1947 : 164), whereas the others as a rule fuse intoasingle, large suborbital. In PA. crenulatathe dorsalmost component or components may remain free, at least in some specimens. The peculiar condition of the dorsalmost infraorbitals and the “‘ accessory sub- orbitals ”’ on the right side in 38110 is also difficult to interpret. It may be due to an ontogenetic disturbance or perhaps to regeneration after an injury. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 399 The preoperculum (Pop, Pls. 8, 9; Pl. 10, figs. 1, 2; Text-fig. 9) is a weakly crescentic bone with a dorsal, more or less vertical limb and an antero-ventrally directed ventral limb, much broader than the dorsal one. Its posterior margin has, ventral to the more or less pronounced angle between the two limbs, a characteristic shallow concavity or indentation. The operculum (Op, Pls. 8,9; Pl. ro, figs. 1, 2; Text-fig. 9) is roughly triangular ; its anterior, vertical margin and its postero-ventral margin are straight and its dorsal margin rounded off. The anterior margin is thickened ventral to the facet for the processus opercularis of the hyomandibular. The suboperculum (Sop, Pls. 8,9; Pl. io, figs. 1, 2; Text-fig. 9) is comparatively large and its antero-dorsal process anterior to the ventral tip of the operculum is remarkably broad and strong. According to 38739, P.1046b and P.4415 the posterior margin of the suboperculum has dorsally a shallow concavity or indentation. The interoperculum (Iop, Pl. 8; Pl. io, figs. 1, 2; Text-fig. 9) has a roughly triangular shape, with the posterior and ventral margins a little concave. Branchistegal rays and gular plate There seems to be a rather large number of branchiostegal rays; they are best exposed in P.4415 (R. Br, Pl. 10, fig. 1), on which 16 rays can be seen, but more small ones may exist in front of the anteriormost of the exposed rays. P.10466 shows parts of a rather large median gular plate (Gu, Pl. 10, fig. 2). Its anterior tip is a little bent upwards, but as the lateral margin is broken off, nothing can be stated regarding its shape. Lower jaw The lower jaw is generally only partly exposed, its dorsal margin being covered by other bones in all specimens investigated except in P.153 (PI. 15, fig. 5), where the lower jaw is isolated and its outline can be determined rather exactly ; its greatest depth is somewhat less than half of its length. The ventral, splenial part of the dentary (De. Spl, Pl. 8; Pl. 10, figs. 1,2; Pl. 15, fig. 5 ; Text-fig. 9) is limited dorsally by a well-marked ridge, and its lateral surface is ornamented with longitudinal ganoin striations and rugosities. The dental part is only gently ascending anteriorly and its lateral, slightly convex surface is not ornamented. In P.153 a few very small teeth can be observed on its margin. Approximately between the first and second third of the length of the lower jaw the dental part ascends abruptly and is thickened ; this thickening obviously corresponds to the dorsally directed thickening of the dentary in Leptolepis. Only the ventral part of the angular (or more precisely angulo-supra-angular, though no limit between the two components can be determined) is generally exposed (Ang, Pl. 8; Pl. ro, figs. 1, 2) ; in P.153 (Ang, Pl. 15, fig. 5) it is, however, entirely exposed although partly crushed and defective. Because of this it is not possible to determine its anterior margin in the dorsal part ; anteriorly it is rounded. Postero-ventrally the lower jaw ends in a conical point, reminiscent of a dermarticular, GEOL. II, 8 35 400 CERTAIN TRIASSIC AND LIASSTIC PHOLIDOPHORIDAE but as no suture can be seen separating it from the rest of the angular, it must be only the hindmost part of that bone. The postero-ventral part of the lateral surface of the angular, obviously exposed when the mouth is closed, is ornamented with ganoin streaks and small tuberculations, for the rest the lateral surface of the bone is smooth. In P.4415 the hindmost part of the right lower jaw is exposed in medial view, showing the articular (Avt, Pl. io, fig. 1; Pl. 15, fig. 8) with its articular fossa for the quadrate and a small but well marked ossified processus coronoideus. No true dermarticular has been observed. Sensory canal system of head The sensory canal system of the head is well developed as canals embedded in the bone tissue with short tubules and pores. The supraorbital sensory canal (soc, Pl.g; Pl. 15, fig. 3; Text-figs. 9, 10) pierces the nasal, the frontal and the parietal. The following description is based on 38110. In the nasal the canal describes an arch from near the anterior end of the lateral margin to near the mid-line of the bone and then passes almost straight back to the posterior tip of the nasal. At the bend it possesses an elevated, rather large pore, and further back there are two more pores or short tubules. There is no evidence of a communication between the supraorbital and the infraorbital sensory canals. From the nasal the canal passes over into the frontal at its concave margin behind the nasal, running at first almost straight backwards and then curving in a postero- lateral direction parallel to the lateral margin of the frontal before continuing postero- medially to the posterior margin of the bone. The number of tubules issuing from the frontal part of the canal seems to vary a little on both sides. On the dorsal side of the anterior straight part of the right canal there is a small pore, and further posteriorly a short tubule seems to issue from the lateral side ; where the canal bends it gives off a mesial series of g tubules of various lengths, and further backwards there seem to be two more, posteriorly directed tubules. On the left side a con- spicuous tubule issues from the lateral side of the anterior straight part of the canal (a corresponding tubule is seen also in 38738) ; more posteriorly the canal gives off at least 7 postero-mesially directed tubules and at least one posteriorly directed tubule ; a short distance anterior to the posterior margin of the frontal there is a conspicuous pore on the dorsal surface of the canal. The parietal contains only a very short part of the supraorbital sensory canal without pores or tubules, ending as a short groove which represents the anterior pit-line. The infraorbital sensory canal (ifc, Text-figs. 9, II) begins with a slightly arched ethmoidal commissure in the rostral, joining the canals of both sides ; no pores or tubules belonging to the ethmoidal commissure can be observed. From the rostral the canal passes over into the antorbital and pierces its anterior broader part, sending out posteriorly an antorbital branch. The antorbital part of the sensory canal system possesses four dorsal pores of which two are on the main CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 401 canal, one on its mesial part and one about dorsal to the point where the antorbital branch comes off, and one is in the anterior and one in the posterior part of this branch. From the antorbital the infraorbital sensory canal passes over into infraorbital 1 (lachrymal) and runs parallel to the anterior and ventral margins of this bone. From its anterior, arched part the canal sends out two very short, forwardly directed tubules, sometimes looking like wide pores, and from the ventral side there issue about 7 postero-ventrally directed tubules. As infraorbital 2 is missing in all specimens available, nothing can be said about the part of the canal belonging to this bone. In infraorbital 3 the number of tubules seems to be subject to rather great indivi- dual variation. In P.1046b (Pl. 10, fig. 2) the somewhat defective infraorbital 3 shows three large tubules, in P.421 and 41857 there are likewise three tubules and in 38110 there are three large, well separated tubules, whereas in P.1046a and 38738 four tubules are situated rather close together ; in 38739 there are two groups each of three small tubules. In the remaining small infraorbitals the sensory canal seems to give off a single, posteriorly directed tubule in each bone. Regarding the course of the infraorbital sensory canal in the dermosphenotic nothing can be stated with accuracy because of the defective state of preservation of this bone. In the dermopterotic the sensory canal runs along the lateral margin of the bone. Dorsally it gives off four short tubules. To what extent this canal belongs to the sensory canal system of the head and to the cephalic division of the main lateral line, respectively, is a question which must be left open. The point at which it gives off the preopercular sensory canal cannot be determined. Cephalic division of main lateral line On the posterior part of the parietal there is a transverse, slightly arched groove with its convexity directed postero-mesially ; this groove may represent the middle pit-line (m.p, Pl. 15, fig. 3 ; Text-fig. 10). A groove mesial to it may be the posterior pit-line (pp, Pl. 15, fig. 3 ; Text-fig. ro). The cephalic division of the main lateral line passes over from the dermopterotic into the extrascapular and continues backwards to the suprascapular, giving off mesially the supratemporal commissure (s. com, Text-figs. 9, 10), but the course of these canals cannot be followed exactly. Judging from the fragmentary extra- scapular on the left side in 38110 (Pl. 9) the main canal gives off laterally a large tubule, and at least four tubules are given off posteriorly from the lateral part of the supratemporal commissure ; its mesial part is missing. In the fragmentary right extrascapular of the same specimen there seem to be two tubules given off laterally from the main canal; the supratemporal commissure and its tubules are lacking. The preopercular sensory canal (Text-fig. 9) pierces the preoperculum in the middle of its length almost parallel to its anterior and posterior margins. In 38739 (PI. 8) it gives off posteriorly 17 tubules, almost evenly distributed over the whole length 402 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE of the canal, five of which are situated in the dorsal limb of the bone, dorsal to the obtuse posterior angle, but as the antero-ventral tip is not entirely exposed it is possible that one more tubule ought to be added to this number. In P.10466 (Pl. 10, fig. 2) there are likewise 17 tubules, four of which are situated in the dorsal limb, but between tubules 7 and 8 reckoned from the antero-ventral tip of the bone there is in addition a short tubule, probably only a branch of tubule 7; tubules 1-7 are situated close together. Specimen 48009 also shows 17 tubules issuing from the morphologically posterior side of the preopercular sensory canal, whereas in P.421 and P.3595 only 15-16 tubules could be determined with accuracy. In 38738 five tubules are clearly visible in the dorsal limb of the preoperculum. In P.1046b and P.4415 the lateral surface of the preoperculum possesses a well- marked groove which may be the posterior portion of the anterior division of the supramaxillary pit-line (ic,, Pl. ro, figs. 1, 2). A postmaxillary pit-line cannot be traced in any of the specimens investigated except in P.421 ; in this specimen the middle of the lateral surface of the preoperculum shows a vertical groove which I interpret as the postmaxillary pit-line. The mandibular sensory canal (mdc, Text-fig. 9) cannot be followed because of the thickness of the splenial part of the dentary, but in P.10460 (PI. ro, fig. 2) many of its short, backwardly directed tubules are discernible, their openings being accen- tuated by ganoin thickenings. On the lateral surface of the angular there is postero-ventrally a short vertical groove in P.153 and P.4415 which must be the oral pit-line (07fy,, Pl. ro, fig. r ; Pl. 15, figs. 5, 8). Exoskeletal shoulder girdle and squamation Very little can be said about the suprascapula and the supracleithrum, which are only fragmentarily preserved in 38110 (Ssc, Sci, Pl. 9) and almost entirely destroyed by crushing in the other specimens investigated ; the outline of the suprascapula and the supracleithrum as given in the attempted restoration (Ssc, Sci, Text-fig. 9) is consequently very doubtful. The cleithrum and the postcleithrum are not sufficiently exposed in the specimens available to allow their description in any detail. The scales are arranged in about 45-50 transverse rows reckoned from the hind margin of the operculum to the middle of the caudal fin ; posterior to the last row there are about 8-9 rows on the body axis dorsal to the lateral line, thus in all about 55-60 rows, which accords well with Woodward’s (1941) statement of about 50-60 transverse rows. On the body there are generally four, sometimes only three, longitudinal rows of enlarged scales, the deepest being about twice as deep as broad and with the posterior margin crenulated ; the number of points on the posterior margin is about 8, seldom up to ro on the largest scales, but it decreases posteriorly in accordance with the decreasing depth of the scales. The scales are comparatively thick on the middle and posterior parts of the body as well as on the tail ; on the anterior part of the body they are, however, thinner and show, as stated by Woodward, a concentric striation. In most specimens investigated CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 403, this striation is visible only in a few transverse rows behind the skull; in P.10466, however (PI. ro, fig. 2), such thin striated scales cover the whole anterior part of the body and extend posteriorly almost to the level of the pelvic fin base. Lateral line The lateral line pierces the suprascapula, and seems to give off three short but rather wide, mesially to postero-mesially directed tubules. In the supracleithrum the course of the sensory line is not clearly visible in any of the specimens available, but it seems to give off three postero-mesially directed tubules. The anterior lateral line scales are rather indistinct, but more posteriorly they are clearly visible, each scale having a distinct pore on its lateral surface. The course of the lateral line is almost straight, along the middle of the body, ending at the base of the middle caudal lepidotrichia. Axial skeleton and paired and unpaired fins The axial skeleton consists of rather thin ring-vertebrae, which are partly visible in some of the specimens available ; in none of them, however, are the vertebrae exposed to such a degree that anything can be made out regarding their shape and number. The pectoral fin is of moderate size. The number of lepidotrichia is rather high ; in P.4415 not less than 26 lepidotrichia can be counted, the ventralmost ones are, however, very small and articulated almost from their bases. In the type specimens P.572 and P.573 as well as in P.1046b and 47461 there seem to be at least 23 or 24 lepidotrichia. The first lepidotrichium is strong and rather densely set with fulcra (Fu, Pl. 8). In the ventral fin the number of lepidotrichia amounts to 15, which number I have counted in P.1046b and P.4415. As in the pectoral fin the last two or three lepido- trichia are rather delicate and easily overlooked. The first two lepidotrichia are rather short and unarticulated, the third one (the first branched lepidotrichium) is densely set with small fulcra. Lateral to the base of the ventral fin there is a large axillary scale and enlarged scales also occur between the fins of both sides. The dorsal fin has, anteriorly, two unarticulated lepidotrichia, the first as a rule rather short, the second longer. The third, long lepidotrichium carries a series of small fulcra on its anterior margin. The total number of lepidotrichia is difficult to determine as in most specimens the last rays are covered by scales or are defective ; in P.1046a there are, however, about 14-15 lepidotrichia, and in P.421 I have counted 15. The number of dorsal radials cannot be stated because of the thickness of the squamation. The anal fin has, as in the dorsal fin, two unarticulated lepidotrichia anteriorly followed by a third long, fulcrated lepidotrichium ; the total number of lepidotrichia is difficult to determine but seems to be at least 11-12. No anal radials are visible. 404 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE The caudal fin may be interpreted as hemi-heterocercal, the scaled lobe dorsal to the lateral line (the reduced body axis) reaching farther posteriorly than the scaled ventral lobe. The dorsal and ventral margins are densely set with small fulcra. HorIZON AND LOCALITY. Lower Lias; Lyme Regis, Dorset. Pholidophoroides limbata (Agassiz) (Pls. 11, 12; Text-fig. 12) 1833 Pholidophorus limbatus Agassiz, 2, 1 : 9. 1843 Pholidophorus limbatus Agassiz, 2, pl. 37, figs. 1-5. 1844 Pholidophorus limbatus Agassiz, 2, 1 : 282. 1895 Pholidophorus limbatus Agassiz; (partim) Woodward : 464, pl. 12, fig. 7. Diacnosis. Pholidophoroides similar to Ph. crenulata but somewhat larger in size, at least up to 180 mm. in total length. Greatest depth of body nearly one-third of standard length or slightly less (ca. 28-32°% of that length). Depth of caudal peduncle one-tenth of standard length or slightly more (10-11%). Maxillary proportionately not quite as deep as in Ph. crenulata. Preopercular sensory canal with about 14 tubules. Scales rather thick in about 40-42 transverse rows reckoned from hind margin of operculum to middle of caudal fin; five to six longitudinal rows of body scales deeper than broad with hind margin strongly though finely serrated. Horotyre. Unknown. Agassiz says (loc. cit., p. 283): “Il en existe des exemplaires dans plusieurs collections d’Angleterre, entre autres au Musée d’Oxford, dans les collections de lord Enniskillen et de sir Philipp Egerton.” None of the specimens from the two last named collections {now in the British Museum (Natural History)] agree with either of the figures (pl. 37, figs. 1, 2) given by Agassiz. I have had no opportunity to visit the Oxford Museum collection. Woodward (1895) remarks: “‘ Type. Distorted specimens of trunk ’’, but gives no information regard- ing the collection to which they belong. P.1047 is chosen as LECTOTYPE. MaTERIAL. The following description is based mainly on P.1047, P.1047a, P.3596, P.4422, 36472, 38531, and 41906. Specimens P.1047) and c, P.3632, 38532 and 27595 also belong to this species but are more defective, and from them only little could be added to the description ; P.4410 is perhaps not referable to Ph. limbata. P.7582, an “‘ imperfect skeleton probably of this species ’’ according to Woodward (1895 : 466) belongs to the species described below (p. 416) as Pholidophoropsis maculata sp. Nn. DeEscriPTION. The total length of the largest specimen, P.1047, is 180 mm., its length from the tip of the snout to the hindmost lateral line scale (standard length) is estimated at 150 mm. The other specimens have a standard length varying be- tween about 140 and 108 mm. The greatest depth of the body is nearly one-third of the standard length or slightly less (about 28-32%), the depth of the caudal peduncle is about 10-11% of the standard length ; the depth of the body and of the CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 405 caudal peduncle are consequently somewhat greater in this species than in Ph. crenulata. The length of the head is about one quarter of the standard length or slightly less (about 23-25°%), about the same as in the type species. The ventral fins are situated approximately midway between the tip of the snout and the hindmost lateral line scale, and the distance between the base of the ventrals and the origin of the anal is about 78-88% of the distance between the bases of the pectoral and ventral fins, about as in Ph. crenulata. A good figure of the entire fish is given by Woodward (1895, pl. 12, fig. 7). All specimens available are unfortunately more or less defective with respect to the snout region and the skull roof, consequently the following description of the cranial bones is rather incomplete and provisional. Presumably it will, however, be sufficient to demonstrate the close relationship between Ph. limbata and the type species of the genus. Exoskeletal skull roof This part of the skull is more or less defective in the specimens available. In specimen P.1047 nothing of it is preserved, in 38531 only the dermosphenotic, the dermopterotic and the extrascapular are partly present, in P.4422 and 36472 some bones of the snout region can be observed, but they are partly defective and dis- located. The premaxillary (Pmx, Pl. 12, figs. 1, 2; Text-fig. 12) is rather well exposed, after preparation, in P.4422. It is an almost square, slightly tumid bone with the lateral surface almost smooth, not tuberculated as in Ph. crenulata. The antero- ventral margin carries a single row of comparatively large teeth, considerably larger than in Ph. crenulata; they are slightly curved and increase in length posteriorly. The rostral (Ro, Pl. 12, fig. 2 ; Text-fig. 12) is to a great extent visible in P.4422 asa thin, somewhat bulging bony plate, partly covering a bone which may be the meseth- moid. The nasal is missing in P.4422, but in 36472 (Na, Pl. 11, fig. 2) there is an oblong bone, obviously pierced by a sensory canal, which I interpret as the right nasal. Its outline cannot, however, be determined in detail. The frontal (Fr, Pl. 11, fig. 2; Text-fig. 12) is preserved, though much crushed and broken off anteriorly, in P.4422 ; in 36472, on the other hand, the anteriormost part of the frontal is preserved. Judging from these two specimens the frontal has on the whole the same shape as in Ph. crenulata. Part of a supraorlital is present in P.4422, but it gives no information regarding the shape of this bone. The dermosphenotic is missing or only present as fragments in all the specimens investigated ; in 38531 (Dsph, Pl. 12, fig. 3) its anterior, pointed part is clearly visible, indicating that its shape is about the same as in Ph. crenulata. The outline of the parietal is impossible to determine. The dermopterotic (Dpt, Pl. 11, fig. 2; Pl. 12, fig. 3; Text-fig. 12) does not seem to differ in general from that of Ph. crenulata; parts of it are exposed in P.4422, 36472, and 38531. 406 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Na / Ant / \ 2 \ Ro Et | eS &) fe. com ly sa! LS om ee ~ aie ar vf ray VA ‘ PS yeti hy 27 | \ I An / \ | 9 he, \ / | orp, / poc | orp Fic. 12. Pholidophoroides limbata (Agassiz). Attempted restoration of head in lateral view. x 4:2. Ang, angular; Ant, antorbital; De. Spl, dentary; Dpi, dermopterotic; Dsph, dermo- sphenotic ; Ext, extrascapular; Fy, frontal; Jfo,—Ifo,, infraorbitals 1 to 7; Jop, inter- operculum ; Mz, maxillary ; Na, nasal; Op, operculum ; Pa, parietal; Pmzx, premaxillary ; Pop, preoperculum ; Ro, rostral; Sbo, suborbital; ‘‘ Sbo’’, ‘“accessory suborbital” ; Smx,, Smx,, anterior and posterior supramaxillaries; So, supraorbital; Sop, suboperculum ; ant. by, antorbital branch of infraorbital sensory canal ; hce,, anterior division of supramaxillary pit-line ; 2fc, infraorbital sensory canal; zfc. com, ethmoidal commissure ; ovp, postmaxillary pit-line ; orp,, oral pit-line ; foc, preopercular sensory canal ; soc, supraorbital sensory canal. The extrascapular (Ext, Pl. 12, fig. 3; Text-fig. 12) is as a rule very defective ; in specimen 38531, however, its lateral part is well exposed, its antero-lateral corner being rounded and its postero-lateral and postero-mesial margins almost straight. Dermal bones of cheek and opercular apparatus The maxillary (Mx, PI. 11, figs. 1,2; Pl. 12, figs. 1, 2; Text-fig. 12) is well pre- served in P.1047 and 36472, its anterior, mesially directed part is, however, not entirely visible but seems to be rather low asin Ph. crenulata. The lateral, posteriorly CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 407 directed part is not as deep as in Ph. crenulata, but otherwise both species agree closely in the most important characteristics, such as the convexity of the dorsal margin below infraorbital 1 and supramaxillary 1, the deeply notched posterior margin and the longitudinal striation of its lateral surface. The posterior part of the convex ventral margin of the maxillary has, as in Ph. crenulata, a delicate, comb-like dentition, but anteriorly there is in P.4422 (Pl. 12, fig. I) a single row of small tuberculations, obviously the bases of somewhat larger teeth which have been broken off. The same arrangement can also be observed on the ventral margin of the left maxillary in 36472 (M~zg, Pl. 11, fig. 2). The two supramaxillanies (Smx,, Smx,, Pl. 11, figs. 1, 2; Text-fig. 12), only well preserved in P.1047 and 36472 (in the last named specimen partly pushed in below the dorsal margin of the maxillary), are also similar to those of Ph. crenulata with regard to shape and the presence of a well-marked striation on the lateral surface. The antero-dorsal corner of supramaxillary 2 does not seem to be drawn out into a larger process. The bones of the infraorbital series are only partly preserved. In specimens P.4422 and 36472 there is a defective bone in the snout region which may be interpreted as an antorbital (Ant, Pl. 11, fig. 2; Pl. 12, fig. 2; Text-fig. 12) ; its shape seems to be about as in Ph. crenulata but its outline cannot be determined exactly. Infraorbital 1 (lachrymal) (Ifo,, Pl. 11, figs. 1, 2; Pl. 12, fig. 2; Text-fig. 12) is present only in P.1047, P.4422, and 36472, in all three specimens only as a fragment ; however, from what can be observed it agrees with that of Ph. crenulata. An infraorbital 2 (Text-fig. 12) is present in P.4422 ; it is elongate antero-posteriorly, its dorsal margin is slightly concave and its ventral margin correspondingly convex. Infraorbital 3 (Ifos, Pl. 11, figs. 1, 2; Pl. 12, fig. 3; Text-fig. 12) is present in P.1047, P.4422, 36472 and 38531. Its outline cannot be followed exactly in all specimens, but as compared with the corresponding bone in Ph. crenulata its posterior part seems to have a greater dorso-ventral extension. Dorsal to infraorbital 3 there is in 38531 (PI. 12, fig. 3) a well-preserved infraorbital 4, and further dorsally there seem to be traces of three more infraorbitals. In 36472 (PI. 11, fig. 2) infraorbital 4 is partly exposed, infraorbital 5 is clearly visible and dorsal to it there are some fragments probably of two more infraorbitals. Con- sequently, as in Ph. crenulata there seem to be seven infraorbitals altogether. The suborbital (Sbo, Pl. 11, figs. 1,2; Pl. 12, fig. 3 ; Text-fig. 12) is partly preserved in P.1047, P.4422, 36472 and 38531. Corresponding to the comparatively great depth of infraorbital 3 in this species, the suborbital is not as deep as in Ph. crenulata. In both P.4422 and 36472 (“ Sbo’’, Pl. 11, fig. 2; Text-fig. 12) a small bone, frag- mentary in P.4422, has been observed between the suborbital and the dermopterotic, which probably corresponds to the “ accessory suborbital ”’ found in some specimens of Ph. crenulata. The preoperculum (Pop, Pl. 11, figs. 1, 2; Pl. 12, fig. 3; Text-fig. 12) has, on the whole, the same shape as in Ph. crenulata, but it is somewhat straighter, the posterior margin being more evenly arched with only a little-marked angle between the dorsal 408 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE and the ventral limbs and almost without the shallow concavity or indentation ventral to the angle that is found in Ph. crenulata. The operculum, the suboperculum and the interoperculum seem not to differ in shape from the corresponding bones in Ph. crenulata. Branchiostegal rays and gular plate Some branchiostegal rays (R. By, Pl. 11, fig. 1, Pl. 12, fig. 3) are preserved in P.1047 and 38531; they show nothing of special interest. A gular plate is not visible in any of the specimens investigated. Lower jaw The lower jaw is partly exposed in P.1047, P.4422 and 36472. The dental part of the dentary has a smooth lateral surface and the lateral surface of the splenial part of the bone is more or less ornamented with a longitudinal ganoin striation as in Ph. crenulata. No dentition can be observed with certainty. Sensory canal system of head Due to the fragmentary preservation of the skull very little can be said about the sensory canal system of the head. In 36472 the anterior part of the swpraorbital sensory canal is indicated in the nasal by an anterior and a posterior opening ; because of the thickness of the bone the course of the canal between the two openings cannot be observed with accuracy. In P.4422 the hindmost part of the sensory canal in the frontal and its short continua- tion in the parietal may be traced. For the rest the supraorbital sensory canal and its tubules and pores cannot be seen. Of the infraorbital sensory canal, only the following few remarks can be given. In P.4422 the ethmoidal commissure (zfc. com, Pl. 12, fig. 2 ; Text-fig. 12) joining the infraorbital sensory canals of both sides can clearly be seen as an arched bony tube which is slightly elevated above the dorsal surface of the rostral. Laterally there seem to be one or two pores, but whether they really are pores cannot be stated with certainty. In the antorbital the infraorbital sensory canal seems to have about the same course as in Ph. crenulata, but because of the defective state of this bone a more detailed description cannot be given. In infraorbital 1 (lachrymal), partly preserved in P.1047, the sensory canal has the same course along the anterior and ventral margins of the bone as in Ph. crenulata ; of the tubules only the ventral one of the two pore-like tubules from the anterior, arched part of the canal and the three anterior tubules from the ventral part of the canal can be seen. In P.4422 and 36472 only a small part of the sensory canal is preserved. In infraorbital 2 the sensory canal runs parallel to the dorsal and ventral margins of the bone. Whether this part of the canal gives off a tubule may be left open. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 409 In infraorbital 3 the sensory canal gives off, posteriorly, three rather long tubules in P.1047 (Pl. 11, fig. r) and four tubules in 38531 (Pl. 12, fig. 3). In Ph. limbata, as in Ph. crenulata, the number and arrangement of the tubules in this bone are subject to individual variation. The canal and its tubules cannot be observed in P.4422 and 36472. In infraorbital 4, clearly visible only in 38531 (Pl. 12, fig. 3), the canal gives off a single, posteriorly directed tubule and the same seems to be true of infraorbital 5 in 36472. In the remaining infraorbitals, both more or less defective, the sensory canal cannot be traced. In the dermosphenotic in 38531 (PI. 12, fig. 3) an antero-dorsally directed branch of the infraorbital sensory canal is clearly visible, giving off at least two short, dorsally directed tubules. As in Ph. crenulata, nothing can be said concerning the various parts of the sensory canal in the dermopterotic and the origin of the preopercular sensory canal. In 36472 and 38531 the course of the sensory canal in the dermopterotic seems to be about the same as in Ph. crenulata. Cephalic division of main lateral line The cephalic division of the main lateral line passes over from the dermopterotic into the extrascapular. In 38531 (Pl. 12, fig. 3; Text-fig. 12) its course in this bone can be followed, in part at least, from near its antero-lateral corner to near its posterior end, mesially giving off the supratemporal commissure (s. com, Text-fig. 12) and laterally three postero-laterally directed tubules. The preserved lateral part of the supratemporal commissure gives off three tubules posteriorly. The preopercular sensory canal (Text-fig. 12) has the same course along the middle of the preoperculum as in Ph. crenulata. Regarding the tubules given off from the posterior side of the canal, thirteen are clearly visible in 38531, but as the pre- operculum is a little damaged at about its mid-point, there may be at least one more tubule. In P.1047 I have counted 14 tubules, but their arrangement is remarkably irregular ; in 36472 at least 12 tubules are visible, but one or two more tubules may be present in the anteriormost hidden part of the preoperculum. The total number of preopercular tubules in Ph. limbata may consequently be estimated as about 14 against about 17 in Ph. crenulata. In P.1047 and 36472 the lateral surface of the preoperculum clearly shows the two grooves representing the posterior portion of the anterior division of the supra- maxillary (horizontal) pit-line and the postmaxillary (vertical) pit-line respectively (hc,, orp, Pl. 11, figs. 1,2; Text-fig.12). Asin Pholidophorus latiusculus an anterior portion of the anterior division of the supramaxillary line is clearly visible on the lateral surface of infraorbital 3. Regarding the mandibular sensory canal, nothing can be made out. In 36472 the lateral surface of the angular possesses a small vertical groove, obviously the oral pit-line (o7p,, Pl. 11, fig. 2). 410 CERTAIN ZRIASSTIC AND LITASSRE PHOLIDOPHORIDAE Exoskeletal shoulder-girdle and squamation The suprascapula and supracleithrum are preserved only in 38531, and then only as fragments which give very little information about their shape. The cleithrum (Cl, Pl. 11, fig. 2; Pl. 12, fig. 3) is, to a certain extent, visible in 36472 and 38531 ; it seems to be proportionately smaller than that of Leptolepis. A deep, scale-like bony plate posterior to the suboperculum and cleithrum may be interpreted as a postcleithrum (Pcl, Pl. 12, fig. 3). The scales are arranged in about 40 to 42 transverse rows counted from the hind margin of the operculum to the middle of the caudal fin ; posterior to the last row there are about 9 rows on the body axis dorsal to the lateral line, thus in all about 50 rows. On the body there are generally five longitudinal rows of enlarged scales, the deepest more than twice as deep as broad (proportionately deeper than in Ph. crenulata) and with the posterior margin finely serrated ; the number of small points on the posterior margin is approximately 15 on the largest scales or about as in pl. 37, fig. 4 of Agassiz (1843). All the scales, including the anterior ones, are com- paratively thick, and I cannot see any concentric striations such as are present in Ph. crenulata. Lateral line Because of the defective state of the suprascapula and supracleithrum nothing can be made out regarding the lateral line in these bones. The anterior lateral line scales are also indistinct, but a few transverse rows behind the operculum the lateral line becomes very distinct and runs almost straight to the base of the middle caudal Tays. Paired and unpaired fins The pectoral fin is similar to that of Ph. crenulata, the highest number of lepido- trichia seems to be 23, perhaps 24, thus a little lower than in the type species ; it must, however, be kept in mind that the number of the small ventral lepidotrichia is difficult to determine. The ventral fin has 12 clearly visible lepidotrichia in 38531, but as in Ph. crenulata there may have been about three more small ones; in P.3596 I have counted 14 lepidotrichia, the two innermost very small. The dorsal fin has the same general shape as in Ph. crenulata. Also regarding the number of lepidotrichia there is no important difference. In 38531 there seem to have been about 14 lepidotrichia, the anterior ones preserved as impressions only ; in the other specimens investigated the dorsal fin is more or less defective. The anal fin also is similar to that in Ph. crenulata; the anterior part of this fin with its two anterior short lepidotrichia followed by the third long and fulcrated lepidotrichium is figured by Woodward (19109, pl. 22, fig. 6). The total number of lepidotrichia is about 13, perhaps up to 15 in P.1047a. In P.4422 the anal radials, 12 in number, are clearly exposed. The caudal fin is hemiheterocercal as in Ph. crenulata. HORIZON AND LOCALITY. Lower Lias ; Lyme Regis, Dorset. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 411 Genus PHOLIDOPHOROPSIS nov. PRELIMINARY DIAGNOSIS. Pholidophoridae of medium size, as far as known similar to Pholidophoroides but differing in the following features. Exoskeletal cranial bones without or with only very feeble ganoin ornamentation. Scales thin and cycloid with fine concentric striations. TYPE SPECIES. Pholidophoropsis caudalis (Woodward). Pholidophoropsis caudalis (Woodward) (Pl. 13, Text-fig. 13) 1844 ? Leptolepis caudalis Agassiz, 2, 2 : 133 (nom. nud.). 1895 Pholidophorus caudalis Woodward : 457, pl. 18, figs. I, 2 (partim). 1941 Pholidophorus (? Pholidophoroides) caudalis Woodward ; Woodward : 90. Diacnosis. Pholidophoropsis of medium size, at least up to 120 mm. in total length. Greatest depth of body slightly more than one quarter of standard length (ca. 26-27% of that length). Depth of caudal peduncle slightly more than one-tenth (ca. 11%) of standard length. Maxillary proportionately not as deep as in Ph. crenulata. Preopercular sensory canal with about 10 tubules. Scales very thin. Hototyre. British Museum (Natural History) No. P.3664. MATERIAL. Besides the holotype, specimens P.3664a (paratype) and 39871 belong to this species and have been used in the following description. DeEscriPTIon. The total length of the holotype is 120 mm.; P.3664a and 39871 have a total length of 95 mm. and 117 mm. respectively, but in both these specimens the tips of the caudal fin are broken, and the maximum length of the last named specimen may consequently have been at least 120 mm.! The standard length or length from the tip of the snout to the base of the middle caudal fin rays is about 100 mm. in the holotype, in the two other specimens 81 and 102 mm. respectively, but P.3664a@ seems to be a little shortened by pressure. The greatest depth of the body is slightly more than one quarter (26-27%) of the standard length and the depth of the caudal peduncle is about 11% of the standard length. The length of the head seems to be exactly one quarter of the standard length in the two larger specimens ; in P.3664a this length is slightly greater because of the shortened standard length. The ventral fins are situated about midway between the tip of the snout and the base of the caudal fin rays, and the distance between the base of the ventrals and the origin of the anal is about 70% of the distance between the bases of the pectoral and ventral fins. As in the species of the genus Pholidophoroides the dorsal fin is situated above the interspace between the ventral and anal fins ; in P.3664a a forward displacement of the dorsal fin seems, however, to have occurred through pressure. Good figures of the entire holotype and of P.3664a are given by Woodward (1895, pl. 18, figs. 1, 2). 1 Woodward (1895 : 457) gives the length of Ph. caudalis as “‘about 0°13’’; This obviously refers to 43055, described below (p. 416) as Pholidophoropsis maculata sp. nov. 412 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE All three specimens are more or less defective and the following description is consequently rather incomplete but sufficient for identifying the species and demon- strating its close affinity with members of the genus Pholidophoroides described above. As a general characteristic of Ph. caudalis it may be mentioned that the exoskeletal bones of the skull, as well as the scales, are rather thin and almost totally lack ganoin ornamentation. Exoskeletal skull roof A premaxilary (Pmx, Pl. 13, fig. 1; Text-fig. 13) is preserved in P.3664a; its denticulated ventral margin is visible, but as the other parts of the bone are hidden below the antorbital, its general shape cannot be stated. The vostval cannot be identified with certainty in any of the specimens available. The nasal (Na, Pl. 13, figs. 1, 2; Text-fig. 13) is partly visible in P.3664a, but this bone too is to a great extent hidden below the antorbital ; a bone fragment anterior to the frontal in the holotype may represent the posterior part of the nasal. From these two fragments it may be concluded that the nasal of this species does not differ very much from the corresponding bone in Ph. crenulata. The frontal (Fr, Pl. 13, figs. 1, 2; Text-fig. 13) seems to be rather broad in its anterior part and its dorsal surface bulges along the anterior part of the supraorbital sensory canal. The posterior, broad part of the frontal is defective and nothing of its posterior margin can be made out. The suture between the frontals of both sides is straight anteriorly but then makes an S-shaped bend to the left and to the right, clearly visible on the holotype. The anterior part of a rather strong supraorbital (So, Pl. 13, fig. 2; Text-fig. 13) obviously similar to that in Ph. crenulata, is present in the holotype. The dermosphenotic and the parietal are not discernible. In the holotype parts of the dermopterotic and extrascapular (Dpt, Ext, Pl. 13, fig. 2; Text-fig. 13) are exposed but are insufficient to allow any detailed description. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pl. 13, figs. 1, 2; Text-fig. 13) is best preserved in P.3664a ; it has, on the whole, the same shape as in Ph. crenulata with the characteristic convexity on the dorsal margin, but the longitudinal striation on the lateral surface is not nearly so pronounced. The posterior margin is partly crushed, but may have been notched in the same way as Ph. crenulata. The main difference from that species, apart from the lesser depth of the bone, lies in the presence of a delicate dentition along the whole ventral margin, not confined to its posterior half only. The two supramaxillaries (Smx,, Smx,, Pl. 13, fig. 1; Text-fig. 13) have the same general shape as in Ph. crenulata, but judging from P.3664a@ in which they are best preserved and well exposed, the antero-dorsal corner of supramaxillary 2 seems to be more markedly drawn out into a short process. Both supramaxillaries show a faint concentric striation on their lateral surface. The bones of the infraorbital series are, for the most part, badly preserved. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 413 Fic. 13. Pholidophoropsis caudalis (Woodward). Attempted restoration of head in lateral view. X65. Ang, angular ; Ant, antorbital; De. Spl, dentary ; Dpt, dermopterotic ; Evt, extrascapular ; Fy, frontal; Ifo,, [fo,—Ifo;, infraorbitals 1 and 3 to 5; lop, interoperculum ; My, maxillary ; Na, nasal; Op, operculum; Pa, parietal; Pmx, premaxillary; Pop, preoperculum; fo, rostral; Smx,, Smx,, anterior and posterior supramaxillaries; So, supraorbital; Sop, suboperculum ; ifc. com, ethmoidal commissure ; poc, preopercular sensory canal; s. com, supratemporal commissure ; soc, supraorbital sensory canal. The antorbital (Ant, Pl. 13, figs. 1, 2 ; Text-fig. 13) is present in P.3664 and P.3664a, but in both the postero-dorsal part is broken off. Judging from the last named specimen it seems, however, that the antorbital in this species is shorter than in Ph. crenulata. Infraorbital 1 (lachrymal) (Ifo,, Pl. 13, fig. 2; Text-fig. 13) can be traced in the holotype only, but the fragment gives no information as to the shape of the bone. Infraorhital 2 is missing in all the three specimens available. An infraorbital 3 (Ifos, Pl. 13, fig. 2 ; Text-fig. 13) is present in the holotype as well as in 39871. It has the ordinary shape with the antero-ventral margin slightly concave and the posterior margin rounded. Of the remaining bones in the infraorbital series, a small infraorbital 4 is present in the holotype (Jfo,, Pl. 13, fig. 2). Two fragments in P.3664a may represent infraorbitals 4 and 5. 414 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE A suborbital bone (Sbo, Pl. 13, fig. 2) is visible in the holotype, but its posterior margin is not distinguishable. Dorsal to this bone there is a small tongue-like bone similar to the small bone or bones dorsal to the suborbital in Ph. crenulata (see p. 398). The preoperculum (Pop, Pl. 13, figs. 1, 2; Text-fig. 13) is more or less defective in the specimens available ; it is best preserved in the holotype and has the same general shape as in Ph. crenulata, but its ventral, antero-ventrally directed limb diminishes markedly in depth towards the tip of the bone. The operculum, the suboperculum and the interoperculum are partly preserved in the holotype and, as far as can be judged, they have the same general shape as the corresponding bones in Ph. crenulata. Lower jaw The lower jaw is visible to a great extent in P.3664a (De. Spl, Ang, Pl. 13, fig. I), but as in most specimens of the two Pholidophoroides species described above, its dorsal margin is covered by the maxillary and therefore neither the entire outline of the lower jaw nor its depth can be determined. The lateral surface of the dentary has a longitudinal crest, obviously corresponding to the well-marked ridge between the dental and splenial parts of the dentary in Ph. crenulata, but ventral to this crest there are no traces of the rich ganoin ornamentation characteristic of this species and of Ph. limbata. Sensory canal system of head The supraorbital sensory canal (soc, Pl. 13, fig. 2; Text-fig. 13) in the nasal is partly visible in P.36644 as a short, laterally curving canal, representing the anterior- most part of this sensory canal. In the holotype no canal can be seen with certainty in the preserved posterior fragment of the nasal. The frontal part of the supraorbital sensory canal has the same general course as in Ph. crenulata and is best exposed in the left frontal of the holotype. Its anterior, straight part is marked by a bulging of the dorsal surface of the bone, and on this part there is a small dorsal pore. At the postero-lateral curve the canal gives off from its mesial side a posteriorly directed tubule and from the lateral side at least two short, laterally directed tubules ; more posteriorly the sensory canal cannot be followed. The main difference from the supraorbital sensory canal in Ph. crenulata is that the tubules are given off mainly from the lateral side of the canal in Ph. caudalis but from the mesial side in Ph. crenulata. The infraorbital sensory canal is clearly visible in the antorbital bone of the holotype as well as in P.3664a ; it gives off a posteriorly directed antorbital branch. For the rest the infraorbital sensory canal can only be traced as a slightly bent canal in infraorbital 3, giving off three postero-ventrally directed tubules in the holotype and in 39871. In the holotype infraorbital 4 shows a single tubule given off from the posterior side of the canal, for the rest the infraorbital sensory canal cannot be followed. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 415 Cephalic division of main lateral line In the holotype the sensory canal in the dermopterotic can be traced parallel to the dorso-lateral margin of the bone, but only a single pore on its dorsal surface can be observed. In the extrascapular the main canal and the supratemporal commissure are rather clearly visible (Pl. 13, fig. 1; Text-fig. 13). No tubules from the main canal are, however, discernible, but the supratemporal commissure gives off from its posterior side at least four backwardly directed tubules, decreasing in length towards the mid-line of the skull. The preopercular sensory canal pierces the preoperculum in the middle of its length, as in Ph. crenulata. The tubules given off from its morphologically posterior side are short and straight ; their number cannot be determined exactly, but they are obviously less numerous than in Ph. crenulata ; in the dorsal limb of the bone only one tubule is visible dorsal to the posterior, ill-defined angle between the dorsal and ventral limbs, but as the upper part of the dorsal limb is not well exposed, there may be one or more tubules in the dorsal limb of the preoperculum. In the ventral limb, well exposed in the holotype, eight tubules are visible. The mandibular sensory canal cannot be seen. Exoskeletal shoulder girdle and squamation Because of the bad state of the material available very little can be determined of the bones belonging to the exoskeletal shoulder girdle ; nothing seems, however, to indicate that they are of special interest as species characteristics. The squamation consists of very thin scales forming, where they are preserved, a continuous thin layer, generally without marked boundaries between the individual scales. In the holotype some scales are, however, clearly visible; they are thin with a delicate concentric striation, and their posterior margins are rounded and not serrated. It must, however, be noted that in P.3664a and 39871 some of the posteriormost scales are markedly thicker than the body scales. A lateral line cannot be traced in any of the three specimens investigated. Axial skeleton and paired and unpaired fins The axial skeleton consists, as in Ph. crenulata, of rather thin ring-vertebrae. The shape and number of the abdominal vertebrae is difficult to state ; the centra of the caudal vertebrae are, judging from P.3664a@, about twice as deep as broad. The best preserved pectoral fin, that of P.3664a, shows 21 lepidotrichia, but it is not quite certain that the series is complete. The first, strong lepidotrichium is fulcrated. The ventral fin has, in the same specimen, 15 clearly visible lepidotrichia ; the first ones are, however, defective, and no fulcra could be seen. The dorsal fin is similar to the corresponding fin in Ph. crenulata, including the presence of fulcra ; in 39871 I have counted 15 lepidotrichia. The radials, clearly visible in this species because of the thin scale covering, in the holotype number at least 14, in P.3664a about 14 and in 39871 thirteen. GEOL. II, 8 36 416 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE The anal fin is likewise similar to that of Ph. crenulata, but the number of lepido- trichia cannot be determined. The number of radials is eleven in both the holotype and P.3664a@ ; in 39871 there must have been more than the g radials still preserved. The caudal fin is hemi-heterocercal as in Ph. crenulata and Ph. limbata, and with the dorsal and ventral margins densely set with small fulcra. REMARKS. The two best preserved specimens of this species, the holotype and P.3664a, could be studied only during short visits to London and therefore it is quite possible that a closer investigation of them will reveal details not mentioned in the description given above. It is my hope, however, that this description will be sufficient for the identification of the species as well as providing a suitable basis for a discussion regarding its affinities and taxonomic position. HoRIZON AND LOCALITY. Lower Lias; Lyme Regis, Dorset. Pholidophoropsis maculata sp. nov. (Pl. 14, Text-figs. 14, 15) 1895 Pholidophorus caudalis Woodward : 457 (partim). 1895 Pholidophorus limbatus Agassiz ; Woodward : 466 (partim : P.7582). Diacnosis. Pholidophoropsis of medium size, at least up to about 160 mm. in total length. Greatest depth of body about one quarter of standard length. Maxil- lary comparatively slender. Preopercular sensory canal with about 14 tubules. External bones of skull with scattered ganoin spots and streaks. Scales compara- tively large, thin, with concentric striae and with posterior margin rounded ; axillary scales with ganoin spots. Hototyre. British Museum (Natural History) No. 43055. MATERIAL. Besides the holotype, P.7582, listed by Woodward under Pholido- phorus limbatus as “‘ Imperfect skeleton probably of this species ’’, shows such striking similarities with the holotype that both undoubtedly belong to one and the same species. It has also been used for the following description, but its defective state of preservation allows only some complementary additions to the description of the holotype. DESCRIPTION. The total length of the holotype is about 127 mm. (PI. 14, fig. I). Woodward’s (1895 : 457) statement that the maximum length of his Ph. caudalis is “‘ about 0-13’ is obviously founded on this specimen. The length from the tip of the snout to the hind margins of the hypurals (standard length) is about 109 mm. The total length of P.7582 cannot be measured ; the skull together with the vertebral column to the posterior margin of the hypurals (standard length) is about 133 mm. and the total length may be estimated as about 160 mm. The greatest depth of the body in the holotype is 27 mm. or almost exactly one quarter of the standard length ; the caudal peduncle is crushed dorso-ventrally and its depth cannot therefoie be measured. The length of the head in the holotype cannot be stated with accuracy CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 417 (the opercular bones seem to have been displaced backwards through pressure) but it may be a little greater than the maximum depth of the body. In P.7582 the length of the head is about one quarter of the standard length. The position of the dorsal, ventral and anal fins is somewhat different from that in the species described above. The base of the ventral fin is situated a little nearer to the tip of the snout than to the hind margins of the hypurals, and the distance between the base of the ventrals and the origin of the anal is about 85% of the dis- tance between the bases of the pectorals and the ventrals. The dorsal fin is situated somewhat further back, beginning about midway between the base of the ventrals and the anal origin and ending above the last named fin. The skull is more or less defective in both specimens investigated ; consequently the following description must be considered as provisional. Exoskeletal skull roof A premaxillary is not preserved in either of the specimens available. In the holotype the rostral seems to be well developed, but its shape cannot be determined in any detail. The nasal (Na, Pl. 14, fig. 2 ; Text-figs. 14, 15), preserved only in the holotype, is a broad, almost leaf-like bone with the antero-lateral margin concave, obviously constituting the postero-mesial limit of the anterior nasal opening. The dorsal surface of the nasal is ornamented with ganoin spots. The frontal (Fr, Pl. 14, fig. 2; Text-figs. 14, 15) has a typical shape, with the broad- est part posterior; its anterior, narrower part is hidden below the nasals in the holotype and defective in P.7582, but seems to be short, perhaps comparatively shorter than in Ph. crenulata. In the holotype the suture between the frontals of both sides is straight anteriorly and then makes an S-shaped bend to the right and to the left. The supraorbital (So, Pl. 14, fig. 2; Text-figs. 14, 15) is only partly preserved in the holotype, but seems to have approximately the same general shape as in PA. crenulata. In P.7582 the supraorbital is lacking. A dermosphenotic is not visible in either of the specimens investigated. The shape of the parietal (Pa, P|. 14, fig. 2 ; Text-figs. 14, 15) cannot be determined, but as the dermopterotic is comparatively broad in the mesio-lateral direction, the exposed dorsal surface of the parietal may be smaller than in Ph. crenulata. The dermopterotic (Dpt, Pl. 14, fig. 2; Text-figs. 14, 15) is clearly visible on the right side in the holotype and on the left side in P.7582, showing that its mesio-lateral breadth is comparatively large. The surface of the dermopterotic, like that of the frontal and the parietal, has an ornamentation of small, partly confluent ganoin spots. The lateral part of a bone, possibly the extvascapular, is visible posterior to the dermopterotic in both specimens, but this interpretation is rather uncertain as no traces of a sensory canal or of tubules can be seen. GEOL, II, 8 36§ 418 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Fic. 14. Pholidophoropsis maculata sp. nov. Attempted restoration of head in lateral view. x 4°2. Ant, antorbital; De. Spl, dentary ; Dpt, dermopterotic ; Fy, frontal; Ifo,, [fo,—Ifo;, infra- orbitals 1 and 3-5; Jop, interoperculum; Mz, maxillary; Na, nasal; Op, operculum ; Pa, parietal; Pop, preoperculum; fo, rostral; Sbo, suborbital; Sop, suboperculum ; ant, by, antorbital branch of infraorbital sensory canal ; fc,, anterior division of supramaxillary pit-line ; zfc, infraorbital sensory canal ; ifc. com, ethmoidal commissure ; ovp, post-maxillary pit-line ; poc, preopercular sensory canal. Dermal bones of cheek and opercular apparatus The maxillary (Mx, Pl. 14, fig. 2; Text-fig. 14) is, in its anterior, mesially directed part rather delicate ; its lateral, posteriorly directed part increases in height back- wards. Its height cannot be measured exactly, but it seems to be about the same as in Ph. caudalis. The convexity of the anterior part of the dorsal margin of the maxillary exists, but it is not as pronounced as in Pholidophoroides species. The ventral margin of the maxillary is not exposed in the holotype and only for a short distance anteriorly in P.7582, but here traces of a delicate dentition are visible, indicating that probably the whole ventral margin is denticulate as in Ph. caudalis. The lateral surface of the posteriorly directed part of the maxillary is richly orna- mented ; in the holotype a few longitudinal striations parallel to the dorsal margin recall the striation in Ph. crenulata, but below them there are numerous short ganoin streaks and spots. Only supramaxillary 1 of the swprvamaxillaries is preserved in the holotype, and it is defective (Smx,, Pl. 14, fig. 2); in P.7582 both supramaxillaries are exposed. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 419 Supramaxillary 1 seems to be relatively broader than in Ph. crenulata, its anterior tip being longer, and the striation on its lateral surface not as well marked as in that species, particularly not in P.7582, where only a few ganoin streaks are visible. Supramaxillary 2 (Smxe2, Text-fig. 14), preserved in a defective state, has about the same shape as in Ph. caudalis, but its lateral surface practically lacks ornamentation. As in the Pholidophoroides species, the supramaxillaries do not overlap the dorsal margin of the maxillary. Fic. 15. Pholidophoropsis maculata sp. nov. Attempted restoration of head in dorsal view, the snout flattened. 4:2. Lettering as in Fig. 14. The bones of the infraorbital series are only partly preserved. The antorbital (Ant, Pl. 14, fig. 2; Text-figs. 14, 15), only visible in the holotype, seems to have a broad triangular shape, not as elongate as in Ph. crenulata. In both specimens infraorbital 1 (lachrymal) (Ifo, Pl. 14, fig. 2) is only preserved as fragments without traces of the sensory canal or of its tubules. Infraorbital 2 is lacking. Infraorbital 3 (Ifo, Pl. 14, fig. 2; Text-fig. 14) is, judging from the exposed part, almost semicircular in the holotype ; in P.7582 its dorsal margin is defective, but also in this specimen infraorbital 3 is by no means as broad as in the Pholidophoroides species. Dorsal to infraorbital 3 in the holotype there are two small bones which obviously represent infraorbitals 4 and 5 (Ifo4, Ifo;, Pl. 14, fig. 2; Text-fig. 14). 420 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Posterior to infraorbitals 3-5 there are, in the holotype, some bony plates which may be fragments of a suborbital bone; the lateral surface of these fragments is ornamented with ganoin spots. In P.7582 there are some bone fragments ventral to the dermopterotic ; they have also been interpreted as parts of a suborbital. The preoperculum (Pop, Pl. 14, fig. 2; Text-fig. 14), best preserved in the holotype, is only slightly arched; it is of almost the same height in the middle and in the ventral parts of the bone ; the dorsal limb seems to taper dorsally, but here the outline of the bone is difficult to follow. The middle and dorsal parts of the preoperculum have a surface ornamentation of lustrous ganoin spots ; such spots are lacking on the dorsal part of the ventral limb, probably indicating that this part of the bone is normally covered by the posterior end of the maxillary. In P.7582 the preoperculum is only partly preserved. In the holotype the operculum (Op, Pl. 14, fig. 2) seems, at first sight, to be very low compared with the corresponding bone in Ph. crenulata, but its postero-ventral part is pressed in below the suboperculum and therefore its outline cannot be deter- mined accurately. In P.7582 the operculum is partly crushed, but its ventral part is well exposed, showing that this bone has the same general shape as in the Pholido- phoroides species. The lateral surface of the operculum is ornamented with ganoin spots. The suboperculum (Sop, Pl. 14, fig. 2; Text-fig. 14), preserved only as fragments in both specimens, has the same general shape as in the Pholidophoroides species ; its antero-dorsal process is well developed. Like the operculum its lateral surface is ornamented with ganoin spots. Only small fragments of the interoperculum (Iop, Text-fig. 14) are exposed in the holotype, in P.7582 its posterior part is exposed ; the lateral surface of this part shows a few ganoin spots. Lower jaw Only parts of the dentary (De. Spl, Pl. 14, fig. 2; Text-fig. 14) are exposed and they give little information regarding its shape. Anteriorly the dental part is, however, only gently ascending, its lateral surface is smooth and no teeth can be seen along its margin. The splenial part is separated from the dental by a well-marked ridge, ornamented with ganoin, and its lateral surface is, at least partly, provided with small ganoin spots and short streaks. Sensory canal system of head The sensory canal system of the head can be traced only in part. The supraorbital sensory canal pierces the nasal as an arched canal from its antero- lateral corner to its posterior end. On the left nasal an elevated pore is visible near the anterior end of the canal, other pores or tubules cannot be seen. In the frontal the canal can only be traced imperfectly, but it seems to have the same course as in Ph. crenulata ; only a few short tubules from the mesial side of the canal are visible. The canal probably ends in the parietal, but its presence there cannot be established. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 421 In the rostral there is an ethmoidal commissure (ifc. com, Pl. 14, fig. 2; Text-figs. 14, 15) joining the infraorbital sensory canals of both sides. This canal begins in the antorbital, in which it gives off posteriorly an antorbital branch. As the frag- mentary infraorbital 1 (lachrymal) has no remains of the sensory canal and as infra- orbital 2 is missing, nothing can be said regarding the part of the canal belonging to these bones. In infraorbital 3 of the holotype the canal is weakly arched and postero- ventrally gives off four diverging tubules with small ganoin spots on their surface ; the anterior tubule is about half the length of the others. In specimen P.7582 the canal and its tubules are indistinct. In each of the infraorbitals 4 and 5 the canal gives off a single, posteriorly directed tubule. The further course of the infraorbital sensory canal is unknown. Cephalic division of main lateral line A deep arched groove on the parietal in the holotype may represent the middle pit-line (mp, Pl. 14, fig. 2). On the lateral margin of the dermopterotic a single pore is visible in both specimens, indicating that the infraorbital sensory canal and the cephalic division of the main lateral line pierce this bone near its lateral margin. The further course of the main lateral line is not visible. The preopercular sensory canal pierces the preoperculum in the middle of its length as in the Pholidophoroides species. From its morphologically posterior side it gives off 13 or 14 tubules in the holotype ; the three anterior tubules are short and straight and are followed by four longer, straight tubules ; posterior to the latter there are two long and slender tubules and posterior to them there can be traced four wider tubules, separated by wider interspaces ; three of these tubules are situated dorsal to the posterior angle of the preoperculum. It cannot be stated with accuracy whether one more tubule exists in the dorsalmost part of the preoperculum. In the defective preoperculum of P.7582 only a few tubules are visible. Little can be said about the mandibular sensory canal except that there is a series of short, backwardly directed tubules on the lateral surface of the splenial part of the dentary. Exoskeletal shoulder girdle and squamation The suprascapula and the supracleithrum cannot be identified. A rather strong cleathrum (Cl, Pl. 14, fig. 2) is partly exposed in P.7582 and the dorsal part of the same bone is visible in the holotype. Posterior to the dorsal part of the cleithrum parts ofarather large postcleithrum (Pcl, Pl. 14, fig. 2) with ganoin spots on its lateral surface are visible in both specimens. The squamation is well preserved in the holotype. The scales are large and remark- ably thin with a delicate concentric striation ; their posterior margin is rounded and not serrated. The middle flank scales seem to be somewhat larger than the others. The axillary scales dorsal to the pectoral fin, also present in P.7582, are ornamented with ganoin spots. No lateral line scales can be distinguished. 422 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE Axial skeleton and paired and unpaired fins The axial skeleton is entirely exposed in P.7582, but the separate ring-vertebrae are partly crushed and partly embedded in the rock and cannot be studied in any detail ; their number cannot be stated with accuracy, but it seems to be about 48. The pectoral fin is not entirely exposed in either of the two specimens available and consequently the number of lepidotrichia cannot be given. The first, strong lepidotrichium is fulcrated (Fu, Pl. 14, fig. 2). The ventral fin can only partly be seen in the holotype, but in P.7582 both ventral fins are well exposed and in the right one 14 lepidotrichia can be counted, the two first ones comparatively short ; the third is densely set with fulcra, also visible in the holotype. As already mentioned by Woodward (1895 : 466) the right basi- pterygium is preserved in P. 7582; the basipterygium can also be traced in the holotype. All lepidotrichia in the dorsal fin are missing from the holotype, but parts of its anterior radials and some impressions indicate the position of the fin. In P.7582 the dorsal fin is rather well preserved with at least 13 lepidotrichia visible, the second one with a series of fulcra along its anterior margin ; anterior to the first clearly visible lepidotrichium there are traces of one or two short lepidotrichia. The first radial is V-shaped and composed of two fused radials. Of the anal fin, 9 clearly visible radials are preserved in the holotype. In P.7582 10 lepidotrichia and 5 radials can be seen, but as the anterior lepidotrichia are only exposed basally, it is impossible to say whether fulcra are present or not. The caudal fin may be hemi-heterocercal as in the Pholidophoroides species, but as the fulcra on the dorsal margin are pressed down over the upper part of the caudal peduncle in the holotype and as the caudal fin is practically wanting in P.7582, nothing can be said regarding the true shape of this fin. REMARKS. It is only with some hesitation that I have created a new species for the specimens 43055 and P.7582. In many respects they are very similar to PA. caudalts (to which species Woodward referred specimen 43055), particularly in having thin scales of the same shape as in the holotype of Ph. caudalis. There are, however, many differences. In Ph. maculata the dorsal fin seems to be situated more posterior- ly than in Ph. caudalis, the exoskeletal cranial bones are thicker in Ph. maculata than in Ph. caudalis and, moreover, richly ornamented with ganoin spots, infraorbital 3 and the preoperculum are differently shaped, and the number and position of the tubules belonging to the preopercular sensory canal are also different in the two species. All these facts seem to me to preclude the identification of 43055 and P.7582 as Ph. caudalis. It must, however, be remembered that our present knowledge regarding these species is rather defective, and a definitive solution to this question needs more and better preserved material. At present I find it most convenient to keep the two types apart as separate species. HORIZON AND LOCALITY. Lower Lias; Lyme Regis, Dorset. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 423 III. DISCUSSION (a) The taxonomic relationship between the genera and species within the family Pholidophoridae s. str. and a preliminary diagnosis of the family The description of the exoskeletal cranial bones in Pholidophorus bechei given above differs in some respects from that given by Miss Rayner (1948). The greatest difference concerns the shape of the antorbital and the course of the supraorbital and infraorbital sensory canals in the snout. The conditions described by Miss Rayner, already doubted by Griffith & Patterson (1963 : 32), can hardly be correct. Although I had no specimen with the exoskeletal bones of the snout preserved in their natural mutual position I think that my attempted reconstruction (Text-figs. I, 2) comes nearer to the true conditions. This interpretation seems also to be corroborated by the arrangement of the corresponding structures in the well-preserved Pholido- phoroides crenulata. Moreover, the material at my disposal clearly shows that the nasals do not meet in the median line but are separated by the anterior parts of the frontals. Apart from minor corrections regarding the shape of some bones such as the premaxillary, supraorbital 1, the dermosphenotic and the preoperculum, it has been possible to add some more details regarding the tubules of the sensory canal system. Our present knowledge of Pholidophorus bechei, the type species of the genus Pholidophorus, is of course far from complete but it seems sufficient to serve as the basis for a discussion of taxonomic affinities, 7.e. which other species may be referred to the genus Pholidophorus in a restricted sense and which species may be considered as belonging to other genera. A species which, as far as could be made out, agrees with Ph. bechez in all features of generic value is Ph. latiusculus from the Upper Trias of Seefeld, Tyrol. Unfortunate- ly the shape of the rostral and antorbital bones with their sensory canals are still unknown, but in other respects the agreement between the two species is so close that they must be considered congeneric. Consequently the diagnosis of the genus Pholidophorus s. str. given on p. 356 is based on these two species. The species Ph. caffii from the Upper Trias of Viciarola, North Italy, considered by most authors as a synonym of Ph. latiusculus, is beyond doubt a distinct species. In many respects strikingly similar to Ph. bechet and Ph. latiusculus, it differs from both in some important respects, above all in the shape of the preoperculum and the preopercular sensory canal. The presence of only a single supramaxillary, if not merely an individual anomaly, is moreover a feature separating it from the two Pholidophorus species mentioned. Its reference to the same genus as those two species would widen the conception of the genus Pholidophorus rather excessively. On the other hand I find it at present inadvisable to create a new genus for a species founded on a single, defective specimen, in most respects similar to the true members of that genus, and therefore I have provisionally attributed it to Pholidophorus. The specimen described above as Ph. cf. pusillus represents a species which, as far as can be judged from the partly incomplete material, seems to agree with Ph. bechei and Ph. latiusculus in all characters of generic value. Consequently it may be placed, 424 CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE at least provisionally, in the genus Pholidophorus s. str. Whether it really belongs to Ph. pusillus Agassiz is a question which at present must be left unanswered. The rather numerous but unfortunately badly preserved specimens from the Lower Lias of Lyme Regis, Dorset, which I have described above under the species name dorsetensis, were referred to Ph. caudalis by Woodward (1895). They have, however, nothing to do with the true Ph. caudalis but show, as far as known, many similarities with Ph. bechei and Ph. latiusculus. The preoperculum, one of the most characteristic bones of those preserved, shows the same notch on its posterior margin as in the two species mentioned and the arrangement of the tubules belonging to the preopercular sensory canal is about the same as in those species, but the canal itself runs still nearer to the anterior margin of the bone, about as in Leftolepis. Further, the extension of the two supramaxillaries over almost the whole dorsal margin of the maxillary recalls a condition similar to that in Leftolepis, and the squamation consists of comparatively thin cycloid scales. The lower jaw is, however, only gently ascending anteriorly, not abruptly as in Leptolepis, and the nasals are com- paratively large. In many respects these specimens are thus similar to the members of the genus Pholidophorus s. str., in others to Leptolepis, but they can, in my opinion, be placed neither in Pholidophorus nor in Leptolepis. Consequently I have pro- posed for the species dorsetensis a new genus, Pholidolepis, the preliminary diagnosis of which is given above (p. 387). The genus Pholidophoroides was created by Woodward for the species Pholidophorus crenulatus and founded mainly on some scale characteristics and on the stoutness of the maxillary (see p. 393). Re-examination of the type species of the genus has revealed many other characteristics, especially a high number of infraorbitals (more than five), a rather broad preoperculum with the preopercular sensory canal running along the middle of the bone and with rather short and straight tubules in its ventral limb, and the presence of only a single supraorbital bone. Consequently, it seems to me that Woodward was quite correct in creating the genus Pholidophoroides. A further species showing, in general, the same characteristics is Pholidophorus limbatus, which therefore must be considered a second species of the genus. Woodward thought that his Ph. caudalis probably belonged to his new genus Pholidophoroides. As far as the partly defective material makes comparison possible, it is quite obvious that caudalis is, in many respects, very similar to other members of that genus ; above all the preoperculum, the preopercular sensory canal and its tubules have the same characteristic shape in caudalis as in Pholidophoroides. There seems, moreover, to be only a single supraorbital. The exact number of infraorbitals is, unfortunately, indeterminable but nothing indicates that their number is more than five ; if this be so there is, in this respect, a difference of taxonomic importance between caudalis and the two Pholidophoroides species. The most striking difference between them is, however, that the squamation in caudalis consists of rather thin cycloid scales, in Pholidophoroides of thick ganoin scales. This difference in squama- tion prevents, in my opinion, caudalis being placed in the genus Pholidophoroides and because of this I have created for it the new genus Pholidophoropsis. CERTAIN DRIASSE®C AND LIASSIC PHOLIDOPHORIDAE 425 As a second Pholidophoropsis species I have, albeit with some hesitation, described two imperfect specimens under the new trivial name maculata. The reasons for the creation of this species are discussed on p. 422. In this connexion it must, however, be remembered that differences in the ganoin covering of the exoskeletal cranial bones may be dependent on ontogenetic age, as described for Ph. bechez, and that the maculata specimens may merely represent older individuals of caudalis. But as the largest caudalis specimen is 120 mm. in length, the smallest maculata 127 mm., the difference in respect of the absence or presence of ganoin spots should not be as pronounced as it actually is if the ganoin spots are a feature characterizing only the largest specimens. The genera and species considered here are more or less closely related and con- stitute a family Pholidophoridae in a restricted sense ; a possible exception is the genus Pholidolepis, the systematic position of which cannot be determined until we have a better knowledge of the genus Leptolepis. This discussion must be deferred until a revision of that genus, now in preparation, is completed. It is of course very probable that many other species will be added to this family as soon as the members of the old genus Pholidophorus become better known. Griffith & Patterson (1963) have in their valuable monograph on Ichthyokentema purbeckensis compared this genus with Pholidophorus as represented by Ph. bechei and Ph. similis Woodward. For this reason I must briefly mention the last named species although I have not yet had the opportunity of studying it myself. According to the description of Ph. similis given by de Saint-Seine (1949) there are so many dissimilarities between that species and the members of Pholidophorus s. str. described above, that it cannot be maintained in this genus. I need mention only the curious shape of the two postorbital infraorbitals (the dorsal one obviously corresponding both to infraorbitals 4 and 5 and to the dermosphenotic in Pholidophorus), the different shape of the maxillary, the suborbital, the preoperculum, the outline of the lower jaw and so on. All these features are atypical for the genus Pholidophorus as represented by Ph. becher and Ph. latiusculus and most of them do not occur in the other pholidophorids described above. However, as I have no personal informa- tion regarding Ph. similis I shall confine myself to these comments. Griffith & Patterson (1963) have convincingly shown that Ichthyokentema pur- beckensis represents a family of its own, Ichthyokentemidae, and have given (p. 6) an exhaustive diagnosis of the new family. For the family Pholidophoridae as understood here no such diagnosis can be given at present because of our imperfect knowledge, particularly regarding the endocranium and many other details. The following diagnosis must therefore be considered a preliminary one. Family PHOLIDOPHORIDAE s. str. PRELIMINARY DIAGNOSIS. Fusiform Pholidophoriformes of small to medium size ; bones of head and scales with or without ganoin covering ; rostral toothless, not separating premaxillaries ; nasals not in contact in the mid-line ; dermosphenotic not elongated dorso-ventrally ; five to seven infraorbitals, infraorbital 3 the largest ; one to two supraorbitals ; two supramaxillaries (except in Ph. (?) caffii?) ; mandible 426 CERTAIN TRIASSIC AND ELASSIC PHOLIDOPHORIDAE gently ascending, with dentary (dentalo-splenial), articular and angular (angulo- supra-angular) ; dentition on premaxillary, maxillary and dentary, only weakly developed on two last named bones; single gular; preoperculum more or less angular; suture between operculum and suboperculum oblique; _ supraorbital sensory canal ending in parietal and having no anastomosis with infraorbital sensory canal; anterior and middle pit-lines on parietal, middle pit-line extending on to dermopterotic ; preopercular sensory canal with moderate to rather long tubuli ; preoperculum with anterior division of supramaxillary (horizontal) and postmaxillary (vertical) pit-lines, former sometimes with anterior portion on infraorbital 3; oral pit-line present on angular; vertebrae, as far as known, with annular centra ; fulcra present on all fins (except in Pholidolepis) ; scales more or less rhomboid with smooth or pectinated hind edge, or cycloid with fine concentric striation ; some longi- tudinal rows of deepened scales on the flank. (b) Some phylogenetic aspects of the genera and species within the family Pholidophondae s. str. The genera and species of the family Pholidophoridae treated above are, as pre- viously stated, more or less closely related. Their phylogenetic relationship is of course difficult to discuss on the basis of our present knowledge, but some comments may be appropriate. The pholidophorids are generally considered to be advanced holosteans, probably the forerunners of the teleostean stage of development within the actinopterygians. One of the most striking differences between the holostean and teleostean stages is the transformation of the thick, ganoin-covered scales into thin cycloid scales. This trend of evolution is clearly demonstrated in the species now under consideration. The Upper Triassic species Pholidophorus latiusculus, Ph. cf. pusillus and Ph. (?) caffia as well as the Lower Liassic Pholidophoroides limbata have thick scales, in Ph. crenulata the anterior body scales are thinner and show, as stated by Woodward (1941), a concentric striation, and in Pholidophorus beche: the younger individuals show a weaker ganoin covering of the exoskeletal cranial bones and scales than do the adult specimens. In Pholidophoropsis maculata the ganoin covering on the exoskeletal cranial bones is confined to small scattered spots and streaks, whilst of the thin cycloid body scales, only some axillary scales carry a few ganoin spots ; thicker ganoin scales exist only at the caudal fin base. In Ph. caudalis no ganoin spots are visible on the exoskeletal cranial bones or axillary scales ; thicker scales occur only at the posterior end of the caudal peduncle as in Ph. maculata. In Pholidolepis dorsetensis a ganoin ornamentation is present at least on the maxillary, the supramaxillaries and the lower jaw, whereas the body scales are thin with a delicate concentric striation. This may indicate that these pholidophorids are at that phylogenetic stage at which the reduction of the ganoin covering of the scales is taking place, and further, that this reduction began in the anterior part of the body and that the posterior part of the caudal peduncle retained the ganoin scales longer. These consecutive stages of scale reduction indicate, however, certainly not a straight phylogenetic series, which is evident from other facts. CERTAIN TRIASSIC AND LIASSIC PHOLIDOPHORIDAE 427 The shape of the preoperculum together with the course of the preopercular sensory canal and the pattern of its tubules seem to be very useful indicators of the relationship between various taxonomic units. In the family Pholidophoridae s. str. the most primitive type is represented by the broad and little curved preoper- culum in Ph. (?) caffii ; the preopercular sensory canal runs in the dorsal limb of the bone somewhat nearer to its posterior than to its anterior margin, in the ventral limb it runs in about the middle ofits length. The tubules given off from the posterior side of the canal are comparatively few, wide and rather short. From this rather primitive preopercular type one can easily derive that of Ph. cf. pusillus ; the general shape of the preoperculum is about the same, but the preopercular sensory canal runs nearer to the middle of the bone and the tubules are more numerous in the ventral part of the bone and are, moreover, longer. A further stage is represented by Ph. latiusculus. Here, the preoperculum is more differentiated in that its dorsal and ventral parts are of a different shape and its posterior margin possesses a shallow notch ; the preopercular sensory canal runs a little nearer to the anterior than to the posterior margin of the bone and its tubules are more numerous; in the ventral part of the bone the tubules are rather long and slightly curved. The preoperculum of Ph, bechet has a fairly characteristic shape ; its anterior margin is slightly concave, its posterior margin is notched and its postero-ventral corner projects backwards. The number of tubules belonging to the preopercular sensory canal is even greater than in Ph. latiusculus and those in the ventral limb of the bone are still longer. The preoperculum of Pholidolepis dorsetensis recalls many of the characteristics of Ph. bechei, but the preopercular sensory canal runs nearer to its deeply concave anterior margin (obviously due to a reduction of the anterior part of the bone) and the tubuli are fewer and relatively shorter. In my opinion the preopercula in the three species, Ph. cf. pusillus, Ph. latiusculus and Ph. bechei, constitute a phylogenetic series which can be derived from the preoperculum in Ph. (?) caffii and from which series the pre- operculum in Pholidolepis can also be derived. In the genera Pholidophoroides and Pholidophoropsis the preoperculum has a somewhat different shape and the tubules of the preopercular sensory canal are straight and rather short, the longest being situated in the middle part of the bone, not in the ventral half. In all the species hitherto known the shape of the preoper- culum is surprisingly uniform. It cannot be derived from Ph. (?) caffit as directly as can that of Ph. cf. pusillus, but probably from a still more primitive type. The genus Pholidophoropsis cannot, however, be directly derived from Pholidophorovdes ; the members of the latter genus have a relatively high number of infraorbitals, seven against the general number of five ; in Pholidophoropsis the number of infra- orbitals cannot be stated with accuracy, but there is nothing to indicate, at least not in Ph. maculata, that it exceeds five. The structure of the preoperculum thus indicates that within the family Pholido- phoridae in the restricted sense given above two diverging lines of evolution may be discerned, the one comprising the genus Pholidophorus s. str., probably giving rise to the genus Pholidolepis, perhaps leading on to the leptolepids, the other comprising the genera Pholidophoroides and Pholidophoropsis. In both these lines a reduction of 428 CERTAIN TRIASSIC AND LLASSIC PHOLIDOPHORT DAE the ganoin covering of the exoskeletal cranial bones and body scales has started independently (as probably in many other lines of evolution), in the Pholidophorus series from scales with a smooth posterior margin, in the Pholidophorotdes series from scales with a pectinated or serrated hind margin. The general phylogenetic evolution can of course be elucidated from other features also, but the material investigated is too fragmentary to illustrate this in great detail. Some indications may perhaps be worth noting. The sensory pit-lines seem to show a tendency towards a successively more super- ficial position and consequently their markings on the underlying bones become successively less pronounced. The supramaxillary (horizontal) and postmaxillary V (pS SS SS SS SS SS 2S FS =i ” | a) Size distribution of Galliaecytheridea postrotunda 15 ite) 5 3 30 50 80 = Length of specimens in mm. SS ES C15 Size distribution of Macrodentina (M)_ ruqulata eS 5 35 (o} uw O} 40 50 60 70 80 Length of specimens in mm, ee 452 BRITISH JURASSIC AND CRETACEOUS OSTRACODA Dimensions in mm. Proportions SS SS SS SSS IL, H I L H I Specimen Number Left valve. . 0°73 O41 — I-00 0°56 — To. 2166 Carapace : . 0°66 0:39 0:29 I'00 0°59 0:44 Io. 2164 Carapace : 2 @oG{e) @sAe) svn I-00 0:58 0:42 To. 2165 REMARKS. These specimens correspond to the holotype, the size distribution as indicated in Text-figure 31, showing a slight difference in length between males and females. Family BRACHYCYTHERIDAE Puri 1954 Genus MACRODENTINA Martin 1940 Subgenus MACRODENTINA Martin 1940 Macrodentina (Macrodentina) rugulata (Jones) (Pl. 2, figs. g—-12) 1885 Cythere vetivugata var. rugulata Jones : 350, pl. 9, figs. 17-20. 1941 Cythere retiyvugata Jones var. yugulata Jones; Anderson : 373, pl. 18, fig. 1. 1956 Dictyocythere (Rhysocythere) rugulata (Jones) Sylvester-Bradley : 18, pl. 4, figs. 1, 2, 5-10, I2-I5. 1958 Macrodentina (Macrodentina) rugulata (Jones) Malz: 18, pl. 6, figs. 83-86. MATERIAL. Thirty-six valves from Bed P.Q. 12 at Poxwell Quarry. Dimensions in mm. Proportions aN SS IL H IL, H Specimen Number Right valve . 5 OWI OPANS I-00 0:63 Io. 2160 Left valve . LOW LOA y I-00 0-66 Io. 2159 Right valve (juv.) . 0:60 0°35 I-00 0-58 Io. 2161 Right valve . On 47 On27, 1:00 0°57 Io. 2162 REMARKS. Specimens of Macrodentina (M.) rugulata are only found in the Roach of Poxwell Quarry. These specimens (see Text-fig. I) are much smaller than those described from the Aylesbury district and Portesham Quarry (Barker 1964a, b). The largest forms have a paramphidont hinge and are therefore thought to be adult. The reasons for this reduction in size must be left unexplained until more specimens have been obtained from the Portland Stone of other localities. Macrodentina (Macrodentina) transiens (Jones) (Pl. 2, figs. 13-15) 1885 Cythere transiens Jones : 349, pl. 9, figs. 13-16. 1956 WDictyocythere (Rhysocythere) tvansiens (Jones) Sylvester-Bradley : 19, pl. 3, figs. 11-13. 1958 Macrodentina (Macrodentina) transiens (Jones) ; Malz: 17, pl. 6, figs. 81, 82. MATERIAL. Fourteen valves and sixty-seven carapaces from various beds on West Weare Cliff. One hundred and seventeen valves from various beds at Poxwell Quarry. BRITISH JURASSIC AND CRETACEOUS OSTRACODA 453 A carapace from Bed P.Q. 12 has the following dimensions: length 0-58 mm., height 0-35 mm., inflation 0-33 mm. These specimens correspond in all details to those described by Sylvester-Bradley 1956. Subgenus DICTYOCY THERE Sylvester-Bradley 1956 Macrodentina (Dictyocythere) retirugata (Jones) (Pl. 1, figs. 1-8) 1885 Cythere retivugata Jones : 350, pl. 9, figs. 21, 23. 1885 Cytheve retivugata var. textilis Jones : 350, pl. 19, fig. 24. 1941 Cythere retivugata var. textilis Jones; Anderson : 374, pl. 18, fig. 3. 1941 Cythere retivugata var. decovata Anderson : 374, pl. 18, fig. 4. 1956 WDictyocythere (Dictyocythere) vetiyugata (Jones) Sylvester-Bradley : 15, pl. 3, figs. 7-10, joe Zip sake, BH Zip amie aos 1077, 1956 Dictyocythere (Dictyocythere) decovata (Anderson) Sylvester-Bradley : 17, pl. 3, fig. I. 1958 Macrodentina (Dictyocythere) decovata (Anderson) Malz ; 25, pl. 6, figs. 87, 88. 1958 Macrodentina (Dictyocythere) textilis (Jones) Malz: 26, pl. 6, figs. 80-91. MATERIAL. Five valves from Bed P.Q. 12 at Poxwell Quarry. Sixty-one valves and three carapaces from various beds at Hounstout Cliff. One hundred and fifty- eight valves and ten carapaces from various beds at Friar Waddon. Thirty-eight valves and four carapaces from Bed W.W. 9 at West Weare Cliff. Dimensions in mm. Proportions it H le L H I Specimen Number Right valve . 0°70 0°45 — 100) O04) Io. 2153 Left valve OF7ONNOr 52.5 —— OOO) O97 == — Carapace _ O29, OoMG) Ovi UG) Orly, Crow lo. 2152 Carapace 5 Orsi Oox“vi7 Oovin I-00 0:58 O-51 To. 2154 REMARKS. These specimens are similar to those from the Aylesbury district except that they are smaller. They are presumably adult since they have the adult hinge. Subgenus POLYDENTINA Malz 1958 Macrodentina (Polydentina) rudis Malz (Pl. 6, figs. 1-8) 1958 Macrodentina (Polydentina) yudis Malz : 31, pl. 4, figs. 57-64. MATERIAL. Three carapaces and thirty-seven valves from Bed H.7, three cara- paces from Bed H.8 at Hounstout Cliff and four valves from Bed W.W.g at West Weare Cliff. DESCRIPTION. Carapace somewhat triangular in lateral outline with a more or less straight dorsal margin which slopes down towards the posterior. Ventral margin very slightly convex. Anterior smoothly curved, meeting the dorsal margin at a 454 BRITISH JURASSIC AND CRETACEOUS OSTRACODA slight angle where there is a slight swelling formed by the anterior hinge teeth and socket. Posterior is narrower than the anterior and is bluntly rounded. The valves are inturned slightly on the dorsal margin. Ventral margin concave to formaconcave ventral surface. Greatest inflation to posterior. In dorsal view the sides are rather flat and the ends are slightly pointed. Left valve larger than right and overreach- ing it all round, most strongly to anterior and posterior. Males are longer than the females. The lateral surfaces are reticulate, the dominant pattern being concentric round the margins and more or less vertical in the central region. Normal pore canals irregularly spaced over the carapace. Radial pore canals not observed. Muscle scars not seen. Hinge paramphidont. There are slight anterior and postero-ventral vestibules. The line of concrescence parallels the outer margin. REMARKS. This is the first recorded occurrence of Macrodentina (Polydentina) yudis in England. Family CYTHERURIDAE Miiller 1894 Genus ORTHONOTACYTHERE Alexander 1933 Orthonotacythere rimosa Martin (Pits) Hes. 7G ek, 12) 1940 Ovthonotacythere vrimosa Martin : 335, pl. 6, figs. 84-86 1961 Orvthonotocythere cf. yrimosa Martin; Martin: 117, pl. 14, fig. 21a—c. MATERIAL. Three valves from Bed P.Q. 7 and two carapaces and twenty valves from Bed P.Q. 12 at Poxwell Quarry. Twelve carapaces and fifty-eight valves from various beds on Hounstout Cliff. Measurements of two specimens are : Dimensions in mm. Proportions Iu, H I iE, H I Specimen Number Carapace é . 0°50 0°27 0:29 1:00 0:54 0:58 lo. 2754 Right valve 0°54 0:33 — 1:00 0°61 — Io. 2753 fragment REMARKS. These specimens correspond in all details to those described from the Aylesbury district but the valves are badly encrusted. Orthonotacythere levis sp. nov. (Pl. 5, figs. 10, 13, 14) DiaGnosis. Orthonotacythere with rounded caudal region and well rounded, smooth lateral surfaces. HoLotyre. Io. 2755. Left valve, from Poxwell Quarry. Length 0-45 mm. height 0-25 mm. BRITISH JURASSIC AND CRETACEOUS OSTRACODA 455 PARATYPE. Io. 2756, from Poxwell Quarry. OTHER MATERIAL. Three right and two left valves from Bed P.Q.12 at Poxwell Quarry. Two right and four left valves from Bed H.7 at Hounstout Cliff. DescriPTion. Lateral outline subtrapezoidal with dorsal and ventral margins subparallel. Anterior margin smoothly and evenly rounded, with the posterior margin obliquely rounded, sharpest towards the dorsal margin. Greatest height and inflation just posterior of centre. Lateral surface covered by many small, faint puncta. Valves swollen towards the ventral margins, particularly near the postero ventral margin. Radial and normal pore canals and muscle scars are not seen. Hinge antimerodont. Inner margin and line of concrescence coincide. REMARKS. This species is distinguished from others of the genus Ovthonotacythere by its more rounded outline and smooth lateral surfaces. Orthonotacythere elongata sp. nov. (Pl. 6, figs. g—11) DiaGnosis. Orthonotacythere with ventral margin sloping upwards to posterior dorsal corner in lateral view. Adults with reticulate ornamentation, juveniles also with spines. HoLotyre. Io. 2760. Left valve from Poxwell Quarry Bed P.Q. 12. Length 0-45 mm., height 0-25 mm. OTHER MATERIAL. One hundred and eight valvesand twenty-five carapaces, adults and juveniles, from various beds at Hounstout Cliff. DESCRIPTION. Carapace somewhat triangular in outline, the dorsal margin being long and straight. Ventral margin slightly irregularly convex and closing with the dorsal margin forming a blunt point. The anterior margin is smoothly rounded but forms a slight angle with the dorsal margin. The greatest inflation is centrally since the dorsal outline is sub rectangular and the greatest height is near the anterior. The lateral surfaces are reticulate with a major ridge running parallel to the dorsal margin just ventral of the centre of the valve and ending before reaching the anterior and posterior borders. Towards the posterior this ridge becomes stronger to form a slight ala and may be at the position of greatest inflation. There is a central deep narrow sulcus from the dorsal margin almost to the venter of each valve. The internals of the valves are not seen but the hinge consists of notched terminal teeth with straight connecting crenulate socket. The line of concrescence appears to be parallel to the inner margin. REMARKS. This species differs from others of Orthonotacythere by its characteristic shape, see Pl. 6. Proportions of adults and juveniles are given in the Text-figure on p. 451. 456 BRITISH JURASSIC AND CRETACEOUS OSTRACODA Genus PROCYTHEROPTERON Ljubimova 1955 Procytheropteron bicosta sp. nov. (Pl. 5, figs. 1-6, 9) DiaGnosis. etosrt $2243 3: Sasstststsasssietssssese Siioees 2: plete sateen st535's2531 ats barsrs seis t pa FeeEpeseitssatetetett ss aaeseass :- SHIT eeretes $2 rsesef SsHBetiieieietste4 efi FS FIH. 3 Base bez pseppesstiiecttet see 228s te: 3 Sti telat ser eterereseeeesetee Ea ee saee cir ape esas eee bsSstorersesy Ssapiditeticsssratecs: opi anecsspigiese: Siete 2 Sresess oteseaee Sores Frees ore. ot: i lelsietaccies sipttesteiatseciactieeess pa tees eieat Sete SIF Siete sisi er, iHieesseiieprerere 32533 i 28 absSieesstisteeacyestaesisas is trssrsstes Saini a Sts aan Hs 3 iH a eitiaits f ee teperh ite hay Ss Sb + 2 ey. ret patie sty HBA ESBS Sette Sespareesesepeseeets sere seer B hashes tes tptete srr tste tet a0 st rt atareasy ipaishtes stots EpaseeStsenes tate caeeance: MARES sa tedettsesesye rs pistaties ste ritst Ssscacsese ssilasiseteceks Peter: picartes pestis: 32) E5332 i te serenade sitetts gists set} sraeefain/Sieio ie: Tersisiiisistersspecvstet ete. ae 553523 Pere rere tesetre je teve seceresrigeesesetirettser afoes Srssny tere srpresp press: irs etatssss pete