xy past He os re ee) NS DRI RNR baat NE EY, Nenana et a if Ege. ra See. * Iie & de rote. wile vg NER Pie. ERS Pay. yey ‘ hglean Sars a ri : : 7 RSA ted tn te 4 arya Partha ik Me 4 toads fi Vina ee cult te Ang Aha fae ta ‘ Saye ty i yaad , J Hore Prostngn thes oe Y pana eve: carer fraayphed ese ere z se : ez mined be tet etl ee 4 4 vaeenaea toga cs a0 Hg a we Wy fe Corer od au 4) a ae ee iat ean Pat’ 238 , tata pele Riya tactee pena iy ie ake ta anya tote Gegrietide Sea Egy PN: wh ir SRO snT 4s Pi Reet ane ta Pies ia ei i a i" UTiaedis Spt Styl hatin aereda anterior soe ered bd a ein ; teal eh rhe wh srg) asec ob aU gt ms, ft : sy LOR Br b eta y ie Reenoe peace Pinna ey thy a Ssicnsy noice ; Vv af red ae a z S VG 2 - 2 = : SS \ w _— » — LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS Saluvugi7_LIBRAR ee ” rape ” z= ” Zz ~~ ia oe = z= 7 Z Hf, z —{ = ' 5 NWN S 5 fof 2 z 5 N = x SA 9° - Oo Gf? a O x " = Wy 2 = =F i Sa ea l= = E | 3 eee ve = > = > = x L a 2 ” Fa ” = ” a NOILNLILSNI _ LIBRARIES SMITHSONIAN INSTITUTION NOILNG tf z= hd > w”) = (dp) J 7) ssa ” Be a = x ra ao an ~ Le a oc ea = eG sd = seh < vc 4 : oc S oO = or a0) — . a = m- — am ae O aw =e O = Oo —_ 29 z ary ; za J Zz al LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS SAZ!1YVUaIT LIBRA « = Ss z i z i | ow 2 ae 2 m7 ° o 3 a = 70 Ee rs) > Ey oa - > * ns af = E = F = - = tn = o Zz on Zz o i NOILNLILSNI SAIYVYUEIT LIBRARIES NOILNS wn ad w Zz one ” z 7) F = x = ee = < z = ah a = ie \ = Zz =| + ma = >" = * 2 ov ca 2 _NVINOSHLINS S3I1YVYEIT LIBRARIES INSTITUTION N iy = 2 “a S ” 3 “4 uv = a = oa oF ra © Ls za of < re ee, < iia * = . te > = a e ty - a co m z z m Z ! | |, NOILALILSNI_ S3IYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILN wn aes z ied Ww = pee wo Ss, MITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3iu¥vVudIT_LIBRARIES ~*~ INSTITUT Saiuvy INSTITUT saluvY INSTITUT NVINOSHLINS S31uVWa NVINOSHLIWS SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSE SMITHSONIAN NVINOSHLIWS WINOSHLINS S3ldYvudil LIBRARIES LIBRARIES SMITHSONIAN NOILNLILSNI LTBRARIES NOILNLILSNI NOILNLILSNI INSTITUTION NOILNLILSNI NVINOSHLINS S3!IUVYGIT LIBRARIES SMITHSC SaIUVYUSIT LIBRARIES S3tuYvuddl) INSTITUTION INSTITUTION INSTITUTION IVINOSHLINS S31YVYEIT LIBRARIES NOILALILSNI NWINOSE NVINOSHLINS S31YVualT rm) z = as Nae = g ye a 2 ONYX 2 a hi a mae CREWS 5 = 1 fff Oo y A 2 56 4,4 ae Or Pa \. aS oO Gos fe ae > = Ss > 77 = | Zz ”) . 7p) = es wn MITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3!1uVvydi Peta ayes peMivnec = > ey oe Zz set ms ” Ws es n a i a AN NS) ee za = oc A oc ee 7 3 B a No : © ae 2 oh 2 Soe = a WWINOSHLIWS _ S34 (uVvudad UF LIBRARI ee Ce AN ST On ~NOILALILSN! NVINOSH ~~ Pee “} Y oO a O yy, aa oO INN > : Ee Giiy, 2 2 SX. Sa IQ = ss = Ub) Es is N ies a Feige ee Fe Sy / _ (op) pee (op) eae MITHSONIAN SMITHS INSTITUTION NOILNLILSNI NVINOSHLINS Saiuvugia_LIBRARIES NVINOSHLINS S3!1Y¥vudl NVINOSHLINS SMITHSONIAN NYINOSHLIWS SMITHSONIAN LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI_ NVINOSE Fa = Ca O ” ae Eas = uv”) NX, = = S f= z LON Re S . x nr w S fie 4 x | N a am. a faa) = \ oS sabi oO me aa oO a , - z J S aa Fae SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYUVWYHYE!IT LIBRARIES SMITHS¢ 7 z iG z us = Va rie = w = ow LY, , wD > a) as 3 = Wy > E ay E 5 a Yi Ms = m x m of oan (op) ‘ = (ep) a NWINOSHLINS S3IYVYAIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSN! NVINOSE . ig =, - : VOLUME 80 = AUGUST 1980 ISSN 0303-2515 ve a ¥ wi A: oat ae (3) | ey , aoe bem ey | te “Us F) ee iy hs eae oF io ' 7: Sa ie ” Me r " {i 7’ ~ ‘OF THE SOUTH AFRIC ~~” MUSEUM CAPE TOWN INSTRUCTIONS TO AUTHORS (a) Centred masthead to consist of Title: informative but concise, without abbreviations and not including the names of new genera or species Author’s(s’) name(s) Address(es) of author(s) (institution where work was carried out) Number of illustrations (figures, enumerated maps and tables, in this order) (b) Abstract of not more than 200 words, intelligible to the reader without reference to the text (c) Table of contents giving hierarchy of headings and subheadings (d) Jntroduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (gz) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of legends for illustrations should be typed separately, their positions indicated in the text. All pages should be numbered consecutively. Major headings of the paper are centred capitals; first subheadings are shouldered small capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further subdivisions should be avoided, as also enumeration (never roman numerals) of headings and abbreviations. Footnotes should be avoided unless they are short and essential. Only generic and specific names should be underlined to indicate italics; ail other marking up should be left to editor and publisher. 4. ILLUSTRATIONS should be reducible to a size not exceeding 12 < 18 cm (19 cm including legend); the reduction or enlargement required should be indicated; originals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and final size. A metric scale should appear with all illustrations, otherwise magnification or reduction should be given in the legend; if the latter, then the final reduction or enlargement should be taken into consideration. All illustrations, whether line drawings or photographs, should be termed figures (plates are not printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items of composite figures should be designated by capital letters; lettering of figures is not set in type and should be in lower-case letters. The number of the figure should be lightly marked in pencil on the back of each illustration. 5. REFERENCES cited in text and synonymies should all be included in the list at the end of the paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable): (a) Author’s name and year of publication given in text, e.g.: “Smith (1969) describes .. .’ ‘Smith (1969: 36, fig. 16) describes...’ “As described (Smith 1969a, 1969b; Jones 1971)’ “As described (Haughton & Broom 1927)...’ ‘As described (Haughton et a/. 1927)...’ Note: no comma separating name and year Pagination indicated by colon, not p. names of joint authors connected by ampersand et al. in text for more than two joint authors, but names of all authors given in list of references. (b) Full references at the end of the paper, arranged alphabetically by names, chronologically within each name, with suffixes a, b, etc. to the year for more than one paper by the same author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a). For books give title in italics, edition, volume number, place of publication, publisher. For journal article give title of article, title of journal in italics (abbreviated according to the World list o, scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part number (only if independently paged) in parentheses, pagination (first and last pages of article). Examples (note capitalization and punctuation) BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan. FIsCHER, P.-H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140. FiscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs Zool. exp. gén. 74: 627-634. ‘ Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Ann, Mag. nat. Hist. (13) 2: 309-320. Konn, A. J. 19605. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bull. Bingham oceanogr. Coll. 17 (4): 1-51. THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270. (continued inside back cover) ANNALS OF THE SOUTH AFRICAN MUSEUM VOLUME 80 ANNALS OF THE SOUTH AFRICAN MUSEUM ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 80 Band August 1980 Augustus CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA THE AMMONITE SUBFAMILY TEXANITINAE COLLIGNON, 1948 By HERBERT CHRISTIAN KLINGER & WILLIAM JAMES KENNEDY Cape Town Kaapstad The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular intervals as material becomes available Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na gelang van die beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000 OUT OF PRINT/UIT DRUK 1, 2(1-3, 5-8), 3(1-2, 4-5, 8, t.-p.i.), 5(1-3, 5, 7-9), 6(1, t.-p.i.), 711-4), 8, 9(1-2, 7), 10(1-3), 11(1-2, 5, 7, t.—p.i.), 15(4—5), 24(2), 27, 31(1-3), 32(5), 33 Copyright enquiries to the South African Museum Kopieregnavrae aan die Suid-Afrikaanse Museum ISBN 0 908407 97 1 Printed in South Africa by In Suid-Afrika gedruk deur The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk., Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA THE AMMONITE SUBFAMILY TEXANITINAE COLLIGNON, 1948 By HERBERT CHRISTIAN KLINGER South African Museum, Cape Town & | WILLIAM JAMES KENNEDY Geological Collections, University Museum, Oxford (With 269 figures) [MS. accepted 14 September 1979] ABSTRACT Ammonites referred to the subfamily Texanitinae Collignon, 1948, are the dominant macrofaunal element in rocks of Upper Coniacian to Middle Campanian age in Zululand, Natal, and Pondoland, and are of extreme interest as far as intraspecific variation, evolutionary trends, apparent endemism and gigantism are concerned. The subfamily is represented by thirty-seven species belonging to the following genera and subgenera: Protexanites s.s., Para- texanites, Plesiotexanites, Reginaites, Texanites, Submortoniceras, Bevahites, Menabites s.s., M. (Australiella) and M. (Delawarella). Subtle differences exist between the Pondoland and Zululand specimens, and these are here separated at subspecific level. Thirteen new species and three new subspecies are described; these are—new species: Paratexanites australis, P. pseudotricarinatum, Plesiotexanites collignoniforme, P. matsumotoi, Reginaites reymenti, R. zulu, Texanites collignoni, T. vanhoepeni, T. postvanhoepeni, T. umzambiense, Bevahites ? enigma, Menabites (Delawarella) gigas, M. (Delawarella) nibelae; new subspecies: Texanites presoutoni presoutoni, T. presoutoni natalense, T. soutoni natalense; and Reginaites? sp. nov.? cf. Plesiotexanites stangeri (Baily). Evolutionary trends are discussed within the Texanitinae as shown by the South African faunas. The geographic distribution of the subfamily shows it to be highly endemic and prolific towards the end of the Santonian and during the Lower and Middle Campanian, shortly before its demise. CONTENTS Introduction 1 Location of specimens 3 Field localities : ; : 3 Dimensions of specimens - i : : : ; 4 3 Suture terminology 3 Annotation of Grime nation 3 Systematic palaeontology : : 4 Evolution and evolutionary frends in Texanitinae . 348 Palaeobiogeography . : : : ; : : . 350 Acknowledgements . ; : : ‘ : ' yee DS) References . : : : : ; : : ; 5 254s INTRODUCTION Ammonites referred to the subfamily Texanitinae Collignon, 1948, are the dominant macrofaunal element in the Upper Coniacian to Middle Cam- panian sediments of southern Africa, both in terms of numbers of individuals 1 Ann. S. Afr. Mus. 80, 1980: 1-357, 269 figs. 2 ANNALS OF THE SOUTH AFRICAN MUSEUM and species, and are of extreme interest as far as intraspecific variation, evolu- tionary trends, apparent endemism and gigantic growth are concerned. Further- more, identification of these faunas allows for detailed stratigraphic correlation with biostratigraphically equivalent strata of Madagascar, and to a very limited extent those of western Europe, the Gulf Coast regions of North America, and Hokkaido, Japan. The authors were struck by the bewildering array of intraspecific variation in the texanitine material at their disposal. Features such as density of ribbing and tuberculation, presence or absence of bifurcations or intercalations, and even relative proportions, which, by current ammonoid classification standards, would be of specific or even generic value, were found to be of relatively minor taxonomic significance if viewed in their proper stratigraphic context. This does not imply that the authors negate the possibility of more than one species being found at the same stratigraphic level, but rather emphasizes that specific identity should be viewed in terms of consanguinity of characters rather than typo- logically. In this respect, the authors were hampered somewhat because a large proportion of the available material is from older collections, primarily that of the late Dr E. C. N. van Hoepen, which lack precise stratigraphic data. Detailed collecting by the authors during several field seasons alleviated this shortcoming to some degree. In addition to the extreme intraspecific variation, a biogeographic separa- tion into a northern Zululand fauna frequenting the deeper water facies of the St Lucia Formation of the False Bay region, and a southern fauna frequenting the shallower water of the transgressive Umzamba Formation of Pondoland, Transkei, becomes apparent. In identifying the South African material, the authors made extensive use of the monographical studies of Collignon (1948), and the Santonian and Campanian parts of his subsequent series Atlas des fossiles caractéristiques de Madagascar (Ammonites) (1966a, 1969, 1970), Young’s (1963) description of the Upper Cretaceous faunas of the Gulf Coast of America, and Matsumoto’s (1970) recent description of the Texanitinae from Hokkaido. Unfortunately, the majority of texanitine species seems to be endemic, thus precluding precise stratigraphic correlation on a global scale. Similarities between the South African and Malagasy material are striking, and Collignon’s biostratigraphic zonation compiled for Madagascar could be employed to good effect. This notwithstanding, a great proportion of the material appears to be new, necessitating the erection of new species. The association of the form genus Spinaptychus Trauth with representatives of the subfamily Texanitinae is firmly established on the basis of more material to supplement previous records by the authors (Klinger 1972, Kennedy & Klinger 1972). Broad outlines of the evolution and evolutionary trends within the Texani- tinae, as displayed by the South African material and aspects of the palaeobio- geography of the subfamily are discussed after the systematic descriptions. CRETACEOUS FAUNAS FROM SOUTH AFRICA 3 LOCATION OF SPECIMENS The following abbreviations are used to indicate the respositories of the material studied : BMNH British Museum (Natural History), London. NMB_ National Museum, Bloemfontein (presently housed in the collections of the South African Museum, Cape Town). OLL OberOsterreichisches Landesmuseum, Linz. SAM South African Museum, Cape Town. SAS Geological Survey of South Africa, Pretoria. ™ Transvaal Museum, Pretoria. UN Geology Department, University of Natal, Durban. FIELD LOCALITIES Details of field localities referred to in the text are provided in Kennedy & Klinger (1975). Full descriptions of the localities are housed in the libraries or collections of the British Museum (Natural History), Geological Survey of South Africa (Pretoria), and the South African Museum. DIMENSIONS OF SPECIMENS Dimensions of specimens are given in millimetres; abbreviations are as follows: D = diameter, Wb = whorl breadth, Wh = whorl height, Wb:Wh = ratio of whorl breadth to whorl height, U = umbilical diameter. Figures in parentheses are dimensions as a percentage of total diameter. SUTURE TERMINOLOGY The suture terminology of Wedekind (1916; see Kullmann & Wiedmann 1970 for a recent review) is followed in the present work: I = internal lobe, U = umbilical lobe, L = lateral lobe, E = external lobe. ANNOTATION OF ORNAMENTATION Terminology of tuberculation followed here is essentially the same as that of Collignon (1948: 55), and Young (1963: 37, text-fig. 6), reproduced here in slightly modified form as Figure 1. Tubercles are numbered sequentially from the umbilical wall ventrally towards the keel; the umbilical tubercle (Ut) is number 1, the lateral tubercle (Lt) 2, the submarginal (St) 3, the marginal (Mt) 4, and the external (Et) 5. This is a genetic annotation. In genera lacking the lateral tubercle, the second tubercle encountered ventrally from the umbilical wall is situated submarginally, and thus annotated as number 3. Similarly, in genera where two or more tubercles arise through division of a single tubercle, the latter is referred to according to its position, e.g. ventrolateral, with the ANNALS OF THE SOUTH AFRICAN MUSEUM Keel ~ External (5) Se IE as eres Submarginal (3) See Lateral (2) ---" Umbilical (1) Dorsal zone of impression Fig. 1. Whorl section of a texanitine ammonite to explain annotation regarding ornamentation used in the text; modified after Collignon (1948: 55) and Young (1963: 37, text-fig. 6). numbers of the tubercles it eventually gives rise to in parentheses, e.g. (3-+4), Rib-counts are indicated as the number of internal (Int.) (umbilical) tubercles versus the number of external (Ext.) tubercles. It was found imprac- ticable to annotate and count ribbing in terms of bifurcating and/or intercalatory ribs as the distinction between these is often slight and generally of subordinate importance in texanitine systematics. SYSTEMATIC PALAEONTOLOGY Phylum MOLLUSCA Class CEPHALOPODA Subclass AMMONOIDEA Zittel, 1884 Superfamily ACANTHOCERATACEAE de Grossouvre, 1894 Family Collignoniceratidae Wright & Wright, 1951 Subfamily Texanitinae Collignon, 1948 The subfamily Texanitinae is a prolific group, ranging in age from ?Lower Coniacian to Middle Campanian, and is best known from Madagascar, the Gulf Coast regions of North America, and Hokkaido, Japan. The subfamily is herein taken to include the following genera and subgenera: CRETACEOUS FAUNAS FROM SOUTH AFRICA 5 Texanites Spath, 1932 Paratexanites Collignon, 1948 Bevahites Collignon, 1948 Submortoniceras Spath, 1921 Menabites s.s. Collignon, 1948 Menabites (Bererella) Collignon, 1948 Menabites (Delawarella) Collignon, 1948 Menabites (Australiella) Collignon, 1948 Protexanites s.s. Matsumoto, 1955 Protexanites (Miotexanites) Matsumoto, 1970 Protexanites (Anatexanites) Matsumoto, 1970 Protexanites (Pleurotexanites) Matsumoto, 1970 Plesiotexanites Matsumoto, 1970 Reginaites Reyment, 1958 Defordiceras Young, 1963 Obvious synonyms of these taxa are discussed in the text. Initially, the subfamily Texanitinae formed a relatively homogeneous group, consisting of unicarinate trituberculate to pentatuberculate taxa, but lately through the inclusion of Reginaites, P. (Pleurotexanites), and Defordiceras, the definition has to be modified to accommodate these tricarinate, interrupted unicarinate, and keel-less forms. Generic subdivision is based mainly on the number of rows of tubercles. Protexanites is trituberculate throughout, except in P. (Anatexanites) where a lateral (2) row of tubercles is added in later stages. Paratexanites is quadri- tuberculate throughout, lacking the lateral (2) row of tubercles. Texanites is pentatuberculate from an early stage, as is Submortoniceras, but in the latter, lateral ornament weakens in later stages of growth. Plesiotexanites links between Paratexanites and Texanites in forming a lateral tubercle after variable stages of growth. Bevahites and Menabites s.\. differ from the other genera in having more tubercles in the external (5) row than in the other rows. In Bevahites the marginal (3) and submarginal (4) tubercles remain more or less approximated throughout, thus differing from Menabites. Reginaites is bituberculate to quadrituberculate, as here interpreted, but is distinctly tricarinate as in Peroni- ceras. This latter similarity is an example of homoeomorphy and not genetic affinity. Defordiceras lacks a keel on the outer whorls. Whether this condition also prevails on the inner whorls is unknown. The suture throughout is relatively simple with a primary suture formula ELU,UjI. The following species are described from South Africa: Protexanites cycni (van Hoepen) Paratexanites australis sp. nov. Paratexanites pseudotricarinatum sp. nov. Paratexanites umkwelanense (Crick) 6 ANNALS OF THE SOUTH AFRICAN MUSEUM Paratexanites sp. aff. P. emscheris (Schliiter) Paratexanites sp. aff. P. serratomarginatus (Redtenbacher) Plesiotexanites stangeri (Baily) Plesiotexanites collignoniforme sp. nov. Plesiotexanites matsumotoi sp. nov. Plesiotexanites ? sp. indet. Reginaites reymenti sp. nov. Reginaites zulu sp. nov. Reginaites ? sp. nov.? cf. Plesiotexanites stangeri (Baily) Texanites collignoni sp. nov. Texanites quadrangulatus Collignon Texanites vanhoepeni sp. nov. Texanites postvanhoepeni sp. nov. Texanites rarecostus Collignon Texanites texanus (Roemer) Texanites umzambiense sp. nov. Texanites presoutoni presoutoni ssp. nov. Texanites presoutoni natalense ssp. nov. Texanites soutoni soutoni (Baily) Texanites soutoni natalense ssp. nov. Submortoniceras woodsi s.\. (Spath) Submortoniceras condamyi s.\. (Collignon) Bevahites ? enigma sp. nov. Bevahites sp. A. gr. ex. B. subquadratus Collignon Bevahites sp. B. gr. ex. B. subquadratus Collignon Menabites (Menabites) boulei Collignon Menabites s.l. sp. juv. indet. M. (Australiella) australis (Besairie) M. (Australiella) falloti (Collignon) M. (Delawarella) delawarensis (Morton) M. (Delawarella) subdelawarensis Collignon M. (Delawarella) sp. aff. M. (D.) jeanneti Collignon M. (Delawarella) gigas sp. nov. M. (Delawarella) nibelae sp. nov. M. (Delawarella) sp. indet. Genus Protexanites Matsumoto, 1955 Type species Ammonites bourgeoisi d’Orbigny, 1850, by original designation of Matsu- moto (1955: 38). Discussion The genus Protexanites was originally erected by Matsumoto (1955) for species ‘similar to Texanites in general aspect . . . typically trituberculate . . . CRETACEOUS FAUNAS FROM SOUTH AFRICA 7 and sometimes quadrituberculate in the later whorls with one lateral tubercles [sic] in addition to the other three’ (Matsumoto 1955: 38). Since then, however, more material has become available and Matsumoto (1970: 227-228) now recognizes four distinct subgenera: Protexanites s.s. Matsumoto, 1955 (type species Ammonites bourgeoisi d’Orbigny, 1850). Anatexanites Matsumoto, 1970 (type species Mortoniceras fukazawai Yabe & Shimizu, 1925). Miotexanites Matsumoto, 1970 (type species Protexanites (Miotexanites) minimus Matsumoto, 1970). Pleurotexanites Matsumoto, 1970 (type species Protexanites superbus Collignon, 1966a). Differences between the various subgenera are distinct. Protexanites s.s. is trituberculate throughout; Anatexanites acquires a lateral (2) tubercle at varying stages of growth; in Miotexanites the development of the ventrolateral tubercle is retarded, and in Pleurotexanites the tubercles of the external (5) row outnumber those of the umbilical (1) or submarginal (3) rows; in addition, the keel, consisting of a line of multiplied tubercles, is interrupted. Only representatives of Protexanites s.s. occur in southern Africa, hence the reader is referred to Matsumoto (1970) for discussions on the limits and affinities of the other three subgenera. Species referred to these latter subgenera by Matsumoto (1970) are: Subgenus Anatexanites P. (Anatexanites) fukazawai (Yabe & Shimizu) (1925: 130, pl. 30 (fig. 1), pl. 31 (figs 1-2, 6-7, non 3), pl. 33 (figs 1-2)) from the Middle Santonian of Kyushu and Hokkaido. P. (Anatexanites) nomii (Yabe & Shimizu) (1925: 131, pl. 32 (figs 1-3)) from the Santonian of Hokkaido. P. (Anatexanites) reymenti Matsumoto, 1970 (=Texanites cf. quattuornodosus in Reyment 1955: 93, pl. 23 (fig. 3), text-fig. 46a) from the Santonian of southern Cameroons. Subgenus Miotexanites Protexanites (Miotexanites) minimus Matsumoto (1970: 246, pl. 33 (figs 1-3), text fig. 8) from the Coniacian or Lower Santonian of Hokkaido. ? Niceforoceras (?) japonicum Matsumoto (1965: 71, pl. 11 (fig. 1) text-fig. 40) from Hokkaido may be an example of Miotexanites. Subgenus Protexanites Matsumoto, 1955 Type species Ammonites bourgeoisi d’Orbigny, 1850, by the original designation of Matsumoto (1955: 38). 8 ANNALS OF THE SOUTH AFRICAN MUSEUM Diagnosis Following the very early smooth stages, ornament consists of single, bifurcating or intercalatory ribs bearing three rows of tubercles only; umbilical (1), submarginal (3), and external (5). The keel may be entire or crenulate. Primary suture quinquelobate with formula ELU,U,I. Discussion Matsumoto (1970: 228) referred the following species to Protexanites s.s.: P. (Protexanites) bontanti (de Grossouvre) (1894: 77, pl. 17 (fig. 2 only)) from the Coniacian of France. P. (Protexanites) bourgeoisi (d’Orbigny, 1850) (De Grossouvre 1894: 73, pl. 13 (fig. 2), pl. 14 (figs 2-5)) from the Senonian of France. P. (Protexanites) canaensis (Gerhardt) (1897: 73, pl. 1 (fig. 2a-c)) from the Lower Senonian, presumably Coniacian, of Venezuela. P. (Protexanites) peroni Matsumoto, 1970 (=Peroniceras czoernigi Peron non Redtenbacher, 1896: 53, pl. 11 (figs 7—-8)) from the base of the Senonian of Djebel Aures, Algeria. P. (Protexanites) shoshonense (Meek) (1876: 449, pl. 6 (figs 3a, c, 6b)) from the Coniacian of Wyoming. P. (Protexanites) strozzii (Desio) (1920: 204, pl. 12(1) (fig. 8)) from the Senonian of Florence, Italy. To this list may be added: P. (Protexanites) eugnamtus (Redtenbacher) (1873: 117, pl. 27 (fig. la-c)) from the Coniacian of the Gosau Beds, Austria. (See Reyment 1958: 45.) Of these species half are either monotypic or poorly known, having been figured once or twice only. These include P. (P.) canaensis and its variety obliquecostata, P. (P.) peroni, P. (P.) strozzii and P. (P.) eugnamtus. Detailed descriptions and figures to illustrate the extensive intraspecific variation and ontogeny of P. (P.) shoshonense and P. (P.) planatus were provided by Reeside (1927) and Young (1963) respectively. Illustrations and descriptions of the types of P. (P.) bourgeoisi and P. (P.) bontanti were provided by Matsu- moto (1966, 1970). : As appears to be the case in other regions, Protexanites is poorly repre- sented in the Coniacian of Zululand, both in numbers and in species. One species only, P. (P.) cycni (van Hoepen) is described below. Due to their relative scarcity throughout the world, current systematics in Protexanites seems to be typologic, rather than based on actual populations, and is extremely conservative when compared with systematic concepts applied to Texanites or Submortoniceras. Specific characters in Protexanites include, apart from relative proportions, density and curvature of ribbing, strength and position of the umbilical (1) and submarginal (3) tubercle, and whether the keel is entire or crenulate. Amongst these, possibly only the latter is of real signifi- cance in Texanites, but it varies according to preservation. CRETACEOUS FAUNAS FROM SOUTH AFRICA 9 Comments on the phylogeny of Protexanites and Paratexanites are given on p. 16. Occurrence Coniacian to Lower Santonian of western and central Europe (France, Spain, Italy, Austria, Czechoslovakia, Rumania), Israel and Sinai, North America (Gulf Coast and Western Interior), South America (Venezuela and Peru), Japan and Saghalien, north Africa (Algeria), South Africa (Zulu- land), and Madagascar. Protexanites (Protexanites) cycni (van Hoepen, 1965) Figs 2-4 Subprionocyclus cycni van Hoepen 1965b: 183, plate I. Subprionocyclus cycni? van Hoepen 19655: 183. Type Holotype is SAS—Z651 by the original designation of Van Hoepen (1965b: 183), here refigured as Figure 2D-F. Material SAS-Z808 and SAS-—Z832, locality 93, near the Hluhluwe Estuary, Zulu- land, St Lucia Formation, Coniacian II to possibly III, the same locality and horizon as the holotype; BMNH-C81431-—C81433, locality 91, close by locality 93, St Lucia Formation, Coniacian IV-V. Dimensions D Wb Wh Wb: Wh U Int. EXE: Z651 43,0 13,0(30) 17,0(40) 0,76 14,0(33,0) — — Z834 62,0 20,0(32,3) 21,5(34,7) 0,93 26,0(41,9) 16 22 Z832 72,0 — 28,0(38,9) — 2S:0C 8:9) en 2 elo <2 Z808 90,0 27,0(30,0) 32,0(35,5) 0,84 BUKUGHEID) DPA ws ‘. -—----~ oo. 4-----., Ne ee ee ne a er ( ee eee eee ! od ] 1.4} 1 o--------+-- —~ =o -~ ~_~ ~ ~ Sa >. SoS —) —_—— Sse ~ Samos) a —-~ =” Sa So -----7- \ aoa Seo ek B Fig. 44. i ais pseudotricarinatum sp. nov. Paratype NMB-D1060. Whorl section and partial suture. Scale bar in millimetres. CRETACEOUS FAUNAS FROM SOUTH AFRICA 59 derived from Paratexanites pseudotricarinatum through compression of the whorl section and addition of a lateral tubercle. The compressed specimen, NMB-D1364 (Fig. 43), here designated Var. E, is thus a potential intermediate form. Paratexanites umkwelanense (Crick) is assumed to occur at more or less the same stratigraphic level, and the two species appear related. P. umkwelanense also has an undulating keel and may, in cases, also have indications of incipient lateral keels. In typical forms, however, adult ornament is much more spinose and conspicuous, and the ventral tubercles are elevated high above the keel. P. pseudotricarinatum sp. nov. superficially resembles P. zeilleri as figured by De Grossouvre (1894, pl. 14 (fig. 1)), but is easily distinguished by the characteristic ventral ornament. Occurrence Coniacian IV-V, and possibly Santonian I of Zululand. Paratexanites sp. aff. P. serratomarginatus (Redtenbacher, 1873) Figs 45-47A Compare: Ammonites serrato-marginatus Redtenbacher, 1873: 110, pl. 25 (fig 2a—-d). Paratexanites (Parabevahites) serratomarginatus (Redtenbacher): Matsumoto, 1970: 260, pl. 36 (figs 1-3), text-fig. 16. Material SAS-Z633, Z814 and Z865, all from locality 91, near the Hluhluwe River Estuary, Zululand, St Lucia Formation, Coniacian IV-V. Dimensions D Wb Wh Wb: Wh U Int. Ext. Z814 57,0 24,5(42,9) 21,0(36,8) Le, 21,0(36,8) 24 24 Z865 70,0 28,0(40,0) 28,0(40,0) 1,00 25,0(35,7) 22) oi, Description Coiling is relatively involute in the early stages, covering the ventral quarter of successive whorls, but becomes more evolute towards the body chamber. On the phragmocone the costal whorl section is subquadrate, as wide as high, or slightly wider than high with maximum width at the umbilical edge. In later whorls the section becomes more inflated, and subtrigonal (Fig. 45B) with no distinct umbilical edge. Ornament, especially on the inner whorls of the three available specimens, differs considerably, ranging from dense, coarse ribbing with spinose ventro- lateral nodes as in SAS—Z814 (Fig. 47A) to dense ribbing lacking strong ventro- lateral ornament as in SAS—Z633 (Fig. 46). On the outer phragmocone whorls, ornament is similar in the three specimens, consisting of strong, radially elongated umbilical tubercles connected by weak, slightly prorsiradiately curved 60 ANNALS OF THE SOUTH AFRICAN MUSEUM \ 4 x = B Fig. 45. Paratexanites sp. aff. P. serratomarginatus (Redtenbacher, 1873). SAS—Z814. Whorl section and partial suture. Scale bar in millimetres. Fig. 46. Paratexanites sp. aff. P. serratomarginatus (Redtenbacher, 1873). SAS-Z633. Specimen with densely costate inner whorls. xX 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 61 ribs to approximated submarginal and marginal tubercles and a slightly more distant row of clavate external tubercles. The keel in all three specimens is undulating, though least noticeable in SAS-Z814, the specimen with strong ornament on the inner whorls. Parts of the body chamber are preserved in the two specimens, and here submarginal (3), marginal (4), and external (5) tubercles are virtually equidistant. The external suture is partly exposed in SAS—Z814 (Fig. 45A). Discussion It is unfortunate that the authors’ material of the species is limited, as it would have been interesting to see whether there are any significant differences between those specimens with coarse early ornament and those with more delicate ornament. For the present all the specimens are referred to one species. As far as the involute coiling and general nature of ornament on the outer phragmocone whorls are concerned, there is strong resemblance to Para- texanites serratomarginatus; especially when comparing SAS-Z814 (Fig. 47A) to Redtenbacher’s figure (1873, pl. 25 fig. 2c—d). This species, however, presents difficulties in interpretation. Of the seven specimens available to Redtenbacher, only two were figured. Subsequent interpretations of this species have differed considerably. Matsumoto (1970: 262 ff.) provided a good discussion on the extent and affinities of the species, and erected a new subspecies, P. serratomarginatus grossouvrei, to accommodate the French Villedieu specimens with undulating keels and stronger submarginal tubercles. The Zululand material resembles the French subspecies closest in possession of the undulating keel, but does not retain the double row of tubercles to such a large diameter. None of the figured specimens of P. serratomarginatus shows satisfactorily whether the submarginal and marginal tubercles remain approximated through- out or whether this feature is restricted only to the phragmocone. This is the main reason for not accepting the subgeneric term Parabevahites, of which P. serratomarginatus is the type species. Until this question is resolved, it is advisable to refer to the Zululand material as P. sp. aff. P. serratomarginatus. Not having seen the specimen, the authors cannot comment on the identity of the large fragment (referred to P. serratomarginatus) mentioned by Collignon (1948: 84), apart from the fact that the locality data appear to be incorrect. Collignon suspected the specimen to have come from the vicinity of Skoenberg, but it is more likely to have been collected in the lower reaches of the Mzinene River (Kennedy & Klinger 1975, locality 73) where Paratexanites species are known to occur. As yet Paratexanites species have not been recorded from the Skoenberg itself by the authors, and this seems unlikely to happen, as the youngest strata exposed there are Lower Coniacian. This species is allied to P. umkwelanense, differing mainly in being more narrowly umbilicate and having more inflated whorls and denser, though not as prominently developed, ornament. ANNALS OF THE SOUTH AFRICAN MUSEUM 62 SapluoxajoAdD °F "TX ‘opounydis ose] oY} 9JON[ “WOT}aS Hoy *€/VIH-SVS “(OL8T “TeIN[YOS) st4ayasiua “gq “ye “ds “S[JOYM JOUUL 9S1eOD YIM UaUTIDEdS “pI8Z-SVS “(ELST “Joyoequo}poy) smoursiDWOIDLAas “q “ye “dS sajiupxajosvg “WY “Ly ‘Sly CRETACEOUS FAUNAS FROM SOUTH AFRICA 63 Occurrence Coniacian IV-V of Zululand. Paratexanites sp. aff. P. emscheris (Schliiter, 1876) Figs 47B-48 Compare: Ammonites emscheris Schliiter, 1876: 155, pl. 42 (figs 8-10). Material SAS-H1A/3, locality 22, Nyalazi River, Zululand, St Lucia Formation, Coniacian IV. Description and discussion A large septate fragment collected from the basal beds exposed at locality 22 (Klinger & Kennedy 1975) is tentatively referred to Schliiter’s (1876) species. The costal whorl section is subrectangular, higher than wide with well-developed submarginal, marginal and external tubercles. The umbilical tubercles are radially elongated, thus differing from Schliiter’s illustration (1876, pl. 42 (figs 8-10)). __ Matsumoto (1970) recently discussed the scope of this species, pointing out that it is not well defined. One of the specimens included under the name Ammonites emscheris by Schliiter (1876: 155), previously described by him (1872: 42 pars., pl. 12 (figs 1-3)) as Ammonites texanus, was regarded by Matsu- moto as being sufficiently different from P. emscheris to merit separate specific rank as Paratexanites rex. In the present specimen the umbilical tubercles are radially elongated as in the latter species, but the whorl section is not as compressed. For the present it is considered best to refer to the available specimen as P. sp. aff. P. emscheris. Occurrence Coniacian IV of Zululand. Genus Plesiotexanites Matsumoto, 1970 Type species Mortoniceras kawasakiti Kawada, 1929, by the original designation of Matsumoto (1970: 267). Diagnosis The inner whorls remain trituberculate (1,3 +4,5) for a considerable period. Eventually the ventrolateral (3-+4) node, which may be very large, divides into two distinct tubercles, submarginal (3) and marginal (4). Simultaneously, or later, a lateral (2) tubercle appears. Tubercles (3) and (4) may remain close together, or separate completely, giving rise to normal Texanites ornament. ANNALS OF THE SOUTH AFRICAN MUSEUM 64 , 1876). SAS-H1A/3. x 1. ff. P. emscheris (Schliter a tes sp i 48. Paratexan ig. F CRETACEOUS FAUNAS FROM SOUTH AFRICA 65 The lateral (2) tubercle may become prominent, remain inconspicuous, or become absorbed in the flanks in later stages of growth. Discussion Matsumoto (1970: 274-280), in erecting Plesiotexanites as a subgenus of Texanites, discussed its affinities extensively, and, apart from some new data and comments on the phylogeny, it requires little elaboration. According to Matsumoto (1970: 274-280) the following species were considered to be representatives of Plesiotexanites: Plesiotexanites kawasakii (Kawada, 1929) (Matsumoto 1970: 280, pl. 38 (figs 1-2), pl. 39 (fig. 1), pl. 40 (figs 1-2), pl. 41 (fig.1), pl. 42 (fig. 4), pl. 43 (figs 1-2), text-figs 21-22) from the Middle Santonian of Hokkaido. Plesiotexanites pacificus Matsumoto (1970: 289, pl. 42 (fig. 2), pl. 45 (figs 1-2), pl. 46 (fig. 4), text-fig. 24) from the Santonian of Hokkaido. Plesiotexanites sanushibense (Yabe & Shimizu) (1925: 132, pl. 33 (figs 3-5)) from the Santonian of Hokkaido. Plesiotexanites schlueteri Matsumoto (1970: 278, text-fig. 20) from an unspeci- fied horizon in northern Germany. Plesiotexanites shiloensis (Young) (1963: 89, pl. 46 (figs 1-4), pl. 54 (figs 4-7), pl. 70 (figs 5-6, 8) text-fig. 24d) from the Upper Santonian of the Gulf Coast of North America. Plesiotexanites stangeri (Baily) (1855: 455, pl. 11 (fig. 2)) from the Lower to Upper Santonian of South Africa, Madagascar, Hokkaido, and the Gulf Coast of North America. Plesiotexanites candelariae (Young) (1963: 102, pl. 56 (figs 1, 3-4), pl. 60 (fig. 8), text-figs 20b, 28af, 29ae, 34af) from the Lower Campanian of the Gulf Coast of North America. Plesiotexanites thomsoni (Jones) (1966: 200, pl. 26 (figs 1-8), text-fig. 2) from the Upper Coniacian or Santonian of California. Plesiotexanites transitorius (Collignon) (1966a: 80, pl. 488 (fig. 1970A-—B)) from the Middle Santonian of Madagascar. Plesiotexanites yezoensis Matsumoto (1970: 294, pl. 44 lif, 2)) from the San- tonian of Hokkaido. Some doubtful contenders mentioned by Matsumoto were: Texanites angolanus Haas (1942: 12, figs 8-10, pl. 15 (fig. 11)) from the Santonian of Angola. Parabevahites dubius Collignon (1966a: 133, pl. 513 (fig. 2026)) from the Upper Santonian of Madagascar. Mortoniceras omeraense Reeside (1927: 38, pl. 42 (figs 3-4), pl. 43 (figs 1-2)) from the Santonian of New Mexico. Submortoniceras (?) aff. soutoni Reyment (1955: 95, pl. 33 (fig. la—b)) from the Santonian of Nigeria. 66 ANNALS OF THE SOUTH AFRICAN MUSEUM To the list of definite representatives of Plesiotexanites may be added: Plesiotexanites matsumotoi sp. nov. Plesiotexanites collignoniforme sp. nov. from the Santonian of Zululand and Pondoland. Plesiotexanites sp. indet. from the Santonian of Zululand, a doubtful contender. The diagnostic characteristics of Plesiotexanites are restricted to the ontogenetic development of ornament. Clearly this presents difficulties when dealing with incomplete material, as Matsumoto (1970: 276) indicated in discussing the affinities of Texanites angolanus and T. ralijaonai, both of which lack the inner whorls. This should, however, in no way prevent the use of the name Plesiotexanites for a group of generally evolute texanitids occupying an intermediate position between Paratexanites or Protexanites and Texanites, and which also probably gave rise to Reginaites. The ontogeny of Plesiotexanites provides clues to its origin. The initial trituberculate stage is like that of Protexanites s.s. The succeeding stages of the various species differ in respect of the stage of appearance of the lateral tubercle and division of the ventrolateral node. Thus, in P. stangeri the ventrolateral node divides before the development of the lateral tubercle, thus mimicking a Paratexanites type of ornament. With the development of the lateral tubercle, the ornament is like that of Texanites s.s. In P. schlueteri the lateral tubercle is retained for a short period only, thus the Texanites ornament here gives way to Paratexanites ornament in the adult stage. In species such as P. thomsoni and P. matsumotoi sp. nov. the lateral tubercle appears before division of the ventro- lateral node, thus producing Protexanites (Anatexanites)-like ornament. According to Matsumoto (1970: 277), Texanites shiloensis Young, which should also be referred to Plesiotexanites, has inner whorls allied to Protexanites (Miotexanites) minimus Matsumoto. P. matsumotoi sp. nov. has inner whorls comparable to those of Para- texanites australis sp. nov., and P. collignoniforme sp. nov. has inner whorls similar to those of Paratexanites pseudotricarinatum sp. nov. or P. umkwelanense (Crick). Available data thus would seem to suggest that the origin of Plesiotexanites is to be found in Protexanites s.1. or in Paratexanites. The South African material of Plesiotexanites is of extreme interest in providing links to both Reginaites and Texanites s.s., and is here accorded full generic rank. Occurrence Definite occurrences of Plesiotexanites are in the Santonian of Japan, Germany, Gulf Coast of America, Madagascar, Zululand, and Pondoland. Other possible occurrences are in the Santonian of Angola, Cameroons and Western Interior of North America. CRETACEOUS FAUNAS FROM SOUTH AFRICA 67 Plesiotexanites stangeri (Baily, 1855) Figs 20B, 49-69 Ammonites stangeri Baily, 1855: 455, p. 11 (fig. 2). Mortoniceras stangeri (Baily): Woods, 1906: 338, pl. 44 (fig. 1). Spath, 1921: 297, text-fig. D-1; 1922: 137, pl. 9 (fig. 2). Mortoniceras stangeri (Baily) var. densicosta Spath, 1922: 138, pl. 5 (fig. 2). Mortoniceras stangeri (Baily) var. sparsicosta Spath, 1922: 138, pl. 5 (fig. 1). Texanites stangeri (Baily): Young, 1963: 88, pl. 45 (figs 1-3), text-fig. 25p. Texanites stangeri (Baily) densicostus (Spath): Young, 1963: 86, pl. 42 (figs 3-4), pl. 43 (figs 2-4), pl. 47 (figs 5-6), pl. 48 (figs 2, 5-6), pl. 71 (figs 1-4), text-figs 25c, e, g—h, 34c. Texanites stangeri (Baily) var. densicosta (Spath): Collignon, 1966a: 72, pl. 484 (fig. 1958), pl. 512 (fig. 2024). Texanites stangeri (Baily) var. sparsicosta (Spath): Collignon, 1966a: 62 pl. 479 (fig. 1951). ?Parabevahites cf. emscheris Collignon, 1966a: 80, pl. 488 (fig. 1966). Texanites stangeri subsp. aff. densicostus (Spath): Wolleben 1967: 1152, pl. 149 (fig. 1). Texanites (Plesiotexanites) stangeri (Baily): Matsumoto, 1970: 285, pl. 41(45) (figs 2-4), text-fig. 23 (97). Kennedy & Klinger, 1973: 102. Texanites (Plesiotexanites) stangeri (Baily) sparsicosta (Spath): Kennedy & Klinger, 1973: 102, pl. 6 (fig 3a—b). Texanites (Plesiotexanites) stangeri (Baily) densicosta (Spath): Kennedy & Klinger, 1973: 102, pl. 5 (fig. 2a—b). Texanites (Plesiotexanites) stangeri (Baily) var. indet. Kennedy & Klinger, 1973: 103. Fig. 49. Plesiotexanites stangeri (Baily, 1855). Copy of Baily’s original figure (1855, pl. 11 (fig. 2)). Baily’s figure is a mirror image and partially restored. See Figs 50 and 51.) 68 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 50. Plesiotexanites stangeri (Baily, 1855). Cast of holotype BMNH-C-73333. x 0,42. Type Holotype is BANH-C73333, the specimen figured by Baily (1855, pl. 11 (fig. 2)) (Fig. 49) from the Umzamba Estuary, Transkei, Umzamba Formation, here refigured as Figures 50-51. Material BMNH-C19443, C19446, C19449-C19450, C19455-C19457, C19462, C35623, C35627, C53144, SAM-K4827, K5544-K5545, K5547-K5551, NMB-D1604, all from locality 1, from the Umzamba Formation at Umzamba Fig. Si. CRETACEOUS FAUNAS FROM SOUTH AFRICA A Plesiotexanites stangeri (Baily, 1855). Cast of holotype BMNH-C-73333. x 0,48. 69 70 ANNALS OF THE SOUTH AFRICAN MUSEUM Estuary, Transkei, Santonian II-III; DM-3505, UN-6557, UN-6564, UN-6570; localities 4 and 5, subsurface deposits at Durban, Umzamba Formation, Santonian III to Campanian I; several unmarked specimens in the Geological Survey Collection, Pretoria, from excavations at locality 6, the Enseleni Game Reserve, St Lucia Formation, Santonian II to Campanian I; similar fragments from locality 14 in road cuttings near Mfolozi, St Lucia Formation, Santonian II to Campanian I; a single fragment SAS—Z1971, locality 74, north-western part of False Bay at Die Rooiwalle, St Lucia Formation, Santonian I-Campanian I; and BMNH-C81533, locality 81, St Lucia Formation, Santonian I-II. Dimensions D Wb Wh Wb: Wh U Int. Ext. C73333 320,0 60,8(19,0) 83,0(26,0) 0,73 179,2(56,0) 39 39 C19443 120,5 — 32,0(26,0) — 49,0(41,0) 24 24 64,0 — 16,6(26,0) — 30,6(47,9) 19 19 32,0 — 9,3(29,1) — 14,6(45,8) 17 19 C19446 135,0 — 39,0(28,8) — 69,0(50,3) 28 28 70,0 — 22,0(31,4) — 34,0(48,5) 20 26 34,0 10,0(29,4) — — 9,4(27,6) 19 Da C19449 211,0 46,6(22,0) 55,7(26,4) 0,83 112,8(53,4) 30 30 112,8 —_ 34,2(30,4) — 40,0(57,1) 23 p24 | C19450 138,6 — 34,6(25,0) — 74,6(53,8) — — 74,6 — 22,6(30,3) —_ 37,3(50,0) 18 32 C19455 137,0 36,0(26,2) 27,7(20,2) 1,29 67,0(48,9) 23 29 67,0 — 24,0(35,8) —_— 33,0(49,2) 19 7a SAM-K5547 139,0 39,0(28,0) 41,0(29,4) 0,95 72,0(51,7) 23 29 72,0 25,0(34,0) 22,0(30,5) 1,1 38,0(52,7) 14 30 SAM-K5550 54,0 16,6(29,6) 15,0(27,7) 1,06 24,0(44,4) 19 21 Description Coiling is very evolute, polygyral, with an average umbilical width of the order of 40 to 50 per cent of the total diameter, but may reach up to 60 per cent in some specimens. The whorl overlap is very little; generally less than 10 per cent of the total whorl height. The species may grow up to a very large size; the holotype is the largest known specimen with a diameter of 320 mm. The majority of the Umzamba specimens measure between 100 to 200 mm. Increase in whorl section is rapid. The innermost whorls have a square to depressed whorl section, but, with increase in diameter, the section becomes more rounded and gradually higher than wide with a rounded umbilical wall and venter (Fig. 52A—B). Ornament is variable due to ontogenetic change and intraspecific variation. Ontogeny In the first, Protexanites stage, which lasts up to c. 40 mm (Fig. 53B-C), ornament is trituberculate, consisting of small, conical umbilical (1) tubercles, and much larger, clavate ventrolateral (3-4) nodes, situated in the umbilical suture and leaning against the umbilical wall of the succeeding whorl. The external (5) tubercles are situated on either side of a prominent central keel, CRETACEOUS FAUNAS FROM SOUTH AFRICA 71 Fig. 52. Plesiotexanites stangeri (Baily, 1855). A. Whorl section of holotype BMNH-C73333, after Matsumoto (1970, text-fig. 23(97)A). B. Whorl section of var. sparsicosta, BMNH-— C73333, after Matsumoto (1970, text-fig. 23(97)B). C. Partial suture of BMNH-C19444, after Spath (1921, fig. D-—1a). which is already developed at a diameter of 7,5 mm. These tubercles are joined to the bases of the large ventrolateral nodes by low, broad connections. They are not aligned with the ventrolateral (3-4) and umbilical (1) tubercles, but are displaced somewhat forward. The external tubercles are very variable, sometimes being elongated spirally to such an extent as to form discontinuous lateral keels (Fig. 54E-F). In the second stage, the ventrolateral tubercles begin to separate into two 2: ANNALS OF THE SOUTH AFRICAN MUSEUM C Fig. 53. Plesiotexanites stangeri (Baily, 1855). A. SAS—Z1971. One of the two specimens of the species thus far found in the False Bay region of Zululand. B-C. SAM-K5889. Juvenile specimen showing pseudotricarinate inner whorls. x 1. different rows through the splitting off on the ventral side of a new row, though still connected by a common base (Fig. 54A—C). This is the ‘Parabevahites’ stage. With increasing diameter the inner of these two rows migrates dorsad from the spiral suture to complete the separation of the two rows. This is the ‘Para- texanites’ stage. In the third stage of development, faint indications of the lateral (2) tubercle start appearing near the dorsal third of the flanks. The moment of appearance and the strength of these lateral tubercles are very variable. On the later part of the phragmocone, and on the body chamber, which consists of one whorl on the holotype (Figs 50-51), bifurcations cease and ribbing becomes single. The whorl section becomes progressively inflated and towards the last part of the body chamber all lateral ornament weakens. CRETACEOUS FAUNAS FROM SOUTH AFRICA 73 Fig. 54. Plesiotexanites stangeri (Baily, 1855). A-C. SAM-—PCZ5686. Specimen illustrating the Paratexanites stage of ontogeny, but with distinct separated external tubercles. D-F. SAM-— K5550. Similar specimen with wavy lateral pseudo-keels. A, C, D-E x 1; B x 2. 74 ANNALS OF THE SOUTH AFRICAN MUSEUM Intraspecific variation Variation in density and prominence of ornament is extensive in this species, and has led to the usage of names such as “sparsicosta’ and “densicosta’, initially adopted by Spath (1922: 138). These varietal names are applicable mainly to density of ribbing on the inner whorls as shown by Spath (1922, pl. 5 (figs 1-2)) and here as Figures 57, 58. The present material shows that virtually all transitions occur from closely ribbed specimens, SAM-—K5447 (Fig. 59), BMNH-C19443 (Fig. 60), via the holotype of the typical form (Figs 50-51) to the holotype of var. ‘densicosta’ (Figs 55-57). Apart from density, differences exist in strength of ornament on later parts of the shell. Fig. 55. Plesiotexanites stangeri (Baily, 1855). BMNH-C19456. Holotype of var. densicosta. (British Museum photograph.) x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 15 Specimens such as BMNH-C19462 (Figs 61-62A) and BMNH-C19449 (Figs 62B, 63) still have strongly developed ribbing and tuberculation on the outer whorls, whereas specimen BMNH-C19459 (Figs 64-65) or BMNH-C19456 (Figs 55-56), the holotype of var. ‘densicosta’, are. virtually devoid of ornament at the same diameters. Density of ribbing on the inner whorls of the latter two specimens clearly shows that strength of ribbing on the outer whorls is inde- pendent of density of ribbing on the inner whorls. The point of appearance of the lateral (2) tubercle is extremely variable, Fig. 56. Plesiotexanites stangeri (Baily, 1855). BMNH-C19456. Holo- type of var. densicosta. (British Museum photograph.) x 1. ANNALS OF THE SOUTH AFRICAN MUSEUM 76 Inner Close-up of 3 icosta. 1855). BMNH-C19456 densi rac ly, whorls of var i texanites stangeri (Ba Plesio Ie Fig dl CRETACEOUS FAUNAS FROM SOUTH AFRICA XS SS RES: inner . Close-up of ta. Xc.3. - SPAFSICOS: Fig. 58. Plesiotexanites stangeri (Baily, 1855). BMNH-C19444 whorls of var 78 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 59. Plesiotexanites stangeri (Baily, 1855). SAM-K5447. x 1. as is its ultimate strength. Lateral tubercles are well developed in BMNH-C19462 (Fig. 61) and C19449 (Fig. 63) compared to the faint lateral tuberculation in BMNH-C19443 (Fig. 60) or BMNH-C19457 (Fig. 66). Discussion This species, together with Texanites soutoni, was the first texanitine recorded from the Cretaceous deposits at Umzamba. Even so, it is rather rare in South African collections and the bulk of the material is to be found in the collections of the British Museum (Natural History). CRETACEOUS FAUNAS FROM SOUTH AFRICA 719 Fig. 60. Plesiotexanites stangeri (Baily, 1855). BMNH-C19443. x 0,75. 80 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 61. Plesiotexanites stangeri (Baily, 1855). BMNH-C19462. x 0,75. CRETACEOUS FAUNAS FROM SOUTH AFRICA Fig. 62. Plesiotexanites stangeri (Baily, 1855). A. BMNH-C19462. B. BMNH-C19449. x 0,9. x Od: 81 82 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 63. Plesiotexanites stangeri (Baily, 1855). BMNH-C19449. x 0,7. The type locality of the species was given as ‘cliffs off the South African Coast near the Umtafuna and Umzambani Rivers’ (Baily 1855: 457). This is currently known as Umzamba Cliff or locality 1 of the authors (Kennedy & Klinger 1975: 281), and is the type section of the authors’ Umzamba Formation. Recent collecting by the authors at the type section showed that the known time range of P. stangeri in that section is Middle to Upper Santonian, with most CRETACEOUS FAUNAS FROM SOUTH AFRICA 83 Fig. 64. Plesiotexanites stangeri (Baily, 1855). BMNH-C19459. x 0,7. occurrences in the Upper Santonian. As far as could be ascertained, there seems to be no stratigraphic difference in occurrence between the varieties densicosta and sparsicosta or the typical form. These findings seem to be sup- ported by Collignon (1966a) and Young (1963), although the authors doubt if the Malagasy and South African biozonations are compatible with those of the Gulf Coast. Collignon records var. densicosta and sparsicosta from the Middle Santonian, and var. densicosta from the Upper Santonian of Mada- 84 ; ANNALS OF THE SOUTH AFRICAN MUSEUM Ray >» Fig. 65. Plesiotexanites stangeri (Baily, 1855). BMNH-C19459. x 0,9. CRETACEOUS FAUNAS FROM SOUTH AFRICA 85 gascar, whereas Young records var. densicosta from the Lower Santonian of the Gulf Coast. In consequence, the authors follow Matsumoto (1970: 288) in here regarding the varietal names as mere convenient morphological terms without taxonomic connotation. The affinities of the species have already been discussed extensively by Matsumoto (1970) and only little need be added. Fig. 66. Plesiotexanites stangeri (Baily, 1855). BMNH-C19457. x 0,8. 86 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 67. Plesiotexanites stangeri (Baily, 1855). BMNH-C19446. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 87 Fig. 68. Plesiotexanites stangeri (Baily, 1855). BMNH-C19450. x 0,75. The external (5) tubercles of some specimens of P. stangeri are extremely elongated, creating the appearance of wavy lateral keels. Spath’s (1921: 297) assertion that ‘Mortoniceras stangeri Baily sp. has tricarinate inner whorls’, however, is a generalization, as the inner whorls of SAM—K5889 (Fig. 53B—C) show. A single specimen, SAM—PCP5687 (Fig. 93C) tentatively referred below to the genus Reginaites, is identical to P. stangeri in all respects except for possession of two distinct continuous lateral keels at a relatively large diameter. 88 ANNALS OF THE SOUTH AFRICAN MUSEUM » Fig. 69. Plesiotexanites stangeri (Baily, 1855). SAM-—K4827. x 0,6. The specimen figured by Woods (1906, pl. 44 (fig. 1)). Baily’s figure of the holotype is reversed and somewhat reconstructed (see Figs 49-51). Occurrence The geographic distribution of P. stangeri in southern Africa is curious. Specimens are relatively abundant at the Umzamba Estuary, while a few speci- mens are known from subsurface deposits at Durban and Richards Bay and in isolated exposures near Umkwelane Hill. This appears to be the northernmost occurrence of the species in South Africa. A single doubtful specimen, SAS-K1971 (Fig. 53A), is known from locality 74 at Die Rooiwalle at the north-western end of False Bay and another, BMNH-C81533, from the Hluhluwe floodplain. To all appearances, P. stangeri seems to be restricted to the shallow water sediments resulting from the south- ward-progressing post-Lower Coniacian diachronous transgression (Kennedy & Klinger 1971), being most abundant in the Middle and Upper Santonian sediments at the Umzamba Estuary. Other reports of this species are from the Middle and Upper Santonian CRETACEOUS FAUNAS FROM SOUTH AFRICA 89 of Madagascar, an unknown horizon, probably Santonian of Hokkaido, Japan, and the Lower Santonian of the Gulf Coast of the United States of America. Plesiotexanites collignoniforme sp. nov. Figs 70-78 Holotype SAS-Z1789 (Figs 70-71), locality 85, on the south-western shores of False Bay, Zululand, St Lucia Formation, Santonian I. Etymology Refers to similarity to Texanites collignoni sp. nov. Material Paratypes are SAS-Z1790a—b, Z1791, Z2109, SAS-H100, H201/43-4, BMNH-C81495-6, C81525-8, C81530-31, all from locality 85, on the south- western shores of False Bay, Zululand, St Lucia Formation, Santonian I. Dimensions D Wb Wh Wb: Wh U Int. EXE: Z1789 — 222,0 58,0(26,1) 67,0(30,2) 0,87 103,0(46,4) 37 27 Z2109 177,0 47,0(26,6) 57,0(32,2) 0,82 80,0(45,2) 31 31 H201/44 127,0 32,0(25,1) 39,0(30,3) 0,88 59,0(46,5) 30 30 Z1791 195,0 49,0(25,1) 59,0(30,3) 0,88 92,0(47,2) 28 28 Description Coiling is very evolute, with an umbilical width of 45 to 47 per cent of the total diameter, and a very shallow dorsal zone of impression. The whorl section is nearly quadrate in the very early stages of growth, but soon becomes higher than wide with a subtrigonal whorl section and little-inflated flanks. Towards the body chamber the flanks become more inflated and the whorl section more rounded (Fig. 72B). SAS-Z1790a has the inner whorls preserved (Figs 73-74), permitting a study of the ontogeny. The innermost whorls are completely smooth up to a diameter of c. 8 mm; thereafter ventrolateral nodes start appearing in the umbilical seam, and soon broad, flat ribs appear, some with incipient looping. At a diameter of 20 mm umbilical tubercles start appearing at the dorsal ends of these ribs. Ribbing is single throughout. Lateral (2) tubercles appear at a diameter of c. 50 mm. Simultaneously, the ventrolateral (3) tubercles become less spinose, and migrate away from the umbilical suture to reveal a fourth row of tubercles. With increasing diameter, ribbing becomes stronger, rursiradiate and slightly sinuous, and tuberculation more prominent. Ribbing continues obliquely over the rounded umbilical wall, but very weakly. The umbilical tubercles (1) are situated high up on the umbilical wall, and tend to migrate up 90 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 70. Plesiotexanites collignoniforme sp. nov. Holotype SAS—Z1789. x 0,6. the ventral part of the flanks. The lateral (2) tubercles are situated near the dorsal third of the flanks, and, though weakly developed, appear most prominent due to the strong development of ribbing in this area. The submarginal (3), margi- nal (4) and external (5) rows of tubercles are situated closer to each other than the first two rows (1 & 2). The marginal and external tubercles tend to be con- nected by a thickened rib in Paratexanites fashion. The intertubercular area on the venter is connected by a swelling, and the faint, undulating keel tends to form a slight tubercle in this area. The external tubercles tend to be connected 91 CRETACEOUS FAUNAS FROM SOUTH AFRICA B fl) ) x 0 Fig. 71. Plesiotexanites collignoniforme sp. nov. Holotype SAS—Z1789 92 ANNALS OF THE SOUTH AFRICAN MUSEUM spirally by very faint ridges, thus creating a reticulate pattern over the venter, as in Paratexanites pseudotricarinatum sp. nov., though weaker. Towards the body chamber, the ribs become thinner, and the lateral tubercles tend to become absorbed, followed by the umbilical and eventually submarginal tubercles. The keel also becomes much weakened on the body chamber. The suture is broad with squat saddles and lobes (Fig. 72A). Discussion Barring slight details in density of ribbing, the species shows very little variation, and is easily distinguished by the evolute coiling, slightly sinuous, rursiradiate ribbing, low, indistinct undulating keel, and faint lateral lines connecting the external tubercles and Paratexanites-like proximity of the external and marginal rows of tubercles. This species is morphologically intermediate between Paratexanites and Texanites in being quinquetuberculate at a relatively small diameter, but in retaining a Paratexanites-like ornament near the ventral part of the shell. Here B Fig. 72. Plesiotexanites collignoniforme sp. nov. Paratype SAS-Z1970a. Whorl section and partial suture line. Scale bar for suture in millimetres. CRETACEOUS FAUNAS FROM SOUTH AFRICA 93 distinction between Texanites s.s. and Plesiotexanites becomes subjective. This species appears to be the source of Texanites collignoni and provides a link between Paratexanites and Texanites. Relative stratigraphic position and similarity of ornament over the venter (undulating keel, reticulate pattern, etc.) leave little doubt as to the origin of the species, namely the plexus of Para- texanites umkwelanense and Paratexanites pseudotricarinatum sp. nov. Similarly, the whorl section, ornament, especially the rounded umbilical wall with the Fig. 73. Plesiotexanites collignoniforme sp. nov. Paratype SAS—Z1970a. ry io Hy bh UH fn Kh fi eS = 1) C O} m On en { \ \ \ ' m —) Sa. Sm. bo OF can (GID) eS) ee SSS GDIGaeE) z z icnomey acne iA | Y Herons, lof fa 10% Pa oH Soacg ii iy ore ee oO iO’ iO’ vow ox ids | ny 8 ap Sak Or, Os Ps SOS Zz SSe Sere A9vIS Ava | Om Dy s Gp oO ©) CD GID | | loyOlunsyew J | WUaWAal “yy | nynz ‘y Fig. 85. Diagram to illustrate probable evolutionary relationship and known ontogenetic changes in South African representatives of genus Reginaites. CRETACEOUS FAUNAS FROM SOUTH AFRICA 109 Discussion Reginaites poses difficulty in interpretation, being poorly represented in numbers. Reyment (1957: 65) introduced the genus as a subgenus of Peroniceras, based on a single specimen from the Asu River, near Agbani, Nigeria, then believed to be of Upper Coniacian age. Apart from the type species, here repro- duced as Figures 87C, 89, Reyment also included Peroniceras leei Reeside 1927: 40, pl. 42 (figs 1-2), pl. 43 (figs 3-4)), a Lower Campanian species, in Reginaites. Young (1963: 92) added another species, Reginaites durhami, from the base of the Lower Campanian, to the genus. Since then, however, new taxa have been erected for species with tri- carinate venters and simple collignoniceratid suture, thus obscuring differences between the Texanitinae and Peroniceratinae, and also within the latter. These taxa are: Fraudatoroceras van Hoepen, 1965 (type species Fraudatoroceras besairiei van Hoepen, 1965a: 36, pl. 27, text-fig. 9a—b) from the Coniacian of Zululand. Cobbanoceras Matsumoto, 1965 (type species Cobbanoceras tanakai Matsumoto, 1965: 220, pl. 37 (fig. 3), pl. 38 (fig. 1), pl. 39 (fig. 1) text-figs 4-5) from the Santonian of Hokkaido and ?Coniacian of California. Reymentites Matsumoto, 1965 (type species Reymentites hataii Matsumoto 1965: 240, pl. 42 (fig. 3), text-figs 19-20) from the Urakawan (Coniacian and Santonian) of Hokkaido. As far as diagnoses are concerned, Fraudatoroceras van Hoepen and Cobbanoceras Matsumoto appear to be synonymous. Both are tricarinate, bituberculate, widely umbilicate and have simple sutures similar to those of Gauthiericeras (see Van Hoepen 1965a: 35-6; Matsumoto 1965: 219-220). If the taxa are, indeed, the same, Cobbanoceras is a junior synonym of Frauda- toroceras, as Van Hoepen’s work precedes that of Matsumoto by four months (27 July 1965 v. 20 November 1965). Full details are to be given in the authors’ pending revision of the Peroniceratinae. According to Matsumoto (1965: 239) Reymentites resembles Reginaites in being trituberculate or quadrituberculate, but differs from the latter in having the tubercles spaced more or less equidistant, rather than having the ventrolateral tubercles (3-+-4) situated close together. This relationship would thus be somewhat analogous to that between Paratexanites and ‘Parabevahites’. The holotype, and, indeed, only individual available of Reymentites hataii, is a wholly septate individual and probably immature, but does show the equi- distant tuberculation. Apart from the latter species, Matsumoto (1965: 239) also referred Reginaites durhami Young to genus Reymentites. The holotype of this species lacks the inner whorls, and the equidistant tuberculation present may well be regarded as the normal result of ontogenetic change in Reginaites as here interpreted. Reymentites is thus provisionally included in the synonymy of Reginaites, contrary to the views of Wiedmann (1978: 670). 110 ANNALS OF THE SOUTH AFRICAN MUSEUM Species which may thus be referred to Reginaites are: Reginaites durhami Young (1963: 92, pl. 39 (fig. 2), pl. 49 (figs 1-2, 4), text- fig. 22bc) from the Lower Campanian of the Gulf Coast of North America. Reginaites gappi Wiedmann (1978: 668, pl. 3 (figs 1-3), text-figs 2c, 3b) from the Upper Santonian of Austria. Reymentites hataii Matsumoto (1965: 240, pl. 42 (fig. 3), text-figs 19-20) from the Urakawan (Coniacian and Santonian) of Hokkaido. Reginaites leei (Reeside) (1927: 40, pl. 42 (figs 1-2), pl. 43 (figs 3-4)) from the Lower Campanian of New Mexico. Reginaites quadrituberculatum Reyment (1957: 65, pl. 11 (fig. la—b), text- fig. 7) from the Upper Coniacian? of Nigeria. To this list may be added two new species: Reginaites reymenti sp. nov. from the Upper Santonian or Lower Campanian of Zululand. Reginaites zulu sp. nov. from the top of the Upper Santonian of Zululand. Specific differentiation between the species is based mainly on the whorl section, presence or absence of lateral (2) tubercles and stage of splitting (if at all) of the ventrolateral tubercle and ornament on the body chamber. Unfortunately, dating of all the species is not satisfactory. Reymentites hataii cannot be dated more accurately than Urakawan (Coniacian and San- tonian). Reginaites leei and R. durhami are of definite Lower Campanian age, whilst the Zululand specimens are not older than Upper Santonian. The holotype of R. quadrituberculatum was collected by R. Blaser in beds above undoubted Coniacian and was thus tentatively dated as Upper Coniacian. According to Reyment (pers. comm. 21 November 1977), however, there is no evidence to gainsay a Santonian age for this species, and in the light of the present know- ledge this would be a more reasonable alternative. Thus, unless we are here dealing with a polyphyletic group, or hetero- chronous homoeomorphic groups, Reginaites seems to be a Santonian to Lower Campanian genus. Similarities between Reginaites and Plesiotexanites are striking. Apart from the presence of lateral keels versus an external (5) row of tubercles, there is no significant difference. This similarity had already been remarked upon by Young (1963: 93). As in other known occurrences, Reginaites does not occur in great numbers in Zululand, but from the specimens available, a phylogenetic sequence as out- lined in Figure 85 could be envisaged. Plesiotexanites matsumotoi sp. nov., with large ventrolateral spines or nodes, appears most suitable as ancestral form in South Africa. Through further lengthening of the already long external tubercles, the tricarinate venter is obtained. The ventrolateral (3-+-4) tubercles remain intact to greater diameters and the lateral (2) tubercle appears later and eventually not at all, thus giving rise to Reginaites zulu sp. nov. via R. reymenti sp. nov. A further possible trend is that the central keel becomes CRETACEOUS FAUNAS FROM SOUTH AFRICA 111 progressively more prominent at expense of the lateral keels and the venter more fastigiate, as in R. zulu. From this it would be possible to connect to Neogauthiericeras. Matsumoto (1965: 240) suggested that Reginaites, Reymentites and Cob- banoceras may be grouped together in a new subfamily. In view of the close similarity and relationship between Plesiotexanites and Reginaites, however, separation at this level is unacceptable, even if this inclusion deviates from the unicarinate character of the rest of the texanitids. Occurrence Stratigraphic data on Reginaites species are still confusing, as discussed above. Current data seem to indicate that the genus is restricted to the Santonian to Lower Campanian of Austria, Nigeria, Gulf Coast of North America, Hokkaido, Zululand, and, questionably, Pondoland. Reginaites reymenti sp. nov. Figs 86, 87A-B, 88 Holotype SAS-Z1986 (Fig. 87A-B), locality 74, Die Rooiwalle, Zululand, St Lucia Formation, Santonian I to Campanian I. ° Etymology Named for Professor R. A. Reyment, Uppsala. Material Paratypes are SAM—PCZ5877, PCZ5887—PCZ5888 from an unlocalized collection, but presumably also from Die Rooiwalle, as is the holotype. Description All the specimens are fragmentary, but together give a good idea of a complete individual. Coiling is evolute, with an umbilical diameter of about 44 per cent. The whorl section in the early whorls is rectangular, wider than high, but becomes more inflated with a rounded venter in later whorls, resulting in a dome-shaped, but still wider than high whorl section. On the earliest whorls preserved, at a diameter of c. 30 mm, ornament consists of small, pointed, conical, to radially elongated umbilical (1) tubercles, and strong, bullate, slightly clavate ventrolateral nodes connected by low, broad, widely-spaced single ribs. The venter has three keels of which the central one is strongest. At larger diameters, the ventrolateral nodes begin dividing into two: a rounded submarginal (3) node and a smaller, clavate marginal (4) tubercle. The first sign of division appears at c. 60 mm diameter. 112 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 86. Reginaites reymenti sp. nov. A-B. Paratype SAM-PCZ5877. C-D. Paratype SAM- PCZ5887. Note the faint lateral tubercle on the upper end of A. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 113 Fig. 87. A-B. Reginaites reymenti sp. nov. Holotype SAS—Z1986. C. Reginaites quadrituber- culatum Reyment, 1957. Holotype BMNH-—C48040. x 1. 114 ANNALS OF THE SOUTH AFRICAN MUSEUM On specimens SAM-PCZ5877 and SAS-—Z1986 (Figs 86A-B, 87A-B), which illustrate development at larger diameters, ribbing becomes prorsi- radiate, the marginal (4) tubercle migrates forward, ahead of the submarginal (3) tubercle on the venter. Further development differs somewhat in these two specimens. At the largest diameter preserved on SAS-Z1986 (Fig. 87A-B), ornament remains trituberculate, i.e. umbilical (1), submarginal (3) and marginal A B Fig. 88. Reginaites reymenti sp. nov. Paratype SAM-PCZ5888. Showing part of body chamber with sharpened ribbing. x 1. 115 CRETACEOUS FAUNAS FROM SOUTH AFRICA 1957. Holotype BMNH-C48040. x 1. Fig. 89. Reginaites quadrituberculatum Reyment, 116 ANNALS OF THE SOUTH AFRICAN MUSEUM (4). On SAM-PCZ5877 (Fig. 86A-B), however, which is of similar size to the outer whorl of the holotype, a small but distinct lateral (2) tubercle appears. In SAM-PCZ5888 (Fig. 88), which probably belongs to the same individual as the former specimen (SAM-—PCZ5887), part of the body chamber is pre- served. Here, all the tubercles tend to become incorporated in the ribbing, which becomes sharper and apparently more distantly spaced. The ribs now are prorsiradiate over the flanks, bend sharply forward between the submarginal (3) and marginal (4) tubercles, and then continue as thin riblets up to the lateral keel. Preservation in this specimen is rather poor but it does appear that the lateral (2) tubercle again disappears on the body chamber, or stays feeble throughout. Discussion This species is characterized by the strong, bullate ventrolateral tubercles on the inner whorls, weak development of the lateral (2) tubercle and tricarinate venter. Development of tuberculation is remarkably like that of Plesiotexanites, especially P. matsumotoi sp. nov., but in this latter species the ventrolateral tubercles remain undivided up to larger diameters, and the lateral (2) tubercle develops more strongly. Furthermore, P. matsumotoi is distinctly unicarinate, being a true texanitid. It would, however, be easy to derive Reginaites reymenti from P. matsumotoi by elongation of the external tubercles to form the lateral keels of Reginaites. The variable development of the lateral tubercle is another feature reminiscent of Plesiotexanites, as observed in P. stangeri (Baily). Reginaites zulu sp. nov. (described below) totally lacks development of the lateral tubercles and remains bituberculate to a much greater diameter. Reginaites leei (Reeside) is imperfectly known, but appears to have weaker ornament. Reyment (1957: 66) states R. /eei to have lateral tubercles, although Reeside (1927: 40) does not mention any. Reginaites quadrituberculatum Reyment, here shown as Figures 87C, 89, has similar ornament on the inner whorls, but has a more compressed whorl section. Reginaites durhami Young has a whorl section much higher than wide throughout, as has Reginaites gappi Wiedmann. Occurrence Probably Upper Santonian or Lower Campanian of Zululand. Reginaites zulu sp. nov. Figs 90-92, 93A-B, 94 Holotype SAS-H126A/9 (Figs 90A-B, 91-92, 93A-B), Bed 1 locality 105, south- eastern shores of False Bay, Zululand, St Lucia Formation, Upper Santonian. 117 CRETACEOUS FAUNAS FROM SOUTH AFRICA “SOTJOUI [IU UI eq speog ‘sdAJO[OY WOoIJ OUT] OINING “g ‘p/VOTIH -SVS odAjyered pure “6/V9Z7IH-SVS ‘odAjojoy 9y} JO Sase}s SAISSoOONS JO UOT}OES [IOUM “WY “AOU “ds mjNZ saywUIsay °06 ‘3A eS q Ol 0 118 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 91. Reginaites zulu sp. nov. Holotype SAS-H126A/9. x 1. Etymology Named after the original inhabitants of Zululand. Material Paratype is a single body chamber fragment, SAS—H126A/4 from the same locality and horizon as the holotype, and possibly belongs to the same specimen. CRETACEOUS FAUNAS FROM SOUTH AFRICA 119 Dimensions D Wb Wh Wb: Wh U H126A/9 140,0 38,7(27,6) 40,0(28,6) 0,96 74,0(52,9) Description Coiling is very evolute, with an umbilical width of 52 per cent. The whorl section varies during ontogeny, changing from subrectangular, higher than wide on the inner whorls, through subsquare with a fastigiate venter in the late phragmocone to distinctly higher than wide on the body chamber (Fig. 90A). The innermost whorls are not preserved, but at a diameter of c. 55 mm, ornament consists of a row of conical umbilical (1) tubercles connected by low, broad, prorsiradiate ribs to bullate, slightly spirally elongated ventrolateral (3-++4) nodes. The venter is distinctly tricarinate, with the central keel strongest. At a diameter of c. 65 mm the ventrolateral (3-+-4) node starts showing signs of division by the formation of a clavate node (4) on the ventral side. With increasing diameter the separation becomes progressively more obvious, and simultaneously, ornament in general becomes weaker, and the venter distinctly fastigiate, as seen at the largest diameter preserved on the holotype. No sign of lateral (2) tubercles is present at this diameter, and it is assumed that they are absent throughout. Rib density is nineteen per whorl. A large body chamber fragment, SAS-H126A/4 (Fig. 94A—B) from the same locality and horizon as the holotype, has a distinct tricarinate keel, and is here regarded as representing the adult stage of the species. The whorl section is distinctly higher than wide, and, barring a few faint, broad undulations on the flanks, devoid of lateral ornament. The suture line is simple, with a long, narrow, bifid asymmetrically lateral lobe (L.) (Fig. 90B). Discussion This species is identified by the apparent lack of a lateral (2) tubercle throughout, presence of a distinctly fastigiate venter in later stages of growth, and loss of ornament on the body chamber. Reginaites zulu is allied to R. reymenti sp. nov. in earlier stages of growth, but differs in later stages of growth in whorl section and ornament. R. durhami is superficially similar in becoming less strongly ornamented in later stages of growth, but has a distinct lateral tubercle on the phragmocone. The ontogenetic change in ornament and whorl section of R. zulu sp. nov. is analogous to that of some species of Yabeiceras, e.g. Y. orientale (Matsumoto et al. 1964: 326, fig. 2a—e). The fastigiate venter on the late phragmocone and the early bituberculate ornament of R. zulu is very much like that of Neogauthiericeras zafimahovai Collignon from the basal Campanian of Madagascar. It could be argued that through retention of the bituberculate stage throughout, and earlier acquisition of a fastigiate venter and loss of the lateral keels, R. zulu could give rise to 120 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 92. Reginaites zulu sp. nov. Holotype SAS-H126A/9. xX 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 121 Fig. 93. A-B. Reginaites zulu sp. nov. Holotype with part of phragmocone added (also shown in Fig. 92C). C. Reginaites? sp. nov.? cf. Plesiotexanites stangeri (Baily, 1855). x 1. 122 ANNALS OF THE SOUTH AFRICAN MUSEUM N. zafimahovai. The respective stratigraphic occurrences of the two species (Upper Santonian and basal Campanian) favour this relationship, but additional material is needed to confirm it. Occurrence Upper Santonian of Zululand. Reginaites ? sp. nov.? cf. Plesiotexanites stangeri (Baily, 1855) Fig. 93C Material SAM-5687 and SAM-K5546, both from Bed 7 locality 1, Umzamba Estuary, Pondoland, Transkei, Umzamba Formation, Upper Santonian. Description and discussion Parts of two whorls, probably belonging to the same individual, are of special interest in that they agree with the general ornament of P. stangeri in all respects, apart from the fact that the venter is definitely tricarinate. On the smaller specimen, SAM-5687, the venter is preserved partially as an internal cast, and partially with the shell preserved (Fig. 93C). In both examples the lateral keels are continuous. On the larger fragment, SAM-—K5546, which is part of a body chamber, the dorsum bears a distinct tricarinate impression of the previous whorl. The venter, unfortunately, is badly damaged and deformed. Part of a lateral keel is preserved at the proximal end, but this shows an undula- tion which may possibly be interpreted as a wavy lateral keel. The present material is insufficient for erection of a new species, if it is, indeed, a representative of Reginaites. The specimen occurs in the same horizon as that of P. stangeri in Pondo- land, thus it does not seem possible to derive the species directly from P. stangeri. In this connection, however, it should be borne in mind that P. stangeri already occurs as early as Mid-Santonian in Madagascar (Collignon 1966a: 62, pl. 479 (fig. 1951)) and Lower Santonian of the Gulf Coast Cretaceous (Young 1963). It is interesting to note that the specimen figured by Collignon (1966a: 62, pl. 479 (fig. 1951)) has very elongated external tubercles, from which the present speci- men may be derived. Ornament of the present specimen is not compatible with that of the other known southern African species of Reginaites, or with P. matsumotoi, from which the latter are supposedly derived. If the specimen is, indeed, a true representative of Reginaites, it would seem to suggest that Reginaites is repre- sented by two distinct, apparently independent lineages in southern Africa (Fig. 266). Occurrence Upper Santonian of Pondoland, Transkei. CRETACEOUS FAUNAS FROM SOUTH AFRICA 123 Fig. 94. Reginaites zulu sp. nov. Paratype SAS—H126A/4. Part of the body chamber. x 0,8. 124 ANNALS OF THE SOUTH AFRICAN MUSEUM Genus Texanites Spath, 1932 Type species Ammonites texanus Roemer, 1852, by the original designation of Spath (1932: 379 footnote). Diagnosis Size of shell variable; may grow to enormous size. Coiling evolute to involute. Ornament consists from a relatively early stage of five rows of tubercles, with the lateral (2) row appearing last in ontogeny. Ornament is generally variable. Discussion In terms of numbers, the genus Texanites is the most important texani- tine taxon in the Upper Cretaceous deposits of southern Africa. Specimens referred to this genus display a bewildering array of intraspecific variation, which causes one to cast serious doubt on current texanitid systematics. In his review of the genus, Matsumoto (1970: 270) listed the following species definitely referable to Texanites: Texanites dichotomus Collignon (1948: 80, pl. 9(3) (fig. 3—-3b), pl. 11(5) (fig. 1-1b)) from the Upper Santonian of Madagascar. Texanites gallicus Collignon (1948: 75, pl. 8(2) (fig. 1—-1a), text-fig. 9) from the Santonian of Madagascar, South America, and Europe. Texanites hispanicus Collignon (1948: 76, pl. 8(2) (fig. 2-2b), text-fig. 11-1 1a) from the Santonian of Spain and Madagascar. Texanites hourcgi Collignon (1948: 78, pl. 7(1) (fig. 1-1b), pl. 10(4) (fig. 1-1a)) from the Middle Santonian of Madagascar. Texanites interpositus Collignon (1948: 82, pl. 7(1) (fig. 3), pl. 11(5) (fig. 2)) from the Upper Santonian of Madagascar. Texanites pseudotexanus (de Grossouvre) (1894: 84) from the Santonian of Germany and Bohemia. Texanites quinquenodosus (Redtenbacher) (1873: 108, pl. 24 (fig. 3a—b)) from the Santonian of Austria, France, Madagascar, and Angola, and tentatively from Hokkaido. Texanites roemeri (Yabe & Shimizu) (1923: 2) from the Santonian of Texas. Texanites texanus (Roemer) (1852: 31, pl. 3 (fig. la—c) non fig. le-d = T. roemeri Yabe & Shimizu) from the Santonian of North America and north Africa. Texanites venustus Collignon (1948: 81, pl. 9(3) (fig. 4-4b)) from the Upper Santonian of Madagascar. In addition to these definite contenders, a number of species based on incomplete material, generally lacking the inner whorls, were referred to the genus: Texanites americanus (Lasswitz) (1904: 32, pl. 8 (fig. 1)) from the Lower to Middle Santonian of the Gulf Coast. CRETACEOUS FAUNAS FROM SOUTH AFRICA 125 Texanites lonsdalei Young (1963: 90, pl. 34 (fig. 1), pl. 51 (figs 3-7), pl. 58 (figs 5-6), text-fig. 22a—d) from the Lower Campanian of Texas and Madagascar. Texanites mikobokensis Collignon (1966a: 122, pl. 507 (fig. 2019), pl. 519 (fig. 2019)) from the Upper Santonian of Madagascar. Texanites oliveti (Blanckenhorn) (1905: 104) from the Santonian of the Middle East. Texanites quadrangulatus Collignon (1966a: 76, pl. 486 (fig. 1961)) from the Upper Santonian of Madagascar. Texanites rarecostus Collignon (1966a: 78, pl. 487 (fig. 1965)) from the Middle Santonian of Madagascar. To this list of doubtful contenders may be added: Texanites angolanus Haas (1942: 12, figs 8-10, 11) from the Santonian of Angola. Texanites omeraense (Reeside) (1927: 38, pl. 42 (figs 3-4), pl. 43 (figs 1-2)) from the Santonian of the Western Interior of North America. Texanites ralijaonai Collignon (1966a: 130, pl. 511 (fig. 2023)) from the San- tonian of Madagascar. New species from South Africa, described below are: Texanites collignoni sp. nov. Texanites postvanhoepeni sp. nov. Texanites presoutoni sp. nov. Texanites umzambiense sp. nov. Texanites vanhoepeni sp. nov. In addition to displaying tremendous intraspecific variation, the Zululand Texanites species connect with Paratexanites and Submortoniceras. Texanites collignoni may be derived from Plesiotexanites collignoniforme sp. nov., which in turn has its origin in Paratexanites. Texanites soutoni 1s derived, probably through 7. presoutoni sp. nov., from T. umzambiense sp. nov., which is closely allied to, though apparently slightly younger than, Plesio- texanites collignoniforme sp. nov. Texanites quadrangulatus and T. rarecostus appear to connect with Paratexanites, without passing through a Plesiotexanites stage, via Texanites vanhoepeni sp. nov., the latter probably derived from Para- texanites umkwelanensis. Texanites soutoni gradually gives rise to Submortoniceras towards the Santonian/Campanian boundary, and Texanites vanhoepeni and T. postvan- hoepeni also show a distinct tendency towards acquisition of Submortoniceras characteristics in stratigraphically younger strata. This takes place through a gradual decrease in umbilical width and loss of lateral ornament. Details are given in the descriptions of the species. Similar trends are shown by Texanites shiloensis and Submortoniceras tequesquitense from the Gulf Coast of North America, which may possibly represent a geographically separated relative of the species group of Texanites soutoni and Submortoniceras woodsi. 126 ANNALS OF THE SOUTH AFRICAN MUSEUM The latter species group also shows what appears to be a distinct geographic differentiation into two groups inhabiting the deep and shallow waters of the St Lucia Formation and Umzamba Formation respectively. Texanites is distinguished from Plesiotexanites by virtue of the early development of the pentatuberculate stage, although, in cases such as T. umzam- biense sp. nov. and P. collignoniforme sp. nov., the distinction between the two taxa becomes subjective. None the less, Plesiotexanites should be maintained separate from Texanites as a distinctive group which provides the source for Reginaites and, in some cases, Texanites. In large phragmocone or body chamber fragments it is impossible to distinguish between Texanites and Plesiotexanites, unless the ventrolateral tubercles are approximated in Paratexanites fashion. As mentioned above, differences between Texanites and Submortoniceras are gradual, hence allocation of border-line cases such as 7. soutoni to either of the genera becomes a matter of opinion, depending on where to draw the line between Texanites and Submortoniceras. Texanites collignoni sp. nov. Figs 95, 96A, 97-98, 99B-C, 100-101 Texanites oliveti non Blanckenhorn: Collignon, 1948: 72, pl. 8(2) (figs 3-5), text-figs 5-8. Texanites oliveti non Blanckenhorn var. spinosa Collignon, 1966a: 68, pl. 482 (fig. 1955). ?Texanites oliveti (Blanckenhorn): Matsumoto & Ueda 1962: 169, pl. 26 (fig. 2), text-fig. 13. Holotype SAM-PCZ5920 (Fig. 98), locality 74, north-western part of Die Rooiwalle, Zululand, St Lucia Formation, Santonian I-II. Etymology Named after the late General Maurice Collignon, Moirans, France. Material Paratypes are SAS—H179/19, locality 6, excavations in the Enseleni Reserve, Zululand, St Lucia Formation, Santonian II-III; SAS—Z866, Z1754, Z1755, BMNH-C81551—C81558, locality 74, north-western part of False Bay at Die Rooiwalle, Zululand, St Lucia Formation, Santonian I-II; SAS-—Z351, from an unknown locality near Mfolozi, probably locality 14, St Lucia Forma- tion, Santonian II to Campanian I; BMNH-C81507—C81508, from foreshore exposures west of Lister’s Point, Zululand, St Lucia Formation, Santonian I-II; BMNH-C81515, from foreshore exposures north of Picnic Point, St Lucia Formation, Santonian I-II; a body chamber fragment NMB-D1353, locality 73, St Lucia Formation, Santonian I, may also be referred to this species; SAM-— PCZ5711, labelled ‘S.A. Cretaceous’ also belongs to the species, and, judging by the matrix, this specimen may have been collected in the vicinity of locality 74 at Die Rooiwalle. 127 CRETACEOUS FAUNAS FROM SOUTH AFRICA "Il X “PSLIZ-SVS odAzered Jo UOr}Ses JJOUM ‘@ ‘SOIOW[IUT UT oINyNS IOJ Ieq BJeOG “SSLTZ-SVS odAyered Jo oul] omNyNg “y “AOU “ds juoUsIjOI SajtuDxXAaT °C6 “SI q V 128 ANNALS OF THE SOUTH AFRICAN MUSEUM Dimensions D Wb Wh Wb: Wh U Int. Ext. H179/19 — 31,6 43,0 0,73 Z866 255,0 66,0(29,3) 86,0(33,7) 0,77 116,0(45,5) 32 32 Z1754 257,0 59,0(22,9) 89,0(34,6) 0,66 117,0(45,5) Mil 33 Z351 213,0 51,0(23,9) 66,0(30,9) 0,77 106,0(49,8) 32 32 FAAS 230,0 56,5(24,6) 79,0(34,3) 0,72 99,0(43,0) —_— es Description Coiling is evolute, with an umbilical width varying between 43 and 49 per cent of the diameter, but usually of the order of 45 per cent. Whorl overlap is minimal, and the dorsal zone of impression consequently small. The whorl section in all the available specimens is higher than wide through- out, but differs in details from one specimen to another, and also throughout ontogeny (Fig. 95B, 96A). The umbilical wall is virtually vertical on the inner whorls, but becomes rounded on the outer whorls, merging gently into the gently inflated flanks. The flanks converge gradually to a rounded venter, narrower than the dorsum. In intercostal section maximum width is towards Fig. 96. A. Texanites collignoni sp. nov. Whorl section of paratype SAS-Z1755. B. Uncorrected sketches of whorl section of Texanites oliveti (Blanckenhorn, 1905) from Israel. x 0,5. CRETACEOUS FAUNAS FROM SOUTH AFRICA 129 the dorsal third of the flanks, coinciding with the position of the lateral tubercle. None of the specimens has the innermost whorls preserved, and the earliest whorls preserved already show pentatuberculate ornament. Typical ornament, as shown by SAS—Z866 (Fig. 97) and SAM—PCZ5920 (Fig. 98), consists of radial to slightly rursiradiate single ribs, each bearing a rounded to radially elongated umbilical (1) tubercle, displaced slightly ventrally Fig. 97. Texanites collignoni sp. nov. Paratype SAS-—Z866. x 0,58. 130 ANNALS OF THE SOUTH AFRICAN MUSEUM from the umbilical edge, a prominent radially elongated, lateral (2) tubercle towards the dorsal third of the flanks (this tubercle juts out furthest laterally though it is not necessarily strongest developed), and progressively clavate submarginal (3), marginal (4) and external (5) tubercles over the rest of the flanks. With increasing diameter, and especially towards the body chamber, the umbilical and the lateral tubercles move ventrally, so that the former is some distance from the umbilical region and the latter nearer midflank. In typical forms ribbing is single, but some specimens feature bifurcations. The keel is undulating; undulations coinciding with lateral ornament. Fig. 98. Texanites collignoni sp. nov. Holotype SAM-—PCZ5920. x 0,5. CRETACEOUS FAUNAS FROM SOUTH AFRICA 131 Fig. 99. A. Texanites oliveti (Blanckenhorn, 1905). Specimen from Judean desert. (Photo- graph Z. Lewy.) B-—C. Texanites collignoni sp. nov. Paratype SAS—H179/19. This specimen remains rectangular to a large diameter. x 1. 132 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 100. Texanites collignoni sp. nov. Paratype SAS-Z1754. x 0,5. Discussion This species is common in the Lower Santonian of Zululand, and the authors (Kennedy & Klinger 1975: 279) initially identified it with Texanites oliveti (Blanckenhorn). Texanites oliveti, however, is a very poorly known species, as the figures provided by Taubenhaus (1920: 30, pl. 5 (figs 3, 5)) are very much reduced. According to Blanckenhorn (1905: 104) the species is identified by: 26 to 28 strong ribs, slightly narrower than the interspaces, simple, single, radial and slightly forwardly curved on the ventral third of the CRETACEOUS FAUNAS FROM SOUTH AFRICA 133 flanks. The undulations on the keel correspond with the ribs. The ribs bear five rows of tubercles. . . . The umbilical tubercles and the two outer rows are strongest developed, while the second (lateral) row is always poorest developed. (Authors’ free translation.) This description fits the material rather well, but Z. Lewy (pers. comm. 1979) has informed the authors that, according to material from Israel, T. oliveti is a typically compressed form, with Wb : Wh ratio being of the order of 0,4 to 0,5 and seems to occur in the Upper Santonian only. A typical specimen from the Judean desert is here figured as Figure 99A, and whorl sections of other Israeli material as Figure 96B. Apart from the fact, thus, that the Zululand and Mala- gasy specimens are older than typical T. oliveti, they are also much more inflated, with Wb : Wh ratios of the order of 0,7. It seems advisable to refer the Zululand and Malagasy material to a new species, JT. collignoni sp. nov., to separate it from the younger, more compressed 7. oliveti. The specimen from the Lower Santonian of the Himenoura Group described by Matsumoto & Ueda (1962: 170) as Texanites oliveti (Blanckenhorn) also probably belongs to this species. Collignon (1948: 72-75) discussed T. collignoni (as T. oliveti) at length, although his material consisted of small fragments only. Apart from the typical form, which has a more or less rectangular section, Collignon recognized two other varieties, var. spinosa and var. triangularis. Var. spinosa has more or less the same whorl section as the typical form, but differs in that the tubercles of the first three rows are rounded, and so strong that the intertubercular areas are as deep as the intercostal spaces. In var. triangularis the ornament is essen- tially the same as in the typical form, but the section is triangular. As far as tuberculation is concerned, the Zululand material compares well with Collignon’s descriptions and figures of the Malagasy material; as far as whorl section is concerned, the Zululand specimens are somewhat different in having a steeper or more rounded umbilical wall. The more complete Zulu- land specimens show that whorl section changes during growth, starting with rectangular section on the inner whorls, eventually becoming subrectangular to ovoid on the later parts of the phragmocone and body chamber. One of the specimens, however, SAS—H179/19 (Fig. 99B) retains the rectangular whorl section to a greater diameter than the other specimens. In most of the specimens ribbing is single on the outer whorls. In SAS- Z1754 (Fig. 100), however, bifurcations are quite common on the outer whorls. Apart from being more compressed than the other specimens, there seems to be no other significant difference, and the specimen is also referred to T. collignoni sp. nov. It appears to connect morphologically with T. soutoni, described below. Differences between T. collignoni and T. texanus are distinct in density of ribbing in typical forms, but become obscure when comparing T. collignoni with T. texanus var. hispanica. Here crenulation of the keel may be of assistance in differentiating between the species. T. quinquenodosus (Figs 102-103) is a species with similar ribbing and coiling, but differs in that tuberculation is generally clavate in all the rows. ANNALS OF THE SOUTH AFRICAN MUSEUM 134 Paratype SAS-Z1755. x 0,6. i sp. NOV. ignoni i Texanites coll ig. 101. F CRETACEOUS FAUNAS FROM SOUTH AFRICA 135 Even though all Redtenbacher’s (1873) specimens were crushed, it appears as though the whorl section is more compressed than in T. collignoni. This raises the possibility that T. oliveti may be a junior synonym of T. quinquenodosus. The rounded umbilical wall, ventrally displaced umbilical tubercle, and undulating keel in T. collignoni are very similar to Plesiotexanites collignoniforme sp. nov. described above (p. 89), and it seems reasonable to derive T. col- lignoni from that species. Schloenbachia (Mortoniceras) sandreczkii Blanckenhorn (1905: 105) may be allied, but is difficult to interpret. In this species only the external row of tubercles is developed, and it seems closer to Submortoniceras than Texanites. Schloenbachia (Mortoniceras) safedensis Blanckenhorn (1905: 106) is another enigmatic species and may possibly represent a compressed specimen of T. oliveti or a closely allied species. Occurrence Lower and Middle Santonian of Madagascar, Santonian I-II of Zululand, Lower Santonian of Japan. [While this article was in press, a publication by Matsumoto & Haraguchi (Matsumoto, T. & Haraguchi, Y. 1978. A new Texanitine Ammonite from Hokkaido. Trans. Proc. Palaeont. Soc. Japan, N.S. 110: 306-318) which has bearing on the discussion of Texanites oliveti, came to the notice of the authors. Matsumoto & Haraguchi suggest that Texanites oliveti is a good index species for the Lower Santonian, but also express doubt whether the specimensdescribed by Collignon from Madagascar are, indeed, conspecific with the Middle East specimens of 7. oliveti, and whether the latter is distinguishable from T. quin- quenodosus. In view of our current knowledge and interpretation of Texanites oliveti, it is not a suitable index species for the Lower Santonian. Pop & Szasz (Pop, G. & Szasz, L. 1973. Santonianul din Reguinea Hategului (Carpatii meridionali). St. cerc. geol., geofiz., geogr., Ser. geologie 18: 463-467) also described and figured a specimen as Texanites oliveti from Bulgaria, which seems identifiable with T. collignoni.] Texanites quinquenodosus (Redtenbacher, 1873) Figs 102-103 Ammonites texanus Roemer: von Hauer, 1858: 10, pl. 2 (figs 4-6). Ammonites quinquenodosus Redtenbacher, 1873: 108(18), pl. 24 (fig. 3a—b). Mortoniceras texanus (Roemer): Ziircher, 1903: 686. Mortoniceras quinquenodosus (Redtenbacher): Yabe & Shimizu, 1923: 30. Texanites quinquenodosus (Redtenbacher): Collignon, 1948: 69(24), text-fig. 2; 1966a: 128, pl. 510 (fig. 2021). Matsumoto 1970: 272. Thomel, 1969: 12(120). Discussion A full review of this species, based on all the original type material, is in preparation by the authors. One of the syntypes, now housed in the 136 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 102. Texanites quinquenodosus (Redtenbacher, 1873). Plaster cast of one of Redtenbacher’s syntypes from the Oberésterreichisches Landes- museum, Linz. x 0,65. CRETACEOUS FAUNAS FROM SOUTH AFRICA 137 Fig. 103. Texanites quinquenodosus (Redtenbacher, 1873). Plaster cast of one of Redten- bacher’s syntypes from the Oberdsterreichisches Landesmuseum, Linz. x 0,65. 138 ANNALS OF THE SOUTH AFRICAN MUSEUM Oberosterreichisches Landesmuseum, Linz (Austria), is here figured photo- graphically (Figs 102-103) for comparison with the South African material. Texanites quadrangulatus Collignon, 1966 Figs 104-108 Texanites quadrangulatus Collignon, 1966a: 76, pl. 486 (fig. 1961). Holotype The specimen figured by Collignon (1966a, pl. 486 (fig. 1961)) from the Upper Santonian of Madagascar, Zone of Pseudoschloenbachia umbulazi. Material SAS-A2580, SAS-—Z1760, SAM-—PCZ5879-PCZ5882, BMNH-C81513- C81514, all from locality 74, False Bay, Zululand, St Lucia Formation, San- tonian I to Campanian I; SAS-H179/18, locality 6, Enseleni Reserve, Zululand, St Lucia Formation, Santonian II to Campanian I; SAS—Z595, locality 14, Mfolozi, Zululand, St Lucia Formation, Santonian IJ-IIJ; BMNH-C81511, locality 84, False Bay, St Lucia Formation, Santonian I. Dimensions D Wb Wh Wb: Wh U Int. Ext. H179/18 — 27,0 27,0 1,0 — = — PCZ5879 125,0 48,0(38,4) 51,0(40,8) 0,94 46,0(36,8) 11x2 — PCZ5880 142,0 54,0(38,0) 54,0(38,0) 1,0 58,0(40,8) 25 26 Z1760 175,0 65,0(37,1) 71,0(40,6) 0,91 61,0(34,9) 13x2 — A2580 245,0 76,0(31,0) 93,0(38,0) 0,82 93,0(38,0) 29 29 Holotype after Collignon 1966a: 76 145,0 50,0(34,0) 50,0(34,0) 1,0 59,0(41,0) — ?30 Description Coiling is relatively evolute, with an umbilical diameter of 35 to 41 per cent of the total diameter. Whorl overlap is minimal with the marginal (4) tubercles clearly exposed in the umbilical seam. Increase in diameter is rapid. The costal whorl section on the phragmocone is distinctly quadrate, sometimes even wider than high in early stages, whereas the intercostal section is more rounded (Figs 106B, 108). Towards the later part of the phragmocone the whorl section becomes more rounded (Fig. 107B), higher than wide, and eventually dome-shaped on the body chamber (Fig. 105A). Ornament on the phragmocone consists of prominent ribs bearing very strong tubercles and separated by wider interspaces. The umbilical (1) tubercles are conical to radially elongated, pointed slightly inwards and with bases extending a little distance down the umbilical wall. The lateral (2) tubercles are situated at midflank or on the dorsal third of the flanks and are conical to clavate, and separated from the umbilical tubercles by a depression on the ribs. The sub- marginal (3), marginal (4), and external (5) tubercles are situated closer to each CRETACEOUS FAUNAS FROM SOUTH AFRICA 139 Fig. 104. Texanites quadrangulatus Collignon, 1966. SAS-A2580. x 0,58. other than the umbilical and lateral ones throughout ontogeny, but especially on the inner whorls. Here the submarginal tubercles may be tremendously enlarged and partially fused to the marginal tubercles, having the appearance of Paratexanites. The external (5) tubercles are very clavate and generally as high as, or higher than, the central keel. Bifurcations occur, but not frequently. Towards the later part of the phragmocone and the body chamber, all ornament becomes subdued (Fig. 104). ANNALS OF THE SOUTH AFRICAN MUSEUM 140 ipa, MO map yy 1966. SAS-A2580. x 0,58. ignon, 105. Texanites quadrangulatus Coll ig. F CRETACEOUS FAUNAS FROM SOUTH AFRICA 141 B Fig. 106. Texanites quadrangulatus Coliignon, 1966. SAM-PCZ5881. Suture line, x 5. Whorl section, x 1. The external suture comprises a blocky, wide E/L saddle, narrow L/U2 saddle and long, narrow lateral lobe (L), somewhat restricted towards the top (Fig. 106A). Discussion The quadrate whorl section on the greater part of the phragmocone, strong ornament and, depending on preservation, strong development of the submarginal tubercle and Paratexanites-like spacing of the submarginal (3), ANNALS OF THE SOUTH AFRICAN MUSEUM 142 TX “6LESZOU-INVS “9961 “WOUsTTIOD smnjnSuviponb sayuoxaL “LOL ‘Std 143 CRETACEOUS FAUNAS FROM SOUTH AFRICA TV Wy SAM-PCZ5881. . A-B 1966 ignon Texanites quadrangulatus Colli 108 Fig. C-E. SAS-H179/18. x 1. 144 ANNALS OF THE SOUTH AFRICAN MUSEUM marginal (4), and external (5) tubercles are all characteristic of the species. T. vanhoepeni is a similar, though older, species. Here, however, the rate of increase of diameter is greater than in 7. quadrangulatus, the whorl section is more rounded in the adult stage, and it lacks the strong ornament. Matsumoto (1970: 272) speculated on the origin of 7. quadrangulatus, suggesting that it could be derived from Protexanites bourgeoisi or Paratexanites serratomarginatus grossouvrel. Present material suggests that it would seem more feasible to derive T. quadrangulatus from Paratexanites pseudotricarinatum or P. umkwelanense, possibly via 7. vanhoepeni. Occurrence Middle and Upper Santonian of Madagascar and Zululand. Texanites vanhoepeni sp. nov. Figs 109-117 Holotype SAS-Z879 (Figs 109-110), locality 83, foreshore exposures at Mason’s Camp, south-western part of False Bay, Zululand, St Lucia Formation, Coni- acian IV. Etymology Named for the late Dr E. C. N. van Hoepen. Material Paratypes are SAS—Z1792, Z2110, H201/33, and BMNH-C81534—-C81535, locality 84, beach exposures in the south-western part of False Bay, Zululand, St Lucia Formation, Santonian I; BMNH-—C81509—-C81510, C81521, C81537, locality 85, Santonian I; SAS—H200/92 from the same horizon and locality as the holotype; NMB-D1364 and D1379, from the same region at locality 86, St Lucia Formation, Coniacian V to Santonian I or II; NMB—D1353, locality 73, the lower reaches of the Mzinene River, St Lucia Formation, Coniacian to Santonian; and BMNH-C81559, locality 74, St Lucia Formation, Santonian I-II. Dimensions D Wb Wh Wb: Wh U Int. Ext. Z879 183,0 69,0(37,7) 71,0(38,8) 0,97 70,0(38,2) 24 24 H200/92 192,0 _ 67,0(34,9) — 79,0(41,1) 28 28 Description Coiling is evolute, with rapidly increasing whorls (Figs 111-112). The whorl section is more or less equidimensional throughout, but changes from subquadrate to rounded during growth, especially at the umbilical edges (Fig. 113A). CRETACEOUS FAUNAS FROM SOUTH AFRICA 145 Fig. 109. Texanites vanhoepeni sp. nov. Holotype, SAS-Z879. x 0,86. 146 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 110. Texanites vanhoepeni sp. nov. Holotype, SAS-Z879. x 0,8. CRETACEOUS FAUNAS FROM SOUTH AFRICA 147 Fig. 111. Texanites vanhoepeni sp. nov. Paratype SAS-H201/33. x 1. None of the specimens has the very innermost whorls preserved to show the ontogeny. Ornament on the phragmocone consists of sparse, single radial ribs bearing prominent tubercles. Rib density is generally of the order of 24, but may be as high as 28. The umbilical (1) tubercles are radially elongated, pointing slightly inwards. The lateral (2) tubercles at the dorsal third of the flanks are usually very prominent and conical to slightly clavate. The submargi- nal (3) at the ventral third of the flanks is clavate and generally smallest. 148 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 112. Texanites vanhoepeni sp. nov. Paratype SAS—H201/33. x 1. The marginal (4) and external (5) tubercles are very clavate, and, in some ases, situated closer to each other than the submarginal (3) and marginal (4). These former tubercles may be strongly developed, and protrude far above the venter. The keel is very low and poorly developed, but distinctly undulating, in some cases forming a row of weakly clavate tubercles along the siphonal line corresponding in location to the ribs. The suture has a very long and thin lateral lobe (L) (Fig. 113B). CRETACEOUS FAUNAS FROM SOUTH AFRICA 149 l J ! I l ! | l 1 | \ ao =- oe B Fig. 113. Texanites vanhoepeni sp. nov. A. Whorl section of para- type SAS—H201/33. B. Partial suture line of paratype SAS—Z2111. x 1. 150 ANNALS OF THE SOUTH AFRICAN MUSEUM On the body chamber ornament remains prominent, but becomes wider spaced (Figs 114-115). Discussion This species is easily distinguished by the rapid increase in whorl diameter, generally sparse costation and undulating keel. Unfortunately, the available material of this species is limited, and accurate stratigraphic data scant. There does, however, seem to be a tendency towards Fig. 114. Texanites vanhoepeni sp. nov. Paratype NMB-D1364. Specimen illustrating wider spacing of ornament towards and on body chamber. x 0,5. 151 CRETACEOUS FAUNAS FROM SOUTH AFRICA Body chamber fragment. ,66 Paratype NMB-D1379 x 0 Fig. 115. Texanites vanhoepeni sp. nov. 152 ANNALS OF THE SOUTH AFRICAN MUSEUM tighter coiling higher up in the sequence, giving rise to a Submortoniceras-like form, Texanites postvanhoepeni sp. nov., described below. This tendency in coiling is illustrated by SAS—H200/92 (Fig. 116) with widest umbilical width at 41 per cent, through the holotype SAS—Z879 (Fig. 109) with umbilical width of 38 per cent, to SAS—Z2100 (Fig. 117) with an umbilical diameter of about 33 per cent. The rounded whorl section in the adult stage, and the undulating keel relate this species to Texanites collignoni, Plesiotexanites collignoniforme, Para- texanites umkwelanense and P. pseudotricarinatum. Stratigraphic data seem to suggest that 7. vanhoepeni may be derived from P. umkwelanense, or some common ancestral form. Apart from the lateral tubercle, the whorl section of the two species is remarkably alike, as is the prominence of the marginal (4) and external (5) tubercles in cases. Unfortunately, the very innermost whorls are not preserved, but at a — a Fig. 116. Texanites vanhoepeni sp. nov. Paratype SAS—H200/92. Specimen with greatest umbilical width. x 0,55. 153 CRETACEOUS FAUNAS FROM SOUTH AFRICA ecimen with smallest Sp e width. x 1 Paratype SAS-Z2110 ] lica p. nov umbi US Texanites vanhoepen Fig. 117. 154 ANNALS OF THE SOUTH AFRICAN MUSEUM diameter of 45 mm lateral tubercles are already present, indicating that this is probably a Texanites and not referable to Plesiotexanites. This suggests that in this case Texanites s.s. was derived directly from Paratexanites without a Plesio- texanites stage as in the case of T. collignoni (p. 126). Texanites collignoni differs in having a completely different whorl section and much slower expanding whorls, as does Plesiotexanites collignoniforme. As far as lateral ornament is concerned, there is close resemblance to Texanites ralijaonai from the Upper Santonian of Bevaho (Belo sur Tsiribihina), Madagascar. In that species, however, increase in whorl diameter is not as rapid as in the present species as Collignon’s (1966a: pl. 512 (fig. 2023)) figure shows. That species also lacks an undulating keel. Texanites rarecostus Collignon from Zululand and Madagascar has similar sparse costation, c. 23-24 per whorl, but has less massive whorls and a very slow rate of whorl increase. Texanites quadrangulatus has proportions and ornament which overlaps with that of the present species to a certain extent, and it is possible that they are genetically related. 7. quadrangulatus, however, generally has less massive whorls, a slower increase in diameter, a smaller lateral tubercle and has the three ventral rows approximated. Occurrence Upper Coniacian to Lower Santonian of Zululand. Texanites postvanhoepeni sp. nov. Figs 118-120 Holotype SAM-PCZ5902 (Fig. 118), locality 85 or 86, foreshore exposures at Mason’s Camp, south-western part of False Bay, Zululand, St Lucia Formation, Coni- acian V to Santonian I and II. Etymology Refers to possible derivation from Texanites vanhoepeni. Material SAS-Z1787 and BMNH-C81522 from the same locality and horizon as the holotype. Dimensions D Wb Wh Wb: Wh Of Int. Ext. PCZ5902 182,0 68,0(38,2) 80,0(43,9) 0,85 59,0(32,4) 20 32 Z1787 220,0 84,0(38,2) 96,0(43,6) 0,87 67,0(30,4) — — Description Coiling is involute, with an umbilical width of 30 to 32 per cent, and very rapidly increasing whorls. The dorsal zone of impression, however, is very CRETACEOUS FAUNAS FROM SOUTH AFRICA 155 shallow, and successive whorls cover each other only up to the middle of the marginal (4) tubercle. The whorl section throughout, as far as visible, is higher than wide, with an overhanging to vertical umbilical wall, flanks parallel up to the lateral (2) tubercle, which marks the greatest width, and then converging slowly to the marginal (4) and eventually the external (5) tubercle (Fig. 120A). On the innermost whorls preserved, ribs bifurcate quite frequently. Small, Fig. 118. Texanites postvanhoepeni sp. nov. Holotype SAM—PCZ5902. x 0,67. 156 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 119. Texanites postvanhoepeni sp. nov. Paratype SAS—Z1787. x 0,67. conical umbilical (1), *lateral (2), submarginal (3), and clavate marginal (4) tubercles are exposed on the flanks. Beyond a diameter of c. 70 mm, all lateral ornament weakens progressively, though never disappearing completely. The suture is incompletely exposed, but shows a subtrifid, long and narrow lateral lobe (L). CRETACEOUS FAUNAS FROM SOUTH AFRICA 157 Fig. 120. Texanites postvanhoepeni sp. nov. Paratype SAS-Z1787. x 0,67. 158 ANNALS OF THE SOUTH AFRICAN MUSEUM Discussion As mentioned above (p. 152), this species appears to be the end member in a series of specimens with progressively tighter coiling. Starting with an evolute form of 7. vanhoepeni sp. nov. with an umbilical diameter of 41 per cent, it terminates with the present species, with an umbilical width of only 30 to 32 per cent. Unfortunately precise stratigraphic data are lacking to fully confirm this assumption, but it would be in line with the general recoiling pattern observed in 7. soutoni and allied species (described below, p. 172). As in the latter, the dividing line here between Texanites and Submortoniceras becomes arbitrary. Because the holotype still has weak lateral tubercles present at the largest diameter preserved, the species is referred to Texanites rather than Submortoniceras. Texanites quadrangulatus has a very similar shape, but differs mainly in having a very prominent quadrate whorl section on the early part of the phrag- mocone, much stronger tuberculation and a slightly wider umbilicus (34-40 %). T. postvanhoepeni is similar to Submortoniceras woodsi, but in the latter the whorl section is more compressed, and coiling looser. Occurrence Lower to Middle Santonian of Zululand. Texanites rarecostus Collignon, 1966 Figs 121-122 Texanites rarecostus Collignon, 1966a: 78, pl. 487 (fig. 1965). Type Holotype by monotypy is the specimen figured by Collignon (1966a: pl. 487 (fig. 1965)) from the Middle Santonian of Beantaly-Souromaraino, Madagascar. Material SAS-Z596, locality 14, Mfolozi, Zululand, St Lucia Formation Santonian II and III to Campanian I; and SAM-—PCZ5878 from an unknown locality, presumably locality 74, the north-western end of False Bay, Zululand, St Lucia Formation, Santonian to Campanian. Dimensions D Wb Wh Wb: Wh U Int. Ext. Z596 120 31,0(25,8) 34,0(28,3) 0,91 5910492) 13x23 Beez 64,0 18,5(28,9) 20,0(31,3) 0,92 30,0(46,9) 13x2 13x2 PCZ5878 102 30,0(29,4) 33,0(32,4) 0,91 50,0(49,0) 11x2 11x2 Holotype after Collignon 128 38,0(30,0) 40,0(31,0) —0,95 62.0(48.0) 23 24 CRETACEOUS FAUNAS FROM SOUTH AFRICA 159 Description Two incomplete specimens with part of the body chamber preserved are tentatively referred to Collignon’s species. Coiling is very evolute with an umbilical width of 46 to 49 per cent of the diameter. Whorl overlap is very small with the marginal tubercles clearly exposed in the umbilical suture. The intercostal whorl section is rounded with maximum width at midflank A B Fig. 121. Texanites rarecostus Collignon, 1966. SAS—Z596. A x 1,05; B x 1,3. 160 ANNALS OF THE SOUTH AFRICAN MUSEUM or at the umbilical edge. Costal whorl section is subquadrangular, slightly higher than wide with maximum whorl breadth at the umbilical (1) tubercle or lateral (2) tubercle, but later migrating towards the submarginal (3) tubercle. SAS-Z596 shows part of the early ontogeny. At a diameter of 14 mm ornament consists of regularly bifurcating thin ribs displaying conical umbi- lical (1) and submarginal (3) tubercles on the exposed flank. At a diameter of c. 32 mm ribs still bifurcate regularly, rib density being about 11 (bifurcating ribs) per half whorl. The submarginal (3) tubercles have now migrated towards the ventral quarter or third of the exposed flanks, exposing the marginal (4) tubercles in the umbilical suture. Lateral (2) tubercles start appearing at a diameter of c. 40 mm, and at this stage the four exposed rows are arranged equidistantly on the flanks. Bifurcations occur less frequently, and all ornament coarsens, resulting in a lower rib density. In the adult stage the umbilical (1) and lateral (2) tubercles are distinctly conical, while the remaining three rows are clavate. The keel is entire, and generally lower than the external tubercles. The suture has massive saddles and a U-shaped lateral lobe (L). Discussion The holotype of the species consists of slightly more than a whorl and is difficult to interpret. The Zululand specimens have comparable relative propor- tions, but seem to have more robust ornament. In addition, SAS—Z596 has distinctly prorsiradiate ribs on the outer whorl compared to rectiradiate on the holotype. On the basis of the limited material available, it is not known whether these differences are of any significance. Differences between J. rarecostus and T. americanus are not very clear. The outer whorl of SAS—Z596 (Fig. 121) is very similar to the holotype of T. americanus (see Lasswitz 1904, pl. 8 (fig. 1)). The inner whorls of the holo- type, however, are not as densely costate as those of the Zululand specimen. Young (1963: 83-4) suggested that Lasswitz’s artist may have erred in drawing the inner whorls of the holotype. Even Young’s material, however, is more coarsely costate on the inner whorls (compare Young 1963: pl. 44 (fig. 2)). On the other hand, if density of ribbing on the inner whorls of Plesiotexanites is taken into consideration, this difference seems very trivial indeed. Of more importance seems to be the difference in whorl section, that of 7. americanus being more inflated than that of T. rarecostus. For the present the two species are retained apart, in cognizance of their close resemblance. Plesiotexanites stangeri is a similarly evolute species, but differs in the late development of the lateral tubercle and in generally lacking the strong ornament on the body chamber of T. rarecostus. Small specimens of the inner whorls of T. vanhoepeni sp. nov. may be mistaken for T. rarecostus in having a similar whorl section, but in more com- plete material the rapid increase in size is apparent. Another allied species is T. quadrangulatus, but in the latter the whorl CRETACEOUS FAUNAS FROM SOUTH AFRICA YS \S\ \\S Fig. 122. Texanites rarecostus Collignon, 1966. SAM-—PCZ5878. x 161 162 ANNALS OF THE SOUTH AFRICAN MUSEUM section is more quadrate at larger diameters, and increase in size is also much more rapid. Occurrence T. rarecostus occurs in the Middle Santonian of Madagascar, Zone of Texanites hourcgi and Santonian of Zululand. Texanites texanus s.\. (Roemer, 1852) Figs 123-125 Texanites texanus (Roemer): Collignon, 1948 : 66(20), text-fig. 1-1b (with synonymy); 1966b: 52, pl. 33. Texanites texanus var. gallica Collignon, 1948: 75(30), pl. 8(2), (fig. 1-1a), text-figs 9-10. Texanites texanus var. hispanica Collignon, 1948: 76(31), pl. 8(2) (fig. 2—2b), text-fig. 11-11a. Texanites texanus texanus (Roemer): Young, 1963: 80, pl. 38 (figs 1-2), pl. 40 (figs 1-3), pl. 41 (fig. 4), text-figs 21g, 22e, 25d. Texanites texanus gallica Collignon: Young, 1963: 82, pl. 38 (figs 3-4). Texanites texanus twiningi Young, 1963: 82, pl. 38 (fig. 5), pl. 39 (fig. 1), pl. 41 (figs 2, 5), pl. 48 (fig. 4). Texanites gallicus Collignon, 1966a: 78, pl. 487 (fig. 1964). Texanites hispanicus Collignon, 1966a: 128, pl. 510 (fig. 2022). Material SAS-H139, locality 14, road cuttings near Mfolozi, Zululand, St Lucia Formation, Santonian IJ-III to Campanian I; and an unnumbered specimen in the offices of the Geological Survey, Cape Town, here referred to as GSO CT, from an unspecified horizon at locality 1, Umzamba Cliff, Pondoland, Transkei, Umzamba Formation, Santonian to Campanian. Dimensions D Wb Wh Wb: Wh U Int. Ext. I T SAS-H139 165 44,5(27,0) 57,0(34,5) 0,78 72,7(44,1) 31 33 114,2 32,8(28,7) 39,8(34,8) 0,82 48,3(42,3) 29 34 GSO CT 223,9 — 71,6(31,9) — 101,5(45,3) 36 36 (unnumbered specimen) Description SAS-H139 and GSO CT are fairly complete specimens with parts of the body chamber preserved. Coiling is evolute, with an umbilical diameter measuring 42 to 45 per cent of the total diameter. Whorl overlap is minimal, covering the successive whorls up to the marginal (4) tubercles. The whorl section throughout is higher than wide, with a sub-ovoid to subrectangular whorl section both in costal and intercostal section. Specimen SAS—H139 has the innermost whorls preserved up to a diameter of 7 mm. Here ornament consists of regularly bifurcating ribs with weak umbilical, and strong ventrolateral, spines. Beyond a diameter of 7 mm ribs bifurcate frequently with occasional single ribs. The ventrolateral row of tubercles progressively migrates away from the umbilical wall to expose a CRETACEOUS FAUNAS FROM SOUTH AFRICA 163 GG Fig. 123. Texanites texanus (Roemer, 1852). Unnumbered specimen from the Cape Town branch of the Geological Survey. Probably referable to subspecies or variety hispanica. 164 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 124. Texanites texanus (Roemer, 1852). SAS—H139. This specimen is best referred to subspecies or variety gallica. x 0,8. third row of tubercles in the umbilical suture. Weak lateral (2) tubercles start appearing at a diameter of c. 20 mm near the dorsal third of the flanks. Beyond that diameter ornament remains more or less the same. The only significant change is that tuberculation becomes more prominent, and bifurcations less frequent, to eventually disappear completely on the body chamber. The ribs are generally straight and slightly rursiradiate throughout. The umbilical (1) tubercles are conical to radially elongated, pointing slightly inwards. The lateral (2) tubercles are conical on the inner whorls, but become prominently elongated in the direction of the ribs in later whorls. The sub- marginal (3) tubercles are conical to radially elongated, whilst the marginal (4) and external (5) tubercles are clavate. CRETACEOUS FAUNAS FROM SOUTH AFRICA 165 GSO CT differs from SAS—H139 in having the lateral tubercle situated virtually at midflank, compared to the dorsal third in the latter. The keel is low and entire. Discussion Collignon (1948) correctly interpreted T. texanus s.s. and removed all European references to this species from the strict synonymy. Apart from the typical form, he recognized two distinct varieties, var. hispanica and var. gallica, but later (1966a) elevated them to specific rank. Young (1963) added another, T. texanus twiningi, as a subspecies. Differences between the four subspecies or varieties are based mainly on density of ribbing and details of ornamentation. T. texanus texanus appears to have single ribs only, numbering about 20 to 21 per whorl. T. texanus twiningi has four to six more ribs per whorl than the nominate subspecies. T. texanus gallica has a sub-ovoid whorl section, and denser costation than the nominate subspecies, and the lateral tubercle is situated on the dorsal third of the flanks. 7. texanus hispanica also has denser costation, but ribbing starts at the umbilical wall and the lateral tubercle migrates ventrally until it is finally situated at midflank. T. texanus s.1. is not a very common species, hence it is difficult to decide whether the four varieties, subspecies or species really merit separation, or whether this is merely another example of morphological classification in the texanitids. If large collections were available, as in the case of P. stangeri or T. soutoni, these differences could possibly all be incorporated in the intra- specific variation of a single species. Young (1963: 82) separated T. texanus twiningi from the nominate subspecies purely on the grounds of stratigraphic separation, stating that ‘if the two forms, T. texanus texanus and T. texanus twiningi occurred together, there would be no need for the new name’. There does, however, seem to be an age difference between the two subspecies, each being utilized to designate a biozone (Young 1963: 22, text-fig. 33). Of the South African specimens, SAS—H139 is best referred to T. texanus gallica, having the lateral tubercle situated at the dorsal third of the flanks. The Pondoland specimen GSO CT has the lateral tubercle situated at midflank on the body chamber, and may be referred to T. texanus hispanica. It should be pointed out, however, that it lacks the extension of ribbing on to the umbilical wall, alleged to be characteristic of T. texanus hispanica. Occurrence Texanites texanus texanus appears to be restricted to Texas, although Peron (1897: 49-50) has also recorded the species from Mansourah in Algeria. Without figures the latter record is questionable. T. texanus twiningi is also restricted to Texas. T. texanus gallica has been recorded from France, Bulgaria, Italy, South America, Texas and Madagascar. T. texanus hispanica is known from Spain and Madagascar. 166 ANNALS OF THE SOUTH AFRICAN MUSEUM B Fig. 125. Texanites texanus (Roemer, 1852). A. SAS-H139. x 1. B. Unnumbered specimen in Cape Town branch of Geological Survey. CRETACEOUS FAUNAS FROM SOUTH AFRICA 167 Texanites umzambiense sp. nov. Figs 126-129, 152A Holotype SAM-PCP5714 (Figs 126-129), Bed C6 south of locality 1, on the southern side of the Umzamba River Estuary, Pondoland, Transkei, Umzamba Forma- tion, Middle Santonian. Etymology Named after the Umzamba River, Pondoland, Transkei. Fig. 126. Texanites umzambiense sp. nov. Holotype SAM-—PCP5714. x 0,59. 168 ANNALS OF THE SOUTH AFRICAN MUSEUM Dimensions D Wb Wh Wb: Wh U Int. Ext. 221,0 59,0(26,7) 66,0(29,7) 0,89 99,0(44,8) 28 28 Description Early whorls up to 26 mm The protoconch is absent, and the initial smooth stage appears to be extremely short or absent (Figs 129, 152A). Ornament consists of strong ventrolateral spines which lean against and into the umbilical wall of the succeeding whorl. On the innermost whorl preserved, these spines are impressed into the umbilical wall to such an extent that the whorl appears angular when viewed laterally. The spines arise from low, broad swellings which cover the whole flank. Spine density on the innermost two whorls is 6 and 11 per whorl respectively. Beyond a diameter of 11 mm the whorl section becomes more angular, with a near vertical umbilical wall and weakly inflated flanks. The broad bases of the ventrolateral spines now begin forming distinct single ribs which cover the flanks of the whorls. Distinct, radially elongated umbilical tubercles start appearing at the umbilical edge. The base of each of these is elongated obliquely forward over the umbilical wall. Rib density at this stage is 21 per whorl, the interspaces being as wide as the ribs. The ventrolateral spines progressively become less prominent, and eventually appear as spirally elongated clavi in the umbilical seam. 26 mm and beyond At a diameter of about 26 mm faint lateral (2) tubercles start appearing at about midflank. Simultaneously the ventrolateral clavi start migrating away from the umbilical suture to reveal the 4th row of spirally elongated tubercles. With increasing diameter the umbilical tubercles become increasingly clavate, pointing obliquely inward. The lateral tubercles eventually migrate towards the ventral third of the flanks, and the third row to the dorsal third, whilst the 4th row remains in the umbilical seam. Rib density at diameters of 62 and 135 mm is 27 and 24 per whorl respectively. On the outer phragmocone whorl complete ornament is visible (Fig. 126). The marginal (4) and external (5) tubercles are situated close together at the ventrolateral edge, and both are prominently clavate. The ventral keel is undulating, being highest in the inter- tubercular area. The preserved body chamber consists of half a whorl. There is no distinct change in ornament at the beginning of the body chamber. On the greater part of the late body chamber, ornament becomes weakened and irregular due to damage to the shell. Discussion Despite being monotypic, this species is easily distinguished from the predominantly younger species belonging to the groups of Texanites soutoni CRETACEOUS FAUNAS FROM SOUTH AFRICA 169 A Fig. 127. Texanites umzambiense sp. nov. Holotype SAM-PCP5714, x 0,74 170 ANNALS OF THE SOUTH AFRICAN MUSEUM and Submortoniceras woodsi (described below). None of these specimens has the marginal and external tubercles situated as close together as the present species, or possesses an undulating keel. The closest ally to, and probably direct ancestor of, this species seems to be Plesiotexanites collignoniforme sp. nov. described above (p. 89). Both species have the marginal and external tubercles approximated, and an undulating keel. 7. umzambiense is here separated from P. collignoniforme mainly because of the earlier appearance of the lateral tubercle (26 mm v. Fig. 128. Texanites umzambiense sp. nov. Holotype SAM-PCP5714. A. x 1. B. Scale bar in millimetres. 171 CRETACEOUS FAUNAS FROM SOUTH AFRICA in millimetres. Holotype SAM-PCP5714. Drawing to illustrate Scale early ontogeny Fig. 129. Texanites umzambiense sp. nov. 172 ANNALS OF THE SOUTH AFRICAN MUSEUM 50 mm), and a more inflated whorl section on the phragmocone. Quite obviously the ontogenetic character of differentiation between Texanites and Plesio- texanites becomes evident here. As noted above (p. 98), this could point to the beginning of the geographic separation of the Zululand and Pondoland texanite faunas. T. texanus hispanica, also reported from the Umzamba Formation above (p. 162), bears superficial similarity in general form, but does not have the external and marginal tubercles as closely spaced and lacks an undulating keel. T. collignoni is related in having an undulating keel, but generally has a much more compressed whorl section and different arrangement of tubercles. T. mikobokensis from the Upper Santonian of Madagascar is another similarly evolute species, but has a much more compressed whorl section and different spacing of lateral ornament (see Collignon 1966a, pl. 509 (fig. 2019)). Occurrence Middle Santonian of Pondoland, Transkei. Group of Texanites soutoni (Baily, 1855)—Submortoniceras woodsi (Spath, 1921) Texanitids of the group centred on the plexus of 7. soutoni and S. woodsi form the bulk of the texanitids in the present collection, and merit special attention, not only because of numerical superiority and stratigraphic import- ance, but because they illustrate (i) what appears to be a partial geographic isolation of two contemporaneous groups within a species with some intermingling, (ii) a striking evolutionary sequence in both populations, spanning the period from Middle Santonian to Early Campanian, starting with a normal evolute Texanites, and ending with an involute Submortoniceras, with no clearcut break, and (iii) a tremendous intraspecific variation, especially as far as ornamentation and suture lines are concerned which causes one to have serious doubts about the validity of current texanitid systematics. The texanitids occur mainly in two areas—the type section of the Umzamba Formation at the Umzamba Estuary, Transkei, locality 1 of the authors (Kennedy & Klinger 1975: 281; Klinger & Kennedy in press) and the south- eastern shores of False Bay, Zululand, at locality 105 (Kennedy & Klinger 1975: 296; herein Fig. 130) continuing the section in the cliffs at locality 74 (Kennedy & Klinger 1975: 283, fig. 8). Minor occurrences connecting the two main areas are at locality 14 (Kennedy & Klinger 1975: 283) near Umkwelane Hill, sub- surface deposits near Richards Bay at locality 6 (Kennedy & Klinger 1975: 282), and similar deposits at Durban at locality 5 (Kennedy & Klinger 1973; 1975: 282). Current evidence seems to suggest that the deposits at False Bay, Zulu- land, represent a deeper water facies than the shallower water transgressive facies of the Umzamba Formation. CRETACEOUS FAUNAS FROM SOUTH AFRICA 173 — = a G Silts with aragonitic concretions Pseudoschloenbachia? sp. 'Heteroceras' amapondense, Hyphantoceras sp. Submortoniceras condamyi Submortoniceras condamyi, Hyphantoceras sp., inoceramids, large indeterminate texanitid Submortoniceras condamyi, INoceramus sp., aragonitic concretions Texanites soutoni natalense, Kossmati- ceras sp., Scaphites sp., inoceramids ES i vas SS Texanites presoutoni natalense, Pseudo- D> cf phyllites latus, phyllites latus, Inoceramus sp. Fig. 130. Stratigraphic section at locality 105 on the south-eastern shores at False Bay. 174 ANNALS OF THE SOUTH AFRICAN MUSEUM The material is extremely variable, but data gained from detailed zonal collecting by the authors, examination of material collected zonally by Gevers, and indications of relative stratigraphic position of specimens in older collec- tions—such as encrustation by modern epizoans indicative of a low strati- graphic position at the type section of the Umzamba Formation—permit partial interpretation of the species centred on the T. soutoni—S. woodsi plexus, both in terms of spatial and temporal variation. At first glance, the texanitine faunas of Pondoland and False Bay appear to be completely different. This apparent ‘difference’ is greatly enhanced by the dissimilar lithologies and associated faunas (the latter to be discussed fully in Klinger & Kennedy in press). Specimens from False Bay generally retain calcareous white shell material in a buff-weathering sandy silt, whereas the Pondoland specimens are mainly preserved in crystalline calcite in a dark greyish-green glauconitic silt. In addition, there is a distinct difference in size between the texanitines of both areas; those from Pondoland are generally gigantic when compared with the normal-sized specimens from False Bay. Detailed stratigraphic collecting shows that there is a distinct trend in both faunas towards acquisition of submortoniceratine shell-form (Figs. 131- 132). True texanitids with umbilical diameters of the order of c. 45 per cent occur near the bases of both sections, and through gradually decreasing umbilical width and whorl breadth, give way to true submortoniceratids towards the higher levels in the sections. Apart from differences in size, real differences are to be found in the mode of acquisition of Submortoniceras ornament, and in the relative width of the lobes of the suture lines. In Pondoland, the submortoniceratine characters are developed in the ‘traditional’ way, i.e. through gradual weakening of ornament from the inside outwards. At False Bay, however, a great proportion of speci- mens has inner whorls with ornament weaker than that of the outer whorls and retains tuberculation on part of the body chamber. It may thus be said that submortoniceratine characteristics are acquired caenogenetically in Pondoland, but proterogenetically at False Bay. As far as suture line is concerned, the False Bay specimens generally have much wider and shallower lobes exposed on the flanks than their Pondoland counterparts. Detailed examination of the associated fauna, however, has shown that the stratigraphic interval exposed at the type section of the Umzamba Formation up to Bed A7-8 (Klinger & Kennedy in press) corresponds approximately to that ex- posed at locality 105 and the higher parts of the section at locality 74 (Kennedy & Klinger 1975: 293, fig. 8). Furthermore, typical, though sometimes dwarfed or smaller, Pondoland faunas occur in the Richards Bay and Umkwelane Hill areas, and even at False Bay, while typical False Bay faunas occur at Umkwelane Hill, indicating an area of intermingling between apparently isolated faunas. The authors are now able to correlate the texanitine faunas of Pondoland and Zululand, and it is now possible to produce a series of specimens of more or less equal involution and age, ranging from completely smooth to coarsely CRETACEOUS FAUNAS FROM SOUTH AFRICA 175 Holotype Jexanites soutoni soutoni Dead-end ? Ko ‘var. inflata var. cOmpressa Texanites soutoni soutoni Texanites presoutoni presoutoni t MIDDLE SANTONIAN — — — — — — UPPER SANTONIAN — — — — — — — —LOWER CAMPANIAN Texanites umzambiense ‘(eee VARIATION = Fig. 131. Diagrammatic illustration showing suggested lines of evolution and intraspecific variation in species of the Texanites soutoni-Submortoniceras woodsi plexus at the type section of the Umzamba Formation. LOWER CAMPANIAN UPPER SANTONIAN — 176 ANNALS OF THE SOUTH AFRICAN MUSEUM woods! natalense forma typica var. inflata Texanites soutoni natalense var. COmpressa Texanites presouton! natalense Fig. 132. Diagrammatic illustration showing suggested lines of evolution and intraspecific variation in species of the Texanites soutoni-Submortoniceras woodsi plexus in the False Bay region of Zululand. CRETACEOUS FAUNAS FROM SOUTH AFRICA 177 ornamented on the inner whorls, all occurring at the same stratigraphic level. The extreme variation at the same stratigraphic level, combined with: the lack of abrupt vertical changes in the population, makes it difficult to differ- entiate satisfactorily between the various groups. As there are no visible abrupt changes in the population, differentiation will have to be arbitrary, and here umbilical width seems to be the most feasible choice. The Pondoland and Zulu- land faunas, where sufficiently different, are separated at subspecific level. On the latter basis, the following morphological taxa may be recognized: U =c. 45% with external and marginal tubercles separated and ribbing pre- dominantly single, especially on the body chamber = 7. presoutoni sp. nov. T. presoutoni presoutoni ssp. nov. Holotype is SAM-—PCP5712 from the Umzamba Formation, Umzamba Cliff at locality 1, presumably Bed Al. T. presoutoni natalense ssp. nov. Holotype is SAS-—H126A/12 from the St Lucia Formation, locality 105, Bed A. U=c. 40% =T. soutoni s.1. T. soutoni soutoni (Baily, 1855). Holotype is the specimen figured by Baily (1855, pl. 11, fig. la—c), here refigured as Figure 143. T. soutoni natalense ssp. nov. Holotype is SAS—H126B/3 from the St Lucia Formation, locality 105, Bed B. In both the above subspecies an inflated and compressed variety may be recognized, both apparently with a tendency towards acquisition of submortoni- ceratine characters, but it appears the compressed variety is a dead-end. (See Figs 131-132.) U =c. 35% = Submortoniceras woodsi s.\. S. woodsi woodsi (Spath, 1921). Holotype is SAM-—5451 from Umkwelane Hill, Zululand, the specimen figured by Spath (1921, pl. 21 (fig. la-d)), here refigured as Figure 186. S. woodsi natalense ssp. nov. Holotype is SAS—Z1875 from the St Lucia Formation, locality 105. U =c. 30% = S. condamyi. S. condamyi condamyi Collignon. Lectotype herein designated is the speci- men figured by Collignon (1948, pl. 18(12) (fig. 2-2b)) from the Lower Campanian of Berere, Madagascar. S. condamyi umzambiense ssp. nov. Holotype is SAM-—PCP5717 from the Umzamba Formation, Umzamba Cliff, locality 1, Bed A7. This framework (see also Figs 131-132) should be seen as a working basis towards classifying the species centred on 7. soutoni-S. woodsi, and not as a rigid, compartmentalized framework. Many of the specimens discussed below could be referred to more than one species or subspecies. This is in part due to the extreme horizontal variation and also the continuous vertical transition from one species to another. In addition, this classification is hampered by the fact that it has to accommodate the holotypes of established species which in 178 ANNALS OF THE SOUTH AFRICAN MUSEUM themselves are already transitional between species (or even genera), e.g. T. soutoni or incomplete, e.g. S. woodsi. Material described by Young (1963) from the Gulf Coast of America also falls within the range of variation of this species group, and has to be accommodated. Texanites presoutoni presoutoni ssp. nov. Figs 133-137, 152C Mortoniceras soutoni (Baily): Woods, 1906: 337, pl. 43 (fig. 1). ?Van Hoepen 1921: 38 pars, pl. 11 only. Holotype SAM-PCP5712 (Figs 133-135), locality 1, basal beds at Umzamba Cliff, Pondoland, Transkei, Umzamba Formation, Upper Santonian. Material SAM-4828, locality 1, basal beds at Umzamba Cliff, Pondoland, Transkei; and BMNH-C19447 from an unknown horizon at the same locality, Umzamba Formation, Upper Santonian. Dimensions D Wh Wb Wb: Wh U Int. Ext. PCP5712 387,0 — 122,0(31,5) — 181,0(46,8) 30 30 SAM4828 230,0 -- 81,0(35,2) —_— 97,0(42,0) 2 Daf Description The material, though limited in numbers, is very varied. Coiling is loose, with an umbilical diameter of 42 to 47 per cent. The whorl section may be quadrate in the very early stages, but later becomes compressed, higher than wide throughout. SAM-—PCP5712 (Figs 133-135, 152C) has the innermost whorls preserved, illustrating the ontogenetic development. After the initial smooth stage, long spinose ventrolateral tubercles appear, leaning against the umbilical wall. Smaller umbilical (1) tubercles appear next, followed by lateral (2) tubercles at a diameter of c. 20 mm. Ornament remains basically the same throughout the rest of the phragmo- cone, though differing considerably in terms of strength and spacing. SAM- PCP5712 (Fig. 133) thus has much coarser ornament on the phragmocone than SAM-4828 (Fig. 136). BMNH-C19447 (Fig. 137) has as sparse costation as SAM-PCP5712 (Fig. 133), but by no means as strongly developed tuberculation. All the specimens have single ribbing throughout—a feature here considered as being ‘primitive’. Towards the body chamber the lateral (2) and submarginal (3) tubercles become effaced, and on the later part all ornament weakens. The suture varies, but has stout saddles E/L and L/U, and long, thin lobes L and U,. CRETACEOUS FAUNAS FROM SOUTH AFRICA 179 Fig. 133. Texanites presoutoni presoutoni sp. nov. Holotype SAM-—PCP5712. x 0,35. ANNALS OF THE SOUTH AFRICAN MUSEUM 180 horls to illustrate ontogeny. Inner W. ites presoutoni presoutoni sp. nov. Holotype SAM-—PCP5712. Close-up of Texan . 134. Fig CRETACEOUS FAUNAS FROM SOUTH AFRICA 181 Fig. 135. Texanites presoutoni presoutoni sp. nov. Drawing of inner whorls based on holotype. Discussion The status of this species is admittedly somewhat artificial, but forms a convenient starting point for describing the species of the T. soutoni-S. woodsi plexus. The specimen figured by Van Hoepen as Mortoniceras soutoni (1921, pl. 11) has ornament on the inner whorls closely resembling that of the holotype of the species, but denser, fiexuous ribbing on the body chamber and an umbilical width of c. 43 per cent. It should probably also be referred to the present species, but the authors have not been able to trace it in the Transvaal Museum for 182 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 136. Texanites presoutoni presoutoni sp. nov. Paratype SAM-4828. x 0,6 The specimen figured by Woods (1906, pl. 43 (fig. 1)). definite identification. Haas (1942: 17) had already suggested that Van Hoepen’s figured specimen be separated from TJ. soutoni as variety densicosta. Apart from the fact that this cannot be accepted as a valid designation, the term densicosta, should it be accepted in formal nomenclature, is extremely mis- leading, as in this case it applies only to the body chamber ornament. The authors’ comments on this specimen (Klinger & Kennedy 1977: 97) are still partially valid, but in suspecting different ontogenetic development the authors CRETACEOUS FAUNAS FROM SOUTH AFRICA gasses = Fig. 137. Texanites presoutoni presoutoni sp. nov. Paratype BMNH-C19447. PE: 183 184 ANNALS OF THE SOUTH AFRICAN MUSEUM were misled by differences in preservation and the extreme intraspecific variation in 7. soutoni. Both Haas (1942: 12) and Young (1963: 85) regarded Woods’s figured specimen (SAM-4828) of Mortoniceras soutoni as not being conspecific with the holotype, but gave no reasons for this. The wide umbilicus, single ribbing throughout and long, narrow lobes in the suture of this specimen are all charac- teristic of the present subspecies. T. presoutoni presoutoni is closely connected to T. soutoni soutoni by speci- mens such as SAM-—PCP5713 and SAM-—PCP5718 (Figs 144-145) with only a few bifurcations and a slightly narrower umbilicus. T. presoutoni s.s. differs from its subspecies in the False Bay region of Zululand, T. presoutoni natalense, mainly in losing tuberculation on the body chamber, whereas the latter remains tuberculate. The Pondoland subspecies attains larger dimensions. * Occurrence Lower part of Upper Santonian of Pondoland, Transkei. Texanites presoutoni natalense ssp. nov. Figs 138, 139B, 140-142 Holotype SAS-H126A/12 (Figs 138, 139B), Bed A locality 105, south-eastern shores of False Bay, Zululand, St Lucia Formation, Upper Santonian. Material Paratypes are SAS-Z1140 and SAM-—PCZ5885, both from the same locality as the holotype. Dimensions D Wb Wh Wb: Wh U Int. Ext. H126A/12 301,0 — 106,0(35,2) — 130,0(43,2) 17 x2 Z1140 226,0 62,0(27,4) 76,0(33,6) 0,81 101,0(44,7) 31 31 PCZ5885 177,0 51,0(28,8) 62,0(35,1) 0,82 75,0(42,4) 28 2H Description Coiling is evolute, with an umbilical width of 42 to 44 per cent of the total diameter. The whorl section is massive, higher than wide throughout with a near vertical umbilical wall, and, in intercostal section, little-inflated flanks converging to a broadly rounded venter. Maximum whorl breadth is near the dorsal third of the flanks in intercostal section, and in costal section at the site of the lateral tubercle, which is very prominent. Ornament on the phragmocone consists of radially elongated umbilical tubercles (1), arising at the umbilical edge, and progressively clavate lateral, submarginal, marginal and external (2—5) tubercles. At larger diameters on the CRETACEOUS FAUNAS FROM SOUTH AFRICA 185 Fig. 138. Texanites presoutoni natalense ssp. nov. Holotype SAS-H126A/12. x 0,53. UMS th B. Texanites . 00 ~wn nS S x x nel co a = 4 Q Na 2 AS aye NAN a Sy ES) fe) Me) Go am 2s & Hm - Se 84 = Ss wm SX ss 28 Se » BS S38 So aS =Z— = wae SF 8 Fig. 142. Texanites presoutoni natalense ssp. nov. Holotype SAS—-H126A/12. Suture line. Scale bar in millimetres. Occurrence Upper Santonian of Zululand. Texanites soutoni soutoni (Baily, 1855) Figs 143-151, 152B, 153-163 Ammonites soutoni Baily, 1855: 455, pl. 11 (fig. 1a-c). Mortoniceras soutoni (Baily): Spath, 1921: 234, pl. 20 (fig. 4); 1922: 136, pl. 7 (fig. 4). Van Hoepen, 1921: 38, pl. 10 (zon pl. 11 =T. presoutoni s.s. sp. nov.). Non Mortoniceras soutoni (Baily): Woods, 1906, 337, pl. 43 (fig. la—b) (=T. presoutoni s.s. sp. nov.). Texanites soutoni (Baily): Matsumoto, 1955: 42, text-fig. 3. Collignon, 1966: 70, pl. 483 (fig. 1956). Klinger & Kennedy, 1977: 89, fig. 9. Non Mortoniceras sp. aff. soutoni Spath, 1922: 235 (=T. soutoni natalense ssp. nov.). ? Mortoniceras aff. soutoni Spath, 1925: 199. ? Texanites cf. soutoni Collignon, 1948: 78, pl. 9(3) (fig. 1a—b). Non Texanites soutoni (Baily): Kennedy & Klinger, 1972: pls 72-73 (with Spinaptychus) (=T. soutoni natalense ssp. nov.). Type Holotype is BMNH-C47261, the specimen figured by Baily (1855, pl. 11 (fig. 1)), here refigured as Figure 143, locality 1, Umzamba Cliff, Pondoland, Transkei, Umzamba Formation, Upper Santonian. Material SAM-—PCP5713-5715, PCP5718, SAM-13090, SAM-7076—-7077, SAS- P747, P1334, P1398, NMB-D1605a-b, BMNH-C19442, C19454, C19461, C47261, all from locality 1, Umzamba Cliff, Pondoland, Transkei, Umzamba Formation, Middle to Upper Santonian; SAS-—Z306 and SAM-PCZ5885, 190 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 143. Texanites soutoni soutoni (Baily, 1855). Photograph of cast of holotype, BMNH-— C47261. x 0,34. Note that Baily’s figure is a mirror image. CRETACEOUS FAUNAS FROM SOUTH AFRICA 191 locality 14, near Mfolozi, St Lucia Formation, Santonian II-III to Campanian J; SAM-PCZ5884, locality 6, excavations near Richards Bay, St Lucia Formation, Santonian II-III to Campanian I; SAM-—PCZ 5903, locality 74, Die Rooiwalle, Zululand, St Lucia Formation, Santonian to Campanian; and possibly SAM-FB from an unknown locality in False Bay, probably also locality 74. Dimensions D Wb Wh Wb: Wh U Int. Ext. PCP5714 231,0 66,0(28,6) 74,0(32,0) 0,89 107,0(46,3) 30 30 PCP5713 200,0 56,0(28,0) 70,0(35,0) 0,80 82,0(41,0) 28 28 SAM-7077 337,0 — 106,0(31,5) — 149,0(44,2) 38 38 P747 357,0 78,0(21,8) 131,0(36,7) 0,59 145,0(40,6) 36 36 P1334 329,0 89,0(27,1) 110,0(33,4) 0,81 140,0(42,6) DIL SRD) D1605b 259,0 — 93,0(35,9) — 105,0(40,5) 30 30 PCP5718 356,0 — 127,0(35,7) — 151,0(42,4) 30 32 Z306 237,0 66,0(27,8) 93,0(39,2) 0,71 91,0(38,4) 26 30 C47261 444,5 114,3(25,7) 152,4(34,3) 0,75 177,8(40,0) 35 37 Diagnosis Texanitid with umbilical width of approximately 40 per cent of total diameter, whorl section higher than wide throughout; with venter narrower than dorsum. Ornament on innermost whorls variable, ranging from virtually smooth to strong, with variable number of bifurcations. Towards the body chamber lateral ornament becomes attenuated. Generally one or two bifurca- tions on outer body chamber whorl. Suture variable. Description As here interpreted, the species is very variable, consequently ample illustrations are provided in lieu of lengthy descriptions. Two varieties within the species may be recognized, both occurring together near the base of the section exposed at Umzamba Cliff: an inflated broad- whorled variety, e.g. SAM-PCP5718 (Figs 144-145), and a more evolute, slender-whorled and compressed variety, e.g. SAM-7077 (Fig. 146), which may here be labelled var. inflata and var. compressa for descriptive purposes. Both extremes, however, are connected by specimens such as SAS—P747 (Figs 147-148) and SAS-P1334 (Fig. 149). Specimens SAM-7077 (Fig. 146), SAS— P747 (Fig. 147) and SAS—-P1334 (Fig. 149) illustrate the variation in strength of ornament in the species. Details of these differences in the ontogeny are discussed below. Stratigraphic data seem to suggest that the holotype of T. soutoni s.s. may be derived from variety inflata through reduction of the umbilical diameter, probably via specimens such as SAM-—PCP5715 (Fig. 150). Variety compressa possibly also evolved in the direction of Submortoniceras via a specimen such as SAM-—7076 (Fig. 151) with attenuated ornament on the outer whorls but a relatively wide umbilicus. This latter trend appears to have been a dead-end (Fig. 131), whereas forms such as the holotype of JT. soutoni s.s. gradually evolved into Submortoniceras woodsi. 192 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 144. Texanites soutoni soutoni (Baily, 1855). SAM-—PCP5718. Variety here referred to informally as var. inflata. x 0,41. Preservation of material at the Umzamba Estuary is generally much better than that of temporally equivalent strata in Zululand, and a number of speci- mens have the inner whorls preserved, permitting a full description of the ontogeny of the species. None of the specimens, however, has the protoconch preserved. Details of the ontogeny are illustrated in Figures. 152-153. Even on the CRETACEOUS FAUNAS FROM SOUTH AFRICA 193 Fig. 145. Texanites soutoni soutoni (Baily, 1855). SAM-— PCP5718. Variety here referred to informally as var. inflata. x 0,48. 194 ANNALS OF THE SOUTH AFRICAN MUSEUM we Fig. 146. Texanites soutoni soutoni (Baily, 1855). SAM-—7077. Variety here referred to informally as var. compressa. X 0,41. innermost whorls differences in strength of ornament are obvious when com- paring SAM-—PCP5713 (Fig. 152B) with SAS—P747 (Fig. 153C). The initial smooth stage is short, except in SAS—P747, where it persists to a diameter of about 10 mm. The ventrolateral tubercles develop first, and in most cases occur as long, curved spines with broad bases leaning against the umbilical wall. In this respect SAS—P747 (Fig. 153C) seems to differ from the rest of the specimens in appearing to lack the strong development of ventrolateral spines. Close investigation of the inner whorls of SAS—P747 (Fig. 153C), however, CRETACEOUS FAUNAS FROM SOUTH AFRICA 195 Fig. 147. Texanites soutoni soutoni (Baily, 1855). SAS-—P747. Specimen connecting between varieties compressa and inflata in having a very compressed whorl section, but not as evolute coiling. x 0,42. 196 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 148. Texanites soutoni soutoni (Baily, 1855). SAS—P747. Specimen connecting between varieties inflata and compressa in having extremely compressed whorls, but relatively narrow umbilicus. x 0,5. CRETACEOUS FAUNAS FROM SOUTH AFRICA 197 Fig. 149. Texanites soutoni soutoni (Baily, 1855). SAS—P1334. Specimen connecting between varieties inflata and compressa, and with very weak lateral ornament. x 0,55. 198 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 150. Texanites soutoni soutoni (Baily, 1855). SAM-—PCP5715. x 0,41. reveals the vestiges of two tubercles and corresponding indentations on the umbilical wall of the succeeding whorl, indicating that spines could have been present and that their absence could be partially due to the state of preservation. Small, rounded umbilical tubercles occur next, followed by a row of lateral tubercles. With increasing diameter the ventrolateral spines in the umbilical suture become smaller and more clavate, and migrate away from the umbilical wall to expose a fourth row of tubercles on the flanks. Thereafter ornament remains basically the same, although details differ, as shown in the Figures. CRETACEOUS FAUNAS FROM SOUTH AFRICA 199 Fig. 151. Texanites soutoni soutoni (Baily, 1855). SAM-7076. Specimen with attenuated ornament, trending towards Submortoniceras, but apparently a dead-end lineage. x 0,42. (See also Fig. 131.) The suture is very variable as Van Hoepen (1921) had already noted. Apart from differences in the length and width of the second lateral saddle (L/U,) already discussed and figured by Van Hoepen (1921), the most obvious and easiest recognizable differences are to be found in the length and incision of the saddles and lobes exposed on the flanks. Data here are scant, but it appears as if there is a tendency towards greater complexity of the suture line higher up in the stratigraphic sequence (Fig. 161) leading towards the formation of the ANNALS OF THE SOUTH AFRICAN MUSEUM 200 Yj 7 Vial to illustrate Holotype SAM-— Texanites pre- imens C Holotype SAM-PCP5712 - NOV. whorls of spec inner toni SAM—PCP5713 ites umzambiense sp i. ton hotographs of A. Texan i presou ir p Texanites soutoni sou souton Stereopa 152 ontogenetic development B Fig PCP5714 201 CRETACEOUS FAUNAS FROM SOUTH AFRICA ites SAM-7077 imens of Texan A ion. ic variat SAS-P747 . inner whorls of spec C —P1398 . SAS B illustrate the ontogenet hotographs of ir p to i ton Fig. 153. Stereopa soutoni sou 202 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 154. Texanites soutoni soutoni (Baily, 1855). SAS—-Z306. A dwarfed specimen from the environs of Umkwelane Hill. x 0,56. narrow-stemmed, deeply incised saddles and lobes of Submortoniceras woodsi and S. condamyi. The lateral lobe (L) appears to become shorter and wider with splayed ends, and to change from subtrifid to more symmetrically bifid. Comparison of the sutures of T. presoutoni s.s. (specimens SAM—PCP5712 and SAM-4828, both from the lower parts of the sequence at Umzamba), through normal representatives of T. soutoni s.s. (SAS-PCP5718, SAS—P747 and NMB- D1605B) and the holotype of T. soutoni s.s. (C47261 and a related form, SAM-—PCP5715) through to Submortoniceras woodsi woodsi and S. condamyi CRETACEOUS FAUNAS FROM SOUTH AFRICA 203 Fig. 155. Texanites soutoni soutoni (Baily, 1855). SAS—Z306. A dwarfed specimen from the environs of Umkwelane Hill. x 0,66. 204 ANNALS OF THE SOUTH AFRICAN MUSEUM umzambiense seems to illustrate the point (Fig. 161). As far as can be ascertained, there seems to be no direct relationship between complexity of suture line and ornament. Discussion Despite being one of the first ammonites to be described from the Cre- taceous deposits at the Umzamba Estuary, this species has been consistently misinterpreted. Both Spath (1922: 136) and Van Hoepen (1921: 38-45) com- mented on the extreme variation of T. soutoni in their collections. Whereas Fig. 156. ? Texanites soutoni soutoni (Baily, 1855). SAM-PCZ5903, from Die Rooiwalle, Zululand. x 0,68. CRETACEOUS FAUNAS FROM SOUTH AFRICA 205 Spath concentrated mainly on density and character of ribbing (presence or absence of bifurcations) and to a lesser extent on the tightness of coiling, Van Hoepen paid close attention to the configuration of the suture lines and came to the conclusion that, in some cases at least, the comparative significance of the suture line at more or less the same age is fiction. Apart from these two authors, little substantial research has been done on this species. The largest known collections of T. soutoni are housed in the collections of the British Museum (Natural History), the South African Museum, Geo- logical Survey of South Africa (Pretoria), and the Transvaal Museum. Exami- nation of these collections, virtually all from the Umzamba Estuary, shows that hardly two specimens are alike. The sediments at Umzamba represent a con- densed sequence (Klinger & Kennedy 1977: 105, and in press). Stratigraphic data are thus essential for meaningful systematics. Unfortunately very little stratigraphic data are available in the literature. Baily (1855: 455) stated of the holotype that Thomas Souton had ‘extracted it . . . from a very hard stratum high up the cliff’. Griesbach (1871: 62 (table)) lists Ammonites soutoni as occur- ring in his Bed d. Plows (1921: 63) recorded a specimen of Mortoniceras soutoni in his Bed 14. Rogers & Schwarz (1902: 41) and Gevers (in Rennie 1930: 253) record Mortoniceras soutoni from the basement beds at Umzamba. Detailed collecting by the authors at the type section of the Umzamba Formation, examination of the material collected zonally by Gevers, and indi- cations of relative stratigraphic position of specimens in older collections, such as encrustation by modern epizoans, indicative of a low horizon at the type section which is inundated by the present sea during high tide, permitted partial explanation of the species plexus centred on T. soutoni as outlined above (Fig. 131). The ‘high horizon’ ascribed to the holotype of T. soutoni by Baily tallies well with the present interpretation that’ submortoniceratine charac- teristics are acquired gradually through decreasing umbilical width and pro- gressive loss of ornament in earlier stages when passing higher up through the sequence exposed at the Umzamba Estuary. The question of whether Baily’s Ammonites soutoni should be referred to Submortoniceras, as advocated by Spath (1953: 52) and Young (1963: 91), or Texanites, following Matsumoto (1955: 42, 1970: 279) or Collignon (1948: 42, 1966a: 70), thus becomes academic, depending on where the boundary between Texanites and Submortoniceras is drawn. According to the present classification, the holotype of Ammonites soutoni is referred to the genus Texanites. A complete, though dwarfed, specimen SAS—Z306 (Figs 154-155) from locality 14 at Umkwelane Hill represents the first recording of this species from Zululand. SAM-PCZ5903 (Fig. 156) and SAM-FB (Fig. 157) are the only two possible representatives of the nominate subspecies from the northern part of Zululand at Die Rooiwalle, locality 74. Specimen SAM-—5491 (Fig. 81) described by Spath (1921: 234) as Mortoni- ceras aff. umkwelanense and considered related to T. soutoni, certainly does not 206 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 157. ? Texanites soutoni soutoni (Baily, 1855). SAM-—FB, a very large specimen, possibly referable to the nominate subspecies from an unknown locality at False Bay, Zululand. x 0,48. 207 CRETACEOUS FAUNAS FROM SOUTH AFRICA in ty i from an unknown local > T1491 x 1 Mozambique. Fig. 158. Texanites sp. aff. soutoni (Baily, 1855) 208 ANNALS OF THE SOUTH AFRICAN MUSEUM belong to this species, but is an example of Plesiotexanites matsumotoi sp. nov. described above. The specimen described by Spath (1921: 235) as Mortoniceras aff. soutoni from Umkwelane Hill, Zululand, belongs to the group of 7. soutoni s.1. but is better referred to TJ. soutoni natalense ssp. nov. on account of the smooth inner whorls. Two very corroded specimens, TM1491 (Fig. 158) and TM1493 (Fig. 159), described by Spath (1925: 199-200) as Mortoniceras aff. soutoni, are very doubtful contenders. Apart from the fact that they were found in ‘Maputoland’ (Mozambique) and presented to the Transvaal Museum by a ‘Sefior Morreira’, no other locality data are available. In view of their poor state of preservation a definite identification is not possible. If they are, indeed, representatives of T. soutoni, they extend the geographic range of the species from Pondoland through Zululand and Mozambique to Madagascar. Judging by similarities in ontogeny, 7. soutoni s.s. can be derived from T. umzambiense sp. nov. or T. presoutoni s.s. sp. nov. (Figs 152-153). Unfortu- nately stratigraphic data are insufficient for determining the exact phylogenetic sequence, but it is assumed here that T. soutoni was probably derived from T. umzambiense via T. presoutoni. Differences between T. soutoni s.s. and T. pre- soutoni sp. nov. s.s. are gradational, being mainly a question of umbilical width and frequency of bifurcations on the phragmocone as noted above (p. 177). Specimens such as SAM—PCP5713 (Fig. 152B) or SAM-13090 (Fig. 160) with robust ornament consisting predominantly of single ribs on the phragmocone, are to be considered transitional between the two species. Similarly, the upper boundary between T. soutoni s.s. and Submortoniceras woodsi is also gradational. Specimens with umbilical diameters in the vicinity of 35 per cent and greatly attenuated ornament on the phragmocone are best referred to Submortoniceras, rather than T. soutoni. The Zululand counterpart of T. soutoni soutoni, T. soutoni natalense ssp. nov., differs mainly in often having attenuated ornament on the inner whorls, a generally stouter whorl section and wider and shallower lobes, but is just as variable and grades laterally into the Pondoland subspecies. In view of the extreme variation, comparisons with species outside southern Africa become tenuous. T. hourcgi Collignon from the Middle Santonian of Madagascar is similar to the more evolute forms of T. soutoni s.s., but is more compressed. When compared with T. soutoni s.s. doubt arises if the possession of regularly bifur- cating ribbing in T. hourcqi is as important a characteristic as implied by Col- lignon (1948: 79). Young (1963: 90) erected a new species, T. Jonsdalei, on the basis of one specimen and a questionable specimen, and included Collignon’s (1948: 78, pl. 9(3) (fig. 1-la)) Texanites cf. soutoni in the synonomy of that species. As interpreted here, Collignon’s specimen may easily be accommodated in the variation of T. soutoni s.s., but not T. lonsdalei because of the much higher than CRETACEOUS FAUNAS FROM SOUTH AFRICA 209 Fig. 159. Texanites sp. aff. soutoni (Baily, 1855). TM1439, from an unknown locality in Mozambique. x 1. 210 ANNALS OF THE SOUTH AFRICAN MUSEUM wide whorl section (see especially Young, pl. 51 (fig. 5)). Young (1963: 91), however, paid more attention to similarities of ornament, and admitted that the Texan specimen ‘may be crushed’. Even if the Texas specimen is crushed, the umbilical width is predominantly of the order of 32 to 35 per cent, with only one recording at 41,5 per cent. This places the species closer to Submortoniceras woodsi, and Young (1963: 92) did, indeed, refer to this species (together with Woods’s specimen of Mortoniceras soutoni (1906, pl. 43 (fig. 1)) and 7. roemeri as being ‘close to the Submortoniceras—Texanites taxonomic boundary’ (Young : eG, — Fig. 160. Texanites soutoni soutoni (Baily, 1855). SAM-—13090, a specimen transitional between T. soutoni soutoni and T. presoutoni presoutoni sp. nov. X 0,65. SUBMORTONICERAS TEXANITES CRETACEOUS FAUNAS FROM SOUTH AFRICA 211 ae tT T. soutoni holotype T. soutoni T. presoutoni Fig. 161. Diagram illustrating apparent trend in suture line towards higher degree of complexity in development from Texanites to Submortoniceras. Whorl section of Texanites soutoni soutoni superimposed. 212 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 162. Texanites soutoni soutoni (Baily, 1855). BMNH-C19461. (British Museum photograph.) x 0,95. 1963: 92). Until the question of the true whorl section of T. lonsdalei is resolved, the species has to be maintained with some doubt, but bearing in mind the affinity with the T. soutoni-S. woodsi plexus. Detailed studies of the ontogeny of the species as outlined here (p. 194) definitely allay Matsumoto’s suspicion (1970: 279) that T. soutoni could possibly have Plesiotexanites inner whorls. Occurrence T’. soutoni s.s. is relatively common in the Upper Santonian of the Umzamba CRETACEOUS FAUNAS FROM SOUTH AFRICA 213 WS Fig. 163. Texanites soutoni soutoni (Baily, 1855). BMNH-C19461. (British Museum photograph.) x 0,95. Formation of Pondoland at the type section, locality 1, and is also known from borehole and other subsurface material near Richards Bay, Zululand (Klinger & Kennedy 1977). Rare occurrences are known from Umkwelane Hill and the False Bay area of Zululand. | The species is very rare in Madagascar (Collignon 1948: 78), and as yet no good specimens have been figured from that area. In view of the other close similarities between Pondoland and Madagascar faunas, e.g. pseudo- schloenbachiids, Hauericeras, etc., this is very surprising. 214 ANNALS OF THE SOUTH AFRICAN MUSEUM Texanites soutoni (Baily) natalense ssp. nov. Figs 164-185 Mortoniceras sp. aff. soutoni Spath, 1921: 235. Texanites soutoni (Baily): Kennedy & Klinger, 1972: pls 72-73. Etymology | Refers to the geographic occurrence. Holotype SAS-H126B/3 (Figs 164-165), Bed B locality 105, south-eastern shores of False Bay, Zululand, St Lucia Formation, Late Santonian. Material Paratypes are SAS-H126/1, H126/2, H126A?/11, H126B/1, H126B/2, H126B/3a, H126E/1, H127B/1, Z1144—Z1147, Z1822, Z1869-Z1871, Z1881, Z1952, Z1952A-E, Z2054, Z2058, Z2199, SAM—PCZ5884, PCZ5886, all from locality 105, south-eastern shores of False Bay, Zululand, St Lucia Formation, Santonian III to Campanian I; and SAM-5492, and SAS-Z353 from an unknown locality near Umkwelane Hill, probably locality 14, St Lucia Forma- tion, Santonian to Campanian. The letter after specimens with catalogue numbers beginning with H126 denotes the horizon at locality 105 (see Fig. 130). Dimensions D Wb Wh Wb: Wh U Int. Ext. H126E/1 91,0 29,0(31,9) 37,0(40,4) 0,78 31,5(34,6) 23 37 Z1146 92,7 30,3(32,7) 39,9(43,0) 0,76 27,6(29,7) 23 35 Z1145 133,0 45,0(33,8) 55,0(41,4) 0,82 50,0(37,6) 19 31 H126B/2 117,0 36,0(30,8) 44,0(37,6) 0,82 43,0(36,8) 13x2 26 Z1952 145,0 45,0(31,0) 57,0(39,3) 0,79 55,0(37,9) 24 28 Z1869 150,0 49,0(32,6) —«62,0(41,3) 0,79 55,0(36,7) 13 oy Z2054 153,5 43,0(28,0) 59,0(38,4) 0,72 59,0(38,4) 24 33 Z1952B 180,0 57,0(31,7) 67,0(37,2) 0,85 69,0(38,3) 13x2 30 Z1844 at 180,0 52,0(28,9) _ 67,0(37,2) 0,78 68,0(37,8) 15x2 30 at 76,0 26,5(34,9) 34,0(44,7) 0,78 27,5(36,2) 14 36 Z1870 180,0 62,0(34,4) 70,0(38,9) 0,89 71,0(9,4) 24 26 H126B/3 187,0 56,0(29,9) 73,0(39,0) 0,77 69,0(36,9) 18 28 Z2199 205,0 58,0(28,3) —-75,5(26,8) 0,77 83,0(40,5) 30 30 Z1822 261,0 75,0(28,7) _ 94,0(36,0) 0,79 LOMOGSaniaeaD 24 PCZ5884 263,0 76,0(28,9) 97,0(36,9) 0,78 97,0(36,9) —_ — Z1881 230,0 62,0(26,9) 79,0(34,3) 0,78 88,0(38,3) 29 — Z353 236,0 63,0(26,7) 90,0(38,1) 0,70 90,0(38,0) 26 38 Diagnosis Extremely variable, but basically moderately involute, with umbilical diameter of about 35 to 40 per cent in the adult stage, in juvenile specimens less. Whorl section higher than wide, with maximum costal width at the lateral (2) tubercle, or, infrequently in juvenile specimens, at umbilical edge. Umbilical wall varies from rounded to near-vertical to overhanging. Ornament varies CRETACEOUS FAUNAS FROM SOUTH AFRICA 215 Fig. 164. Texanites soutoni natalense ssp. nov. Holotype SAS—H126B/3. x 0,75. from weak to strong. Some specimens tend to develop a smooth Submortoni- ceras-like ornament on the inner phragmocone whorls, followed by normal Texanites ornament on the outer phragmocone whorls. The umbilical (1) tubercles point inward, the lateral (2) tubercles are clavate on the inner whorls, but tend to become radially elongated on the outer whorls, and may be very prominent. The submarginal, marginal and external (3-5) tubercles are clavate throughout. Ribs on body chamber may curve slightly forward over the flanks, with oblique, prorsiradiate extensions over the umbilical wall. Bifurcations and 216 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 165. Texanites soutoni natalense ssp. nov. Holotype SAS-H126B/3. x 0,9. CRETACEOUS FAUNAS FROM SOUTH AFRICA 217 Fig. 166. Texanites soutoni natalense ssp. nov. Paratype SAS-Z2199. x 0,67. intercalations occur frequently on the inner whorls, but becume scarcer on the outer whorls. Suture is of the normal collignoniceratid pattern, but with wide, open lateral lobe (L) and U,, tending to become minutely frilled in some speci- mens (Fig. 185). Description Texanitids occur abundantly along the south-eastern shores of False Bay, Zululand, at locality 105, a steep cliff-face approximately 10 m high (see p. 173, 218 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 130). A few specimens from Umkwelane Hill are also conspecific with this assemblage. The material is extremely variable and hardly two specimens are alike. Differences are restricted mainly to strength of ornament at comparable diameters, and relative proportions. For descriptive purposes, three morpho- logical varieties may be recognized: a typical form, an inflated form, and a compressed form. Differences between these three varieties may be slight and boundaries gradational, and, lacking precise stratigraphic control, are probably predominantly isochronous. In lieu of long descriptions the extent of variation will be illustrated as far as possible. Forma typica Apart from the holotype, SAS—H126B/3 (Figs 164-165), specimens to be referred to the typical form are SAS—Z2199 (Figs 166-167A), SAS—H126B/1, SAS-H126B/2 (Fig. 168), SAS—H126B/3a (Fig. 169), SAS—Z1952 (Figs 170-171), - SAS-Z1144 (Fig. 172) and SAM-5492 (Fig. 173). These specimens all have inner whorls with weaker ornament than the outer whorls, but details differ considerably. Thus SAS—H126B/3a (Fig. 169) has relatively coarser and more prominent ornament on the inner whorls at comparable diameters than the holotype, SAS-—H126B/3 or SAS—H126B/2, even though they all occur within the same horizon at locality 105. SAS—Z1144 differs from the rest of the speci- mens in having abundant weak, bifurcating ribs on the inner whorls. One fragment, with part of the body chamber preserved, SAS—H126/2 (Fig. 174), with weakened but still visible ornament, may possibly belong to the typical form. None of the specimens has the complete body chamber preserved. Var. inflata Specimen SAS-Z1870 (Figs 175-176), and to a lesser extent SAS—Z1822 (Figs 177-178) differ from the rest of the material in being much more inflated and in having a smaller umbilical diameter, and may be referred to as var. inflata. This variety also differs from the typical form in lacking smooth or weaker ornamented inner whorls. Tuberculation here persists on part of the body chamber preserved in SAS—Z1879 (Fig. 175). Transitions to the typical form occur via SAS—H126B/2 (Fig. 168). Var. compressa Two specimens, SAS—Z1881 (Figs 179-180B) and SAS—Z353, differ from the others in being more laterally compressed and in having weaker ornament throughout, and may be referred to as var. compressa. As far as preservation permits examination, a smooth inner stage appears not to be present. Juvenile specimens A number of juvenile specimens, SAS—Z1146-Z1147, SAS—Z1871, SAS-— Z1952E, SAS-—H126E/1, SAS-H127B/1, and SAM-PCZ5886, and, possibly, CRETACEOUS FAUNAS FROM SOUTH AFRICA 219 A B Fig. 167. A. Texanites soutoni natalense ssp. nov. Paratype SAS—-Z2199. x 0,76. B. Texanites presoutoni natalense ssp. nov. Paratype SAM-PCZ5885. x 0,72. 220 ANNALS OF THE SOUTH AFRICAN MUSEUM NMB-D1320 are also referable to this species plexus, although positive refer- ence to any of the above varieties is not possible. Variation in respect of strength of ornament and whorl section is as considerable as in the adult population. Specimens SAS—H127B/1 (Fig. 183B) and SAS—H126E/1 (Fig. 181B) suggest that the specimens with more compressed Submortoniceras-like whorls occur higher up in the sequence than those with more inflated whorl sections. Two sets of specimens with similar relative proportions at equal diameters, SAS-Z1146-Z1147, SAS-Z1871, SAS-H126E/1, and SAS-—Z1952E, SAS- H127E/1, SAM-PCZ 5886 respectively, again illustrate the variation in strength of ornament (Figs 181-183), NMB-—D1320 is more inflated than any of the other juvenile specimens, and may possibly belong to another species. Fig. 168. Texanites soutoni natalense ssp. nov. Paratype SAS—H126B/2. Typical form. x 0,89. CRETACEOUS FAUNAS FROM SOUTH AFRICA 221 Interrelationships of varieties The majority of specimens in the present collection belong to the typical form, whereas the other two varieties are poorly represented. Due to lack of sufficient stratigraphic data, the temporal relationship of the different varieties is not very clear. We know only that the typical Submortoniceras forms are youngest, and that those forms with smooth Submortoniceras-like inner whorls Fig. 169. Texanites soutoni natalense ssp. nov. Paratype SAS—H126B/3a. Typical form. x 0,82. 222 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 170. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1952. Typical form. x 1. first occur in Bed B at locality 105. Suggested lines of development are outlined in Figure 132. Submortoniceras woodsi natalense ssp. nov. may be derived protero- genetically from 7. soutoni natalense forma typica; the inner smooth stage is retained for progressively longer periods of growth, accompanied by a simul- taneous reduction of umbilical width and compression. Var. inflata appears to be sterile. As in T. soutoni soutoni, it seems possible to derive a widely umbilicate, CRETACEOUS FAUNAS FROM SOUTH AFRICA 223 Fig. 171. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1952. Typical form. x 1. 224 ANNALS OF THE SOUTH AFRICAN MUSEUM compressed submortoniceratine form such as SAS—Z1762 (Fig. 184) from var. compressa. As in the nominate subspecies, this compressed lineage also seems to be a dead-end. Discussion This subspecies is very variable, and should be interpreted in terms of consanguinity of characteristics predominating in the majority of specimens rather than narrowly defined characteristics present only in the holotype. The stratigraphic occurrence of typical specimens in Bed B at locality 105 (Fig. 130), immediately overlying Bed A containing T. presoutoni natalense, may suggest genetic relationship between the two species, as here assumed (Fig. 132). Similarly, the relatively higher stratigraphic occurrence of Sub- mortoniceras woodsi natalense again seems to support genetic relationship between the two. T. soutoni natalense ssp. nov. is of age similar to the predominantly Pondo- land subspecies 7. soutoni s.s. It is associated with an ammonoid fauna con- sisting of Hauericeras gardeni, Pseudoschloenbachia umbulazi and varieties, ‘Heteroceras’ amapondense and Pseudophyllites latus (Fig. 130), similar to that of Umzamba Cliff (Klinger & Kennedy in press) but in much reduced numbers. Coarsely ornamented forms lacking the inner whorls, as SAS—Z1952 (Fig. 170), are indistinguishable from inflated forms of 7. soutoni s.s. Current evidence seems to suggest that 7. soutoni s.s. flourished in the shallow water of the transgressive sequence ranging from Umkwelane Hill in the north to Pondoland in the south, whereas 7. soutoni natalense inhabited the deeper part of the sedimentary basin in the False Bay area. Intermingling took place in the vicinity of Umkwelane Hill. In this regard T. soutoni soutoni and T. soutoni natalense may qualify for subspecific status in the biological sense. Major differences between T. soutoni soutoni and T. soutoni natalense are: the presence of a relatively smooth Submortoniceras-like stage on the inner whorls of some specimens, followed by more coarsely ornamented Texanites- like ornament on the outer whorls; generally smaller size and the presence of a wide, shallow lateral and umbilical lobe; and more inflated whorl section in T. soutoni natalense. The juvenile specimens described above (Figs 181-183), may be compared with a number of figured texanitid species, but specific differentiation at these diameters is impossible, e.g.: Submortoniceras woodsi (Spath) (see below). Submortoniceras tequesquitense Young (see especially Young 1963, pl. 51 (fig. 1), pl. 28 (fig. 1)). Texanites texanus gallica (in Young 1963, especially pl. 38 (fig. 3)). Texanites hourcgi Collignon (Collignon 1948, pl. 7(1) (fig. 1)) ete. The adult stages of these species, however, usually reveal the differences with T. soutoni natalense. Vp CRETACEOUS FAUNAS FROM SOUTH AFRICA lform. X 0,89. ica Typ Fig. 172. Texanites soutoni natalense ssp. nov. Paratype SAS-Z1144 226 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 173. Texanites soutoni natalense ssp. nov. Paratype SAM-—5492. Typical form. This is the specimen described by Spath (1921: 235) as Mortoniceras sp. aff. soutoni. X 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 227 Fig. 174. Texanites soutoni natalense ssp. nov. Paratype SAS—H126/1. Specimen with part of the body chamber preserved; probably belonging to the typical form. Note continuation of pentatuberculate ornament on body chamber section. x 0,57. 228 ANNALS OF THE SOUTH AFRICAN MUSEUM SAS-Z1881 (Fig. 179), here named var. compressa, is remarkably similar to the holotype of Texanites shiloensis (see Young 1963, pl. 46 (fig. 1)) in having a more or less Submortoniceras-like ornament on the outer whorl. As shown below, relative proportions are similar. Dimensions D Wb Wh U Int. Ext. UT-1986 275,0 — (32,0) (40,0) 31 31 (T. shiloensis) Z1881 230,0 62,0(26,9) 79,0(34,3) 88,0(38,3) 30 — Fig. 175. Texanites soutoni natalense ssp. nov. Paratype SAS-Z1870. Variety inflata. x 0,68. CRETACEOUS FAUNAS FROM SOUTH AFRICA 229 Fig. 176. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1870. Variety inflata. x 0,68. 230 ANNALS OF THE SOUTH AFRICAN MUSEUM Yj ‘YW - Fig. 177. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1822. Variety inflata. x 0,52. The holotype of T. shiloensis has a measured diameter of 275 mm, and the only apparent difference between the two, apart from size, is the finer, denser ribbing with frequent intercalations on the inner whorls. In addition to being very similar to the present material, T. shiloensis is of further interest in that it grades vertically into Submortoniceras tequesquitense. As far as relative size is concerned, 7. shiloensis is close to the compressed variety of T. soutoni S.S., €.g. specimen SAM-7077 (Fig. 146), but in the weakened ornament is closer to T. soutoni natalense, e.g. SAS—Z1881 (Fig. 179). CRETACEOUS FAUNAS FROM SOUTH AFRICA 231 Given sufficient material, it may eventually prove possible to refer T. shilo- ensis and S. tequesquitense to the group of T. soutoni-S. woodsi at subspecific level, as a North American population. Occurrence Upper Santonian of Zululand. Genus Submortoniceras Spath, 1921 (= Butticeras Anderson, 1958 ?=Jimenesites Carrasco, 1967 ?= Antsirasirella Collignon, 1966 ?= Ankilizatella Collignon, 1970) Type species Mortoniceras woodsi Spath, 1921, by original designation of Spath. Diagnosis Coiling generally involute, with whorls increasing rapidly in size. Penta- tuberculate from a relatively early stage, but there is a distinct tendency towards loss of lateral ornament on the outer whorls. Whorl section generally com- pressed, especially in the later species. Discussion Intraspecific variation in Submortoniceras species is bewildering, even by Texanites standards as applied in the South African material. The large number of seemingly endemic species probably results from the difficulty encountered in trying to compare species from different areas. The species below are those referred to Submortoniceras in the literature. A great number of these are probably synonymous, as will be discussed below. S. angusteumbilicatum Collignon (1948: 105, pl. 16 (figs 1-1b, 2—2a)) from the Middle Campanian of Madagascar. S. behamotrense Collignon (1970: 46, pl. 626 (figs 2318—2320)) from the Middle Campanian of Madagascar. S. candelariae Young (1963: 102, pl. 56 (figs 1, 3-4), pl. 60 (fig. 8), text-figs 20b, 28af, 29ae, 34af) from the Lower Campanian of the Gulf Coast of North America. S. chicoense (Trask) (1856: 92, pl. 2 (fig. 1)) from the Lower Campanian of California, Texas, and James Ross Island Group. S. collignoni Shimizu, 1935 (=M. woodsi in Collignon 1932: 35, pl. 5 (fig. 2-2b) text-fig. 19) from the Middle Campanian of Madagascar. S. condamyi Collignon (1932: 36, pl. 5 (fig. 3-3b), text-fig. 20) from the Lower Campanian of Madagascar and South Africa. S. debile Collignon (1948: 103, pl. 19(13) (fig. 3-3a)) from the Middle Cam- panian of Madagascar. S. dubium Collignon (1948: 97, pl. 18(12) (fig. 3-3b)) from the Middle Cam- panian of Madagascar. 252 ANNALS OF THE SOUTH AFRICAN MUSEUM S. elimatum Collignon (1948: 104, pl. 20(14) (fig. 2-2a)) from the Lower Cam- panian of Madagascar. S. franiattae Collignon (1970: 43, pl. 626 (fig. 2316)) from the Middle Campanian of Madagascar. S. imlayi Collignon (1948: 101, pl. 20(16) (fig. 3-3a)) from the Middle Cam- panian of Madagascar. S. johannisludovici Collignon (1970: 47, pl. 628 (fig. 2321)) from the Middle Campanian of Madagascar. S. mariscalense Young (1963: 104, pl. 59 (fig. 3), pl. 60 (figs 1, 4-6), text-fig. 14bf) from the Lower Campanian of the Gulf Coast of North America. S. pauper Collignon (1948: 103, pl. 18 (fig. 2-2a)) from the Lower and Middle Campanian of Madagascar. S. piveteaui Collignon (1948: 102, pl. 7(1) (fig. 6-6b), pl. 19(13) (fig. 2-2a)) from the Middle Campanian of Madagascar. S. punctatum Collignon (1948: 99, pl: 18(12) (fig. 1-1a)) from the Lower Cam- panian of Madagascar. S. rarecostum Collignon (1948: 101, pl. 20(14) (fig. 3-3a)) from the Middle Campanian of Madagascar. S. renniei Collignon (1948: 106, pl. 17 (fig. 1)) from the Lower Campanian of Madagascar. S. rugetae Collignon (1969: 189, pl. 593 (fig. 2242)) from the Lower Campanian of Madagascar. S. sancarlosense Young (1963: 100, pl. 55 (figs 1-4), pl. 62 (fig. 3), text-figs 20g, 27d) from the Lower Campanian of the Gulf Coast of North America. S. spathi Collignon (1948: 106, pl. 20(14) (fig. 1-la)) from the Middle Cam- panian of Madagascar. S. tenuicostulatum Collignon (1948: 96, pl. 19(13) (fig. 1-1b)) from the Lower and Middle Campanian of Madagascar. S. tequesquitense Young (1963: 97, pl. 28 (fig. 1), pl. 42 (figs 1-2), pl. 44 (figs 4-5), pl. 51 (figs 1-2), pl. 52 (figs 1-4), pl. 57 (fig. 4), pl. 70 (fig. 1), text-figs 12b, 28b) from the Lower Campanian of the Gulf Coast of North America. S. uddeni Young (1963: 105, pl. 59 (figs 1-2, 4-9), pl. 60 (figs 2-3, 7, 9-10), text-figs 14de, 28c) from the Lower Campanian of the Gulf Coast of North America. S. vanuxemi (Morton) (1830: 244, pl. 3 (figs 3-4)) from the Campanian of North America. S. vandaliaense Young (1963: 102, pl. 55 (figs 6-7), text-fig. 26a) from the Lower Campanian of the Gulf Coast of North America. S. woodsi (Spath) (1921: 223, pl. 21 (fig. la—d)) from the Lower Campanian of Madagascar and South Africa. ‘Butticeras’ buttense Anderson (1958: 272, pl. 53 (figs 3—-4)) from the Lower Campanian of the Pacific Coast of America. ‘Butticeras’ studleyi Anderson (1958: 272, pl. 51 (figs 3—-4)) from the Lower Campanian of the Pacific Coast of America. CRETACEOUS FAUNAS FROM SOUTH AFRICA 233 Fig. 178. Texanites soutoni natalense ssp. nov. Para- type SAS-Z1822. Variety inflata. x 0,6. 234 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 179. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1881. Variety compressa. x 0,59. Anderson (1958: 272) introduced the genus Butticeras, with type species Butticeras buttense, for a group of small ammonites in which the keel is lost in maturity, the venter becomes rounded, and all lateral tuberculation is lost except the umbilical and external rows. Matsumoto (1959: 132) speculated that Butticeras may represent an extreme form of Submortoniceras chicoense, though not definitely committing himself. Consequently, Butticeras should be regarded as a junior synonym of Submortoniceras. This appears to be reasonable pro- cedure, as subdivision of such a variable genus should be limited to the absolute CRETACEOUS FAUNAS FROM SOUTH AFRICA 235 minimum. On this basis it would also be advisable to include the monotypical subgenus S. (Antsirasirella) Collignon (type species S. (Antsirasirella) stellata Collignon (1969: 201, pl. 599 (fig. 2252)) in the strict synonymy of Submortoni- ceras. S. (Antsirasirella) stellata has extremely reduced lateral ornament, similar to Butticeras, but has a flattened venter in the adult stage, thus differing from the latter. For the present it is probably best to consider S. (Antsirasirella) as merely representing an extreme development of the attenuation of ornament in Submortoniceras. The monospecific subgenus Texanites (Ankilizatella) Collignon (type species T. (Ankilizatella) ankilizatensis Collignon 1970: 50-51, pl. 629 (figs 2322-2325)) from the Middle Campanian of Madagascar is probably better placed in Submortoniceras than in Texanites. The type (and only) species has a compressed disc shape and smooth outer whorls as in Submortoniceras, but a wide umbilicus as in Texanites. The compressed section and lack of ornament on the outer whorls are more characteristic of Submortoniceras than of Texa- nites. As far as umbilical width is concerned, Ankilizatella possibly connects with Submortoniceras s.s. via S. behamotrense, S. renniei and S. mariscalense; especially the former. Submortoniceras is easily distinguished from most texanitine genera by virtue of the generally tight coiling and attenuation of ornament on the outer whorls. Difficulties arise, however, in the Middle Campanian in distinguishing between Submortoniceras and Menabites (Delawarella), especially when the inner whorls are lacking. Matsumoto (1959: 125-6) preferred to refer M. (Dela- warella) as a subgenus to Submortoniceras rather than to Menabites, but later (Matsumoto 1970: 299) retracted this view. From a phylogenetic point of view, however, this former procedure cannot be accepted (see also Young 1963: 39). M. (Delawarella) typically has a multiplied row of external tubercles, charac- teristic of Menabites s.1., and is often trituberculate to a large diameter. How- ever, when lateral ornament is attenuated, it is difficult to see whether the greater number of external tubercles is due to the original multiplication of Menabites s.l., or due to abundant bifurcations and intercalations of Submortoniceras. Young’s (1963: 76) key to the classification of the Texanitinae is of little use in this case. Despite this example of isochronous homoeomorphy, Submortoni- ceras and Delawarella should be maintained separate, because of different phylogenies and ontogenetic development. Both Young (1963: 39) and Matsumoto (1970: 239) agree in deriving Submortoniceras from Texanites, a view corroborated by the South African material. As described above, S. woodsi is derived gradually from T. soutoni through reduction of umbilical diameter and effacement of the ornament on the outer whorls. Submortoniceras appears to be a dead-end in texanitine development, and the last Submortoniceras species occur in the Middle Cam- panian of Madagascar. As is seen from the list of species referred to the genus, the majority seem to be endemic to the regions from which they were first described. None of the 236 ANNALS OF THE SOUTH AFRICAN MUSEUM SS : \ ~ hep Pi Gj My WH hy, UY); Fig. 180. Texanites soutoni natalense ssp. nov. A. Paratype SAS-—Z1144. Typical form. x 0,6. B. Paratype SAS—Z1881. Var. compressa. X 0,69. CRETACEOUS FAUNAS FROM SOUTH AFRICA 237 Fig. 181. Texanites soutoni natalense ssp. nov. Juvenile paratypes. A. Z1147. B. SAS-H126E/1. x 1. 238 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 182. Texanites soutoni natalense ssp. nov. Juvenile paratype SAS—Z1146 with much attenuated ornament. x 1. species described from Madagascar is known to occur definitely in North America and vice versa. This causes one to think that Submortoniceras is either restricted to certain geographic areas with little or no intermingling between the populations, but highly successful and populous as far as number of species are concerned, or it is an extremely successful group with limited speciation potential but wide dispersion and tremendous intraspecific variation. Analysis of the South African material and work by Matsumoto (1959) seem to suggest the latter. In his examination of S. chicoense, Matsumoto (1959) was able to recognize three morphological variants, forma «, B and y, which, by traditional systematic procedure, would probably qualify for separate specific rank. He found, however, that there was a continuous gradation from one form to the other, and that the different types may even occur in the same nodule. In addition, Matsumoto tentatively compared the Californian specimens with some of the numerous species of Submortoniceras described by Collignon (1948) from Berere, Madagascar. S. piveteaui resembles S. chicoense forma a, S. imlayi is close to forma B, and S. angusteumbilicatum and S. spathi are close CRETACEOUS FAUNAS FROM SOUTH AFRICA 239 Fig. 183. Texanites soutoni natalense ssp. nov. Juvenile paratypes. A. SAM-—PCZ5886. B. SAS-H126B/1. C. SAS—Z1952. x 1. 240 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 184. Texanites soutoni natalense ssp. nov. Paratype SAS—Z1762. Compressed, widely umbilicate variety probably derived from var. compressa (see also Fig. 132). x 0,48. CRETACEOUS FAUNAS FROM SOUTH AFRICA 241 Fig. 185. Texanites soutoni natalense ssp. nov. Suture lines of paratypes. A. SAS-Z1881. B. SAS-Z1146. C. SAS-Z358. D. SAS—Z1871. E. SAS—Z1147. to forma y of S. chicoense. Even though Matsumoto would not definitely commit himself to synonymizing the species without having seen the Mada- gascar species in their stratigraphic context, it is gratifying to see that the ‘endemic’ gap between the North American and Madagascar species can be narrowed. : The South African material is as variable as S. chicoense, and is here referred to the two existing species S. woodsi and S. condamyi. As will be seen in the systematic descriptions below, ornament and suture are extremely variable. Specimens with smooth inner whorls occur at the same stratigraphic level as heavily ornamented ones, and are connected by transitions. In the South African material, ornament is partially associated with geographic location, suggesting that ornament alone is of very little systematic value in the classi- fication of Submortoniceras species. There appears to be a definite trend in the South African Submortoniceras population towards progressive tighter coiling and compression of the whorls. At present the significance of this is not clear, 242 ANNALS OF THE SOUTH AFRICAN MUSEUM but the authors would suggest that it may be connected in some way to a slightly deeper water habitat or more active way of life. A similar trend is found in the Gulf Coast deposits of North America where Texanites shiloensis grades upward into Submortoniceras tequesquitense (see Young 1963: 90). Consequently, degree of involution is here considered of maximum value in classifying Submortoniceras species. This all suggests that the number of species listed above may be greatly reduced, provided sufficient detailed stratigraphic data and material become available. Occurrence Submortoniceras species occur profusely in the Lower and Middle Cam- panian of Madagascar, and in slightly smaller numbers in the Upper Santonian— Lower Campanian of South Africa, the Lower Campanian of the Pacific Coast and Gulf Coast areas of North America, and sporadically in British Columbia. A single specimen of S. chicoense was reported from the Sub-Antarctic James Ross Island Group. Young (1963: 39) refers Redtenbacher’s (1873) Ammonites propoetidum to the genus Submortoniceras, but the authors would rather place this species in Gauthiericeras, as was done by Reyment (1958), or in Zuluiceras Van Hoepen. Submortoniceras woodsi s.1. (Spath, 1921) Figs 186-199, 200B Mortoniceras woodsi Spath, 1921: 232, pl. 21 (fig. la—d). Non Besairie, 1930: 637, pl. 67 (fig. 1) (=S. angusteumbilicatum Collignon). ?Non Collignon, 1932: 35, pl. 5 (fig. 2-2b) (=S. collignoni Shimizu). Mortoniceras vanuxemi (Morton): Spath, 1921: 308, pl. 23 (fig. 4a—4b). Submortoniceras woodsi (Spath): Collignon, 1948: 43. Kennedy & Klinger, 1973: 103, pl. 2 (fig. 3), pl. 3 (figs 2-3), pl. 5 (fig. 3). Type Holotype is SAM-5451 (Fig. 186) from an unknown horizon at Umkwe- lane Hill near Mtubatuba, Zululand, St Lucia Formation, Santonian to Campanian. Material SAS-Z1875, Z1138, Z1893, SAM-—PCZ5884, PCZ5883, all from locality 105, south-eastern shores of False Bay, St Lucia Formation, Upper San- tonian to Lower Campanian; SAM-PCZ5896 and BMNH-C81489, from Umkwelane Hill, near locality 14, St Lucia Formation, Upper Santonian to Lower Campanian; SAM-—PCZ5897, locality 6, Richards Bay, St Lucia Formation, Upper Santonian to Lower Campanian; SAM-—PCP5716, Bed A3 locality 1, Umzamba Formation, Upper Santonian to Lower Campanian; BMNH-C81487, locality 74, Santonian III; and BMNH-C81488 from the same locality, St Lucia Formation, Campanian I. 243 CRETACEOUS FAUNAS FROM SOUTH AFRICA 1 yo af 4 1 ! / 43,562.17 6.= S.renniei 7.= S.collignoni 34 38 18 22 26 30 UMBILICAL WIDTH °/,——> Fig. 211. Diagram to illustrate overlapping umbilical diameters of various Submortoniceras species. CRETACEOUS FAUNAS FROM SOUTH AFRICA DiS nately, S. mariscalense is monotypic, thus precluding detailed comparison between the two species. Occurrence Upper Santonian to Lower Campanian of Pondoland and Zululand, and Lower Campanian of Madagascar. Genus Bevahites Collignon, 1948 Type species Bevahites bevahensis Collignon by original designation of Collignon (1948). Diagnosis Shell of variable size; generally small, with compressed or subsquare whorls. The pentatuberculate stage is attained rapidly. Adult ornament is charac- terized by the proximity of the submarginal (3) and marginal (4) tubercles, and multiplication of the external (5) tubercles. Due to bifurcations many more external than internal rows of tubercle occur. Discussion Bevahites is poorly represented in the present collections which offer little basis for discussion of the genus. Originally Parabevahites was included in Bevahites as a subgenus, but most workers now follow Wright (1957: L432) in referring Parabevahites to Paratexanites as a subgenus, and even consider including Parabevahites in the synonomy of Paratexanites as is done here (see p. 14). Bevahites is normally easy to distinguish from the other texanitid genera because of the proximity of the submarginal (3) and marginal (4) tubercles, and also because of the multiplication of the external row of tubercles. In this latter respect there is close comparison to Menabites s.s. Both have more external (5) tubercles than umbilical (1) ones, but those of Bevahites are also partially due to extensive bifurcations. It is difficult to distinguish between Bevahites species which lack the pronounced doubling of the submarginal (3) and marginal (4) tubercles, and species of Menabites s.s. which attain the pentatuberculate stage of development at an early stage. From the list supplied below, it is clear that Bevahites is virtually restricted to Madagascar, with only two reports from the Gulf Coast of North America and specifically unidentifiable material from France (Fabre-Taxy 1963: 19-20). To this may now be added a new species and two doubtful contenders from the South African Cretaceous. Species referred to Bevahites in the literature are: B. bevahensis Collignon (1948: 84, pl. 11(5) (fig. 3-3b)) from the Upper San- tonian of Madagascar and Gulf Coast of North America, possibly ranging up to the Lower Campanian (including Collignon’s ‘mutation’ intermedia). 274 ANNALS OF THE SOUTH AFRICAN MUSEUM B. dichotomus Collignon (1948: 93, pl. 7(1) (fig. 4), pl. 17(11) (fig. 1-1b)) from the Lower Campanian of Madagascar. B. costatus Collignon (1948: 90, pl. 7(1) (fig. 5), pl. 15(9) (fig. 1-1b)) from the Lower Campanian of Madagascar (including ‘mutations’ crassicostata Collignon, 1948: 91, pl. 15Q) (fig. 2-2b) from the Lower Campanian of Madagascar; costulata Collignon, 1948: 92, pl. 16(10) (fig. 22a) from the Middle Campanian of Madagascar; rarecostata Collignon, 1948: 92, pl. 16(10) (fig. 1) from the Middle Campanian of Madagascar; inornata Fig. 212. Submortoniceras condamyi (Collignon, 1948). SAM-PCP5883. x 0,8. CRETACEOUS FAUNAS FROM SOUTH AFRICA 275 Fig. 213. Submortoniceras condamyi (Collignon, 1948). Suture lines of specimens referable to B. B. subspecies umzambiense. A. PCP5715. B. NMB-D1191. Collignon, 1948: 93, pl. 16(10) (fig. 3-3b) from the Middle Campanian of Madagascar). . costatus Collignon coahuilaensis Young (1963: 96, pl. 47 (figs 1-4), pl. 71 (fig. 5), text-fig. 34b) from the Lower Campanian of the Gulf Coast of North America. . curvicostatus Collignon (1948: 87, pl. 13(7) (fig. 1-1b)) from the Lower Campanian of Madagascar. . lapparenti Collignon (1948: 88, pl. 14(8) (fig. 1-1b)) from the Middle Cam- panian of Madagascar. . quadratus Collignon (1948: 86, pl. 12(6) (fig. 3-3b)) from the Lower and Middle Campanian of Madagascar. . romani Collignon (1948: 95, pl. 17(11) (fig. 2—2a)) from the Middle Cam- panian of Madagascar. . subbevahensis Collignon (1948: 85, pl. 11(5) (fig. 4-4a)) from the Upper Santonian of Madagascar. subogivalis Collignon (1948: 87, pl. 13(7) (fig. 2-2b)) from the Middle Cam- panian of Madagascar. subquadratus Collignon (1948: 89, pl. 13(7) (fig. 3—3a, pl. 14(8) (fig. 2—2a)) from the Lower and Middle Campanian of Madagascar. Occurrence Bevahites occurs mainly in Madagascar, from the Upper Santonian to Middle Campanian and Upper Santonian of the North American Gulf Coast, and imprecise horizons in the Santonian and Campanian of Zululand. Specifi- cally unidentifiable material was reported from the Lower Santonian of France (Fabre-Taxy 1963). 276 ANNALS OF THE SOUTH AFRICAN MUSEUM Bevahites? enigma sp. nov. Fig. 214 Holotype SAS-H68/2 (Fig. 214), locality 103, hill slopes near the Nyalazi River Estuary, Zululand, St Lucia Formation, Campanian I or II. Etymology Refers to uncertain generic allocation. Material The holotype only. Dimensions D Wb Wh Wb: Wh U Ur(1) LZt(2) St(3) Mrt(4) Et(5) 90,0 42,0(46,7) 37,0(41,1) 1,13 30,0(33,3) 19 ? 23 29 36 Description The specimen is septate throughout, but lacks the very innermost whorls. The inner whorls preserved have a prominently depressed whorl section, much wider than high, but with increasing diameter the whorl section becomes more rounded, with a slightly outward-leaning umbilical wall, a prominent umbilical edge merging into little-inflated flanks which join imperceptibly with a broadly rounded to flattened venter. At a diameter of 35 mm, ornament consists of a rounded umbilical (1) tubercle, a very prominent, radially elongated ventrolateral (submarginal) tubercle, closely situated indistinct marginal (4) tubercles and prominently rounded external (5) tubercles. The external (5), and in some cases already the submarginal (4), tubercles outnumber the submarginal (3) ones due to bifurca- tion or intercalation of ribbing near the venter at the site of the submarginal (3) tubercle. With increasing diameter, the submarginal (3) tubercles become weaker, and the distance between the latter and the marginal (4) tubercles becomes greater. Lateral (2) tubercles start appearing at a diameter of c. 45 mm on the dorsal quarter of the flanks as very indistinct little nodes. With further increase in diameter, all tuberculation weakens and becomes subordinate to ribbing, and becomes equidistantly spaced over the flanks in the process. The ratio of umbilical (1) to submarginal (3), marginal (4) and external (5) tubercles varies tremendously due to bifurcations and intercalations, as illus- trated diagrammatically in Figure 214D. Bifurcations or intercalations may occur at the umbilical (1), the submarginal (3) or the marginal (4) tubercles, giving rise to the odd ratio of tubercles as provided in the list of dimensions. Thus, one umbilical (1) tubercle may correspond to up to four external (5) tubercles. CRETACEOUS FAUNAS FROM SOUTH AFRICA 2TH Fig. 214. Bevahites ? enigma sp. nov. Holotype SAS-H68/2. x 1. 278 ANNALS OF THE SOUTH AFRICAN MUSEUM The keel is entire and rather prominent on the early parts of the phragmo- cone, but tends to weaken on the last part preserved, and also shows slight undulations. The suture is imperfectly preserved, showing a wide, bifid lateral lobe (L). Discussion This species is readily recognized by the stout whorl section and by the irregularly bifurcating and intercalating ribbing, so much so that it requires specific rank, even though monotypic. The generic affinity of the species, how- ever, is enigmatic, hence the name. The relative late persistence of a quadri- tuberculate stage places this species close to Bevahites, whereas the eventual equidistant pentatuberculate ornament is more in line with that of Menabites. For the present it is considered advisable to refer the species to Bevahites with a mark of interrogation, as it appears to fit best in the group of Bevahites sp. gr. ex. quadratus of Collignon (1948: 86) which includes the nominate species B. quadratus, B. costatus, B. subogivalis, and B. lapparenti. None of these species, however, has as narrow an umbilical width or as irregularly dicho- tomizing ornament. The closest ally amongst the former is probably B. lappa- renti, which has a similar strongly developed ventrolateral tubercle in the early stages and which later splits into two. As far as irregular bifurcations and intercalations are concerned, analogies are to be found in B. bidichotomus, but the latter has a much more compressed whorl section than the Zululand species. Bevahites sp. A gr. ex. B. subquadratus Collignon, 1948 Fig. 215 Material NMB-D1185A and D1185B from an unknown locality in the Morrisvale area, probably the authors’ (Kennedy & Klinger 1975) locality 147 of Santonian Age. Description The two specimens differ considerably in terms of strength of ornament, but are both characterized by a compressed whorl section with maximum whorl breadth at the lateral tubercle. NMB-D1185A (Fig. 215A) has smooth inner whorls up to a diameter of c. 15 mm. At a diameter of c. 30 mm virtually complete Bevahites ornament is present, though very weak, but the submarginal (3) tubercle appears absent at that stage. With increasing diameter, however, the ornament becomes stronger, especially the lateral (2) and external (5) tubercles. NMB-D1185B (Fig. 215B-D), on the other hand, already has a distinct lateral ornament at 15 mm diameter, consisting of an umbilical (1) tubercle, 279 CRETACEOUS FAUNAS FROM SOUTH AFRICA Migs: vinnie yyy , 1948. A. NMB-D1185A. ignon xe subquadratus Coll NMB-D1185B. gr. ex. B. B-D. .A Fig. 215. Bevahites sp 280 ANNALS OF THE SOUTH AFRICAN MUSEUM connected by a stout rib to a prominent ventrolateral spine leaning against the umbilical wall. At 30 mm diameter the total ornament is visible, showing the typical joined submarginal (3) and marginal (4) tubercles and proliferation of the external (5) tubercles. The lateral (2) tubercle has apparently just appeared at this diameter, and is still weakly developed. Ornament strengthens with increasing diameter, and at a diameter of 50 mm the submarginal (3) and marginal (4) tubercles are completely separated. Ornament at this stage consists of radially elongated umbilical (1), very strong lateral (2), weak clavate sub- marginal (3) and marginal (4), and strong clavate external (5) tubercles. Unfortunately the outer whorl of NMB-—D1185B is crushed and eroded, but it shows the adult ornament consisting of strong umbilical, very strong lateral, and very weak marginal and submarginal and, again, strongly clavate external tubercles, and a very weak, narrow keel. The suture is not exposed in either of the specimens. Discussion Due to lack of precise stratigraphic data, it is not known if we are here dealing with one or two species, one with smooth inner whorls, and another with coarser-ornamented inner whorls, a situation perhaps comparable with that found in Texanites soutoni s.l. or Plesiotexanites stangeri. According to the data in the late E. C. N. van Hoepen’s catalogue, the material was found near the confluence of the Msinduzi and Mkuze Rivers, which is probably the equivalent of, or close to, the authors’ (Kennedy & Klinger 1975) locality 147 which was tentatively dated as Santonian. For the above reasons it is considered best to refer to the material merely as Bevahites sp. A gr. ex. B. subquadratus, which includes the nominate species, B. subquadratus, B. costatus (including mutants crassicosta, costulata, rarecosta and inornata), B. bidichotomus and B. romani. The inner whorls of B. subquadratus (see Collignon 1948, pl. 7(1) (fig. 5)) are similar to the coarsely ornamented inner whorls of NMB-D1185A (Fig. 215A), but ribbing is far more sinuous. Despite slight compression of the Zululand specimens, the holotype of B. subquadratus (Collignon 1948, pl. 14(8) (fig. 2-2a)) also appears more inflated. B. costatus, including Collignon’s (1948) ‘mutations’ crassicostata, costulata, rarecosta and inornata has a similar compressed whorl section, but in all cases lacks the strong development of external tubercles, and rather tends to have a more rounded venter. B. romani is related to the Zululand material in the eventual separation of the submarginal tubercles, but differs mainly in strength of the different rows of tubercles and in the late development of the lateral tubercle. Occurrence Probable Santonian of Zululand. CRETACEOUS FAUNAS FROM SOUTH AFRICA 281 Bevahites sp. B. gr. ex. B. subquadratus Collignon, 1948 Fig. 216 Material NMB-D1067 from an unknown locality near the north-western end of False Bay, Zululand, probably at locality 74 towards the top of the section measured by Kennedy & Klinger (1975: 293, text-fig. 8). Description and discussion The specimen consists of about a quarter of a whorl, is septate throughout and very sparsely ornamented. The whorl section is much higher than wide, with a vertical umbilical wall, slightly outward bulging flanks up to midflank, and then converging to a narrow, flat venter with a prominent row of clavate external (5) tubercles (Fig. 216B). Lateral ornament is very weak, consisting mainly of radially elongated umbilical (1) tubercles and about twice as many clavate external (5) tubercles. Lateral (2) tubercles merely appear as low nodes, and submarginal (3) and marginal (4) tubercles are absent for all practical purposes. The sutures are crowded, but appear to be very incised. The whorl section is obviously that of the group of B. subquadratus, but on the basis of this limited material positive identification is not really possible. The whorl section is reminiscent of that of the B. costatus specimen figured by Collignon (1948, pl. 15(9) (fig. 2a)), but has a more flattened venter and better developed ventral tubercles. This specimen differs from the previously described specimens of the group B. subquadratus in having a more compressed whorl section, and also in lacking the strong lateral ornament of the latter. Occurrence Locality data are imprecise, but probably Santonian to Lower Campanian. Genus Menabites Collignon, 1948 Type species Menabites menabensis Collignon, 1948, by original designation of Collignon, 1948. Diagnosis Menabites is identified by the relatively long persistence of the tritubercu- late stage, sometimes apparently to maturity, and also by the multiplication of the tubercles of the external (5) row, numbering two to three times as many as the marginal (4) ones. Discussion The whorl section, nature of ornamentation and persistence of the diag- nostic trituberculate stage are very variable, so much so that, apart from 282 ANNALS OF THE SOUTH AFRICAN MUSEUM Menabites s.s., three other subgenera are recognized: Bererella, Australiella and Delawarella. Menabites s.s. contains those forms in which the trituberculate stage passes relatively rapidly into a final pentatuberculate texanitine stage. Bererella (type species M. (B.) bererensis Collignon, 1948) contains widely umbilicate, flat forms with few and widely-spaced ribs. Australiella (type species M. (A.) australis (Besairie)) contains those forms with wide whorl sections, sometimes strongly inflated and enormous lateral (2, 3, 4) tubercles, and may apparently retain this trituberculate stage to large diameters or throughout life. Delawarella (type species M. (D.) delawarensis Collignon) contains forms with numerous ribs, closely spaced, often flexuous, and with attenuated orna- ment and a narrow umbilicus. Of all the described species and subspecies of Menabites s.1., about half appear to be restricted to Madagascar; the remaining known reports are from the Gulf Coast and adjacent areas of North America, Zululand and Hokkaido. As in Submortoniceras, it is difficult to decide whether this distribution pattern reflects real geographically separated species groups in each area, or, rather, the individual systematicist’s approach towards classification. Menabites s.\. is very easy to distinguish from the majority of other texani- tids on account of the multiplication of the external row of tubercles, but also on stratigraphic grounds, being apparently the last texanitids to survive of a very successful lineage dating from the Upper Coniacian. Differentiation between the four subgenera as outlined above sometimes becomes difficult, as characteristics often overlap, and the diagnostic differences appear to be based on ontogenetic features. Further differences arise in distinguishing between Menabites s.s. and Bevahites, both of which have a multiplied external row of tubercles. The basic distinguishing feature between Menabites s.s. and Bevahites is the longer persistence of the trituberculate stage in the former, and, to a lesser extent the proximity of the submarginal (3) and marginal (4) tubercles in Beyahites, although application of this criterion sometimes becomes subjective as the Zululand material shows. Unless the inner whorls are preserved to reveal details of ontogeny, it is difficult to distinguish between late representatives of Submortoniceras, e.g. S. johannisludovici Collignon, and smooth forms of Menabites (Delawarella). The Zululand specimens of Menabites are very large, with diameters of up to 700 mm in one M. (Delawarella) species. This, combined with the general lack of the early inner whorls, makes it difficult to identify these specimens with the much smaller species described from Madagascar. Difficulties occur especi- ally in the interpretation of M. (Australiella) species. Apart from the fact that all the Zululand representatives of this subgenus are highly variable, they grow to much larger diameters, and in doing so pass from the typically trituberculate Australiella stage to a pentatuberculate Menabites or Bererella stage. Unfortu- CRETACEOUS FAUNAS FROM SOUTH AFRICA 283 Fig. 216. Bevahites sp. B. gr. ex. B. subquadratus Collignon, 1948. NMB-D1067. x 1. 284 ANNALS OF THE SOUTH AFRICAN MUSEUM nately, the authors have insufficient data on the Madagascan material to find out whether these Australiella species were, in fact, all adult or immature. Phylogenetic relationships between Menabites s.1. and the other texanitine genera, and even between the subgenera within Menabites, are not quite clear. In Madagascar, Menabites s.s. starts appearing in the upper part of the Lower Campanian, Zone of Menabites boulei and Anapachydiscus arrialoorensis, but is replaced by representatives of Australiella, Delawarella and Bererella in the Middle Campanian. Data on the first appearance of Menabites in Zulu- land are scant, but the abundant occurrence of Australiella and Delawarella at the south-western tip of the Nibela Peninsula seems to suggest the same temporal sequence. This, however, is at variance with the views held by Matsumoto (1970: 298) and Young (1963: 107). Their views are supported by the alleged occurrence of Australiella in sediments of Santonian age. Locality data on Australiella austi- nensis Young are confusing (Young 1963: 116), and Young considered it to be ‘probably Lower Santonian, but could be lowest Campanian’(!). Australiella sp. aff. A. pattoni in Matsumoto (1970: 300) also lacks a firm dating and was dated as ‘probably Santonian’ (Matsumoto 1970: 300). On the basis of the Malagasy and Zululand material, the authors would rather be inclined to derive Australiella, Delawarella and Bererella from Menabites s.s. The origin of Menabites s.l. is obscure, and the Zululand material provides no direct information. Young (1963: 107) and Matsumoto (1970: 289) are inclined to derive Menabites (or Australiella) from Protexanites. Protexanites (Pleurotexanites) from the Mid-Santonian of Madagascar only has three rows of tubercles, of which the external ones are multiplied as in Menabites s.1., and is thus a possible ancestral form. Another possibility, which closes the time gap between Mid-Santonian and Early Campanian, is Bevahites. Through reduction of the lateral tubercle and strengthening of the double ventrolateral tubercle, the initial trituberculate situation as in early Menabites s.s. ontogeny can be derived. Clearly, no satisfactory answer to this question will be found unless based on detailed collecting. Venzo’s (1936: 88, pl. 3 (fig. 12)) Mortoniceras vinassai is definitely not a Menabites, but probably a Forresteria. The locality, ‘Riva sinistra del basso Umsinene’ corresponds to the authors’ (Kennedy & Klinger 1975) locality 72 or 73 of Upper Coniacian or Lower Santonian age. Species referred in literature to subgenus Menabites are: M. (Menabites) belli Young (1963: 106, pl. 54 (fig. 1), pl. 58 (fig. 2), pl. 70 (figs 2-4, 7), text-fig. 15a) from the Lower Campanian of the Gulf Coast of North America. M. (Menabites) boulei Collignon (1948: 14, pl. 22 (fig. 1-1b)) from the Lower and Middle Campanian of Madagascar. CRETACEOUS FAUNAS FROM SOUTH AFRICA 285 M. (Menabites) breistrofferi Collignon (1948: 19, pl. 24 (fig. 1-1b)) from the Campanian of Madagascar. M. (Menabites) cottreaui Collignon (1948: 16, pl. 23 (fig. 2-2b)) from the Middle Campanian of Madagascar. M. (Menabites) dabolavaensis Collignon (1948: 20, pl. 24 (fig. 3-3a)) from the Lower and ?Middle Campanian of Madagascar. M. (Menabites) densinodosus (Renz) (1936: 8, pl. 2 (fig. 1-1a)) from the Lower Campanian of the Gulf Coast of North America. M. (Menabites) gignouxi Collignon (1948: 15, pl. 22 (fig. 2-2b)) from the Campanian of Madagascar. M. (Menabites) internodosus (Renz) (1936: 10, pl. 3 (fig. 2—2b)) from the Lower Campanian of the Gulf Coast of North America. M. (Menabites) jacobi Collignon (1948: 13, pl. 20 (fig. 3-3b)) from the Middle Campanian of Madagascar. M. (Menabites) lenoblei Collignon (1948: 19, pl. 24 (fig. 1-1b)) from the Cam- panian of Madagascar. M. (Menabites) latissimus Collignon (1969: 192, pl. 594 (fig. 2244)) from the Lower Campanian of Madagascar. M. (Menabites) mazenoti Collignon (1948: 11, pl. 20 (fig. 2—2a)) from the Middle Campanian of Madagascar. M. (Menabites) menabensis Collignon (1948: 7, pl. 17 (figs 3a—b, 44a), pl. 18 (fig. 1-1b)) from the Lower Campanian of Madagascar. M. (Menabites) menabensis var. lata Collignon (1948: 9, pl. 18 (fig. 2—2b)) from the Lower Campanian of Madagascar. M. (Menabites) multiformis Collignon (1948: 12, pl. 7 (fig. 8-8b), pl. 19 (fig. 1-la)) from the Middle Campanian of Madagascar. M. (Menabites) paucituberculatus Collignon (1948: 21, pl. 24 (fig. 2—2a)) from the Middle Campanian of Madagascar. M. (Menabites) rarecostus Collignon (1969: 194, pl. 595 (fig. 2246)) from the Lower Campanian of Madagascar. M. (Menabites) savornini Collignon (1948: 9, pl. 7(1) (fig. 9-9a), pl. 19 (figs 2—2a, 3—3a), pl. 20 (fig. 1), pl. 21 (figs 1-1b, 2—2b)) from the Middle Campanian of Madagascar. M. (Menabites) tsirananai Collignon (1969: 196, pl. 595 (fig. 2247)) from the Lower Campanian of Madagascar. Menabites s.1. walnutensis Young (1963: 109, pl. 58 (figs 1, 4), text-figs 20ef, 26k) from a loose block, presumably Lower Campanian of the Gulf Coast region of North America. Species referred to subgenus M. (Delawarella) in the literature are: M. (Delawarella) bassae Collignon, 1948 (Basse 1931: 41, pl. 6 (figs 1-4); see also Collignon 1948: 36, pl. 31 (fig. 3)) from the Middle Campanian of Madagascar. 286 ANNALS OF THE SOUTH AFRICAN MUSEUM M. (Delawarella) campaniensis (De Grossouvre) (1894: 84, pl. 13 (figs la—b, 3a—b)) from the Middle Campanian of France, and from the Lower Cam- panian of the Gulf Coast of North America. M. (Delawarella) danei Young (1963: 114, pl. 57 (fig. 6), pl. 62 (figs 1-2), pl. 64 (figs 1, 5), pl. 65 (figs 1-2), pl. 66 (figs 3-4), text-figs 24c, 33b) from the Lower Campanian of the Gulf Coast of North America. M. (Delawarella) delawarensis (Morton) (1830: 244, pl. 2 (fig. 4)) from the Middle Campanian of Delaware, New Jersey and Texas. M. (Delawarella) jeanneti Collignon (1948: 34, pl. 30 (fig. 3-3b), pl. 31 (fig. 1-1b)) from the Middle Campanian of Madagascar. M. (Delawarella) jeanneti var. compressa Collignon (1948: 36, pl. 31 (fig. 2—2a)) from the Middle Campanian of Madagascar. M. (Delawarella) roedereri Collignon (1948: 38, pl. 32 (fig. 1-1b)) from the Middle Campanian of Madagascar. M. (Delawarella) sabinalensis Young (1963: 112, pl. 54 (fig. 2), pl. 63 (figs 1, 3-4), text-figs 20c, 2le, 26c) from the Lower Campanian of the Gulf Coast of North America. M. (Delawarella) subdelawarensis Collignon (1948: 32, pl. 29 (figs la—b, 2a—b)) from the Middle Campanian of Madagascar. M. (Delawarella) subdelawarensis var. lenticularis Collignon (1948: 43, pl. 30 (fig. 1-la)) from the Middle Campanian of Madagascar. The following species have been referred to subgenus (or genus) Australiella in the literature: M. (Australiella) australis (Besairie) (1930: 638, pl. 64 (fig. 2—2a)) from the Middle Campanian of Madagascar. M. (Australiella) australis ingens Collignon (1970: 62, pl. 633 (fig. 2334)) from the Middle Campanian of Madagascar. M. (Australiella) antsirasiraensis Collignon (1948: 28, pl. 28 (fig. 3-3a)) from the Middle Campanian of Madagascar. M. (Australiella) austinensis Young (1963: 115, pl. 64 (figs 3-4), pl. 65 (fig. 6), pl. 67 (figs 4-6), text-fig. 28e) from an imprecise horizon in the Gulf Coast, reported as Senonian and ‘probably Lower Santonian, but could be lowest Campanian’ (Young 1963: 116). M. (Australiella) besairiei (Collignon) (1938: 75, pl. 4 (fig. 2—2b)) from the Middle Campanian of Madagascar. M. (Australiella) coronata Collignon (1970: 56, pl. 631 (fig. 2328)) from the Middle Campanian of Madagascar. M. (Australiella) moreti Collignon (1948: 28, pl. 28 (fig. 2-2b)) from the Middle Campanian of Madagascar. M. (Australiella) pattoni Young (1963: 116, pl. 65 (figs 4-5), pl. 66 (figs 1-2, 5-6), pl. 68 (figs 1-3, 6), text-figs 24b, 26h, 33ac, 34dg) from the Lower Campanian of the Gulf Coast of North America, and, doubtfully, from an unknown horizon, possibly Santonian, of Japan. CRETACEOUS FAUNAS FROM SOUTH AFRICA 287 M. (Australiella) souqueti Collignon (1970: 59, pl. 623 (fig. 2332)) from the Middle Campanian of Madagascar. | M. (Australiella) subaustralis Collignon (1948: 27, pl. 27 (fig. 3-3a)) from the Middle Campanian of Madagascar. M. (Australiella) tintanti Collignon (1970: 57, pl. 631 (fig. 2330)) from the Middle Campanian of Madagascar. M. (Australiella) welderi Young (1963: 117, pl. 65 (fig. 3), pl. 68 (figs 4-5, text- fig. 25kn) from the Lower Campanian of the Gulf Coast of North America. Species referred to subgenus Bererella in literature are: M. (Bererella) aculeata Collignon (1948: 23, pl. 26 (fig. 2-2a)) from the Middle Campanian of Madagascar. M. (Bererella) atropsi Collignon (1969: 200, pl. 598 (fig. 2250)) from the Lower Campanian of Madagascar. M. (Bererella) bererensis Collignon (1948: 22, pl. 25 (fig. 1-1b)) from the Lower Campanian of Madagascar. M. (Bererella) bererensis mut. multicostata Collignon (1948: 198, pl. 27 fig. 1-1b)) from the Lower Campanian of Madagascar. M. (Bererella) bererensis compressa Collignon (1969: 200, pl. 598 (fig. 2251)) from the Lower Campanian of Madagascar. M. (Bererella) bifidata Collignon (1948: 25, pl. 26 (fig. 3)) from the Middle Campanian of Madagascar. M. (Bererella) rochi Collignon (1948: 25, pl. 27 (fig. 2—2b)) from the Middle Campanian of Madagascar. M. (Bererella) tenuis Collignon (1948: 24, pl. 25 (fig. 2-2b)) from the Middle Campanian of Madagascar. Occurrence Menabites s.1. occurs profusely in the Lower and Middle Campanian of Madagascar and Zululand, and in reduced numbers in the Gulf Coast and adjacent regions of North America and also Hokkaido, where they allegedly occur as early as Santonian. A single species is known from the Campanian of France. Menabites (Menabites) boulei Collignon, 1948 Fig. 217 Menabites boulei Collignon, 1948: pl. 22 (fig. 1-1b); 1969: 192, pl. 594 (fig. 2243). Type Holotype is the specimen figured by Collignon (1948, pl. 22 (fig. 1-1b)) by original designation. Material SAM-PCZ5690, locality 100, north of Nkundusi, Zululand, St Lucia Formation; originally considered to be Lower Santonian, but, on the basis of this identification, now dated as Lower Campanian. 288 ANNALS OF THE SOUTH AFRICAN MUSEUM — oo Fig. 217. Menabites (Menabites) boulei Collignon, 1948. SAM-PCZ5690. x 1. Dimensions D Wb Wh Wb: Wh U Int. Ext. 101,0 33,0(32,6) 39,0(38,6) 0,85 36,0(35,6) 11x2 18x2 Description Unfortunately the innermost whorls are lacking, but parts of them and their impressions indicate the presence of a large, ventrolateral Menabites node, confirming the generic allocation. The umbilical width is 35 per cent of the total diameter, and the dorsal zone of impression very shallow, with whorl overlap scarcely to the marginal (4) tubercle. The whorl section is basically compressed, higher than wide, with slightly inflated flanks converging to a flattened venter. Ornament in the species is very strong, consisting of conical, equally strong umbilical (1) and lateral (2) tubercles, very strong conical to radially elongated CRETACEOUS FAUNAS FROM SOUTH AFRICA 289 submarginal (3) tubercles, and very weak clavate marginal (4) and stronger clavate external (5) tubercles. Bifurcations or intercalations occur at the submarginal (3) tubercle, but not regularly, as single ribs occur as well, resulting in a ratio of umbilical to external tubercles of slightly less than 1 : 2. The suture is poorly exposed, but features a large lateral lobe (L). Discussion This species is easily characterized by the strength of the ornament on the outer whorl. M. (M.) boulei and M. (M.) menabensis have similar inner whorls, but differ on the outer whorls by the sparser ornament of M. (M.) boulei and also the strength of the lateral (2) tubercle in the latter as compared to that in M. (M.) menabensis. M.(M.) cottreaui has comparable ornament at similar stages, but has a much more compressed whorl section as compared to M. (M.) boulei. M. (M.) lenoblei has also been compared to the present species, and differs mainly in the effacement of ornament on the later whorls, especially towards the venter. Furthermore, the tubercles tend to be radially elongated. M. (M.) dabolavaensis is close to M. (M.) boulei, but generally has less strong, but denser ribbing. In addition, the ratio of external to internal tubercles in M. (M.) dabolavaensis is greater, there being up to three times as many external than internal tubercles. Occurrence Menabites (M.) boulei is the index fossil for the uppermost Zone of the Lower Campanian in Madagascar. Zone of M. boulei and Anapachydiscus arrialoorensis, and thus allows for correct dating of locality 100 in Zululand which the authors previously (Kennedy & Klinger 1975: 295) had erroneously attributed to Lower Santonian. Menabites s.1. sp. juv. indet. Fig. 218B—C Material SAM-PCZ5921 from the south-western shores of False Bay north of Nkundusi at locality 100, St Lucia Formation, Lower Campanian. Description and discussion A fragment of a septate whorl and the corroded impression of a small Menabites may probably be referred to the subgenus Australiella. At a diameter of 10 mm the umbilical (1), ventrolateral (2, 3, 4) and external (5) tubercles are already prominently developed, and the keel appears entire. At 35 mm diameter, ornament consists typically of small, pointed umbilical (1) tubercles, very large nodose ventrolateral (2, 3, 4) tubercles and about two to three times as many obliquely elongated clavate external (5) tubercles. The keel is well developed and entire. 290 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 218. A. Menabites (Australiella) australis (Besairie, 1930). NMB—D1336B. B-C. Mena- bites s.l. sp. juv. indet. SAM-PCZ5921. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 291 Specific identification of this fragment is impossible, but it is included to show the early trituberculate ornament. Similar juvenile specimens have been illustrated by Collignon (1948, pl. 7(1) (figs 8-9)) to illustrate the ontogenetic development of Menabites. Menabites (M.) boulei occurs at the same locality and the inner whorls could possibly belong to that species. Occurrence Lower Campanian of Zululand. Menabites (Australiella) australis (Besairie, 1930) Figs 218A, 219-229, 230B, 231-240 Mortoniceras australe Besairie, 1930: 638, pl. 64 (fig. 2—2a). Mortoniceras cf. australe Besairie, 1930: 638. Menabites (Australiella) australis (Besairie): Collignon, 1948: 26, pl. 28 (fig. 1-1a). Australiella australis (Besairie): Collignon, 1970: 54, pl. 630 (fig. 2326). Type Holotype is the specimen figured by Besairie (1930, pl. 64 (fig. 2—2a)) from the Campanian of Menabe, Madagascar. Material SAM-PCZ5898, Bed 3 locality 110, on the Nibela Peninsula, Zululand, St Lucia Formation, Campanian Il; BMNH-C81475, C81478, Bed 4 at the same locality, Campanian III; SAM-—PCP5689, SAS-Z191, Z705, Z1847, NMB-D1336A-D1336B, all from the same locality as SAM—PCZ5898 at various unrecorded horizons at this section; BMNH-—C81476, C81482, C81485, locality 109, on the Nibela Peninsula, St Lucia Formation, Campanian II. Dimensions D Wb Wh Wb: Wh U Int. Lt Ext. PCZ5898 233,0 83,0(35,6) 83,0(35,6) 1,00 83,0(35,6) 20 20 34 D1336A 100,0 53,0(53,0) 34,0(34,0) 1,56 41,0(41,0) — 14 31 at 140,0 68,0(48,6) 50,0(35,7) 1,36 55,0(39,3) 20 17 35 at 182,0 81,0(44,5) 62,0(34,1) 1,31 71,0(39,0) 24 21 37 D1336B 86,0 48,0(55,8) 34,0(39,5) 1,41 —_ — 15 29 at 120,0 64,0(53,3) 52,0(43,3) 1,23 — — 16 31 at 147,0 78,0(53,1) 67,0(45,6) 1,16 51,0(34,7) — 17. 31 Z705 135,0 55,0(40,7) 46,0(34,1) 1,20 53,039,3) — — — Z191 196,0 80,0(40,8) 65,0(33,2) 1,23 76,0(38,8) 22 21 39 After Collignon, 1948: I 49 28(57) 22(45) 1,27 16(33) —_- —- — II 62 43(70) 27(44) 1,59 22(35) —- —- — III 90 54(60) 42(70) 1,28 34(38) —- —- — IV (type) 92 56(61) 44(48) 1,27 35(38) —_- —- — Description This species, as here interpreted, may grow to a very large size; one speci- men, SAM-PCZ5898 (Figs 219-220), is still septate at a diameter of 315 mm. 292 ANNALS OF THE SOUTH AFRICAN MUSEUM ws = S * Fig. 219. Menabites (Australiella) australis (Besairie, 1930). SAM—PCZ5898. Largest known specimen at 315 mm diameter, and still septate. Also shows transition from Australiella to Menabites or Bererella adult stage. x 0,39. CRETACEOUS FAUNAS FROM SOUTH AFRICA 293 Fig. 220. Menabites (Australiella) australis (Besairie, 1930). SAM-PCZ5898. x 0,4. (See also Fig. 219.) 294 ANNALS OF THE SOUTH AFRICAN MUSEUM C Fig. 221. Menabites (Australiella) australis (Besairie, 1930). NMB—D1336B. x 1. Changes in whorl section during ontogeny. (See also Figs 222-229.) CRETACEOUS FAUNAS FROM SOUTH AFRICA 295 Coiling is relatively loose with an umbilical width of the order of 35 to 40 per cent of the diameter. The whorl section and details of ornament change continually through the ontogeny (Figs 221-229). Unfortunately, none of the specimens has the very innermost whorls preserved, but at a diameter of c. 80 mm, the whorl section is already distinctly rectangular, much wider than high, with a very shallow dorsal zone of impres- sion, little inflated, virtually parallel flanks and a slightly rounded venter (Fig. 227). Ornament at this stage consists of small, conical umbilical tubercles, connected in most cases by low, thick ribs to enormous bullate to spatulate ventrolateral nodes, which in turn are connected by very indistinct riblets to two or three rounded or obliquely elongated external tubercles. Some of the umbilical tubercles appear intercalatory, not being associated with any ribs. The keel is low and broad. Striking at these diameters are the enormous ventro- lateral nodes and the curious, occasional intercalatory umbilical nodes. With increasing diameter, the whorl section becomes more inflated, and the ventrolateral nodes become smaller and narrower and eventually show signs of division, giving rise to a new row of tubercles on the ventrolateral side. During this process, the lateral and ventrolateral tubercles tend to become absorbed in the ribs, which are now well defined. The diameter at which division of the ventrolateral tubercle into two takes place is variable, ranging from c. 120 mm in NMB—-D1336B to 150 mm in NMB-D1336A. Last to form is the lateral tubercle (2), dorsolaterally of the original ventro- lateral spine. By this time the whorl section is well rounded, though still wider than high. With increasing diameter ribbing becomes bolder and tends to absorb the tuberculation to some extent. The available specimens are all phragmocones, the largest being 315 mm in diameter. The suture has blocky saddles and lobes with minutely frilled elements (Fig. 231B). Discussion The Zululand material is of extreme interest in showing virtually the whole ontogenetic development, changing from prominently trituberculate through quadrituberculate to pentatuberculate. The authors are not entirely sure that their interpretation of the species is wholly correct, due to great differences in size between the Madagascar and Zululand specimens. It is not clear from Besairie’s (1930: 638) or Collignon’s (1948 : 26-27, 1970: 54) descriptions of M. (A.) australis whether these specimens are still septate, or adult. The Madagascar specimens, if adults, are all very small, c. 80 mm—150 mm in diameter, when compared to the Zululand speci- mens, which are still septate at diameters in excess of 300 mm. The inner whorls of the Zululand specimens, with depressed, subrectangular whorl section and enormous ventrolateral spines are, however, like typical Madagascar M. (A.) australis. This raises the question of whether Madagascar M. (A.) australis which are, in fact, trituberculate throughout, have arisen caenogenetically from 296 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 222. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336B. Stage preceding that of Figure 223. x 1. a population similar to the much larger Zululand material, or whether this species has been interpreted only in terms of incomplete material in Mada- gascar. The holotype of M. (A.) australis, according to Besairie’s figures, measures about 150 mm in diameter. None of the Zululand specimens is still unequivocally trituberculate at that diameter, although specimens such as NMB-D1336 (Fig. 229), with a very indistinct fourth row of tubercles, are very close to the holotype. Besairie (1930: 638) had referred to the Zululand material as Mortoniceras cf. australis, stating that it differed from M. australis only by the possession of a doubled row of tubercles in lieu of a single row. In view of the normal wide range of intraspecific variation in the Texani- tinae, it is probably correct to refer the Zululand material to Besairie’s species, and to regard the holotype of M. (A.) australis as an immature specimen. Differences between M. (A.) australis and the majority of the other M. (Australiella) species described from Madagascar, e.g. M. (A.) falloti, M. (A.) besairiei, M. (A.) subaustralis, M. (A.) moreti, M. (A.) antsirasiraensis, M. (A.) tintanti, and M. (A.) coronata are difficult to formulate, as they seem CRETACEOUS FAUNAS FROM SOUTH AFRICA 297 to grade into each other both in a vertical and in a lateral sense. The majority of these species occur at the same stratigraphic level in Madagascar, and differ mainly in details of whorl section and appearance or attenuation of ornament. In view of the normal wide range of intraspecific variation, the authors doubt if all these names should be retained. Australiella australis ingens was separated by Collignon (1970: 62) from the nominate subspecies on the basis of the strong development of the ventro- Yj Fig. 223. Menabites (Australiella) australis (Besairie, 1930). NMB-—D1336B. Stage following that of Figure 222, and preceding that of Figure 224. x 1. 298 ANNALS OF THE SOUTH AFRICAN MUSEUM MSiltidts, Fig. 224. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336B. Stage following that of Figure 223, and preceding that of Figure 225. x 0,9. CRETACEOUS FAUNAS FROM SOUTH AFRICA 299 Fig. 225. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336B. x 0,8. 300 ANNALS OF THE SOUTH AFRICAN MUSEUM lateral spines, the rounding of the venter and attenuation of the keel and external row of tubercles. None of the Zululand specimens shows this development and no data can be added. Relationships with the closely allied M. (A.) falloti described below (p. 315) are not quite clear, but it seems that M. (A.) falloti grades upward in the exposures into M. (A.) australis. M. (A.) falloti has a more rounded whorl Fig. 226. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336B. Lateral view of completely assembled specimen shown in Figures 222-225. x 0,69. CRETACEOUS FAUNAS FROM SOUTH AFRICA 301 Fig. 227. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336A. x 1. (See also Figs 218A, 228-229.) 302 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 228. Menabites (Australiella) australis (Besairie, 1930). NMB-D1336A. x 1. 303 CRETACEOUS FAUNAS FROM SOUTH AFRICA , 1930). rie i (Besa 8 > x 0 NMB-D1336A ites (Australiella) australis Menab Fig. 229. 304 ANNALS OF THE SOUTH AFRICAN MUSEUM (oo) © Pg OO”0Sg D 6 Fig. 230. Menabites (Australiella) species. Whorl sections. A. M. (A.) falloti. B. M. (A) australis, C. M. (A.) besairiei. D. M. (A.) moreti. CRETACEOUS FAUNAS FROM SOUTH AFRICA 305 B Fig. 231. Menabites (Australiella) australis (Besairie, 1930). SAS—Z1847. Showing whorl section and partial suture. x 1. 306 ANNALS OF THE SOUTH AFRICAN MUSEUM section (Fig. 230A) and is pentatuberculate at a smaller diameter than M. (A.) australis. Single specimens each of M. (A.) falloti, M. (A.)-australis, and M. (D.) delawarensis occur respectively in Beds 1, 2 and 3 at locality 110 (see Kennedy & Klinger 1975, fig. 9), suggesting that M. (A.) australis could be derived proterogenetically from M. (A.) falloti. Definite conclusions, however, must await more detailed collecting. M. (A.) tintanti is a related species which also progresses beyond the tri- Fig. 232. Menabites (Australiella) australis (Besairie, 1930). SAS-Z191. Specimen approxi- mating the limits of M. (A.) moreti in being weaker ornamented and more rounded. x 0,7. CRETACEOUS FAUNAS FROM SOUTH AFRICA 307 tuberculate stage to quadrituberculate, and eventually pentatuberculate, though the lateral tubercle is weak. Dimensions of the holotype given below are virtually identical to the Zululand specimen NMB-D1336B at the same diameter. D Wb Wb Wb: Wh U M. (A.) tintanti after Collignun .. 120 60(50) 47(39) 1,28 45(38) M. (A.) australis NMB-D1336B . . 120 64(53) 52(43) 1,23 — Fig. 233. Menabites (Australiella) australis (Besairie, 1930). SAS-Z191. Specimen approxi- mating the limits of M. (A.) moreti in being weaker ornamented and with a more rounded whorl section. xX 0,65. 308 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 234. Menabites (Australiella) australis (Besairie, 1930). SAS-Z705. Specimen close to M. (A.) coronata due to depressed, coronate whorl section x 1. (See Fig. 235A.) Unfortunately, the holotype of M. (A.) tintanti was figured in lateral view only, but it seems to occupy a morphologically intermediate position between M. (A.) falloti and M. (A.) australis, in a sequence leading from a pentatubercu- late stage through quadrituberculate to eventually trituberculate at comparable diameters. Given more material it would most probably be possible to include M. (A.) tintanti in the synonymy of one of the two species. M. (A.) subaustralis has the same ornament and whorl section as M. (A.) australis at small diameters, but after 60 mm all ornament weakens, thus differing from the latter species. 309 CRETACEOUS FAUNAS FROM SOUTH AFRICA x 1. B. Menabites (Dela- irie, 1930). SAS—Z705. Specimen close ion. coronate whorl sect is (Morton, 1831) SAS—Z2100. x 0,72. b) Fig. 235. A. Menabites (Australiella) australis (Besa warella) delawarens to M. (A.) coronata due to depressed 310 ANNALS OF THE SOUTH AFRICAN MUSEUM M. (A.) besairiei also seems somewhat transitional between M. (4.) falloti and M. (A.) australis, but has a distinct octagonal whorl section (Fig. 230C) compared to the rounded section of M. (A.) falloti or the rectangular depressed section of M. (A.) australis. Transitional forms such as SAM—PCZ5689 (Figs 239-240) and SAM-PCZ5685 (Fig. 241) connect the species. M. (A.) moreti has inner whorls very much like M. (A.) australis, but has a Fig. 236. Menabites (Australiella) australis (Besairie, 1930). SAS-Z1847. x 0,8. CRETACEOUS FAUNAS FROM SOUTH AFRICA 311 Fig. 237. Menabites (Australiella) australis (Besairie, 1930). SAS—Z2101. x 0,84. more rounded, narrower whorl section (Fig. 230D) than the latter, and has differently evolving ornament in that the ventrolateral spine weakens and becomes absorbed in the ribbing. SAS—Z191 (Fig. 232) is very much like M. (A.) moreti, in being more inflated and weaker ornamented than typical M. (A.) australis, pointing to the close relationship between the two species. M. (A.) antsirasiraensis lacks the depressed whorl section and loses the ventrolateral spine at diameters between 70 and 80 mm. 312 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 238. Menabites (Australiella) australis (Besairie, 1930). SAS-Z2102. x 0,9. CRETACEOUS FAUNAS FROM SOUTH AFRICA 313 M. (A.) coronata has a whorl section (Collignon 1970, pl. 631 (fig. 2328)) comparable to typical M. (A.) australis though not quite as depressed rectangular. Here the ventrolateral spines are not as strong and spatulate and about four external tubercles occur per spine. The whorl section of SAS- Z705 (Figs 234-235A) is virtually identical to that of the holotype of M. (A.) coronata, but it lacks the proliferation of external tubercles. Fig. 239. Menabites (Australiella) australis (Besairie, 1930). SAM-PCZ5689. Transitional to M. (A.) besairiei. x 0,6. 314 ANNALS OF THE SOUTH AFRICAN MUSEUM g Yi “py , Fig. 240. Menabites (Australiella) australis (Besairie, 1930). SAM- PCZ5689. With whorl section transitional to M. (A.) besairiei. x 0,8. CRETACEOUS FAUNAS FROM SOUTH AFRICA 315 M. (A.) austinensis was regarded as being close to M. (A.) australis by Young (1963), but was said to differ on account of having weaker ventrolateral clavae and stronger umbilical tubercles. These differences could all probably be incorporated in the intraspecific variation of M. (A.) australis as here interpreted, but the stratigraphic position of the species is perplexing. The species is based on two specimens, both with dubious locality data. Young (1963: 116), how- ever, concludes that the species is either from ‘Formation B or the Desau Chalk. Formation B seems the most likely. . . .” This would place the species in either Lower Santonian or Lower Campanian, probably Lower Santonian, and would be the oldest record of the subgenus. Until the question on the exact stratigraphic position is resolved, it is considered advisable to maintain the species apart from M. (A.) australis. M. (A.) pattoni differs from M. (A.) australis in lacking the depressed whorl section in the early stages. Retention of the trituberculate stage to a great diameter, and the presence of occasional intercalatory umbilical tubercles in M. (A.) australis, is very reminiscent of the situation as found in Plesiotexanites matsumotoi sp. nov. from the Middle Santonian of the Umzamba Formation. Apart from a more evolute mode of coiling in the latter, it also lacks the proliferation of external tubercles characteristic of Menabites. This seems to be just another example of hetero- chronous partial convergence with little genetic relationship. Occurrence Middle Campanian, Zone of Delawarella subdelawarensis and Australiella australis in Madagascar, and Campanian II-III of Zululand. Menabites (Australiella) falloti (Collignon, 1932) Figs 230A, 241-243 Mortoniceras falloti Collignon, 1932: 37, pl. 5 (fig. 1-1b); 1938: 76. Type The holotype is the specimen figured by Collignon (1932; pl. 5 (fig. 1-1b)) from the Senonian of Zululand, collected by P. Fallot in 1929 and presently housed in the Palaeontology Collections of the Geological Institute, Nancy. Material SAM-PCZ5870, Bed 1 locality 110, foreshore exposures at the south- western tip of Nibela, Zululand, St Lucia Formation, Campanian II; SAS- Z843, Z1547, Z2097, all from an unknown horizon at locality 109 or 110 at the south-western tip of the Nibela Peninsula, St Lucia Formation, Cam- panian II; BMNH-C81443-C81444, C81477, locality 109, St Lucia Formation, Campanian II. 316 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 241. Menabites (Australiella) falloti (Collignon, 1932). SAM- PCZ5685. With whorl section transitional to M. (A.) besairiei. xan: CRETACEOUS FAUNAS FROM SOUTH AFRICA 317 Dimensions D Wb Wh Wb: Wh U Int. L. Ext. Holotype after Collignon 90 34 (37,8) 35,0(38,9) 0,97 30,0(33,3) 22 118,0 46,0(38,9) 47,0(39,8) 0,98 35,0(29,6) 14—-— — Z843 124,0 54,0(43,5) 45,0(36,3) 1,20 — — Z2097 176,0 63,0(35,8) 61,0(34,7) 1,03 58,0(32,9) Description Coiling is moderately involute with an umbilical width of the order of 30 to 33 per cent of the total diameter. Whorl overlap is up to the fourth row of tubercles and the dorsal zone of impression is very shallow. The whorl section is rounded, as wide as high in the early stages of growth (Fig. 243) but later becomes higher than wide (Fig. 230A); in some cases extremely so. None of the Zululand specimens has the innermost whorls preserved, but at a diameter of c. 70 mm all five rows of tubercles are already present. These are: a conical umbilical (1) tubercle connected to a thickened rib which con- tains the second, third and fourth rows of tubercles, and of which the third row is most prominent. At the fourth row faint bifurcations become noticeable, ending in spirally elongated external (5) tubercles which protrude far beyond the venter on either side of a broad, weak central keel. In some cases it appears as if the bifurcations already originate at the third tubercle. Ornament generally becomes stronger towards the later part of the phrag- mocone, but weakens again towards the body chamber. The suture has broad saddles and lobes with minutely frilled terminations. Discussion This species is distinguished by the more or less rounded whorl section in the early stages of growth and by the early appearance of the pentatuberculate stage. Differences between M. (A.) falloti and M. (A.) australis, and even M. (Delawarella) delawarensis are gradational, as discussed above. Subgeneric identity of the species is subjective. It may be argued that because of the relative early appearance of the pentatuberculate stage it should be referred to Menabites s.s. Because of the close resemblance to the holo- type of the subgenus M. (Australiella), however, it is here considered advisable to refer the species to subgenus Australiella rather than Menabites s.s. It may be noted here that Collignon also appeared to be in some doubt as to the subgeneric allocation in his 1948 monographic description of the Texanitinae. In comparing the species with M. (A.) australis (Collignon 1948: 27) he 1eferred to it as Menabites (Australiella) falloti, but in his species list of the Texanitinae in the same volume (Collignon 1948: 44) refers it to Bererella. Specimen SAS—Z843 (Figs 242-243) is closest to the holotype, which was most probably collected at the same locality by P. Fallot in 1929, during the Zululand excursion of the XV Session of the International Geological Congress. 318 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 242. Menabites (Australiella) falloti (Collignon, 1932). SAS-Z843. x 1. Differences between M. (A.) falloti and M. (A.) australis, described above, are to be found mainly in the longer retention of the trituberculate stage and in the depressed rectangular whorl section of the latter species in the early Stages. Although sufficient precise stratigraphic data are not available, it appears possible to derive M. (A.) australis from M. (A.) falloti caenogenetically as discussed above. Specimen SAM-—PCZ5870 differs from the rest of the material in being much larger and in having a much more compressed section on the inner whorls, though the latter may be partially due to secondary crushing. Ornament on this specimen also weakens on the outer whorls, and may possibly connect CRETACEOUS FAUNAS FROM SOUTH AFRICA 319 Fig. 243. Menabites (Australiella) falloti (Collignon, 1932). SAS-Z843. x 1. 320 ANNALS OF THE SOUTH AFRICAN MUSEUM with M. (Delawarella) gigas sp. nov., described below. SAM-PCZ5685 (Fig. 241) has inner whorls typical of M. (A.) falloti, but at a diameter of c. 180 mm, has an octagonal whorl section, very similar to that of M. (A.) besairiei. It is easily distinguished from the latter, however, on account of the early appearance of the pentatuberculate stage. M. (A.) tintanti is a similar, large species, and was compared to M. (A.) falloti by Collignon (1970: 58); it differs mainly in retaining a quadrituberculate stage to a large diameter. SAS—Z2097 appears transitional to M. (A.) tintanti in having a very weakly developed lateral tubercle throughout and robust ribbing on the outer whorl. As discussed earlier, M. (A.) tintanti is intermediate between M. (A.) falloti and M. (A.) australis, and, given more material, may probably be synonymized with either of the two species. The inner whorls of M. (A.) falloti resemble those of M. (Delawarella) delawarensis, which occurs at approximately the same stratigraphic level, and they may possibly be derived from a common ancestor. In larger specimens, however, M. (Delawarella) delawarensis is easily distinguished by the loss of lateral ornament. Occurrence Middle Campanian of Zululand and Madagascar, though apparently rare in Madagascar. Menabites (Delawarella) delawarensis (Morton, 1830) Figs 244-247A Ammonites delawarensis Morton, 1830: 244, pl. 2 (fig. 4); 1834: 37, pl. 2 (fig. 5). Whitfield, 1892: 252, pl. 42 (figs 6-8). Mortoniceras delawarensis (Morton): Weller, 1907: 837, pl. 103 (fig. 1). Grabau & Shimer, 1910: 227, fig. 1508. Menabites (Delawarella) delawarensis (Morton): Collignon, 1948: 29. Reeside 1962: 131, pl. 72 (figs 1-3), pl. 73 (figs 1-5), pl. 74 (fig. 2). Delawarella delawarensis (Morton): Young, 1963: 111, pl. 55 (fig. 5), pl. 61 (figs 1-6), pl. 63 (fig. 2), text-figs 15e, 20d, 25b, 26bfg, 27c, 29f. Type Holotype is the specimen figured by Morton (1830, pl. 2 (fig. 4)), housed in the Philadelphia Academy of Sciences. Material SAM-PCZ5905, Bed 3 locality 110, south-western tip of the Nibela Peninsula, Zululand, St Lucia Formation, Campanian II; SAS—Z2097-Z2098, both from the same locality as above, but of unknown horizon. BMNH-C81438- C81439, locality 109, St Lucia Formation, Campanian II. Dimensions D Wb Wh Wb: Wh U Int. L. Ext. Z2098 145,0 60,0(41,4) 60,0(41,4) 1,0 44,0(30,3) 18 18 34 Z2097 70,0 31,0(44,3) 30,0(42,8) 1,03 24,034.22) — — — CRETACEOUS FAUNAS FROM SOUTH AFRICA 321 Description The material is not very well preserved, lacking the inner whorls, and finer details of ornament are obscured by crystalline matrix. Coiling is moderately evolute, the umbilicus comprising 30 to 40 per cent of the diameter. The whorl section in all three specimens is virtually equi- dimensional hexagonal, with a near-vertical umbilical wall, flat flanks parallel up to the ventrolateral nodes on the inner whorls and then converging to a flattened venter (Fig. 247A). Fig. 244. Menabites (Delawarella) delawarensis (Morton, 1830). SAS-Z2098. x 1. 322 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 245. Menabites (Delawarella) delawarensis (Morton, 1830). SAS-Z2098. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 323 On the inner whorls, ribs are feebly prorsiradiate up to the middle of the flanks where they pass through a large node, and then generally bifurcate and continue weakly up to the clavate external (5) nodes. At a diameter of c. 30 mm the ventrolateral node starts dividing into two on the ventral side, giving rise to a smaller clavate marginal (4) row of tubercles. The lateral (2) tubercles appear later, at c. 40 mm diameter, but these latter never really become prominent. On the largest specimen, SAS—Z2098 (Fig. 244), the five rows of tubercles are spread out evenly over the flanks, and the submarginal (3) row is the most prominent, being distinctly pointed. Towards the larger end of this specimen tuberculation on the flanks weakens, and ribbing becomes dominant. The external tubercles, however, become stronger with increasing diameter and stand out as prominent serrate ridges. The keel, at this stage, shows low undula- tions corresponding in number to the external tubercles. The suture is not well exposed in either of the specimens. Discussion This species was discussed extensively by Reeside (1962) and Young (1963) and not much can be added to their comments. The species is easily recognized by the very strong ornament when com- pared to other species of /. (Delawarella), and may in this respect be considered A B C Fig. 246. Menabites (Delawarella) delawarensis (Morton, 1830). SAS-Z2097. x 1. 324 ANNALS OF THE SOUTH AFRICAN MUSEUM B Fig. 247. A. Menabites (Delawarella) delawarensis (Morton, 1830). A. Whorl section of SAS-Z2907. x 2. B. Menabites (Delawarella) subdelawarensis Collignon, 1948. Suture line of SAS-Z2100. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 325 transitional between Menabites s.s. and M. (Delawarella). M. (Delawarella) roedereri is similar in this respect in having strong orna- ment, but differs mainly in having a more compressed whorl section. Collignon (1948: 32) separated M. (Delawarella) subdelawarensis (described below) as a Malagasy representative of this American species, differing mainly in having weaker ornament, and also a more compressed whorl section. These reasons appear sufficient for separation of the two species. Unfortunately, none of the Zululand specimens has the body chamber preserved, but the magnificent specimen illustrated by Reeside (1962, pl. 74 (fig. 2)) shows the low, broad ribbing on the flanks. Occurrence Campanian of Zululand, Gulf Coast region, and Delaware and New Jersey regions of North America. Menabites (Delawarella) subdelawarensis Collignon, 1948 Figs 247B-251 Menabites (Delawarella) subdelawarensis Collignon, 1948: 32, pl. 29 (figs 1-2). Menabites (Delawarella) subdelawarensis var. lenticularis Collignon, 1948: 34, pl. 30 (fig. 1). Delawarella subdelawarensis Collignon, 1970: 64, pl. 634 (fig. 2335). Type Holotype is Collignon’s (1948, pl. 29 (fig. 2)) specimen by original designation. Material SAM-PCZ5901, Bed 3 locality 110, south-western tip of the Nibela Penin- sula, Zululand, St Lucia Formation, Campanian III; SAS—Z2100 from an unknown horizon at the same locality; BMNH-—C81443-C81444, locality 109, St Lucia Formation, Campanian II. Dimensions D Wh Wb Wb: Wh U Int. L. Ext. Z2100 221,0 84,0(38,0) 103,0(46,6) 0,81 59,0(26,7) 23 — — PCZ5901 340,0 119,0(35,0) 155,0(45,6) 0,76 .110,032,4) — — 42 Description Unfortunately, the specimens are in a poor state of preservation, but partially permit examination of specific characteristics. _ The umbilical width of the very large specimen, SAM-—PCZ5901 (Figs 248-250), which is still septate at the larger end, is probably a little too large due to fracture of the last part of the whorl. An umbilical width of 26 per cent as in the smaller specimen seems more realistic. The whorl section is massive 326 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 248. Menabites (Delawarella) subdelawarensis Collignon, 1948. SAM-—PCZ5901. Gigantic specimen still fully septate. x 0,34. throughout, higher than wide, with a dorsal zone of impression which embraces preceding whorls up to the lateral tubercle, and a vertical to inward or outward sloping umbilical wall merging abruptly with the flanks, which remain parallel up to the lateral tubercle at about midflank, and then converge obliquely to a very broad venter. In SAS-—Z2100 (Fig. 251) the typical Menabites trituberculate ornament is still visible at a diameter of about 40 mm, but thereafter very quickly becomes pentatuberculate and rather weak. Ornament after that consists of relatively weak ribs bifurcating occasionally at the umbilical edge, each bearing five CRETACEOUS FAUNAS FROM SOUTH AFRICA 327 Fig. 249. Menabites (Delawarella) subdelawarensis Collignon, 1948. SAM—PCZ5901. Gigantic specimen still fully septate. Opposite lateral view of Fig. 248. x 0,34. . rows of tubercles of which the umbilical (1) and external (5) ones are strongest. The lateral (2) tubercles are very weak, and the submarginal (3) and marginal (4) ones only slightly stronger. All five rows of tubercles are spaced equidistantly over the flanks. In this smaller specimen, SAS—Z2100 (Fig. 251), ribs become bolder and more widely spaced at the largest diameter, whilst the whorl section also appears to become more inflated. In the larger specimen, SAM-—PCZ5901 (Figs 248-250), all ornament weakens at the largest diameter with no distinct strengthening of the ribs. 328 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 250. Menabites (Delawarella) subdelawarensis Collignon, 1948. SAM-—PCZ5901. Gigantic specimen still fully septate. x 0,46. (See also Figs 248-249.) CRETACEOUS FAUNAS FROM SOUTH AFRICA 329 Fig. 251. Menabites (Delawarella) subdelawarensis Collignon, 1948. SAS-Z2100. x 0,63. The keel throughout is broad and low, and eventually becomes very indistinct at large diameters. The suture line is blocky with minutely frilled terminations, though details are somewhat obscured by weathering and calcitic encrustation (Fig. 247B). The lateral lobe (L) in SAS—Z2100 is wide and asymmetrically trifid, and deeper than the external lobe. 330 ANNALS OF THE SOUTH AFRICAN MUSEUM Discussion Comparison of the Zululand material with that described from Mada- gascar by Collignon (1948, 1970) is difficult due to great differences in size, the largest described specimen from Madagascar (Collignon 1970: 68) being only 166 mm in diameter compared to the smallest Zululand specimen at 221 mm diameter. Nevertheless, the massive whorls, and generally coarse ornament of the Zululand specimens compare well with the Madagascar material. The closest ally to M. (D.) subdelawarensis is undoubtedly M. (D.) jeanneti, which also occurs in two forms; a normally inflated form and a compressed form. Major differences between the two species according to Collignon (1948: 36) are that the whorls of M. (D.) jeanneti are less massive, the ribs straighter, finer and denser, and the lateral lobe (L) is bifid instead of trifid. The Zululand material seems to suggest that M. (D.) jeanneti could possibly be derived from M. (D.) subdelawarensis through tighter coiling and prolifera- tion of ribbing, but unfortunately insufficient material is as yet available to fully substantiate this assumption. The inner whorls of the larger specimen, SAM—PCZ5901 (Figs 248-250), are remarkably like those of M. (D.) gigas sp. nov. which occurs at a slightly lower stratigraphic level at the Nibela Peninsula, but in the latter increase in whorl size is much more rapid. Occurrence Campanian III of Zululand, and the Mid-Campanian Zone of Delawarella subdelawarensis and Australiella australis of Madagascar. Menabites (Delawarella) sp. aff. jeanneti s.1. Collignon, 1948 Figs 252-255 Compare: Menabites (Delawarella) jeanneti Collignon, 1948: 34, pl. 30 (fig. 3-3b) pl. 31 (fig. 1-1b). Menabites (Delawarella) jeanneti var. compressa Collignon, 1948: 36, pl. 31 (fig. 2—2a). Delawarella jeanneti Collignon: Collignon, 1970: 64, pl. 634 (fig. 2336). Dimensions D Wb Wh Wb: Wh U Int. L. Ext. PCZ5691 143,0 56,0(39,2) 68,0(47,6) 0,82 36,0(25,2) 23 —— Material SAM-PCZ5691, Bed 7 locality 110, south-western tip of the Nibela Penin- sula, Zululand, St Lucia Formation, Campanian III; and SAM-FB from the collection of S. H. Haughton from an unknown horizon and locality in the False Bay region of Zululand, but probably the same locality as above. CRETACEOUS FAUNAS FROM SOUTH AFRICA 331 Fig. 252. Menabites (Delawarella) sp. aff. jeanneti s.1. Collignon, 1948. SAM-—PCZ5691. x 0,78. Description Coiling is involute with an umbilical diameter of about 25 per cent. The whorl section is compressed with a prominent dorsal zone of impression, a vertical to overhanging umbilical wall with a well-defined umbilical edge and weakly inflated flanks with maximum width at the site of the lateral tubercle. None of the specimens has the innermost whorls preserved, but at a diameter of 45 mm ornament is already very much weakened with barely per- ceptible lateral and submarginal tubercles. The umbilical tubercles are radially 332 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 253. Menabites (Delawarella) sp. aff. jeanneti s.1. Collignon, 1948. SAM-—-PCZ5691. x 0,85. CRETACEOUS FAUNAS FROM SOUTH AFRICA : 333 elongated and point strongly inward. With increasing diameter ribbing remains weak and prorsiradiate with barely perceptible lateral (2) and submarginal (3), and slightly stronger marginal (4), but progressively stronger umbilical (1) and external (5) tubercles. Numerous intercalations and bifurcations occur, generally at the site of the submarginal tubercle, but also occasionally at the umbilical tubercle. This results in there being about twice as many external as internal tubercles. Specimen SAM-PCZ5691 (Figs 252-253) has part of the early body chamber preserved. Here tuberculation weakens even more, and the ribs become wider spaced. The suture is only partially exposed in the present material. The keel, as far as can be seen, is never very prominent. Fig. 254. Menabites (Delawarella) sp. aff. jeanneti s.1. Collignon, 1948. SAM-FB. x 1. 334 ANNALS OF THE SOUTH AFRICAN MUSEUM Discussion This species is identified by the fine, dense ribbing throughout, and the very poor development of the lateral and submarginal tubercles. Collignon (1948: 34-36), in erecting M. (D.) jeanneti, distinguished between a typical form and a more compressed form, var. compressa, the difference in whorl breadth between the two forms being approximately 5 per cent. The larger Zululand specimen, SAM-PCZ5691 (Figs 252-253), with a whorl breadth of 39 per cent of the diameter, appears closer to the typical form of M. (D.) jeanneti than to the compressed variety. The main difference between the Zulu- land material and the Malagasy specimens of M. (D.) jeanneti, lies in the fact that, in the adult stage, the latter lose ornament on the venter and take on the appearance of a pachydiscid. In the Zululand specimens the venter remains Fig. 255. Menabites (Delawarella) sp. aff. jeanneti s.1. Collignon, 1948. SAM-FB. x 1. CRETACEOUS FAUNAS FROM SOUTH AFRICA 335 more or less flat, and the external tubercles are prominent, hence it is advisable to refer to the material as M. (D.) sp. aff. jeanneti rather than a definite reference. The specimen described below as M. (Delawarella) sp. indet. has similar weak ornament on the flanks, but lacks the strong development of the umbilical and external tubercles, and, in addition, has a distinct crenulate keel. M. (Delawarella) nibelae sp. nov. (discussed below) differs from the present species in being slightly more narrowly umbilicate and in having a more inflated whorl section. The closest ally to the present species is M. (D.) subdelawarensis, from which it is possibly derived through slight reduction of umbilical width, com- pression of the whorls and fining of ornament. Other species of M. (Delawarella) such as M. (D.) gigas sp. nov. and M. (D.) delawarensis, all differ on account of the much more prominent ornament. Occurrence Campanian of Zululand; in the Mid-Campanian Zone of Delawarella subdelawarensis and Australiella australis of Madagascar. Menabites (Delawarella) gigas sp. nov. Figs 256-258 Holotype SAM-PCZ5900 (Fig. 256), Bed 1 locality 110, south-western tip of the Nibela Peninsula, Zululand, St Lucia Formation, Campanian II. Etymology Pertaining to great size. Material SAS-Z1951 and SAS-A1867, both from the same locality as the holotype but from an unknown horizon, can also be referred to this species. Dimensions D Wb Wh Wb: Wh U PCZ5900 c.700 — c.270(37) — c.240(34) Description Unfortunately none of the specimens has the very innermost trituberculate whorls preserved. Coiling is involute, with a very rapid increase in whorl size, and an umbilical width of the order of 30 to 35 per cent of the total diameter. The whorl section throughout is higher than wide, but is more inflated on the phragmocone than on the body chamber. The umbilical wall is vertical to overhanging. Ornament on the phragmocone is basically of the Menabites type, very 336 ANNALS OF THE SOUTH AFRICAN MUSEUM 0 100 a) Fig. 256. Menabites (Delawarella) gigas sp. nov. Holotype SAM-—PCZ5900. x 0,23. Scale bar in millimetres. CRETACEOUS FAUNAS FROM SOUTH AFRICA 337 reminiscent of M. (Australiella) falloti (see p. 315), which occurs at the same stratigraphic level. Ornament here consists of low, single ribs, each bearing a distinct umbilical (1) tubercle, but low, rounded and indistinct lateral (2), submarginal (3), and marginal (4) tubercles. The body chamber, which is partially preserved on the holotype (Fig. 256) and SAS-Z1951 (Fig. 257), is enormous, comprising at least half a whorl. All lateral ornament is here effaced, and on the holotype this even affects the umbilical tubercles. The suture is imperfectly known, but includes a very wide, asymmetric bifid lobe (L) (Fig. 258). Fig. 257. Menabites (Delawarella) gigas sp. nov. Paratype SAS-Z1951. x 0,45. 338 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 258. Menabites (Delawarella) gigas sp. nov. Paratype SAS- Z1951. Showing partial suture. x 0,95. Discussion This species is very easily recognized by virtue of its large size, lateral compression and complete lack of ornament on the body chamber. The inner phragmocone whorls are very similar to those of M. (Australiella) falloti, and the holotype of the species was recovered from the same horizon (Bed 1 locality 110) as M. (A.) falloti, specimen SAM-—PCZ5870. This seems to indicate that, at that particular stratigraphic interval, Menabites (Australiella) and M. (Delawarella) had already evolved into two distinct lineages. It is also interesting to note the very large size of the Zululand specimen. This, to the authors’ knowledge, is the largest texanitid recovered so far. M. (D.) jeanneti resembles M. (D.) gigas in losing virtually all texanitid ornament on the outer whorls (see Collignon 1948, pl. 31 (fig. 1-1b)), but differs in having a more inflated whorl section. M. (D.) subdelawarensis, another species known to grow to enormous size in Zululand (p. 325), also differs in having a more inflated whorl section and generally much finer ornament. M. (A.) australis, which may also grow to enormous size, is easily dis- tinguished by the long persistence of the trituberculate stage and also by the more robust ornament. The very large body chamber compared to the relatively small phragmo- cone indicates that buoyancy control in this species must have been very delicate. Occurrence Campanian II of Zululand. CRETACEOUS FAUNAS FROM SOUTH AFRICA 339 Menabites (Delawarella) nibelae sp. nov. Figs 259-261 Holotype SAS-Z2101 (Figs 259-261), locality 110, south-western tip of the Nibela Peninsula, Zululand, St Lucia Formation, probably Campanian iII. Etymology Named after the Nibela Peninsula at the northern side of Hell’s Gates, connecting False Bay to Lake St Lucia. Material Only the holotype. Dimensions D Wb Wh Wb: Wh U Int L Ext 22101 164,0 67,0(40,8) 81,0(49,4) 0,83 37,0(22,6) 25 25 39 Description Coiling is involute, with a very narrow umbilicus, measuring only 22 per cent of the total diameter. The whorl section in the adult stage is higher than wide (Fig. 261) with a vertical umbilical wall, and little inflated flanks converging to a rounded venter. Inflation of the whorl increases with increasing diameter. Where visible, ornament is subdued throughout, and consists of small radially elongated umbilical (1) tubercles, a hardly visible lateral (2) and sub- marginal (3) tubercle, poorly developed marginal (4) tubercle but prominent external (5) tubercles, all situated on poorly developed radial to prorsiradiate ribs. The ribs may be single, but the majority generally bifurcate at the sub- marginal tubercle or arise as intercalatories at the marginal tubercles, giving rise to many more external than internal tubercles. The keel is initially broad and strong, but becomes indistinct towards the later part of the phragmocone and on the body chamber. Discussion This species combines some of the characteristics of M. (D.) delawarensis, M. (D.) subdelawarensis and M. (D.) jeanneti, pointing to the close relationship between the various species. The wide spacing of ribbing on the outer whorls is reminiscent of M. (D.) delawarensis, the whorl section is that of M. (D.) subdelawarensis, whereas the virtual absence of lateral (2) and submarginal (3) tubercles is reminiscent of M. (D.) jeanneti. There is certain resemblance to the specimen described below as M. (D.) sp. indet. (p. 342), but in the latter the whorl section is more inflated at comparable diameters, and the keel is crenulate, compared to entire in the present species. 340 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 259. Menabites (Delawarella) nibelae sp. nov. Holotype SAS—Z2101. x 0,88. This is the M. (Delawarella) species with the narrowest umbilical width, the next tightly-coiled species being M. (Delawarella) jeanneti with an umbilical width of 24 to 29 per cent compared to 22 per cent in the former. Examination of the whorl sections of the two species and details of ornament over the venter at comparable diameters (see Collignon 1948, pl. 30 (fig. 3), pl. 31 (fig. 1)) clearly shows the differences. The whorl section of M. (D.) jeanneti is much CRETACEOUS FAUNAS FROM SOUTH AFRICA 341 Fig. 260. Menabites (Delawarella) nibelae sp. nov. Holotype SAS-Z2101. x 0,94. 342 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 261. Menabites (Delawarella) nibelae sp. nov. Holotype SAS-—Z2101. Showing whorl section. x 1. more compressed, and at larger diameters tends towards weakening of the external tubercles over the venter. The exact horizon from which this specimen was collected at locality 110 is not known, but as far as the authors can recall, it was found high up in the sequence, above the level of abundant menabitids. This could possibly indicate yet another trend towards recoiling in the texanitids, possibly starting with a relatively evolute form such as M. (A.) falloti through M. (Delawarella) jeanneti to the present species. Occurrence Campanian of Zululand. Menabites (Delawarella) sp. indet. - Fig. 262 Material SAM-PCZ5692, locality 110, south-western tip of the Nibela Peninsula, Zululand, St Lucia Formation, Campanian II. Description A single septate specimen consisting of approximately a quarter of a whorl is referable to M. (Delawarella), although specific identification is not possible. CRETACEOUS FAUNAS FROM SOUTH AFRICA 343 Fig. 262. Menabites (Delawarella) sp. indet. SAM-PCZ5692. x1. . 344 ANNALS OF THE SOUTH AFRICAN MUSEUM The whorl section is ovoid, wider than high with a prominent dorsal zone of impression, a vertical umbilical wall and gently rounded flanks converging to a broadly rounded venter. The keel is broad and low, and distinctly undulating, though it is not quite clear if the undulations correspond to the external tubercles. Save slight undulations, all lateral ornament is absent, with only umbilical (1) and external (5) tubercles present. The umbilical nodes are rounded to slightly radially elongated. The external tubercles are clavate and obliquely aligned in a forward direction. The ratio of external to umbilical tubercles is of the order of three to one. The suture is beautifully exposed as seen in Figure 262. Discussion Despite the fragmentary nature of this specimen, it differs from all known Zululand specimens on account of the broad, undulating keel and rapid loss of lateral ornament. M. (Delawarella) subdelawarensis (see expecially Collignon 1948, pl. 29 (fig. 1)) has a comparable whorl section, but lacks the undulating keel and strong attenuation of ornament. M. (Delawarella) delawarensis has an undu- lating keel in later stages of growth, but is easily distinguished on account of the strong ornament. M. (Delawarella) jeanneti is also somewhat similar in attenuation of lateral ornament, but has a more compressed whorl section at a comparable diameter (see Collignon 1948, pl. 30 (fig. 3)). Hopefully further collection will permit definite identification of this species. Occurrence Campanian II of Zululand. Genus Spinaptychus Trauth Figs 263-265 Type species Aptychus spinosus Cox, 1926. Description and discussion The association of the aptychus form-genus Spinaptychus Trauth with the genus Texanites was discussed on an earlier occasion by the authors (Kennedy & Klinger 1972: 394-399; Klinger 1972: 105-109). Two texanitine specimens, one of them referred tentatively to Texanites soutoni, were available to the authors, and it was concluded that the association of Spinaptychus with Texa- nites is real. Since then, three more aptychi have been found, either in body chambers of, or associated with, texanitids, thus further establishing the rela- tionship between Spinaptychus and Texanitinae. CRETACEOUS FAUNAS FROM SOUTH AFRICA 345 Fig. 263. Spinaptychus. SAM-K5551. Fragment of Spinaptychus in body chamber of Plesiotexanites stangeri. X 1. 346 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 264. Spinaptychus. A. Silicone rubber mould of SAM—PCZ5876 (see Fig. 265). B. SAM-PCZ5875. x 1. 347 CRETACEOUS FAUNAS FROM SOUTH AFRICA SAM-PCZ5876. In body chamber fragment of Texanites soutoni «1,15. natalense ssp. nov. Spinaptychus. 265 Fig. 348 ANNALS OF THE SOUTH AFRICAN MUSEUM SAM-K5551 (Fig. 263), from locality 1 at Umzamba Cliff, Pondoland, is a body chamber fragment of Plesiotexanites stangeri with part of one valve of the aptychus preserved. The concentric striae clearly point to Spinaptychus. SAM-—PCZ5875 (Fig. 264B) is a complete internal mould of the right valve of a Spinaptychus from the Santonian/Campanian of locality 6 at the Enseleni Reserve. The accompanying ammonoid fauna consists predominantly of Pseudoschloenbachia umbulazi, Hauericeras gardeni and Texanites gr. ex. soutoni, and Plesiotexanites stangeri. Of these, only T. soutoni is large enough to accom- modate such a large aptychus. SAM-PCZ 5876 (Fig. 265) is a virtually complete internal mould of the right valve of a Spinaptychus in the body chamber of Texanites soutoni natalense, from the Upper Santonian/Lower Campanian sediments at locality 105 on the south-eastern shores of False Bay. Unfortunately none of the specimens is sufficiently preserved for detailed studies, but the sheer size and relative position of the aptychi still lead the authors to regard Spinaptychus as possibly an operculum as well as a jaw- apparatus as has been suggested for some uncalcified forms. especially anaptychi (see Lehmann 1967, 1970, 1971, 1972, 1975). Spinaptychus is now definitely known to occur with T. soutoni and P. stangeri. As T. soutoni is transitional to Submortoniceras, it is to be expected that the latter genus also bears spinaptychi. EVOLUTION AND EVOLUTIONARY TRENDS IN TEXANITINAE Details of the suggested evolution of the Texanitinae as indicated by the South African material, and discussed in various sections of the systematic descriptions, are outlined in Figure 266. These differ somewhat from Matsu- moto’s (1955, 1965, 1970), Collignon’s (1948), and Young’s (1963) ideas, but should be seen as complementary to their findings rather than revolutionary. The origin of the Texanitinae is still not resolved. The earliest genus appears to be Protexanites s.s., but data on the origin of the latter appear conflicting. Matsumoto (1955: 39, 1970: 230) favours deriving Protexanites s.s. from Subprionocyclus, and the authors concur in the case of species such as P. bontanti, P. canaensis, and P. cycni, but have their doubts in the case of more evolute, coarser-ornamented species such as P. bourgeoisi and P. shoshonense. Some of the South African material seems to suggest that it would be possible to derive Paratexanites via Protexanites, or both, from the aberrant Peroni- ceratid ‘Fraudatoroceras’, but more material is needed to solve this problem satisfactorily. The root of texanitine radiation in South Africa appears to lie in Para- texanites, especially the species P. australis, P. pseudotricarinatum, and P. umkwelanense. Thus, from these, Protexanites (Pleurotexanites), Plesio- texanites, and probably also part of Texanites s.s. may be derived. It is possible CRETACEOUS FAUNAS FROM SOUTH AFRICA 349 CONIACIAN SANTONIAN CAMPANIAN I] II M.(A.) australis M.(A) falloti M.(D.) gigas Protexanites cycni M.(D) delawarensis M.(D.) subdelawarensis M.(D.) aff. jeanneti MD.) nibelae snjoAd0u0ldgns & Menabites Fraudatoroceras Bevahites Tadd Neogauthiericeras SEIIDIUOIOY Paratex. australis Paratex. pseudotricarinatum P collignoniforme ‘ T.umzambiense T. presoutoni P umkwelanense T. soutoni S.woodsi S. condamyi P aff. serratomarginatus P aff. emsheris Fig. 266. Suggested lines of evolution within Texanitinae as inferred from South African material. 350 ANNALS OF THE SOUTH AFRICAN MUSEUM to derive Bevahites from Pleurotexanites, which in turn may give rise to Mena- bites s.1. This is basically the same view held initially by Matsumoto (1955: 43, table 1), but lately he (Matsumoto 1970: 299) and Young (1963: 107) favour the alternative of deriving Menabites (Australiella) from Protexanites. Matsu- moto (1970: 299) also dismisses Pleurotexanites as an ancestor of M. (Australi- ella), and considers these two taxa to be sibling offshoots of Protexanites s.s. This view seems to be based on the alleged occurrence of two species of Australiella, A. austinensis and A. sp. aff. pattoni, in strata of Santonian age. Both these species lack precise stratigraphic data, and the authors prefer deriving Australiella, Delawarella, and possibly Bererella via Menabites s.s. from Bevahites, a view supported by the stratigraphic occurrence of these taxa in Madagascar and Zululand. Plesiotexanites may be regarded as a Paratexanites with a variably developing lateral tubercle. Through fusing of the external (5) tubercles in Plesiotexanites, the tricarinate venter of Reginaites may be derived. Develop- ment in Reginaites tends towards discarding of the lateral (2) Plesiotexanites tubercle, and possibly also a sharpening of the venter and gradual weakening of the lateral keels, possibly resulting in Neogauthiericeras. Plesiotexanites also probably gives rise to part of Texanites, e.g. T. collignoni, and possibly T. umzambiense and T. texanus. Submortoniceras is derived from Texanites through reduction of the umbilical diameter in the South African material, and supports Young’s (1963: 90) views concerning the relationsip between Texanites shiloensis and Submortoniceras tequesquitense. Part of Texanites appears to be derived directly from Paratexanites without passing through a Plesiotexanites stage, e.g. T. vanhoepeni and T. postvanhoepeni. One distinct evolutionary trend noticeable in the South African texanitids is the tendency towards reduction of umbilical width and ornament in younger forms. A striking example of this is the sequence starting with T. umzambiense, through 7. presoutoni and T. soutoni to Submortoniceras woodsi and S. condamyi (Fig. 131). Another possible sequence is that of M. (Delawarella) subdelawarensis through M. (D.) sp. aff. jeanneti to M. (D.) nibelae. In addition, there appears to be a reduction of overall size in these two lineages, but here the trend is not very distinct. Through reduction of the umbilical width and ornament, and also overall size and length of the body chamber, the shell becomes more streamlined and consequently hydrodynamically stable. This seems to indicate a more active mode of life, and, possibly, a deeper water habitat. PALAEOBIOGEOGRAPHY The palaeobiogeographic distribution of the Texanitinae during the Late Coniacian to Early Santonian, Santonian s.l., and Early to Mid-Campanian, as shown in Figures 267-269. is basicallv the same as that shown by Collignon (1948: 52). CRETACEOUS FAUNAS FROM SOUTH AFRICA 351 O PROTEXANITES @ PARATEXANITES 60° A TEXANITES 50° 40° 36 20 10° 0 10° 20) 30) 40° 50 60 Fig. 267. Palaeobiogeography of Texanitinae during Coniacian to Early Santonian time (Map after Smith et al. 1973, fig. 7.) The geographic, as well as the biological origin (see p. 348) of the Texani- tinae is obscure, and detailed stratigraphic collection is necessary to substantiate Collignon’s assumption that radiation started in western Europe. Current data show that Protexanites and Paratexanites species already occur in western and central Europe, North Africa, the Gulf Coast region of North America, Hok- kaido, and Zululand by Late Coniacian time. Wiedmann (1960), however, reports Protexanites bourgeoisi, Protexanites cf. bontanti, and Protexanites sp. nov. from the Lower Coniacian of the Vascogotic region of the Iberian Peninsula, which may substantiate Collignon’s views. The rate of dispersion thus seems to have been very rapid. During Santonian time, four main regions of development were sheath established: western and central Europe and the Middle East; the Gulf Coast and adjacent areas of North America; Hokkaido and environs; and Mada- gascar and South Africa; with minor centres in north Africa (Algeria), west Africa (Nigeria and Cameroons), south-western Africa (Angola), and northern South America (Venezuela and Peru). In terms of numbers and diversity, the centre of maximum development appears to be Madagascar and South Africa. 352 ANNALS OF THE SOUTH AFRICAN MUSEUM O PLESIOTEXANITES @ REGINAITES A TEXANITES 1 DEFORDICERAS W@ BEVAHHES ® SUBMORTONICERAS of 7@ MENABITES 8 ® PARATEXANITES “SQ - Fig. 268. Palaeobiogeography of Texanitinae during Middle to Late Santonian time. (Map after Smith ef al., 1973, fig. 7.) Even though the majority of species in Santonian time appear to be endemic to these four main regions, faunal exchange did take place on a limited scale, as shown by the distribution of Plesiotexanites stangeri (Madagascar, South Africa, Hokkaido, and Gulf Coast). During Early and Middle Campanian times, endemism and development of the Texanitinae seems to have reached its peak. Again Madagascar and South Africa seem to have been the centre of development as exemplified by the large number of Menabites s.1. and Submortoniceras species, even if excessive taxonomic splitting is taken into account. In Europe, texanitine species decreased numerically, and this region was now of subordinate importance to North America and Hokkaido. Faunal interchange now was even more limited than during Santonian time, and restricted to a few species, e.g. M. (Delawarella) delawarensis (Zululand and North America), M. (Delawarella) campaniensis (North America and Europe), Submortoniceras chicoense (North America and Subantarctic Islands). The demise of texanitids towards the end of Middle Campanian time appears to have been virtually instantaneous, following a period of virtual maximum development and occasional gigantism in Zululand, and no texani- tids are known from the Upper Campanian sediments. CRETACEOUS FAUNAS FROM SOUTH AFRICA 353 @ REGINAITES O SUBMORTONICERAS @ BEVAHITES A MENABITES 60) Fig. 269. Palaeobiogeography of Texanitinae during Early to Middle Campanian time. (Map after Smith e¢ al., 1973, fig. 7.) ACKNOWLEDGEMENTS We are indebted to many persons and the institutes to which they are attached for assistance in completing this study. First and foremost we wish to express our gratitude to the late General M. Collignon, (Moirans) for assisting us with many of the identifications and replying patiently to our numerous queries, and we dedicate this work in his honour. For further assistance in the form of constructive advice, material, litera- ture and access to their collections we are grateful to Prof. Dr Jost Wiedmann (Ttibingen), Mr C. W. Wright (Oxford), Dr M. K. Howarth and Mr D. Phillips (British Museum, Natural History), Prof. R. A. Reyment (Uppsala), Dr Z. Lewy (Jerusalem), Dr H. Kohl (Oberésterreichisches Landesmuseum, Linz), Prof. K. Young (Austin), Drs C. K. Brain and E. Vrba (Transvaal Museum), Mr P. J. Rossouw (Geological Survey, Pretoria), and Dr J. N. Theron (Geological Survey, Cape Town). For access to restricted areas and assistance in the field we are indebted to the Natal Parks, Game and Fisheries Board, The Department of the Interior, 354 ANNALS OF THE SOUTH AFRICAN MUSEUM Transkei, and the Geological Survey, Pretoria. Mr Johannes Nonyane provided invaluable assistance during fieldwork in 1970 and 1971. Assistance with illustrations by L. Scott, N. Eden, C. Hunter and V. Branco of the South African Museum is gratefully acknowledged. Financial assistance to H. C. Klinger was provided by the South African Council for Scientific and Industrial Research; fieldwork by W. J. Kennedy in 1970 was financed by the Sir Henry Strakosh Bequest, and further research by a grant from the Natural Environment Research Council (United Kingdom). REFERENCES ANDERSON, F. M. 1958. Upper Cretaceous of the Pacific Coast. Mem. geol. Soc. Am. 71: 1-378. BalILy, W. H. 1855. Description of some Cretaceous Fossils from South Africa; collected by Capt. Garden, of the 45th Regiment. Q. J/ geol. Soc. Lond. 11: 454-465. BASSE, E. 1931. Monographie paléontologique du Crétacé de la Province de Maintirano, Mada- gascar. Gouv. Gén. Madagascar et Depend., Service des Mines, Madagascar. BESAIRIE, H. 1930. Les rapports du Crétacé Malgache avec le Crétacé de 1’Afrique australe. Bull. Soc. géol. Fr. 30: 107-117. BLANCKENHORN, M. 1905. Geologie der naheren Umgebung von Jerusalem. Z. dt. Palast. V. 28: 75-120. CARRASCO, V. B. 1967. Jimenensites, una nueva amonita del Campaniano inferieur de Mexico. Ingenieria Petrolera 7: 15-21. COLLIGNON, M. 1932. Fossiles du Crétacé supérieur du Menabe. Annls Paléont. 21: 1-54 (35-86). COoLLIGNON, M. 1938. Ammonites Campaniennes et Maestrichtiennes de l’ouest et du sud de Madagascar. Annls géol. Serv. Mines Madagascar 9: 1-65. COLLIGNON, M. 1948. Ammonites néocrétacées du Menabe (Madagascar). 1. Les Texanitidae. Annls géol. Serv. Mines Madagascar 13: 7-60 (63-116). COLLIGNON, M. 1966a. Atlas des fossiles caracteristiques de Madagascar (Ammonites), XIV (Santonien). Tananarive: Service Géologique. COLLIGNON, M. 1966b. Les Cephalopodes Crétacés du bassin cotier de Tarfaya. Notes Mem. Serv. Mines Carte géol. Maroc. 175: 10-148. COLLIGNON, M. 1969. Atlas des fossiles caracteristiques de Madagascar (Ammonites) XV (Campanien inférieur). Tananarive: Service Géologique. COLLIGNON, M. 1970. Atlas des fossiles caracteristiques de Madagascar (Ammonites) XVI (Campanien moyen et supérieur). Tananarive: Service Geologique. Cox, L. R. 1926. Aptychus spinosus sp. n. from the Upper Chalk. Ann. Mag. nat. Hist. 9(17): 573-580. Crick, G. C. 1907. Cretaceous fossils of Natal. 3. Rep. geol. Surv. Natal Zululand 3: 163-250. Desio, A. 1920. La creta nel bacino di Firenze. Palaeontogr. ital. 26: 187-243. FABRE-TAXxy, S. 1963. Faunes du Coniacien et du Santonien de Provence. 1. Les ammonites du bassin de Beausset (Var.). Annls Paléont. 49: 1-29 (99-126). GERHARDT, K. 1897. Beitrag zur Kenntniss der Kreideformation in Venezuela und Peru. N. Jb. Min. Geol. Paldont. Beil. Bd. 11: 65-117. GRABAU, A. W. & SHIMER, H. W. 1910. North American index fossils. New York: A. G. Seiler & Co. GRIESBACH, C. L. 1871. On the Geology of Natal in South Africa. Q. JI geol. Soc. Lond. 27: 53-72. GROSSOUVRE, A. DE 1894. Recherches sur la Craie Supérieure. II. Paléontologie. Les Ammonites de la Craie Supérieure. Mém. Carte géol. dét. France (1893): 1-264. Haas, O. 1942. Some Upper Cretaceous ammonites from Angola. Am. Mus. Novit. 1182: 1-24. HAUER, F. von 1858. Uber die Cephalopoden der Gosauschichten. Beitr. Geol. Osterr. 1: 7-14. HOoeEPEN, E. C. N. vAN 1921. Cretaceous Cephalopoda from Pondoland. Ann. Transy. Mus. 8: 1-48. CRETACEOUS FAUNAS FROM SOUTH AFRICA 355 Hoepen, E. C. N. VAN 1965a. The Peroniceratinae and allied forms of Zululand. Mem. geol. Surv. S. Afr. 55: 1-67. HoerpPEN, E. C. N. VAN 19655. New Ammonites from Zululand. Ann. geol. Surv. S. Afr. 4: 183-191. HowarTH, M. K. 1966. Ammonites from the Upper Cretaceous of the James Ross Island Group. Bull. Br. Antarct. Surv. 10: 55-69. Hyatt, A. 1900. Cephalopoda. In: ZiTTEL, K. A. von 1896-1900. Textbook of Palaeontology. translated by C. R. Eastman: 502-604. London: Macmillan. JONES, D. L. 1966. New Upper Cretaceous ammonite, Protexanites thomsoni from California. J. Paleont. 40: 199-203. KawabA, M. 1929. On some new species of Ammonites from the Naibuchi district, South Saghalien. J. geol. Soc. Tokyo 36: 1-6. KENNEDY, W. J. & KLINGER, H. C. 1971. A major intra-Cretaceous unconformity in eastern South Africa. J. geol. Soc. 127: 183-186. KENNEDY, W. J. & KLINGER, H. C. 1972. A Texanites—Spinaptychus association from the Upper Cretaceous of Zululand. Palaeontology 15: 394-399, KENNEDY, W. J. & KLINGER, H. C. 1973. In: KENNEDY, W. J., KAUFFMAN, E. G. & KLINGER, H. C. Upper Cretaceous invertebrate faunas from Durban, South Africa. Trans. geol. Soc. S. Afr. 76: 95-111. KENNEDY, W. J. & KLINGER, H. C. 1975. Cretaceous faunas from Zululand and Natal, South Africa. Introduction, Stratigraphy. Bull. Br. Mus. nat. Hist. (Geol.) 25: 265-315. KLINGER, H. C. 1972. The possible association of Spinaptychus with the genus Texanites. Ann. geol. Sury. S. Afr. for 1968/9 7: 105-109. Kuincer, H. C. & KENNEDY, W J. 1977. Upper Cretaceous ammonites from a borehole near Richards Bay, South Africa. Ann. S. Afr. Mus. 72: 69-107. KLINGER, H. C. & KENNEDY, W. J. In press. The Umzamba Formation at its type section, the Umzamba Estuary, Pondoland, Transkei. Ann. S. Afr. Mus. KULLMANN, J. & WIEDMANN, J. 1970. Significance of sutures in phylogeny of Ammonoidea. Paleont. Contr. Univ. Kans. 47: 1-32. LAsswITz, R. 1904. Die Kreide-Ammoniten von Texas. Geol. paldont. Abh. (N.F.) 6: 232-259. LEHMANN, U. 1967. Ammoniten mit Kieferapparat und Radula aus Lias-Geschieben. Paldont. Z. 41: 38-45. LEHMANN, U. 1970. Lias-Anaptychen als Kieferelemente (Ammonoidea). Paldont. Z. 44: 25-31. LEHMANN, U. 1971. Jaws, radula and crop of Arnioceras (Ammonoidea). Palaeontology 14: 338-341. LEHMANN, U. 1972. Aptychen als Kieferelemente der Ammoniten. Paldont. Z. 46: 34-48. LEHMANN, U. 1975. Uber Nahrung und Ernahrungsweise von Ammoniten. Paldont. Z. 49: 187-195. Matsumoto, T. 1955. Evolution of the Peroniceratidae. Trans. Proc. palaeont. Soc. Japan (N.S.) 18: 37-44. Matsumoto, T. 1959. Upper Cretaceous Ammonites of California II. Mem. Fac. Sci. Kyushu Univ. (D) Special Volume 1: 1-172. Matsumoto, T. 1965. A Monograph of the Collignoniceratidae from Hokkaido Part II. Mem. Fac. Sci. Kyushu Univ. (D) 16: 209-243. MATSUMOTO, T. 1966. Notes on Ammonites bourgeoisi D’Orbigny from the Upper Cretaceous of France. Trans. Proc. palaeont. Soc. Japan (N.S.) 61: 201-206. Matsumoto, T. 1970. A Monograph of the Collignoniceratidae from Hokkaido Part IV. Mem. Fac. Sci. Kyushu Univ. (D) 20: 225-304. Matsumoto, T. & HIRANO, H. 1976. Colour Patterns in some Cretaceous ammonites from Hokkaido. (Studies of Cretaceous Ammonites from Hokkaido and Saghalien). Trans. Proc. palaeont. Soc. Japan (N.S.) 102: 334-342. Matsumoto, T., OBATA, I., MAEDA, S. & SATO, T. 1964. Yabeiceras (Cretaceous ammonite) from Futaba, Northeast Japan. Trans. Proc. palaeont. Soc. Japan (N.S.) 56: 322-331. Matsumoto, T. & UEDA, Y. Jn UEDA, Y. 1962. The type Himenoura Group. Mem. Fac. Sci. Kyushu Univ. (D) 12: 129-178. MEEK, F. B. 1876. Jn: MEEK, F. B. & HAYDEN, F. V. 1876. A report on the invertebrate Cre- taceous and Tertiary fossils of the Upper Missouri Country. U.S. Geol. Surv. Terr. 9: 1-629. 356 ANNALS OF THE SOUTH AFRICAN MUSEUM Mor Ton, S. G. 1830. Synopsis of the organic remains of the ferruginous sand of the United States. Am. J. Sci. 18: 243-250. OrsBiGNy, A. D’ 1850. Prodrome de Paléontologie Stratigraphique Universelle des Animaux Mollusques et Rayonnes. 2. Peron, A. 1896-7. Les ammonites du crétacé supérieur de l’algérie. Mém. Soc. géol. Fr. Paléont. 6: 1-24. REDTENBACHER, A. 1873. Die Cephalopodenfauna der Gosauschichten in den nordéstlichen Alpen. Abh. geol. Bundesanst. Wien 5: 91-149. REESIDE, J. B. 1927. Cephalopods of the Eagle Sandstone and related Formations in the Western Interior of the United States. Prof. Pap. U.S. geol. Surv. 151: 1-87. REESIDE, J. B. 1962. The Cretaceous fossils of New Jersey, part 2. Bull. New Jers. Div. Geol. 61. RENNIE, J. V. L. 1930. New Lamellibranchia and gastropoda from the Upper Cretaceous of Pondoland (with an appendix on some species from the Cretaceous of Zululand). Ann. S. Afr. Mus. 28: 159-260. RENZ, H. H. 1936. Neue Cephalopoden aus der oberen Kreide von Rio Grande del Norte (Mexico und Texas) mit einer Einflihrung von Walther Staub. Abh. schweiz. paldont. Ges. 57: 1-16. REYMENT, R. A. 1955. The Cretaceous Ammonoidea of southern Nigeria and the southern Cameroons. Bull. geol. Surv. Nigeria 25: 1-112. REYMENT, R. A. 1957 Uber einige wirbellose Fossilien aus Nigerien und Kamerun, West- afrika. Palaeontographica 109: 41-70. REYMENT, R. A. 1958. Neubeschreibung der Redtenbacher’schen Ammoniten Originale aus den Gosauschichten. Stockholm Contr. Geol. 2: 1-49. Roemer, F. A. 1852. Die Kreidebildungen von Texas und ihre organischen Einschliisse. Bonn: Adolph Marcus. Rocers, A. W. & SCHWARZ, E. H. L. 1902. General survey of the rocks of the southern parts of the Transkei and Pondoland including a description of the Cretaceous rocks of eastern Pondoland. Rep. geol. Commn. Cape Good Hope 1901: 25-46. SCHLUTER, C. 1867. Beitrag zur Kenntniss der jiingsten Ammoneen Deutschlands. Bonn. SCHLUTER, C. 1871-1876. Die Cephalopoden der oberen deutschen Kreide. Palaeontographica 21: 1-24 (1871), 25-120 (1872), 24: 121-262 (1876). SHimizu, S. 1935. The Upper Cretaceous cephalopods of Japan. Part 1. J. Shanghai Sci. Inst. 2: 159-226. SMITH, A. G., BRIDEN, J. C. & Drewry, G. E. 1973. Phanerozoic world maps. Spec. Pap. Palaeont. 12: 1-42. SPATH, L. F. 1921. On Cretaceous Cephalopoda from Zululand. Ann. S. Afr. Mus. 12: 217-321. SPATH, L. F. 1922. On the Senonian ammonite fauna of Pondoland. Trans. R. Soc. S. Afr. 10: 113-147. SPATH, L. F. 1925. On Upper Albian Ammonoidea from Portuguese East Africa. Ann. Transv. Mus. 11: 179-216. SPATH, L. F. 1932. A monograph of the Ammonoidea of the Gault. Part 9. Palaeontogr. Soc. [Monogr.] 84: 379-410. SPATH, L. F. 1953. The Upper Cretaceous Cephalopod fauna of Grahamland. Sci. Rep. Falkld. Isl. Dep. Surv. 3 : 1-160. TAUBENHAUS, H. 1920. Die Ammoneen der Kreideformation Paldstinas und Syriens. Z. dt. Palast. V. 43: 1-58. THOMEL, G. 1969. Sur quelques ammonites Turoniennes et Sénoniennes nouvelles ou peu connues. Annls Paléont. 55: 1-11 (108-119). TRASK, J. B. 1856. Description of new species of ammonite and baculite from the Tertiary rocks of Chico Creek. Proc. Calif. Acad. Sci. 1: 85-86. VENZO, S. 1936. Cefalopodi del Cretacea medio-superiore dello Zululand. Palaeontogr. ital. 36: 59-133. WEDEKIND, R. 1916. Uber Lobus, Suturallobus und Inzision. Zentbl. Miner. Geol. Paldont. 1916: 185-195. WELLER, S. 1907. A report of the Cretaceous paleontology of New Jersey. Geological Survey New Jersey, Paleontology Series 4: 1-871. WHITFIELD, R. P. 1892. Gastropoda and Cephalapoda of the Raritan clays and greensand marls of New Jersey. Monogr. U.S. geol. Surv. 18: 1-402. CRETACEOUS FAUNAS FROM SOUTH AFRICA 357 WIEDMANN, J. 1960. Crétacé supérieur de l’Espagne et du Portugal et ses Céphalopodes. Estudios Geol. 20: 107-148 (reprinted in 1964). WIEDMANN, J. 1978. Eine palaogeographisch interessante Ammonitenfaunula aus der alpinen Gosau (Santon, Becken von Gosau, Oberdésterreich). Eclog. geol. Helv. 71/3: 663-675. Woops, H. 1906. The Cretaceous fauna of Pondoland. Ann. S. Afr. Mus. 4: 275-350. WOLLEBEN, J. A. 1967. Senonian (Cretaceous) Mollusca from Trans-Pecos Texas and north- eastern Chihuahua. J. Paleont. 41: 1150-1165. WRIGHT, C. W. 1957. In: Moore, R. C. (Ed.) Treatise on invertebrate paleontology, Part L, Mollusca, Cephalopoda, Ammonoidea. Geological Society of America and University Kansas Press. WRIGHT, C. W. & WRIGHT, E. V. 1951. A survey of the fossil Cephalopoda of the Chalk of Great Britain, Palaeontogr. Soc. [Monogr.] 1951: 1-40. YABE, H. & SHIMIZU, S. 1923. A note on the genus Mortoniceras. Jap. J. Geol. Geogr. 2: 27-30. YABE, H. & SHIMIZU, S. 1925. Japanese Cretaceous ammonites belonging to Prionotropidae. Sci. Rep. Tohoku Imp. Univ. 7: 125-138. Youna, K. 1963. Upper Cretaceous ammonites from the Gulf Coast of the United States. Univ. Texas Publs. 6304: 1-142. ZITTEL, K. A. VON 1884. Handbuch der Palaeontologie (Palaeozoologie). i. Abt., ii Band, Lief. iii. Cephalopoda. Munich & Leipzig: Oldenbourg. ZURCHER, P. 1905. Communication. Bull. Soc. géol. Fr. 5: 686. pa “4. . SYSTEMATIC papers must conform to the International code of zoological nomenclature particularly Articles 22 and 5i). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be ollowed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. OV., Syn. nov., etc. . An author’s name when cited must follow the name of the taxon without intervening unctuation and not be abbreviated; if the year is added, a comma must separate author’s lame and year. The author’s name (and date, if cited).must be placed in parentheses if a pecies or subspecies is transferred from its original genus. The name of a subsequent user of Scientific name must be separated from the scientific name by a colon. Synonymy arrangement should be according to chronology of names, i.e. all published cientific names by which the species previously has been designated are listed in chronological rder, with all references to that name following in chronological ordez, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A ucula (Leda) bicuspidata Gould, 1845: 37. eda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b). ucula largillierti Philippi, 1861: 87. : eda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9. Note punctuation in the above example: comma separates author’s name and year semicolon separates more than one reference by the same author full stop separates references by different authors figures of plates are enclosed in parentheses to distinguish them from text-figures dash, not comma, separates consecutive numbers Synonymy arrangement according to chronology of bibliographic references, whereby he year is placed in front of each entry, and the synonym repeated in full for each entry, is ot acceptable. In describing new species, one specimen must be designated as the holotype; other speci- ens mentioned in the original description are to be designated paratypes; additional material ot regarded as paratypes should be listed separately. The complete data (registration number, pository, description of specimen, locality, collector, date) of the holotype and paratypes ust be recorded, e.g.: olotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach ort Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and date. . SPECIAL HOUSE RULES apital initial letters a) The Figures, Maps and Tables of the paper when referred to in the text e.g. ©... the Figure depicting C. namacolus .. .’; ‘. . . in C. namacolus (Fig. 10)...’ b) The prefixes of prefixed surnames in all languages, when used _in the text, if not preceded by initials or full names e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian unctuation should be loose, omitting all not strictly necessary eference to the author should be expressed in the third person Roman numerals should be converted to arabic, except when forming part of the title of a book or article, such as “Revision of the Crustacea. Part VIII. The Amphipoda.’ pecific name must not stand alone, but be preceded by the generic name or its abbreviation to initial capital letter, provided the same generic name is used consecutively. ame of new genus or species is not to be included in the title: it should be included in the abstract, counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts. HERBERT CHRISTIAN KLINGER & WILLIAM JAMES KENNEDY CRETACEOUS FAUNAS FROM ZULULAND AND NATAL, SOUTH AFRICA THE AMMONITE SUBFAMILY: TEXANITINAE COLLIGNON, 1948 INSTITUTION NOILALILSNI SSIIYVEEIT. LIBRARTES LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI = Wd & wl = re Ly al 2 Ee n nr Su fy x A —4 WY fo 2 < Se 2 < 7 = co = a = ra) O = S) — O ae iz =i 2 a Zz ah IOSHLINS SAJIYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNILILSNI we — Pas S a Z re SNR 5 2% = a 5 = SS Sa = > ae > i ral QE : 2 : = 2 io) os ui Zz D = vs FHSONIAN 2) = = ae wn O = > z SMITHSONIAN NVINOSHLIWS SMITHSONIAN NVINOSHLIWS QQ FA SMITHSONIAN NVINOSHLIW = w —_— Ww — é ed ae : z Y fyg “aS rea =|. : : de a : : ac er “x Yy oz ac Cc e 5 o Ul = 5 a Bs] FA =i z “ INSTITUTION NOILALILSNINVINOSHLINS S3IYVYUGIT LIBRARIES SMITHSONIA s Z a: 5 a a\h ee = = 5 = = L 3 =) aad =) 2 = a i = re 2 Z PO) aie a z NOSHLINS SAIUVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSHLIM a 7 ee < za, 2 = Zz = 5 L fy z : : : 2 YO O ae 1 = O Vi Zz = = = ZT aad > = > = ~ > PZ Wu) za YW”) 2 ee = -HSONIAN INSTITUTION NOILMLILSNI_NVINOSHLINS S3!YVYGIT LIBRARIES LIBRARIES SMITHSONIAN NOLLALILSNI EL ERAR LES NOILALILSNI LI BRAR IES NOILALILSNI fOSHLINS SZIYVYUEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNILILSNI eS Ss ro oO — hone 9 = oe z oD Ree Se 5 a > 2 =) io So Xe > > a ie ca i= 4 NN - tis i as = 2 ie OD z a = o . (HSONIAN INSTITUTION NN ee toe an < = 2) Os z = tt fra = a is SS = O EM ‘bp fe) oe GO UNNON 5 ‘hj ae 2 = = Z “yy = 2 Ee Z = > = > = a nee n ree n 2 a Seyi JOSHLINS S3IYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIA 7p) = w = = oe w = @ = a Fa Ss NE: z =. i fp 5 ZX SE 2 of Uf 3 < = HE FQ oc = “x Yyy a4 2 4 YS Ns 5 Ge = 5 SO TRE Mama Fz aes Zz pe z ~ LIBRARIE NOILNLILS! NOILNLILS LIBRARIE NOILNLILSH LIBRARIE V¥d!l)l LIBRARIES baa INSTITUTION NOILNLILSNI S31uvVudI c Zi z ER = w Oo =] a8 oO Ps) be Beef I= w ed 2. 7Y 4 = aay 3 a . YG Ee z = = i- “GHG - = = ‘ 2 i Z ts ie SHTUTION NOILALILSNI NVINOSHLINS SJluyvudydil LIBRARIES. SMITHSONIAN NVINOSHLINS S3ZIuUVYEIT a Z z z K. Z = BS 5 i fan / fy Zz WS 5 = » SS = tthe fig 5 Oy, S \Y Ae =z 5 2 ty Ub a hile ye \\ “ D o) TS 2 Vy =a “fy E ne = : = YW) Ww) +. Pad W) Vudit LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI = y) / > Ww = Y n a a” = Ww ud = z = - = : y << = o = a = ac nce a rs) as rs) = e a 5 z oy z peek UTION _ NOILALILSNI NWINOSHLIWS [SSIS VEatT LIBRARIES SMITHSONIAN INSTITUTION | O aha eee < = ae Oe 5 : 5 : : : | = “4 = = a jor we { z i es a = a Vudii x u ee pi NOU ee eM NVINOSHLIWS |S 3 byYvudid Wr, = gate < < = yy, Yyy = 4 Zz = ee a Aft 4 O Ns rt O ae iat S| <= 2 \\ © rm) Oo an) wn ZE ‘VSS e) ef eRe: ee O E Wy é E Z, = Z. 1 = “SS > = >" = > 7) a z ” ca z ” Zz UTION NOILNLILSNI_NVINOSHLINS S3luVvudi LIBRARIES SMITHSONIAN INSTITUTION a > = ‘ us G - z | gy, =? : yo A = a = e D e rs} o. Oo : ay z= | Zz as | ze ; IVWYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVvudgty : : : 5 : 5 = oe O ie oO > -xZf ff > = > § = a Wy = ss 2 iz = ND “fl m ple m g Ww WM = ie 2) . = ITUTION VOSHLINS S3SIYVYEIT LIBRARIES SMITHSONIAN & Ww wa = ie 2) z NY . = = Vass | = = XY = 3 Ze = Zz : aE a . ~ a 4 as S| = = = Ss Fa — = = ; > > = pee = a > Zz w hb hae wn Fie as yWUdII INSTITUTION NOILONLILSNI NVINOSHLINS S3ZI1YvVaugit NOILALILSNI LIBRARIES NOILNLILSNI LIBRARIES NOILOLILSNI Pea ati im Mattia wi SRO rE eT ae eR AL Pee hebraaeh SMITHSONIAN INSTITUTION LIBRARIES Rann s4ee, rey raee crt Ratigte’? % are DAE Eye RP ET ER Bet by! 5 3 t ora ALHe MPS 4 : 4 Bek : ‘ ; : ark: a x 6 Tat fony ott iy Lae ye woe : Pak « . sh ¥ae) F ' nay Be Ce rae * rice ¢ . t Lah. Wate f ce ay A Ot ee ae ee ALP RES Puy es Ray of N Ae fora’ aoe Vetaseu 3 9088 01206 6726 Riplhrten Rens dy ants by ne od REGS eet ob. OF ad Pizns tS Yin sud Vai Petes bites theta echned tas EOL wots PET a porte Shee esae ey Whe Wes aheley Pere Eero aw: Di ae Soy Er eee es rugs HE ri, bee! ore ipepie beh ws 9 oh eagle . es erm nepe ns eye ey: Ora tata ewe Myarete, Ure ee wes ee wee TG reece enn: Sak ib es epee ~ ar cba nleseih ch vole Lp wet ey Whee Py OS hang! sey eas CY ee ce ee EY rane pds hy Pvc OER i ek OR ok aa + ere re verge. ea wa Gt ee Ne yey ee Re yi tes Bel We RAPA BRET Sin ane Pe Aes ‘ 2 a an ae Re Wee te cn oa : ! tye potas Oe. 3) i hie ie we ” pe : . : Jie © W be ray ey ; Natt ee eae: ete iy ch hh heh Rett e ee? + ats \ : : p LEE evietoace it BH ee iets Bipe Fane PPE Uae INET Br ere seh Pa Te chat ea) AS Veeeey HOES rigs bs ee re ce re vespabe My We net WP ESM P LENIN, Or Ursa CAN by ae Mg ays ‘ went Nsom wet, bas ims Deere ays PABA AR bevy dite ‘ ss Nan V : bs lea eee Bee seyret ytury ¢ Wy, oo PO ee. y ‘ oe : , Wud reyes Bey Wee BNEW ete BRS ae f a etre ‘ : Cares ae aie Styne phe Sey 4Fy tha aan apes | yi ah ace : POO I OTe AE RT Ue Ebi ed b cask ts Cer einen aa “eek } had i! ure oo : sraseipteroroesonnenes HPO OS HOY bry! Nee al rae 4 So DBR hy steep 7 ; te bp i ’ ; f Z SPW eer, leech idk ba id dal SET Md kn Ce ae na Sear j él , : ‘ F bo te seta Wie Stele ere Saat Fad oh Meh ECA ELE oh TA f Lived ; Ds hg a ‘ KE SP ge Ore by we \ "4 , 8 neonate Sees Sakon he Wea hes raid arity rie eure ; Sa aoe Sad Ci din add Bie Be ee he Pry e iF ‘s y ’ 4 ‘ Pretorya gh * PREVA Meh sidaL dD : dey Bebe Meg a ‘ hen fibibn cance Pee oe wy RY, Otis tee : we SAN . pleat 7 : SVereaway ly rer pee Pave per ey luc ebay t PH eriacdbe pihey ¢ ee SA - ‘ (a pkg hth els ais PSE AEDST EC APYPUY EMI) Ee) Fi aoe tebe f oa vi vk He (i Sti th ore Degraw 59. a0vbas bey Gry 4 Paeaepaping < big a ie ‘ : ral orbtaoleuree ioe vas tibia Aa SVE ee dy ayy ae api LA Oh : te . i hide tila Re dies Ek de I areas at iw BONG pivey % ; vel a ee? ht pry ie at ie eg etite: Wea ay ay B95 ree ci yet 7 : ‘ ‘ bee I EE een lyan sobre eee HT A iebey t 4 ; ‘ \ : we iserevatercia idee eoitn 8 48 : be : ‘ vo p ee i a ~* Ws r Cera ’ Lptteh ho vacak Ls vw nasetor ae usw by Fea Li . magnet ae eat cane id Saber CLD) wt ; 405) tee) Ope Vary ; : ( I el ORL t ‘ } vt, iy ra ‘ Bey iy “i Patt og) Oe Sey . eevee ‘ , PI ABD Pune. ‘ * 4 { Pare tek ph Py, eet i abet yy : avers ey : OER a Ok Ck ie OE A ee : PR: : EB a sik) vie &, d ' ah my , Oa Mae We D ie serra % ey moe Ura iW, : i Ss i UAE fe} ire eee a Ay; { pers wns fF ; Lhe ‘ Latte ater oe . or, 4 i IPE SV Sy v hy ; tA Ber WAL hae Hers Wiese Ages yet Wiehe : hg Oi ies