VOLUME 69 PART 6 APRIL 1976 i * il ~ me, f ite | Ss pe = — of Mey ee" ISSN 0303-2515 INSTRUCTIONS TO AUTHORS 1. MATERIAL should be original and not published elsewhere, in whole or in part. When accepted, copyright becomes the property of the Trustees of the South African Museum. 2. LAYOUT should be as follows: (a) 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) Introduction (e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents (f) Summary, if paper is lengthy (g) Acknowledgements (h) References (i) Abbreviations, where these are numerous 3. MANUSCRIPT, to be submitted in triplicate, should be typewritten, double spaced with 2,5 cm margins all round. Tables and 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. Only generic and specific names should be underlined to indicate italics; all other marking up should be left to editor and publisher. 4, ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18 cm (19 cm including legend); originals larger than 35 x 47 cm should not be submitted; photographs should be final size. A metric scale should appear with all illustrations, otherwise magnification or reduc- tion should be given in the legend. 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 marked 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, 19695; Jones 1971)...’ “As described (Haughton & Broom 1927)...’ ‘As described (Haughton et al. 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 of 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. 19606. 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 ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM Volume 69 Band April 1976 April Part 6 Deel " RASSSn, SB: B.B.9-9) S Ae a Xx Souig nov M8 ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE (REPTILIA, SAURISCHIA) by P. M. GALTON & M. A. CLUVER 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 Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM word uitgegee in dele op ongereelde tye na beskikbaarheid van stof Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad OUT OF PRINT/UIT DRUK 1, 21, 3, 5-8), 3(1-2, 4-5, t.—p.i.), 511-3, 5, 7-9), 6(1, t.—p.i.), 70-4), 8, 911-2), 10(1), 11(1-2, 5, 7, t.-p.i.), 15(5), 24(2), 27, 31(1-3), 33 Price of this part/Prys van hierdie deel R3,70 Trustees of the South African Museum © _ Trustees van die Suid-Afrikaanse Museum 1976 ISBN 0 949940 86 0 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 ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE (REPTILIA, SAURISCHIA) By P. M. GALTON University of Bridgeport, Bridgeport, U.S.A. & M. A. CLUVER South African Museum, Cape Town (With 13 figures and 3 tables) [MS accepted 22 October 1975] ABSTRACT A complete description of the skeleton of the type specimen (SAM-990) of the pro- sauropod dinosaur Gyposaurus capensis Broom is given. Comparison with other prosauropods indicates that this dinosaur represents a valid South African species of Anchisaurus Marsh, 1885 from the Upper Triassic of North America. It is considered that the infra-order Pro- sauropoda should be divided into three families, viz. Anchisauridae, Plateosauridae and Melanorosauridae. It is proposed that the family Anchisauridae be restricted to prosauropods with relatively slender feet, and that broad-footed forms previously assigned to the Anchi- sauridae be transferred to the Plateosauridae. The Family Anchisauridae is therefore con- sidered to include the genera Anchisaurus, Efraasia, Thecodontosaurus and several indetermi- nate species. The genera Ammosaurus, Aristosaurus, Massospondylus (including Gryponyx, Aetonyx and Dromicosaurus), and Lufengosaurus (including Yunnanosaurus) are included in the Plateosauridae, while Arctosaurus and Ischisaurus are referred to the suborder Theropoda. ‘Thecodontosaurus’ gibbidens, Spondylosoma absconditum and Teleocrater alphos are placed within the Thecodontia, and Tanystropheus primus and T. latespinatus in the order Lacertilia. Thus constituted, the family Plateosauridae becomes the dominant and most widespread prosauropod family, while the Anchisauridae, in contrast, is known from a geographically and numerically restricted fossil record. CONTENTS PAGE Introduction , : : : 122 Redescription of An enisaunes ener (Broom) : : z 124 Vertebral column . : : : 5 : : ; 124 Pelvic girdle . : : : : : : : : 126 Hind limb : : ; : . 129 Slender and broad- footed Prosauropods ; ; : a - b3il Systematic discussion . , : ; ‘ : ? 132 The Family Anchisauridae : : ? : : d 132 Anchisauridae nomina dubia ; F : 141 Prosauropod species incorrectly assigned to Anchisauridae ‘ 143 Prosauropoda nomina dubia Z g : 150 Non-prosauropod species incorrectly assiened to Anchisauridae : 150 Notes on the Families Anchisauridae and Plateosauridae : 153 Summary . 2 : ‘ : ; ; ; : : 155 Acknowledgements. 3 : : , : : ; 155 References : 3 5 : : : F , : 156 Abbreviations. 3 : 2 P : ‘ ; : 158 IBZ Ann. S. Afr. Mus. 69 (6), 1976: 121-159, 13 figs, 3 tables. 122 ANNALS OF THE SOUTH AFRICAN MUSEUM INTRODUCTION Broom (1906) described a specimen (SAM-990) from the Cave Sandstone, Stormberg Series (Upper Triassic) of Ladybrand, Orange Free State, South Africa, and referred it to the prosauropod taxon Hortalotarsus skirtopodus Seeley, 1894. The holotype of Hortalotarsus skirtopodus was referred to the genus Thecodontosaurus Riley & Stutchbury, 1836, by Huene (1906) as T. skirtopodus and, because SAM-990 differed in several aspects from Theco- dontosaurus, Broom (1911) made it the holotype of Gyposaurus capensis. Broom (1906) noted several resemblances between SAM-990 and Anchisaurus Marsh, 1885, from the Upper Triassic of North America, so he referred Gypo- saurus capensis (SAM-990) to the Family Anchisauridae Marsh, 1885. Galton (1973, in press) provisionally accepted the validity of Gyposaurus capensis but noted that the holotype should be carefully compared with Anchisaurus polyzelus (Hitchcock) to determine whether or not these species are generically distinct. A comparison of photographs of SAM-990 with a specimen of Anchi- saurus polyzelus (YPM 1883) showed that Gyposaurus is a junior synonym of Anchisaurus and that SAM-990 should be redescribed. Huene in several papers between 1906 and 1932 made important contri- butions to an understanding of the Family Anchisauridae (as Thecodonto- sauridae Lydekker, 1890). Charig et al. (1965) considerably enlarged the family by referring to it genera of Triassic theropods (Family Gryponychidae = ‘Palaeosauridae’) based solely on postcranial material which was_ indis- tinguishable from that of prosauropods. The generic list of Romer (1966: 370) includes these changes and the comprehensive list given by Steel (1970) faithfully but rather uncritically records all the genera and species referred to the family. A taxonomic revision of the Family Anchisauridae is necessary for several reasons: 1. Several of the suggested and generally accepted synonymies are probably incorrect. In assessing these the recognition of slender- and broad-footed types (Galton 1971, 1973, in press) is useful. 2. The skeletal anatomy of prosauropods is remarkably uniform, so taxa should be based on specimens which include either most of the specimen or bones which are diagnostically different from those of other prosauropods. Because of the limited number of skeletal variations it is necessary to diagnose taxa on the basis of a combination of characters which, if each was taken in isolation, would not be diagnostic. Unfortunately several of the taxa listed by Steel (1970) are based on specimens which are generically and_ specifically indeterminate. 3. The group has been over-classified and insufficient account has been taken of the range of individual variation possible within a dinosaurian species; this can be quite extensive, as shown by the prosauropod Lufengosaurus huenei Young (see Rozhdestvensky 1966) and the ornithopod Hypsilophodon foxii Huxley (see Galton 1974). 4. A few of the taxa included within the family are not prosauropods. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 123 Fig. 1. Anchisaurus capensis (Broom). Type specimen (SAM-990) showing vertebral column (top), pubis, ischia and pes (middle) and ilium, femur and fibula (bottom). 124 ANNALS OF THE SOUTH AFRICAN MUSEUM Summaries of certain aspects of the revision of the family Anchisauridae have been published earlier (Galton 1971, 1973, in press). The first author (P. M. G.) is responsible for all sections except the description and illustration of SAM-990, which is the work of the second author (M. A. C.). REDESCRIPTION OF ANCHISAURUS CAPENSIS (BROOM) The specimen (SAM-990) is preserved in a soft sandstone matrix (Fig. 1). The bone is generally not well preserved and is inclined to crumble during preparation. Much of the specimen (including most of the vertebral column) is preserved as bone impression, and details of certain of these portions were obtained from silicone rubber positives. Chief portions of the preserved skeleton are 17 vertebrae (including, probably, dorsals, sacrals and caudals), an incom- plete left pubis, the right illum and pubis and both ischia, and the right femur, fibula and pes. Several other fragmentary bones, mostly seen as impressions, are scattered through the block: anteriorly a portion of the right scapula blade can be made out, while impressions of 13 ribs lie ventral to the dorsal vertebrae. VERTEBRAL COLUMN (Fig. 2) Altogether 17 vertebrae are preserved, some very incompletely. Eleven of these are in articulation and consist of a number of dorsals and possibly two sacrals. A space separates the last of these from the first of the posterior group, which have been displaced to the right of the anterior series. The space is sufficient to accommodate three vertebrae of the size of those on each side of it, and there is thus the possibility that this gap was originally filled by three sacral vertebrae. However, as will be shown below, it is more likely that a parting of the vertebral column between two sacral vertebrae occurred prior to fossilization of the specimen. The anterior three vertebrae, imperfectly seen, are not in natural articula- tion, although still in relatively close association with each other. The centrum of the second vertebra is opisthocoelous and slightly convex anteriorly, while the third centrum is concave anteriorly and posteriorly. The fourth vertebra has a procoelous centrum and is markedly convex posteriorly. This convex rear meets the apparently convex anterior surface of the fifth vertebra’s centrum, which also appears to be opisthocoelous. The sixth vertebra is provided with a biconcave centrum, the neural spine is broad in lateral view, and its postero- dorsal edge overhangs the postzygapophysis so that a posterior embayment is formed above the postzygapophysis. The seventh vertebra resembles the sixth and the succeeding eighth in the shape of the neural spine, and both the seventh and eighth vertebrae nave biconcave centra, similar to that of the sixth vertebra. The eighth vertebra shows the neural spine clearly, and probably represents the condition which existed in the less complete sixth and seventh vertebrae. The spine, posteriorly situated, is antero-posteriorly lengthened and fairly low. The neural spine of the ninth vertebra is shorter antero-posteriorly. The ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 125 Fig. 2. Anchisaurus capensis (Broom). SAM-990. A. Vertebral column in lateral view. Broken edges shown in dashed outline, hatched areas seen in section only. « 0,5. B. Mould of centra of dorsal vertebrae. 126 ANNALS OF THE SOUTH AFRICAN MUSEUM tenth vertebra has the only complete neural spine and there is a shallow notch above the postzygapophysis. The ninth and tenth vertebrae are possibly pro- coelous, with a convexity on the posterior articular surface. The posterior group of six vertebrae lies slightly to the right of the anterior row, and appears to be in near-natural association with the right ilium. The spine of the most anterior vertebra (the twelfth in the column as preserved) is fairly high and slightly rounded anteriorly; there is only a slight posterior notch above the postzygapophysis. The spine resembles that of the tenth (and eleventh?) vertebra of the anterior row fairly closely. The thirteenth vertebra has a high and narrower spine, posteriorly inclined, and a chevron can be seen extending back and down from below its centrum. The fourteenth and fifteenth vertebrae are similar, as far as can be seen. Remnants of chevrons are seen between the centra of vertebrae fourteen to seventeen. Identification of vertebral types Caudal vertebrae are fairly clearly represented by nos. 13 to 17 in the rear series, with narrower, obliquely inclined spines and a series of chevrons. The last two vertebrae (10 and 11) of the anterior row have narrower and possibly higher spines than the preceding ones, and they resemble the first member of the posterior group. From this it can be argued that the tenth, eleventh and twelfth vertebrae of the column as a whole are sacrals, separated by the dis- integration of the pelvic girdle. Vertebra 12 is in fairly natural association with the right ilium, and 10 and 11 are close to the ilium of the left side. The left ilium has been displaced sideways and forward relative to the right side, and this could account for the gap in the series. Moreover, no recognizable loose vertebral portions can be identified in the surrounding matrix. In Efraasia (Galton 1973) and Plateosaurus (Huene 1926) there are fifteen dorsal vertebrae so, if the above interpretation is correct, the specimen as preserved probably includes dorsal vertebrae 7 to 15, sacral vertebrae | to 3, and caudal vertebrae | to 5. Dimensions of the vertebrae, numbered according to their above identi- fication, are given in Table 1. PELVIC GIRDLE The pelvic girdle is represented by the right ilium, both ischia, and an incomplete left and almost complete right pubis. The right side of the pelvis (Figs 3, 5) is preserved in almost natural association. The ilium is well pre- served, but both pubis and ischium are incomplete distally, and do not make any clear contact with each other. The areas of articulation between ischium, pubis and ilium are imperfectly preserved and could not be determined. Tlium The ilium (Figs 3A, 5) of the right side is uncrushed and complete except for a portion of the dorsal crest. It is characterized by long anterior and posterior ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 127 TABLE 1. Dimensions of Vertebrae (mm). Anchisaurus capensis (Broom) SAM-990 Length Height of Length of between pre- spine above centrum Maximum and post- Dost- Length of Vertebra ventrally height zygapophysis zygapophysis spine 7th dorsal . ; 2 — — == si: 8th dorsal . 5 29 = — aes ae 9th dorsal . ; +30 — = aoe si 10th dorsal . ; 30 — — we ma 11th dorsal . : 31 — = ps pees 12th dorsal . ; 31 40 — 13 30 13th dorsal . ; 33 42 40 16 32 14th dorsal . : 33 40 48 14 30 15th dorsal . : 32 38 44 13 20 Ist sacral . : Sil 43 39 19 18 2nd sacral . ‘ — — oss has ss 3rd sacral . : — — 35 18 Dill lst caudal . : 18 SI 34 2) 13 2nd—Sth caudals . +18 — = =o eu processes, and a pre-acetabular process considerably longer than the post- acetabular process. The anterior process lies slightly external to the more posterior surface of the bone, and extends as far forwards as the anterior edge of the pre-acetabular process. The slender finger-like form of the anterior process (Fig. 10B) is similar to that of Anchisaurus polyzelus (Fig. 10A) and Ammosaurus (Galton 1971) and in contrast to the small triangle of other pro- sauropods (Fig. 10C—D). A prominent pre-acetabular buttress is developed, arising from close above the tip of the pre-acetabular process and flaring out laterally before merging with the body of the ilium at the base of the postacetabular process. The body of the ilium is expanded to a certain extent above the buttress. The postacetabular process is considerably shorter than the pre-acetabular, and takes no part in the formation of the buttress. Above the postacetabular process the ilium is continued posteriorly as a short crest, medial to the base of the posterior process. The maximum length of the ilium is 130 mm. Pubis The proximal part of the pubis (Fig. 3A) is incomplete, and the areas of articulation with the ilium and, to a lesser extent, the ischium are not fully preserved. The ventral edge of the bone is deeply notched below the proximal end, and this appears to be a natural condition. Below this embayment, which represents an open obturator foramen, the pubis curves medially and forwards to terminate as a horizontal plate with an average width of 24 mm. The length of the pubis, as preserved, is 145 mm. For prosauropods an open obturator foramen is described to date only 128 ANNALS OF THE SOUTH AFRICAN MUSEUM ant. proc. A pub. ped. Fig. 3. Anchisaurus capensis (Broom), SAM-990. A. Pelvic girdle in right lateral view. B. Ischia in ventral view. > 0,5. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 129 in SAM-990 (Fig. 10H) and Anchisaurus polyzelus (Fig. 10G, K); in all other prosauropods it is enclosed ventrally as in Efraasia (Fig. 10J, N). A similar open obturator foramen is present in most theropods (for Allosaurus and Ceratosaurus see Gilmore 1920). Romer (1923) noted that as a result of the more vertical orientation of the archosaurian femur there is a trend amongst archosaurs to reduce that part of the pelvis equivalent to the central portion of the pubo-ischiadic plate of primitive reptiles. The loss of the ventral border of the obturator foramen in Anchisaurus (Fig. 10G, H, K) and most therapods probably represents the loss of that portion of the m. pubo-ischio-femoralis externus 2 which originated ventral to the acetabulum in most prosauropods and in all sauropods (see Romer 1923: fig. 2, Camarasaurus). Ischium Both ischia (Fig. 3), with a preserved length of 136 mm, are present, but are incomplete posteriorly. In each the widened proximal portion is curved outwards and carries two embayments, one above a ventral hook-shaped keel and the other, less clearly defined, lying more dorsally. The proximal portions of the ischia are separated by an ovoid space, but the shafts are closely appressed and form a dorsally open trough. HIND LIMB Femur The femur (Fig. 4A), which is seen in dorsal (anterior) view, is broad and fairly powerful but, with a preserved length of 194 mm, it is incomplete proximally and distally. The proximal end, as shown by what is still preserved, was inclined fairly sharply inwards. Below the proximal end the femur is strongly built and raised to a smooth crest, which runs from proximo-laterally to disto-medially where it merges into the flat distal end. No condyles are preserved, and only an indication of the base of the fourth trochanter can be made out, high in the upper half of the bone. Fibula The right fibula (Fig. 4B) lies in its natural position between the femur and the pes and is seen in lateral (external) view. The proximal half is stoutly built and leads to a crest formed in the middle of the bone, directed sideways and forwards. The distal part of the bone is slender and tapers off to the articu- lation with the pes. Both proximal and distal ends are incomplete and the preserved length of the bone is 176 mm. Pes The pes (Figs 4C, 6, 11K) is seen in ventral view, the bones being either eroded or indicated by bone impressions. The calcaneum is incomplete laterally, and probably extended out slightly farther than shown. The astragalus is transversely elongated, with a rounded anterior surface curving back to the 130 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 4. Anchisaurus capensis (Broom), SAM-990. A. Right femur in dorsal (anterior) view. B. Right fibula in lateral view. C. Right pes in ventral view. 0,5. narrow lateral corner of the bone. Both calcaneum and astragalus are pre- served as impressions, both approximately 15 mm long. Two small distal tarsal elements are preserved, probably nos 3 and 4. No. 3 is no more than a bony nodule, while 4 is more robust with posterior and medial surfaces at right angles to each other and a convex anterior surface facing metatarsals IV and V. The considerable space between the astragalus and the proximal ends of metatarsals I and II suggests that the two distal tarsals are laterally displaced. All five metatarsals are preserved, in several cases as somewhat imperfect impressions. There is the usual overlapping of the proximal ends with each metatarsal slightly underlying its medial fellow. Metatarsal V is short and narrow but provided with a strong and wide base. Metatarsal IV is about twice as long and of more or less equal width over its entire length. Meta- ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 131 tarsal III is the longest and most robust of the series, while II is slightly shorter than IV. The distal ends of metatarsals III and IV are squarely truncated but metatarsal II terminates in an oblique surface, so that the medial inclination of the row of phalanges seems to be a natural one. This could be true, too, of the short metatarsal I, although this element is incomplete distally. A small fragment of bone in front of the fifth metatarsal probably repre- sents a vestigial phalanx. The form and degree of preservation of the phalanges are apparent from Figures 4C and 6, and from Table 2. TABLE 2. Anchisaurus capensis (Broom), SAM—990. Dimensions of pes (in mm). Metatarsal Phalanx length length (proximal to distal) Metatarsal I 44 Digit I 30 34 Il 76 II 30 22 31 Ul 86 Il 30 17 14 30 IV 75 IV 24 16 12 12 21 Vv 48 V = SLENDER- AND BROAD-FOOTED PROSAUROPODS When the feet of anchisaurids and plateosaurids are drawn so that digit II of the manus (Fig. 7) or digit III of the pes (Fig. 8) are reduced to unit length then two groups are distinguishable, those with slender feet (Figs 7A, C—D, I, 8D-G, 81) and those with broad feet (Figs 7B, E-H, J-Q, 8A—C, H, J-S). ™ ‘ = Wy - 2. Sy, a Fig. 5. Anchisaurus capensis (Broom), SAM-990. Stereophotograph of right ilium. 132 ANNALS OF THE SOUTH AFRICAN MUSEUM The difference is clearest for the manus (digits II to IV) and, where both fore- and hind feet are known, the manus and pes are both slender (Figs 7C, E, I, 8D) or broad (Figs 7B; 8H); 7E, 8K; 7G, 88; 75, 8N; 7K, 80; 7Q78@)s ihe difference between slender and broad feet is not growth related since there are small prosauropods with broad feet (Figs 7B, 8A—C, H). These appear to be juveniles of species that grew much larger (Figs 7B, H, K, 8H, O-P—all Ammosaurus major, see Galton 1971) and all the larger prosauropods are broad-footed. The prosauropod families Plateosauridae (Figs 7-8) and Melano- rosauridae (see Bonaparte 1972a: figs 62, 70; Raath 1972: figs 9f, 10a—b) are all broad-footed, whereas the Anchisauridae as currently classified include both slender and broad-footed forms (Figs 7-8). ‘Gyposaurus’ capensis (Figs 3C, 5, 7G, 10K) and Thecodontosaurus antiquus (Figs 8F, 11G) are both slender- footed species but broad-footed species have been incorrectly referred to both genera. SYSTEMATIC DISCUSSION Order SAURISCHIA Suborder SAUROPODOMORPHA Infra-order PROSAUROPODA Family Anchisauridae Marsh, 1885 Diagnosis Smaller forms, skull lightly built, shallow posterior half of lower jaw with articulation in line with tooth row, manus and pes slender. Genus Anchisaurus Marsh, 1885 (includes Megadactylus Hitchcock, Amphisaurus Marsh, Gyposaurus Broom) Diagnosis Centra of dorsal vertebrae low, broad bases to neural spines of anterior caudal vertebrae, ilium with long anterior process, pubis with open obturator foramen and a relatively narrow distal part that is not apron-like. The characters of the skull, neck and manus listed below (p. 133) for Anchisaurus polyzelus may also be diagnostic of the genus, but these regions are not known in A. capensis. Anchisaurus polyzelus (Hitchcock, 1865) Megadactylus polyzelus Hitchcock, 1865: 40, pl. 9 (fig. 6). Cope, 1870: 122A-G, pl. 13 (preocc.). Amphisaurus polyzelus Marsh, 1882: 84 (preocc.). Anchisaurus polyzelus Marsh, 1885: 169; 1892: pl. 16 (fig. 3) pl. 17 (fig. 6); 1896: 147, pl. 3 (figs 4-5). Lull, 1915: 119, figs 14-17; 1953: 99, figs 12-14a. Galton, 1971: 782, fig. 7C; IWS sity AVES VEG Ibe Oy 12s Ay Anchisaurus colurus Marsh, 1891: 267; 1892: 543, pl. 15, 16 (figs 1, 2); 1893: 169, pl. 6; 1896: 148, pl. 2 (figs 1-3), pl. 3 (figs 1, 2), pl. 4. Huene, 1906: 6, figs 1-6, pls 1-3; 1914b: 69, figs 1-11; Lull, 1912: 414, figs 2-3; 1915: 130, figs 18-21, pls 4, 10. Thecodontosaurus polyzelus Huene, 1906: 19, figs 10, 10a; 19145: 75, figs 23-24; 1932: 116. Yaleosaurus colurus Huene, 1932: 119, pl. 14 (fig. 1), pl. 54 (fig. 3). Lull, 1953: 107, figs 15-18, pl. 4. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 133 Types Incomplete skeleton (AM 41/109) from Longmeadow Sandstone, upper part of Newark Series (Upper Triassic) of Springfield, Massachusetts, U.S.A. (A. polyzelus). Almost complete skeleton (YPM 1883) from Portland beds, upper part of Newark Series near Manchester, Conn., U.S.A. (A. colurus). Diagnosis + 9 maxillary teeth, 16 dentary teeth, basipterygoid processes very small, cervical vertebrae elongate, metacarpal I broad, digits If and III of manus subequal in length, ungual I of pes smaller than ungual II. Discussion Megadactylus Hitchcock being preoccupied, Marsh (1882) replaced it with Amphisaurus (also preoccupied) and then Anchisaurus (Family Anchisauridae also proposed). Huene (1906) referred the material of Megadactylus polyzelus Hitchcock to Thecodontosaurus as T. polyzelus (Hitchcock) and used the Family Thecodontosauridae (originally proposed by Lydekker, 1890: 246) to replace Anchisauridae. Marsh (1891) made YPM 1883 the holotype of a new species of Anchisaurus, A. colurus, but he did not indicate how it differed from A. poly- zelus (AM 41/109). Huene (1906) suggested that Anchisaurus polyzelus resembled Thecodontosaurus and differed from Anchisaurus colurus in several features, discussed here together with others noted later by Huene (1907-08, 1932) (for full discussion see Galton, in press): 1. Shortness of cervical vertebrae. Huene (1932) noted that AM 41/109 resembled Thecodontosaurus (Fig. 11A) in the shortness of the anterior cervical vertebrae, which are elongate in YPM 1883 (Figs 9H, 11B). How- ever, this comparison was based on misidentification of part of the neural arch of a dorsal vertebra (Galton in press, fig. 3a—c) as a cervical vertebra (Huene 19144: fig. 23a). 2. Shortness of dorsal vertebrae. Huene (1906) originally noted that AM 41/109 differed from Thecodontosaurus and resembled YPM 1883 in having very elongate dorsal vertebrae. Later Huene (1914) figured an extremely short centrum of AM 41/109 as that of a dorsal vertebra. Although not stated, this implied that the dorsal vertebrae of AM 41/109 are extremely short as in Thecodontosaurus. The isolated centrum figured by Huene (19146) could not be located but judging from the proportions it was probably part of an anterior caudal vertebra. The proportions of an isolated neural arch (Galton in press, fig. 3a—c) and of a centrum (Galton in press, fig. 5c) show that the dorsal vertebrae of AM 41/109 were probably elongate (i.e. the centra were low) as in YPM 1883 (Fig. 9K). 3. Slenderness of neural spines of anterior caudal vertebrae. This comparison by Huene (1906) was based on misleading figures given by Cope (1870) and Marsh (1893, 1895, 1896). Cope (1870: pl. 8 (fig. 7): see Lull 1953: 134 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 6. Anchisaurus capensis (Broom), SAM-990. Stereophotograph of right pes in ventral view. fig. 12a) did not indicate that the neural spines of the anterior caudal vertebrae of AM 41/109 were damaged and incomplete (Fig. 9P); originally the neural spines were broader and not as narrow as in the vertebrae referred to Thecodontosaurus (Fig. 9S) by Huene (1907-08: pl. 77 (fig. 4); 19146: fig. 40). Most of the supposed differences between AM 41/109 and YPM 1883 were either the result of misinterpretation (1-3 above, different position of fourth trochanter of femur) or the result of differences in preservation (form of radius, metacarpals, tibia, fibula: see Galton in press). YPM 1883 does differ from AM 41/109 in having a proportionally long centrum to the last dorsal vertebra (but this might be sacral vertebra 3) and a proportionally smaller manus with a less trenchant first ungual phalanx (Fig. 7A, C). However, these differences probably represent individual variations within a species because individuals of the ornithopod dinosaur Hypsilophodon foxii show a much wider range of morphological variation (see Galton 1974). YPM 1883 should be referred to Anchisaurus polyzelus because, on the basis of available material, it cannot be distinguished from AM 41/109 by any characters of taxonomic significance and, in addition, AM 41/109 does not show any unique resemblance to Theco- dontosaurus. Consequently, Anchisaurus colurus Marsh is a junior synonym of Anchisaurus polyzelus (Hitchcock) and, as A. colurus is the type species of the 135 ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE ‘(I161) woorg wolf ‘snunr.yfo xduodday “CE “79S HNN ‘Sl4isnjod xduojay “q “(116[) Woolg woul ‘szysnjod xduojapy ‘CG *(80-LO6T) sUeN_, Aq sours snanvsojp1a] Se poinsy “ds sninvsoajvy]d ‘N (TES) Suan Aq xXv{v snanvsdyavg se pansy “ds sninvsoavj_ "W *(TL6]) UOIeED AdIJe ‘sisuarsiag -Uajia SNANDSOIID] "T (Lp61) 8UNOX Aq snusvu “T se poinsy “lauany snanvsosuafnT “y *(L16]) Woo1g Woy ‘Isar.ivy snjdpuodsossvyy *f* L997I SNWS “2ousousvlp vispo.] “1 (Ty61) SUNOX Aq Jsuony snanvsounuunx se poinsy ‘iauany snanvsosuafny “HY *(GOT6T) UadeoH UeA Aq IuMo4g ‘WW se poinsy “saiioy snjdpuodsosspopy “A “(TL6T) UOYeD “sof “Jo snanvsounup “FZ “SElT WdA ‘Sunbyuv snanvsojuoposay °C *€88l Wd ‘snjazdjod snanvsiyoup *D “(Tp6t) BunoX Aq sisuauis snanvsod&yH se pansy “Jauany snanvsosuafnT *| “6OI/Ib WV snjazdjod snanpsiyoup “WW ‘(N-M ‘H) oeplinesoaje[g pue (O-O ‘f “4 ‘C-V) oepunesiyouy sarruey oy} UsaMj0q (9961) JOWOY Aq PeplAIp ole vIOUDs dSOy IT, “WD ¢ = 9[eOs “YISUS] WUN J] USIP OF UMBIP []e “(|) MOIA [LIYUIA UT JO] IO MIA [VSIOP UT IYSII Joyye ‘spodoinesoid Jo snuewl oy} Jo uostedwo|D */ “S14 136 ANNALS OF THE SOUTH AFRICAN MUSEUM genus Yaleosaurus (Huene 1932: 122), Yaleosaurus is a junior synonym of Anchisaurus Marsh, 1885. Anchisaurus capensis (Broom, 1911) Hortalotarsus skirtopodus (non Seeley, 1894) Broom, 1906: 201, pl. 3. Gyposaurus capensis Broom, 1911: 293. Type Partial skeleton (SAM-990) from the Cave Sandstone, Stormberg Series (Upper Triassic) of Ladybrand, Orange Free State, South Africa. Diagnosis Ungual | largest on pes. Discussion Because of its nature, SAM-990 can be distinguished from Anchisaurus polyzelus (YPM 1883) oniy by the relative size of ungual | of the pes—large in SAM-990 and small in YPM 1883. Anchisaurus capensis (Broom) is dis- tinguishable from all prosauropods other than A. polyzelus (Hitchcock) by the following combination of characters: 1. Centra of posterior dorsal vertebrae (about the tenth) are proportionally le low so that the ratio of central length to height CG) is 2,1 (Figs 2; 916): 2. Anterior process of ilium is elongate (Figs 3, 5, 10B). . Subacetabular part of pubis is emarginated ventrally so that the obturator foramen is open (Figs 3, 10H). 4. The pes is slender (Figs 4C, 6C, 11K). 5. Broad bases to neural spines of anterior caudal vertebrae (Figs 2, 9Q). oS) The form of these elements in the species of Anchisauridae is summarized in Table 3 to facilitate comparisons. The systematic position of other species incorrectly referred to ‘Gyposaurus’ are discussed below (pp. 141, 143, 147). TABLE 3. Comparison of species attributed to ‘Gyposaurus’ with other anchisaurids. Dorsal Neural Proximal Centra: spine of Anterior pubis| length| anterior process of obturator height caudal ilium foramen Pes ‘“Gyposaurus’ capensis - : Pal wide long shallow/open slender ‘Gyposaurus’ erectus 1,4 wide short deep/closed broad “Gyposaurus’ sinensis 1,1 — short deep/closed broad Anchisaurus polyzelus : Dl wide long shallow/open slender Thecodontosaurus antiquus ed narrow short = slender Efraasia diagnostica 25) narrow short deep/closed slender ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 137 Genus Efraasia Galton, 1973 Diagnosis Basipterygoid processes of medium length, cervical vertebrae elongate, centra of dorsal vertebrae low, narrow bases to neural spines of anterior caudal vertebrae, slender metacarpal I, digit I! of the manus robust and appreciably longer than digit III, ilium with short triangular anterior process, pubis with closed obturator foramen with an apron-like distal part, ungual | largest in pes. Efraasia diagnostica (Huene) Thecodontosaurus diagnosticus Fraas, 1913: 1098 (nomen nudum). Palaeosaurus (?) diagnosticus Huene, 1932: 52, 73, figs 1, 2, 7-8, pls 4-6. Palaeosauriscus diagnosticus Charig, 1967: 712. Efraasia diagnostica Galton, 1973: 247, figs I[A-E, 2-15, 16A, 17C-D. Syntypes An almost complete skeleton (SMNS 12667) (Berckhemer 1938) together with additional material (SMNS 12668) from the Stubensandstein (Upper Triassic) of Pfaffenhofen, Wiirttemberg, West Germany (see Galton 1973). Diagnosis As for genus. Discussion As shown by the manus (Fig. 71) and the pes (Figs 8D, 11M), Efraasia diagnostica undoubtedly represents a slender-footed prosauropod. Apart from that of Anchisaurus polyzelus (YPM 1883, Fig. 12B; Huene 1906: pl. 1), this is the only reasonably complete and well-preserved skeleton (Fig. 12A; Berck- hemer 1938) of a slender-footed prosauropod described to date. Efraasia diagnostica (Figs 71, 8D, 91, M, 10D, J, N, 11M) resembles Anchisaurus capensis in several features (Table 3) but differs in three important respects, viz. the ilium has a short triangular anterior process (Fig. 10D), the subacetabular part of the pubis is deep with a complete obturator foramen (Fig. 10J, N), and the bases of the neural spines of the anterior caudal vertebrae are narrow (Fig. OR). Genus Thecodontosaurus Riley & Stutchbury, 1836 Diagnosis At least 21 dentary teeth (in holotype, Fig. 9B); from referred specimens without teeth diagnosis tentatively expanded as follows: elongate basiptery- goid processes, cervical vertebrae proportionally short compared with other anchisaurids, high centra to dorsal vertebrae, narrow base to neural spines of anterior and caudal vertebrae, high placed deltopectoral crest on proximal third of humerus, manus with slender metacarpal I and digits II and III subequal in length, short triangular anterior process to ilium. 2 SCRE rr le! S : =e : Z SS Z = . See = = = em = OES pa a ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 139 Thecodontosaurus antiquus Morris Thecodontosaurus Riley & Stutchbury, 1836: 398; 1840: 352, pl. 29 (figs 1-2). Thecodontosaurus antiquus Morris, 1843: 211. Type Incomplete dentary with teeth from the Magnesian Conglomerate (Upper Triassic) near Bristol, England. Diagnosis As for genus. Discussion Riley & Stutchbury (1836, 1840) did not give a specific name for Theco- dontosaurus and this oversight was rectified by the proposal of T. antiquus Morris, 1843. Thecodontosaurus was the first genus of prosauropod to be described so it is unfortunate that there is no articulated association between teeth of the type and the postcranial material referred to the genus by Seeley (1895a) and Huene (1907-8, 19145). Indeed, the only articulated bones referred to Thecodontosaurus are a few short sequences of vertebrae (Huene 1907-8: figs 214, 218-220) and a fore limb with scapula, cervical vertebra and dorsal ribs (Fig. 11A—G). The description and skeletal reconstruction of Thecodonto- saurus antiquus given by Huene (1932: 116, pl. 54 (fig. 1)) are based on many specimens. However, the postcranial remains from Bristol indicate the presence of a slender-footed prosauropod (Fig. 11A—G) and it is reasonable to refer this material to Thecodontosaurus antiquus. Species of Thecodontosaurus from other parts of the world (see next section and pp. 145, 147, 152, 153) are incor- rectly referred to this genus. Fig. 8. Comparison of the pes in various prosauropods, either right in dorsal view or left in ventral view (A, C, E, G, I,), all drawn to digit III unit length, scale lines represent 5 cm. These genera are divided by Romer (1966) between the families Anchisauridae (B, C, E-I, L, N, Q) and Plateosauridae (M, O, P, R, S). A. Ammosaurus major, YPM 209. B. Aristosaurus erectus, from Van Hoepen (1920a). C. Thecodontosaurus browni, from Huene (1932). D. Efraasia diagnostica, SMNS 12668. E. Anchisaurus polyzelus, YPM 1883. F. Thecodontosaurus antiquus, metatarsal II], from Huene (1907-08). G. Anchisaurus capensis, SAM-990, H. Lufengosaurus huenei, figured as Gyposaurus sinensis by Young (1941). I. Hortalotarsus skirtopodus, figured as Thecodontosaurus skirtopodus by Huene (1906). J. Ammosaurus major, YPM 208. K. Ammo- saurus cf. major, from Galton (1971). L. Massospondylus harriesi, figured as M. browni by Yan Hoepen (1920b). M. Plateosaurus gracilis, from Berckhemer (1938). N. Massospondylus harriesi, from Broom (1911). O. Lufengosaurus huenei, figured as Yunnanosaurus magnus by Young (1947). P. Lufengosaurus huenei, figured as Yunnanosaurus robustus by Young (1951). Q. Gryponyx africanus, from Broom (1911). R. Plateosaurus robustus, from Huene (1932). S. Plateosaurus sp., figured as Pachysaurus wetzelianus by Huene (1932). 140 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 9. Comparison of skulls and vertebrae of prosauropods, scale = 2,5 cm, A, B, D-F x 0,5. A. Skull of Anchisaurus polyzelus, YPM 1883, lateral view. B. Thecodontosaurus anti- quus, left partial dentary in lateral view, from Riley & Stutchbury (1840). C. Skull of Plateo- saurus in lateral view, from Romer (1966). D. Massospondylus harriesi, left partial lower jaw in lateral view, from Haughton (1924). E-G. Braincases in ventral view: E. Anchisaurus polyzelus, YPM 1883. F. Thecodontosaurus antiquus, YPM 2192. G. Efraasia diagnostica basisphenoid, SMNS 12667. H—J. Third cervical vertebra in lateral view: H. Anchisaurus polyzelus, YPM 1883. I. Efraasia diagnostica, SMNS 12667. J. Lufengosaurus huenei, figured as Gyposaurus sinensis by Young (1941). K—O. dorsal vertebrae (tenth to twelfth) in lateral view: K. Anchisaurus polyzelus, YPM 1883. L. Anchisaurus capensis, SAM-990. M. Efraasia diagnostica, SMNS 12667. S. Thecodontosaurus antiquus, from Huene (1907-08). ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 14] Anchisauridae nomina dubia Hortalotarsus skirtopodus Seeley, 1894 Hortalotarsus skirtopodus Seeley, 1894: 411, figs 1-3. Thecodontosaurus skirtopodus Huene, 1906: 44, figs 72-78, pls 13, 14. Haughton, 1924: 370. Huene, 1932: 117. Haughton & Brink, 1954: 35. Gyposaurus skirtopodus Charig, 1967: 712. Type Incomplete hind limb in the Albany Museum, from the Cave Sandstone, Stormberg Series, of Barkly East Division, Cape Province, South Africa. Discussion The assignment of Hortalotarsus skirtopodus to Thecodontosaurus by Huene (1906) was based in part on the characters of isolated bones found at localities different from that of the type specimen. The pes of the type (Fig. 81) is obviously that of a slender-footed prosauropod, and Huene (1906) could not distinguish it from the pes of Thecodontosaurus. However, this pes (Fig. 81) is also indistinguishable from those of Anchisaurus capensis (Fig. 11K), Anchi- saurus polyzelus (Fig. 11L) and Efraasia diagnostica (Fig. 11M). This specimen is generically and specifically indeterminate so Hortalotarsus skirtopodus Seeley is a nomen dubium. Thecodontosaurus browni (Seeley, 18956) Massospondylus browni Seeley, 18956: 118, figs 13-14. Thecodontosaurus browni Huene, 1906: 141, pl. 12 (figs 7-8); 1932: 118. Broom, 1911: 293. Haughton, 1924: 370. Type Limb bones (BMNH R3302) from the Red Beds, Stormberg Series of Telle River, Herschel, Cape Province, South Africa. Discussion Seeley (18954) noted that the proportions of the phalanges of the pes are very similar to those of Hortalotarsus (Fig. 8H) so this is probably another generically and specifically indeterminate specimen of a_ slender-footed prosauropod. Thecodontosaurus minor Haughton, 1918 Thecodontosaurus minor Haughton, 1918: 468; 1924: 376, fig. 21. Thecodontosaurus browni: Huene, 1932: 118. Type Left tibia, a cervical vertebra and a portion of a left ilium (SAM-3451) from the Red Beds, Stormberg Series, from road-cutting at Naude’s Nek, Pitsing, Maclear District, Cape Province, South Africa. Discussion The tibia is slender so this may be a slender-footed prosauropod but, on 142 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 10. Comparison of ilium and pubis of prosauropods. Scale = 2,5 cm. A—F. Lateral view of left ilium: A. Anchisaurus polyzelus, YPM 1883. B. Anchisaurus capensis, SAM-990. C. Lufengosaurus huenei, figured as Gyposaurus sinensis by Young (1941). D. Efraasia diagnostica, SMNS 12667. E. Aristosaurus erectus, fromVan Hoepen (1920a). F. Thecodontosaurus antiquus, from Huene (1907-08). G—J. lateral view of left pubis: G. as A. H as B. Las C. J as D. K-N. Ventral view of right pubis (L, M) or dorsal view of left pubis (K, N): K as A, G. Las C, I. M as E. N as D, J. the basis of available material, this specimen is a generically and specifically indeterminate prosauropod and Thecodontosaurus minor is a nomen dubium. Thecodontosaurus macgilivrayi (Seeley) Agrosaurus macgilivrayi Seeley, 1891: 161, figs 1-6. Thecodontosaurus macgilivrayi Huene, 1906: 147, figs 86-90; 1932: 52. Type Tibiae, radius, an ungual and tooth (BMNH 49984) from York Peninsula, Queensland, Australia. Discussion Huene (1932: 52) subsequently referred Agrosaurus macgilivrayi to the Theropoda (Coelurosauria) as do Romer (1956, 1966), Steel (1970) and White (1973). However, the latero-distal surface of the tibia is notched (Fig. 11H—J) (Huene 1906: fig. 86a, d-e) to receive the central ascending process of the astragalus in typical prosauropod fashion. This material undoubtedly represents ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 143 a prosauropod but is generically and specifically indeterminate, so Agrosaurus macgilivrayi is a nomen dubium; it may represent a slender-footed prosauropod. PROSAUROPOD SPECIES INCORRECTLY ASSIGNED TO ANCHISAURIDAE Family Plateosauridae Marsh, 1895 Diagnosis Larger forms, skull massively built, deep posterior half to lower jaw with articulation offset ventral to line of tooth row, manus and pes broad. Ammosaurus major Marsh, 1889 Anchisaurus major Marsh, 1889: 331, fig. 1. Ammosaurus major Marsh, 1891: 267; 1892: 545, pl. 16 (fig. 4), pl. 17 (fig. 3); 1896: 150, pl. 3 (figs 3, 6). Huene, 1906: 15, pls 5-9; 1907-08: 303-04, figs 297-298; 1914a: 13; 19145: 74, figs 20-22; 1932: 26. Lull, 1915: 148, figs 24-25; 1953: 123, figs 19-20. Galton, 1971: 786, figs 9, 11A. Anchisaurus solus Marsh, 1892: 545; 1896: 149. Huene, 1914b: 72, figs 12-19. Lull, 1915: 144, figs 22-23; 1953: 120. Anchisaurus (?) solus Huene, 1906: 14, pl. 4. Ammosaurus solus Huene, 1932: 27, pl. 49 (fig. 1). Type Pelvis and hind limbs (YPM 208) from the Portland Beds, upper part of Newark Series near Manchester, Connecticut, U.S.A. (Ammosaurus major). Almost complete skeleton (YPM 209) from the same locality and horizon (Anchisaurus solus). Discussion Ammosaurus has long been regarded as a primitive theropod dinosaur but it is considered as an anchisaurid by Steel (1970) and by Galton (1971), who provides a detailed discussion of the taxonomic position of this genus (Galton in press). The pes of the holotype (Fig. 8J) and of the referred specimens (Fig. 8A, K) plus a referred manus (Fig. 7E) are of the broad type. Ammo- saurus is a broad-footed prosauropod characterized by the following combina- tion of characters: centra of dorsal vertebrae low, slender sacral rib 3, elongate anterior process to ilium, subacetabular part of the ischium emarginated ventrally (Galton, in press). Aristosaurus erectus van Hoepen, 1920a Aristosaurus erectus yan Hoepen, 1920a: 82, figs 1-6, pls 9-10. Haughton 1924: 379, Haughton & Brink, 1954: 33. Gyposaurus capensis Huene, 1932: 123, pl. 54 (fig. 2). Gyposaurus erectus Charig, 1967: 712. Steel, 1970: 49. Type An almost complete skeleton as slab and counterpart (TM 130) from the Cave Sandstone, Stormberg Series, near Roosendal, Senekal District, Orange Free State, South Africa. 144 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 11. Thecodontosaurus, Agrosaurus and comparisons of the anchisaurid pes. Scale = 5 cm, A-J + x 0,45. A-G. Thecodontosaurus antiquus, YPM 2195, from Durdham Down, Bristol, England. Matrix indicated by stipple, broken bone by diagonal shading: A. Right side of anterior cervical vertebra. B. Right scapula in lateral view. C. Left humerus in medial view. D. As C, in anterior view. E. Left ulna in proximal, lateral and distal views. F. Left radius in lateral view. G. Left manus in lateral or dorsal view. H-J. Agrosaurus macgillivrayi, distal end of left tibia in lateral view (H), anterior view (1), and distal view (J), all from Huene (1906). K-—M. Anchisaurid pes, drawn to digit III unit length: K. Anchisaurus capensis, SAM-990, compare with Fig. 5. L. Anchisaurus polyzelus, YPM 1883. M. Efraasia diagnostica, SMNS 12668. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 145 Discussion Romer (1956, 1966), Charig (1967), Steel (1970) and White (1973) follow Huene (1932, 1956) in regarding Aristosaurus as a junior synonym for Gypo- saurus but this is unlikely because, in contrast to the situation in Gyposaurus capensis (Table 3): 1. In photographs of the skeleton of Aristosaurus erectus in Van Hoepen (1920a: pls 9-10) and in the reconstruction (Fig. 12D) given by Huene (1932) the dorsal vertebrae are proportionally higher with a central length to height ratio of about 1,4 for dorsal 10. 2. The anterior process of the ilium is short (Fig. 10E). 3. The obturator foramen of the pubis is closed ventrally (Fig 10M). 4. The hind feet of Aristosaurus erectus appear to be of the broad type. There is no reason why Aristosaurus erectus should be referred to the genus Anchisaurus. Aristosaurus erectus appears to be a valid taxon of broad- footed prosauropod, but further preparation and illustration of the holotype is needed. Genus Massospondylus Owen, 1854 (includes Leptospondylus Owen, Pachyspondylus Owen, Aetonyx Broom, Gryponyx Broom, Dromicosaurus Van Hoepen) Massospondylus carinatus Owen, 1854 Massospondylus carinatus Owen, 1854: 97. Seeley, 1895b: 102, figs 1-12. Huene, 1906: 36, figs 43-70, pls 13-16. Broom, 1911: 241. Haughton, 1924: 383. Huene, 1932: 124. Leptospondylus capensis Owen, 1854: 97. Pachyspondylus orpenii Owen, 1854: 97. Type Isolated bones from the Red Beds, Stormberg Series of Beaucherf, Harri- smith, Orange Free State, South Africa. The holotype in the Museum of the Royal College of Surgeons in London was destroyed during World War II but casts of this material are in the National Museum of Southern Rhodesia, Bulawayo (J. Attridge, pers. comm.). Massospondylus harriesi Broom, 1911 Massospondylus harriesi Broom, 1911: 299, pls 15-17. Haughton, 1924: 384, figs 21-29. Huene, 1932: 125. Massospondylus browni (non Seeley, 1895h): Van Hoepen 19204: 118, pls 17-22. Aetonyx palustris Broom, 1911: 304, figs 20-23. Haughton, 1924: 404, fig. 30. Huene, 1932: 91. Gryponyx africanus Broom, 1911: 294, figs 1-9. Haughton, 1924: 417, figs 36-38. Huene, 1932: 88, pl. 7 (figs 1-4). Gryponyx taylori Haughton, 1924: 420, fig. 39. Huene, 1932: 90. Dromicosaurus gracilis Van Hoepen, 1920b: 103, figs 8-21, pls 13-16. Haughton, 1924: 405. Thecodontosaurus dubius Haughton, 1924: 377. 146 ANNALS OF THE SOUTH AFRICAN MUSEUM Types Massospondylus harriesi: Bones of the fore limb (SAM-3394) from the top of the Red Beds, Stormberg Series of Foutanie, Fouriesburg, Orange Free State, South Africa. Aetonyx palustris: Partial skeleton (SAM-2768, 2769, 2770) from the top of the Red Beds (but listed as from Cave Sandstone by C. E. Gow in Anderson & Anderson 1970), Stormberg Series from Foutanie, Fouriesburg, Orange Free State, South Africa. Gryponyx africanus: Pelvis and hind limb, right and left manus, vertebrae (SAM-3357-9) from the top of the Red Beds (but listed as from the Cave Sandstone by C. E. Gow in Anderson & Anderson 1970), Stormberg Series of Foutanie, Fouriesburg, Orange Free State, South Africa. Gryponyx taylori: Pelvic girdle and sacral vertebrae (SAM-3453) from the top of the Red Beds (but listed as from Cave Sandstone by C. E. Gow in Ander- son & Anderson 1970), Stormberg Series of Fouriesburg, Orange Free State, South Africa. Dromicosaurus gracilis: Partial skeleton (TM 123) from Red Beds of Naaupoort Nek, Bethlehem, Orange Free State, South Africa. Thecodontosaurus dubius: Larger portion of a skeleton (SAM-3712) from the Cave Sandstone, Stormberg Series of Ladybrand, Orange Free State, South Africa. Discussion On the basis of the phalanges Seeley (1895b) stated that Massospondylus carinatus had a broad hind foot and Huene (1906) separated this genus from Plateosaurus mainly because of its Thecodontosaurus-like tibia. Massospondylus was the first genus of broad-footed prosauropod to be described from South Africa. The manus (Fig. 7F—-G, J) and the pes (Fig. 8L, N) of Massospondylus harriesi (SAM-3394) are obviously of the broad type. The material (Figs 7G, 8L, 12E) described by Van Hoepen (19204) as Massospondylus browni should be referred to this species (Haughton 1924). The manus of Aetonyx palustris Broom (Fig. 7O) is of the broad type and the pes is similar, as indicated by the measurements given by Huene (1932: 92). J. Attridge (pers. comm.) regards Aetonyx palustris as a junior synonym for Massospondylus harriesi. The manus (Fig. 7Q) and pes (Fig. 8Q) of Gryponyx africanus are of the broad type and Gryponyx africanus is probably a junior synonym for Massospondylus harriesi. Gryponyx taylori Haughton is a nomen dubium because the material is generically and specifically indeterminate; it probably represents another specimen of Massospondylus harriesi. The manus and pes of Dromicosaurus gracilis Van Hoepen are not pre- served but were probably of the broad type because Dromicosaurus was regarded as being closely allied to Aetonyx and Massospondylus by Van Hoepen (19205), Haughton (1924) and Huene (1932). J. Attridge (pers. comm.) regards Dromico- ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 147 saurus gracilis as a junior synonym for Massospondylus harriesi. The type specimen of Thecodontosaurus dubius Haughton has never been figured, but Huene (1932: 92) referred it to Aetonyx palustris so it 1s presumably a broad- footed form; J. Attridge (pers. comm.) refers this specimen to Massospondylus harriesi. Stratigraphically and geographically, Massospondylus harriesi is the most ubiquitous prosauropod in southern Africa. Plateosaurus gracilis (Huene, 1907) Thecodontosaurus (?) hermannianus Huene, 1907-08: 216, fig. 236, pl. 144 (fig. 1). Plateosaurus gracilis Huene, 1932: 303. Type Right maxilla with teeth from the Stubensandstein (Upper Triassic) of Heslach, in Stuttgart, West Germany. Lufengosaurus huenei Young, 1941a Lufengosaurus hueni Young, 1941a: 1, figs 1-25, pls 1-6; 1947: 41; 1951: 50, fig. 11, pl. 12. Rozhdestvensky, 1965. Gyposaurus sinensis Young, 1941b: 205, pls 1-9; 1948: 91, pls 1-5; 1951: 49. Yunnanosaurus huangi Young, 1942: 64, figs 1-17; 1951: 56. Lufengosaurus magnus Young, 1947: 2, figs 1-14. Yunnanosaurus robustus Young, 1951: 58, figs 12-14, pls 7-10. Types Several incomplete skeletons from the lower Lufeng Series (Upper Triassic, Rhaetic) of Lufeng, Yunnan, China. Discussion Rozhdestvensky (1966) restudied 70 specimens from the Lufeng Series and decided that Lufengosaurus huenei, L. magnus, Yunnanosaurus huangi, Y. robustus and Gyposaurus sinensis of Young were all differentiated only on size-related characters and are conspecific (as Lufengosaurus huenei Young, 1941a). Rozhdestvensky (1966) noted that, judging from the original diagnosis of Young (19414), Gyposaurus sinensis is hardly distinguishable from G. capensis, but that without visual comparisons or more detailed descriptions it is impos- sible to decide the relationship between these two species. Galton (1973, in press) noted that ‘Gyposaurus’ sinensis was incorrectly referred to the genus Gyposaurus, and may represent a new genus or be based on juveniles of either Lufengosaurus or Yunnanosaurus. ‘Gyposaurus’ sinensis should not be referred to the genus Gyposaurus (or Anchisaurus) because of the following anatomical features (Table 3): 1. Centra of posterior dorsal vertebrae are proportionally high with a central length to height ratio of 1,1 (Fig. 90). 2. Anterior process of ilium is short and triangular (Fig. 10C). 3. Subacetabular part of pubis is broken but originally this region was 148 ANNALS OF THE SOUTH AFRICAN MUSEUM Fig. 12. Skeletal reconstructions of prosauropods. Scale = 20 cm. Tails of A and B diagram- matically folded over. A. Efraasia diagnostica, SMNS 12667, 12668, from Galton (1973). B. Anchisaurus polyzelus, YPM 1883, AM 41/109, from Galton (1973). C. Lufengosaurus huenei, based on figures of individual bones given by Young (19415) as Gyposaurus sinensis. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 149 . eh ‘eg ae Se PC Ree | LE aati ~~ et: WES ace > Fig. 12. (cont.) D. Aristosaurus erectus, modified from Gyposaurus capensis of Huene (1932). E. Massospondylus harriesi, modified from Thecodontosaurus browni of Huene (1932), based on specimen described as Massospondylus browni by Van Hoepen (1920a). deep with a complete obturator foramen (Young 1941b: 222; 1948: 96) (Figs 10I, L). 4. Pes (Fig. 8H) and manus (Fig. 7B) are broad. Young (19414) did not cite the papers of Broom (1906, 1911) so the assign- ment as Gyposaurus sinensis was probably based on the skeletal reconstruction of Aristosaurus erectus given by Huene (1932) as Gyposaurus capensis. However, the skeleton of ‘Gyposaurus’ sinensis (Fig. 12C) differs greatly in several aspects from that of Aristosaurus erectus (Fig. 12D) and this is especially true for the form of the neck vertebrae (Fig. 9J) and fore limb. It should be noted that the skeletal reconstruction of ‘Gyposaurus’ sinensis given by Young (19415, pl. 9) bears practically no resemblance to a reconstruction (Fig. 12C) based on figures of the bones of the same specimen given by Young (19410). The correct- ness of the contention of Rozhdestvensky (1966) that Gyposaurus sinensis 150 ANNALS OF THE SOUTH AFRICAN MUSEUM Young, 1941b is a junior synonym of Lufengosaurus huenei Young, 1941a is shown by the proportionally short neck and proportionally massive manus of the two types. Prosauropoda nomina dubia Gryponyx transvaalensis Broom, 1912 Gryponyx transvaalensis Broom, 1912: 82, figs 3-4. Van Hoepen, 1920b: 102. Haughton, 1924: 420. Huene, 1932: 91, pl. 7, fig. 5. Type Ungual | of the manus and a metatarsal (in the Transvaal Museum) from the Bushveld Sandstone (Cave Sandstone), Stormberg Series of Wiepe 1258, northern Transvaal, South Africa. Discussion This material is probably prosauropod but is generically and specifically indeterminate. NON-PROSAUROPOD SPECIES INCORRECTLY ASSIGNED TO ANCHISAURIDAE Order SAURISCHIA Suborder THEROPODA Arctosaurus osborni Adams, 1875 Arctosaurus osborni Adams, 1875: 177. Lydekker, 1889: 352. Type Isolated cervical vertebra (NMI 62 1971) from Heiberg Formation (Upper Triassic) of north-west extremity of Cameron Island, Bathurst Group, Arctic Archipelago, Canada. Discussion Arctosaurus was described as reptilian by Adams (1875) but subsequently Lydekker (1889) referred it to the family Anchisauridae. Regarded as a turtle by Huene (1906) and White (1973) but referred to the prosauropod family Melanorosauridae (as Plateosauravidae) by Huene (1956) and to the Anchi- sauridae (as Thecodontosauridae) by Romer (1966). The region of the dia- pophysis is slightly damaged (Fig. 13B, F) but from the adjacent curves of the neural arch (Fig. 13E) it is obvious that the diapophysis was very small and, as a result, this vertebra is from the anterior part of the series and is probably either the third or fourth cervical vertebra. It is proportionally very much shorter than the equivalent vertebrae of Anchisaurus (Figs 9H, 12B), Efraasia (Figs 9I, 12A) and Plateosaurus (Huene 1926). The only prosauropods with cervical vertebrae proportionally as short are Thecodontosaurus (Fig 11A) and Lufengosaurus (Figs 9J, 12C) but in both cases the vertebrae are proportionally much lower, the neural spines are not so well developed and there is no pleuro- coel (exaggerated in Arctosaurus because of crushing). Arctosaurus osborni is ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 15] | | | Fig. 13. Arctosaurus osborni Adams, holotype NMI G2 1971, anterior cervical vertebra in: A. Anterior view. B. Left lateral view. C. Posterior view. D. Ventral view. E. Dorsal view. F. Stereo photograph of left side, compare with B. Scale = 2,5 cm. Broken bone indicated by diagonal shading. not a prosauropod and, on the basis of the general form-of the vertebra and the presumed presence of a pleurocoel, this specimen is tentatively regarded as Theropoda incertae sedis as listed by Steel (1970). Ischisaurus cattoi Reig, 1963 Ischisaurus cattoi Reig, 1963: 10, figs 4B, 5. Colbert, 1970: 27. Bonaparte, 1972b: 673, fig. 22. Type Two incomplete skeletons from the Ischigualasto Formation (lower Upper Triassic) of Argentina. Discussion Ischisaurus is listed as an anchisaurid by Steel (1970) and as Saurischia incertae by Bonaparte (1972b: 674), who notes that ‘the suggested affinities SZ ANNALS OF THE SOUTH AFRICAN MUSEUM with the Coelurosauria appears as one of the possible relationships’. Romer (1966) and Colbert (1970) list Zschisaurus as a coelurosaurian theropod. Order THECODONTIA ‘Thecodontosaurus’ gibbidens Cope, 1878 Thecodontosaurus gibbidens Cope, 1878: 177. Huene, 1921: 571, figs 14-15. Type Isolated teeth from the Upper Triassic of Pennsylvania, U.S.A. Discussion The isolated teeth are almost circular rather than oval in cross-section as in prosauropods (Thecodontosaurus antiquus Riley & Stutchbury, 1840; Anchi- saurus, YPM 1883; Plateosaurus. AMNH 6810), so these teeth are provisionally referred to the Ornithischia (Galton, in press). Spondylosoma absconditum Huene, 1935 Spondylosoma absconditum Huene, 1935: 247, pl. 30, figs 1-13. Charig, 1967: 712. Colbert, 1970: 19. Bonaparte, 19726: 674. Type Scapula, humerus, femur, tibia (all incomplete) and eight vertebrae from the Santa Maria Formation (Upper Triassic) of Brazil. Discussion Huene (1935, 1942) regarded Spondylosoma as a saurischian but did not make a more specific assignment for this genus. Romer (1956, 1966) referred Spondylosoma to the Anchisauridae (as Thecodontosauridae) as did Colbert (1970) and Charig (1967), who noted at the same time the possibility of its being a prestosuchid pseudosuchian. Bonaparte (19725: 674) notes that ‘there are doubts regarding its assignment to the Saurischia, or even to Prosauropoda. Unfortunately there are not sufficient diagnostic pieces to define better its taxonomic position’ and Spondylosoma is listed as Saurischia incertae. Teleocrater alphos (Haughton, 1932) Thecodontosaurus (?) alphos Haughton, 1932: 662, fig. 19. Teleocrater alphos Charig, 1967: 712. Type Two cervical vertebrae (SAM-10654) from the Manda Formation (Upper Triassic, Anisian) of Tanzania. Discussion Charig refers this material to the pseudosuchian Te/leocrater. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 153 Subclass LEPIDOSAURIA Order LACERTILIA Huene (1932, not 1931 as cited by Steel 1970) considered that the follow- ing Middle Triassic (Muschelkalk) species were based on specimens which should be referred to the eosuchian Tanystropheus, a lacertilian according to Wild (1974); the non-prosauropod nature of these species is also discussed by Colbert (1970): Tanystropheus primus (Huene, 1907-08) Thecodontosaurus primus Huene, 1907-08: pl. 42 (figs 8-9). Tanystropheus primus Huene, 1932: 6. Tanystropheus latespinatus (Huene, 1907-08) Thecodontosaurus (?) latespinatus Huene, 1907-08: figs 237-245. Tanystropheus latespinatus Huene, 1932: 6. NOTES ON THE FAMILIES ANCHISAURIDAE AND PLATEOSAURIDAE The infra-order Prosauropoda is currently divided into three families: Anchisauridae (= Thecodontosauridae), Plateosauridae and Melanorosauridae (see Romer 1956; Colbert 1964; Charig et al. 1965; Bonaparte 1972a). Post- cranially the separation is clearest between melanorosaurids and non- melanorosaurids (Romer 1956: 617; Bonaparte 1972a: 160). Galton (1971, 1973) suggests that the range of morphological variation is insufficient to warrant the retention of two families of non-melanorosaurid prosauropods. However, the skulls of Anchisaurus (Fig. 9A) and Plateosaurus (Fig. 9C) are very different and, because of this, Galton (in press) now considers that they should not be included in the same family. Fortunately the genera concerned are the basis for the first two valid prosauropod family names to be proposed: Anchisauridae Marsh, 1885, and Plateosauridae Marsh, 1895. In only one case (Anchisaurus polyzelus, YPM 1883) is a well-preserved skull found in natural association with a skeleton of a slender-footed pro- sauropod. Consequently the referral of Efraasia and Thecodontosaurus to the Family Anchisauridae is tentative. Contrary to the impression given by Huene (1932: fig. 7; 1956: fig. 10), the skull of Efraasia is very incomplete but, as noted by Galton (1973), Efraasia is an ideal ancestor for the more recent Anchisaurus. The holotype of Thecodontosaurus antiquus 1s an incomplete dentary but, judging from what is preserved (Fig. 9B), the complete lower jaw was probably more like that of Anchisaurus (Fig. 9A) than that of Plateo- saurus (Fig. 9C). Although considered unlikely, the discovery of additional material may show that the restriction of the family Anchisauridae to slender- footed forms is artificial. However, the criterion is practical and with it most taxa and specimens of non-melanorosaurid prosauropods are readily referable to either the Family Anchisauridae or the Family Plateosauridae. Plateosaurus engelhardti Meyer, 1837 from the Keuper (Upper Triassic) 154 ANNALS OF THE SOUTH AFRICAN MUSEUM of Germany is the earliest taxon of plateosaurid to be described, but since then a multitude of taxa have been erected for material from the Triassic of Germany (Steel (1970) lists 9 species as Plateosaurus, 8 species as Gresslyo- saurus). However, this material is probably extremely overclassified and all the taxa of European plateosaurids listed by Steel (1970: 53-56) should be provisionally regarded as junior synonyms of Plateosaurus engelhardti Meyer, 1837. All the European plateosaurid material can probably be referred to (at the most) three species of Plateosaurus, but a restudy of all the holotypes is needed to determine the other valid species. Lufengosaurus (see Young 194la, b, 1942, 1947, 1951) and Plateosaurus (see Huene 1907-8, 1926, 1932) possess the features listed above as charac- teristic of plateosaurids. The skulls of Ammosaurus (see Galton, in press) and Aristosaurus (see Van Hoepen 1920a) are not well enough preserved to tell anything about the form of the skull. However, the holotype of Massospondylus harriesi includes a lower jaw (Fig. 9D), the posterior part of which is deep with the articulation offset ventral to the line of the tooth row. J. Attridge is studying two skulls of Massospondylus harriesi (SAM- K388 and K1314) and has found that both skulls show the features listed above (pers. comm.). Massospondylus is the most ubiquitous prosauropod in southern Africa and its previous classification as an anchisaurid made the prosauropod fauna of Africa unique, because in other areas with abundant, well-preserved pro- sauropod skeletons, plateosaurids are the most common form. However, Cox (1973: 213) notes that ‘it is clear that land connections between all the continents existed for much, at least, of the Triassic’. As regards prosauropods the presence of Anchisaurus in North America and South Africa and the presence of Plateosaurus in Germany and South America (Casamiquela 1964; Bonaparte 19725) indicated that this was the case for the continents on either side of the Atlantic. With the transfer of broad-footed forms, previously listed under the Anchisauridae, to the Plateosauridae, this family becomes the dominant and cosmopolitan prosauropod family of the world. In marked contrast, the Anchi- sauridae have an extremely restricted fossil record (total of about 10 articulated specimens for North America, Europe and South Africa) with no remains discovered to date from Asia (Young 1951; Rozhdestvensky 1966) or South America (Bonaparte, pers. comm.). Haughton (1924) noted that the Stormberg Series of South Africa was deposited under conditions of progressively increasing aridity, and it is interesting that skeletal remains of melanorosaurids occur in the lowermost levels (Passage Beds, Charig et a/. 1965; basal Red Beds, Haughton 1924), most plateosaurid skeletons occur higher in the Red Beds, and those of anchisaurids are found in the overlying Cave Sandstone (Charig et al. 1965; Haughton 1924: Haughton & Brink 1956). Charig et al. (1965) report the presence of small tridactyl footprints in the Passage Beds and in the lower Red Beds and note (p. 204) that these ‘. . . may indicate the movement of thecodontosaurids from one upland region to another via a lowland area’. The world-wide rarity of anchisaurid skeletal remains is presumably because ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 155 these species occupied the drier upland areas, which are rarely represented in the fossil record. SUMMARY The holotype (SAM-990) of the prosauropod dinosaur Anchisaurus capensis (Broom, 1911) is characterized by the following combination of characters: low centra to posterior dorsal vertebrae, broad bases to neural spines of anterior caudal vertebrae, ilium with long anterior process, pubis with open obturator foramen, and with slender pes with first ungual phalanx the largest. SAM-990 differs from the North American Anchisaurus polyzelus (Hitchcock) only in the large size of the first ungual of the pes, and SAM-990 is referred to Anchisaurus Marsh, 1885 as Anchisaurus capensis (Broom). Taxa which have been incorrectly referred to the genus Gyposaurus (= Anchisaurus) are Gyposaurus erectus (Van Hoepen) (= Aristosaurus erectus Van Hoepen), Gyposaurus sinensis Young (= Lufengosaurus huenei Young) and Gyposaurus skirtopodus (Seeley) (= Hortalotarsus skirtopodus Seeley, nomen dubium). The Family Anchisauridae is restricted to those species with slender feet, viz. Anchi- saurus polyzelus, A. capensis, Efraasia diagnostica, and Thecodontosaurus antiquus. Taxa with broad feet previously classified as anchisaurids (Aristo- saurus, Ammosaurus, Gyposaurus sinensis (as Lufengosaurus huenei), Masso- spondylus (including Aetonyx, Dromicosaurus, Gryponyx africanus, G. taylori, Thecodontosaurus dubius), Yunnanosaurus (= Lufengosaurus) are transferred to the family Plateosauridae, the dominant and cosmopolitan family of prosauropods. ACKNOWLEDGEMENTS P. M. Galton is grateful to the following for the loan or use of specimens from their respective institutions, listed in alphabetical order: Drs W. Coombs, Jr and A. E. Wood, Amherst College Museum, Amherst, Massachusetts; Drs E. H. Colbert and E. S. Gaffney, American Museum of Natural History, New York; Dr A. J. Charig, British Museum (Natural History), London; C. E. O’Riordan, National Museum of Ireland, Dublin; Dr K. D. Adam, Staatlichen Museum fiir Naturkunde in Stuttgart; Drs J. H. Ostrom and E. L. Simons, Peabody Museum, Yale University, New Haven, Connecticut. Galton thanks Mr J. Attridge, Birkbeck College, University of London; Dr D. Baird, Princeton University, New Jersey and Dr J. Bonaparte, National University of Tucuman, Argentina for helpful discussions and personal communications as cited above, and Drs J. H. Ostrom and E. L. Simons of Yale University for the continued use of the research facilities of the Division of Vertebrate Paleontology of the Peabody Museum of Natural History. The photographs were taken by Mr N. J. Eden, Department of Palaeontology, South African Museum (Figs 1, 5, 6) and Mr Alan Coleman, Yale Peabody Museum (Fig. 13F). 156 ANNALS OF THE SOUTH AFRICAN MUSEUM Support for this research was provided to the senior author by the Connecticut Research Commission (Grant no. RSA 680) while at Yale University and by a Faculty Research Grant from the University of Bridgeport. REFERENCES Apams, A. L. 1875. On a fossil saurian vertebra, Arctosaurus osborni, from the Arctic regions. —Proc. R. Ir. Acad. 2: 177-179. ANDERSON, H. M. & ANDERSON, J. M. 1970. A preliminary review of the biostratigraphy of uppermost Permian, Triassic and lowermost Jurassic of Gondwanaland.— Palaeont. afr. 13 (Suppl.): 1-22. BERCKHEMER, F. 1938. Wirbeltierfunde aus dem Stubensandstein des Strombergs.— Naturwiss. Monatsh Stuttgart 51: 188-198. BONAPARTE, J. 1972a. Los tetrapodos del sector superior de la Formacion Los Colorados, La Rioja, Argentina. (Triasico Superior) I Parte.—Op lilloana 22: 1-183 BONAPARTE, J. 1972b. Annotated list of the South American Triassic tetrapods. Proceedings, Second Gondwana Symposium. I.U.G.S. Commission on Stratigraphy, South Africa, 1970 665-682. Bonpb, G. 1955. A note on dinosaur remains from the Forest Sandstone (Upper Karroo).— Occ. Pap. Rhod. Mus. 20: 795-800. Broom, R. 1906. On the South African dinosaur (Hortalotarsus).— Trans. S. Afr. phil. Soc. 16: 201-206. Broom, R. 1911. On the dinosaurs of the Stormberg, South Africa.— Ann. S. Afr. Mus. 7: 291-308. Broom, R. 1912. On the remains of the theropodous dinosaur from the Northern Transvaal.— Trans. geol. Soc. S. Afr. 14: 82-83. CASAMIQUELA, R. M. 1964. Sobre el hallazgo de dinosaurios tridsicos en la Provincia de Santa Cruz. Argent. austral. 35: 10-11. CHARIG, A. J. 1967. Archosauria. In: The fossil record. A symposium with documentation: 708-718. London: Geological Society of London. Cuaric, A. J., ATTRIDGE, J. & Crompton, A. W. 1965. On the origin of the sauropods and the classification of the Saurischia.— Proc. Linn. Soc. Lond. 176: 197-221, CouserT, E. H. 1964. Relationships of saurischian dinosaurs.— Am. Mus. Novit. (2181): 1-24. COLBERT, E. H. 1970. A saurischian dinosaur from the Triassic of Brazil.—Am. Mus. Novit. (2205) 51-39: Cope, E. D. 1870. Synopsis of the extinct Batrachia, Reptilia and Aves of North America.— Trans. Am. phil. Soc. 14: 1-252. Copg, E. D. 1878. On some saurians found in thé Triassic of Pennsylvania, by C. M. Wheatley. — Procy Am: phil. Soc. Wis Wide Cox, C. B. 1973. Triassic tetrapods.—Jn: HALLAM, A. ed. Atlas of biogeography: 213-223. London: Elsevier Scientific Publishing Company. FRAAS, E. 1913. Die neusten Dinosaurierfunde der Schwabischen Trias.— Naturwissen- schaften (1) 45: 1097-1100. GALTON, P. M. 1971. The prosauropod dinosaur Ammosaurus, the crocodile Protosuchus, and their bearing on the age of the Navajo Sandstone of northeastern Arizona.—J. Paleont. 45: 781-795. GALTON, P. M. 1973. On the anatomy and relationships of Efraasia diagnostica (Huene) n. gen., a prosauropod dinosaur (Reptilia: Saurischia) from the upper Triassic of Ger- many.—Paldont. Z. 47: 229-255. GALTON, P. M. 1974. The ornithischian dinosaur Hypsilophodon from the Wealden of the Isle of Wight.— Bull. Br. Mus. nat. Hist. (Geol.) 25: 1-125c. GALTON, P. M. (in press). Prosauropod dinosaurs (Reptilia: Saurischia) of North America.— Postilla. GrorFrroy, E. F. S.-H. 1833. Divers mémoires sur les grands sauriens trouvés 4a l’etat fossile vers les confins maritimes de la Basse Normandie, attribués d’abord au crocodile, puis determines sous les noms de Jeleosaurus et Steneosaurus. Mém. Acad. Sci. Inst. Fr. 12: 3-138. ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 157 GiLmorg, C. W. 1920. Osteology of the carnivorous Dinosauria in the United States National Museum, with special reference to the genera Antrodemus (Allosaurus) and Ceratosaurus. — Bull. U.S. natn. Mus. 110: 1-159. HauGuton, S. H. 1918. A new dinosaur from the Stormberg Beds of South Africa.— Ann. Mag. nat. Hist. 2: 468-469. HAuGHTON, S. H. 1924. The fauna and stratigraphy of the Stormberg series.— Ann. S. Afr. Mus. 12: 323-497. HauGutTon, S. H. 1932. On a collection of Karroo vertebrates from Tanganyika Territory. QO. Jl. geol. Soc. Lond. 88: 634-671. HAUGHTON, S. H. & Brink, A. S. 1954. A bibliographic list of Reptilia from the Karroo Beds of Africa.—Palaeont. afr. 2: 1-187. Hitcucock, E. 1865..A supplement to the ichnology of New England. Boston: Wright and Potter. HueENE, F. von. 1906. Uber die Dinosaurier der aussereuropaischen Trias.— Geol. paléont. Abh. (N.F.) 8: 99-156. HueENE, F. von. 1907-1908. Die Dinosaurier der Europaischen Triasformation mit Beruck- sichtigung der aussereuropdischen Vorkomnisse.— Geol. paldont. Abh. (Suppl.) 1: 1-419. HUueEng, F. von. 1914a. Saurischia et Ornithischia Triadica (“Dinosauria’ Triadica). In: Fos- silium Cat. (Animalia) 4: 1-21. HUuENE, F. von. 1914b. Nachtrage zu meinen fruheren Beschreibungen triassischer Saurischia. —Geol. paldont. Abh. (N.F.) 13: 69-82. HUuENE, F. von. 1921. Reptilian and stegocephalian remains from the Triassic of Pennsylvania in the Cope Collection.— Bull. Am. Mus. nat. Hist. 44: 561-574. HUuENE, F. von. 1926. Vollstandige Osteologie seines Plateosauriden aus dem Schwabischen Keuper.— Geol. paliont. Abh. (N.F.) 15: 134-180. HUENE, F. von. 1932. Die fossile Reptil-Ordnung Saurischia, ihre Entwicklung und Geschichte. — Monogrn Geol. Paldont. (1) 4: 1-361. HUuENE, F. von. 1935-1942. Die Fossilen Reptilien des Sudamerikanischen Gondwanalandes. Munich: Beck’sche Verlagsbuchhandlund. HUuENE, F. von. 1956. Palaontologie und Phylogenie der niederen Tetrapoden. Jena: Fischer. LuLL, R. S. 1912. The life of the Connecticut Valley.— Am. J. Sci. 33: 397-422. LuLL, R. S. 1915. Triassic life of the Connecticut Valley.— Bull. Conn. St. geol. nat. Hist. Surv. 24: 1-285. LuLL, R. S. 1953. Triassic life of the Connecticut Valley. Revised edition.— Bull. Conn. St. geol. nat. Hist. Sury. 81: 1-331. LYDEKKER, R. 1890. Catalogue of fossil reptilia and amphibia. Part 4. London: British Museum (Natural History). Mars, O. C. 1882. Classification of the Dinosauria.— Am. J. Sci. 23: 81-86. Marsh, O. C. 1885. Names of extinct reptiles.— Am. J. Sci. 37: 331-336. Marsh, O. C. 1889. Notice of new American Dinosauria.— Am. J. Sci. 37: 331-336. Marsh, O. C. 1891. Notice of new vertebrate fossils.— Am. J. Sci. 42: 265-269. Marsu, O. C. 1892. Notes on Triassic Dinosauria.— Am. J. Sci. 43: 543-546. Marsh, O. C. 1893. Restoration of Anchisaurus.— Am. J. Sci. 45: 169-170. Marsh, O. C. 1895. On the affinities and classification of the dinosaurian reptiles.— Am. J. Sci. 50: 483-498. Marsu, O. C. 1896. The dinosaurs of North America.— Rep. U.S. Geol. Surv. (1): 133-244. Meyer, H. von. 1837. Briefliche Mitteilung an Prof. Bronn tiber Plateosaurus engelhardti.— Neues Jb. Miner. Geogr. Geol. Petrefakt. 1837: 316. Morris, J. 1843. A catalogue of British Fossils. London: British Museum (Natural History). OweEN, R. 1854. Descriptive catalogue of the fossil organic remains of Reptilia contained in the Museum of the Royal College of Surgeons of England. London: British Museum (Natural History). RAATH, M. A. 1972. Fossil vertebrate studies in Rhodesia: a new dinosaur (Reptilia, Saurischia) from near the Trias-Jurassic boundary.— Arnoldia (Rhod.) 30: 1-37. ReIG, O. A. 1963. La presencia de dinosaurios saurisquios en los ‘Estratos de Ischigualasto’ (Mesotriasico Superior) de las Provincias de San Juan y la Rioja (Republica Argentina). —Ameghiniania 3: 3-20. 158 ANNALS OF THE SOUTH AFRICAN MUSEUM Rivey, H. & StutrcHBury, S. 1836. A description of various fossil remains of three distinct saurian animals discovered in the autumn of 1834, in the Magnesian Conglomerate on Durdham Down, near Bristol.— Proc. geol. Soc. 2: 397-399. Ritey, H. & StuTcHBURY, S. 1840. A description of various fossil remains of three distinct saurian animals, recently discovered in the Magnesian Conglomerate near Bristol.— Trans. geol. Soc. Lond. 5: 349-357. Romer, A. S. 1923. The pelvic musculature of saurischian dinosaurs.— Bull. Am. Mus. Nat. Hist. 48: 605-617. Romer, A. S. 1956. Osteology of the Reptiles. Chicago: University of Chicago Press. Romer, A. S. 1966. Vertebrate paleontology. (3rd ed.) Chicago: University of Chicago Press. ROZHDESTVENSKY, A. K. 1966. Growth changes in Asian dinosaurs and some problems of their taxonomy (translation of 1965).— Int. Geol. Rev. 8: 782-793. SEELEY, H. G. 1891. On Agrosaurus mcgillivrayi, a saurischian reptile from the north east — coast of Australia.—Q. Jl. geol. Soc. Lond. 47: 164-165. SEELEY, H. G. 1894. On Hortalotarsus skirtopodus, a new saurischian fossil from Barkly East, Cape Colony.— Ann. Mag. nat. Hist. 14: 411-419. SEELEY, H. G. 1895a. On Thecodontosaurus and Palaeosaurus.— Ann. Mag. nat. Hist. 15: 144-153. SEELEY, H. G. 1895b. On the type of the genus Massospondylus, and on some vertebrae and limb-bones of M. (?) browni.— Ann. Mag. nat. Hist. 15: 102-125. STEEL, R. 1970. Handbook of Palaeontology. Part 14. Saurischia. Portland: G. Fischer. VAN HoepeEN, E. C. N. 1920a. Contributions to the knowledge of the reptiles of the Karroo Formation. 5. A new dinosaur from the Stormberg Beds.— Ann. Transy. Mus. 7: 77-92. VAN HOoepPEN, E. C. N. 1920b. Contributions to the knowledge of the reptiles of the Karroo Formation. 6. Further dinosaurian material in the Transvaal Museum.— Ann. Transv. Mus. 7: 93-141. Waite, T. E. 1973. Catalogue of the genera of dinosaurs.— Ann. Carneg. Mus. 44: 117-155. WILb, R. 1974. Die Triasfauna der Tessiner Kalkalpen. XXIII. Tanystropheus longobardicus (Bassani).— Schweiz. palaeont. Abh. 95: 1-162. YOUNG, C.-C. 1941a. A complete osteology of Lufengosaurus huenei Young (gen. et sp. nov.). Palaeont. sin. (C) 7: 1-53. YOUNG, C.-C. 1941b. Gyposaurus sinensis (sp. nov.), a new Prosauropoda from the Upper Triassic Beds at Lufeng, Yunnan.— Bull. geol. Soc. China 21: 205-253. YOUNG, C.-C. 1942. Yunnanosaurus huangi (gen. et sp. nov.), a new Prosauropoda from the Red Beds at Lufeng, Yunnan.— Bull. geol. Soc. China 22: 63-104. YOUNG, C.-C. 1947. On Lufengosaurus magnus (sp. nov.) and additional finds of Lufengosaurus huenei Young.— Palaeont. sin. 12: 1-53. YOUNG, C.-C. 1948. Further notes on Gyposaurus sinensis Young.— Bull. geol. Soc. China 28: 91-103. YOUNG, C.-C. 1951. The Lufeng saurischian fauna in China.—Palaeont. sin. 13: 1-96. ABBREVIATIONS acetabulum acet anterior process of ilium ant proc astragalus as centrum Cc caudal vertebra ca calcaneum cal chevron ch dorsal vertebra dor neural spine ns pleurocoel - pl postzygapophysis poz prezygapophysis prz ANCHISAURUS CAPENSIS (BROOM) AND A REVISION OF THE ANCHISAURIDAE 159 pubic peduncle pub ped sacral vertebra sac Amherst College Museum, Amherst, Massachusetts AM American Museum of Natural History, New York AMNH British Museum (Natural History), London BMNH National Museum of Ireland, Dublin NMI South African Museum, Cape Town SAM Staatlichen Museum fiir Naturkunde in Stuttgart SMNS Yale Peabody Museum, New Haven, Connecticut YPM 6. SYSTEMATIC papers must conform with the International code of zoological nomenclature (particularly Articles 22 and 51). Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed by the appropriate Latin (not English) abbreviation, e.g. gen. n., sp. n., comb. n., syn. n., etc. An author’s name when cited must follow the name of the taxon without intervening punctuation and not be abbreviated; if the year is added, a comma must separate author’s name and year. The author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is transferred from its original genus. The name of a subsequent user of a 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 scientific names by which the species previously has been designated are listed in chronological order, with all references to that name following in chronological order, e.g.: Family Nuculanidae Nuculana (Lembulus) bicuspidata (Gould, 1845) Figs 14-15A Nucula (Leda) bicuspidata Gould, 1845: 37. Leda plicifera A. Adams, 1856: 50. Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (figs 8a—b). Nucula largillierti Philippi, 1861: 87 Leda 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 the year is placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable. In describing new species, one specimen must be designated as the holotype; other speci- mens mentioned in the original description are to be designated paratypes; additional material not regarded as paratypes should be listed separately. The complete data (registration number, depository, description of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.: Holotype SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Elizabeth (33.51S, 25.39E), collected by A. Smith, 15 January 1973. Note standard form of writing South African Museum registration numbers and of date. 7. SPECIAL HOUSE RULES Capital initial letters (a) The Figures, Maps and Tables of the paper when referred to in the text e.g. ©... the Figure depicting C. namacolus... *...1in 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 Yon Huene _— but _-F. von Huene (c) Scientific names, but not their vernacular derivatives e.g. Therocephalia, but therocephalian Punctuation should be loose, omitting all not strictly necessary Reference 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.’ Specific 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. “11 00 0 P. M. GALTON & M. A. CLUVER ANCHISAURUS CAPENSIS (BROOM) | AND A REVISION OF THE ANCHISAURIDAE (REPTILIA, SAURISCHIA)